Fine-grained ship recognition in remote sensing imagery is essential for maritime applications. However, its development is hindered by two challenges: 1) the limited granularity of existing ship detection datasets, and 2) the disturbance of complex maritime conditions as well as the arbitrary ship orientations and distributions. To address the first issue, we annotated a large-scale fine-grained ship instance detection dataset (LAFI), comprising 48,717 ship instances worldwide with 49 categories. To tackle the challenges of marine disturbance and diverse ship status, we proposed a controllable generative knowledge-driven ship detection framework (COSD). It employs a controllable diffusion model guided by ship-marine textual prompt to generate millions of synthetic images that not only preserve ship structures but also cover diverse sea and weather conditions for robust pretraining. The pretraining stage then utilizes masked reconstruction to learn component-level cues under occlusion, clutter, fog, and illumination changes. Furthermore, a heterogeneous feature alignment decoder is designed to align multi-modal metrics of orientation and distribution features in the latent space, allowing for accurate representation of diverse ship status. Extensive experiments on two benchmark datasets showed that our method respectively increased 0.011 and 0.030 mean average precision (mAP@50) over SOTA methods, particularly in scenarios involving small, densely packed and arbitrary oriented ships.

Benefiting from long-term time series and large spatial coverage, Sentinel-2 has been widely used in urban tree cover retrieval. However, mixed pixel effects in Sentinel-2 imagery make it challenging to accurately identify urban tree covers. To address this problem, Sub-Pixel Mapping (SPM) is developed to reconstruct a high-resolution urban tree cover from medium-resolution imagery. While deep-learning-based SPM seeks fine-grained patterns solely within medium-resolution feature spaces and spatiotemporal fusion-based SPM leverages additional high-resolution imagery from different times at the same location, both face limitations: the former lacks detailed spatial constraints, and the latter struggles with acquiring geographically aligned imagery. To address these challenges, this study proposes a Geographic Prior guided Sub-pixel Mapping (GPSPM) approach for urban tree cover reconstruction. The geographic prior is grounded in the scaling law of geography, a fundamental principle of spatial heterogeneity stating that high-resolution imagery contains far more detailed features (e.g., small tree parcels) than lower-resolution imagery. These fine-grained features enhance SPM by providing robust cross-scale spatial prior based on a “teacher-student” domain adaptation training framework. Besides, considering the geometric feature discrepancy and long-tail distribution exists across different geographic scales, cross-scale image mosaicking and resampling strategy are further developed. Experiments on public urban tree cover dataset demonstrate that the proposed method improves the Intersection over Union (IoU) of urban tree cover by approximately 5% compared to traditional unsupervised SPM and shows significant improvements in spatial detail quality.

Landsat satellites have provided high-quality Earth observations for more than 40 years, which significantly benefits much research on agriculture, environment, ecology, and so on. However, its low temporal resolution (16-day) and disturbances such as cloud contamination, prevent its usage in some scenarios. Therefore, reconstructing the Landsat image series is always an important topic. Currently, the approaches for reconstructing that are massive, which mainly include the filling-based methods and fusion-based ones. However, their scalability and applicability in large-scale applications are limited. To address this problem, based on the Google Earth engine (GEE), which is a powerful cloud platform, this article introduces a GEE-based fusion and filling model (GFFM) for generating high-quality synthetic Landsat surface reflectance time series data. This model adopts a pixel-wise regression technique to fuse the Landsat and Moderate Resolution Imaging Spectroradiometer (MODIS) data, providing a synthetic image series first. Then, a harmonic analysis is used to densify the Landsat image series. Finally, we utilize a Bayesian model average (BMA) as a weight function to integrate and adjust the previously obtained data to acquire the final seamless image series. We compare the proposed GFFM with some state-of-the-art fusion and filling approaches on various datasets. The experimental results demonstrate that the GFFM not only outperforms these fusion and filling approaches on different datasets, but also shows more robustness in cases of less and cloudy input data.

With the widespread application of remote sensing images in military and civilian fields, remote sensing object detection (RSOD) has become an important research direction. However, limited generalization and complex background interference have long been persistent challenges that hinder the development of RSOD. To address these issues, we propose a Pre-trained Scene-aware Object Detection Network (PSODNet). Firstly, we design an Enhanced Object Network (EON), which leverages a multi-head pretraining strategy to jointly train data from diverse sources, thereby expanding the scale of dataset and improve the generalization ability. Secondly, we introduce scenario-object relationship module to learn a multi-scale relationship map between objects and scenes, which is used to constrain the solution space of object detection, thereby enhancing performance in complex scenarios. Lastly, by using Label Smoothing Loss, PSODNet leverages mutual information of label to prevent extreme distributions of classification probabilities and reduce the risk of overfitting. In the experiment part, PSODNet was pretrained on multiple datasets and then fine-tuned on three datasets for validation. Results on three public datasets demonstrate that PSODNet outperforms existing models in detection performance by up to 2.4%, achieving a maximum mAP of up to 95%. Through visual interpretation of the relationship map, we found that PSODNet is able to bridge the semantic relevance between objects and scenes, demonstrating its potential in object detection in complex scenarios. Code is available at: https://github.com/creature-compound/PSODNet.

Comprehending large-scale tree cover dynamics is paramount for effective environmental management and policy formulation. This study introduces a comprehensive and cost-effective framework for mapping China's Annual Tree Cover (2000–2022) at a 30m spatial resolution. The methodology, amalgamating time-series MODIS and Landsat imagery with a global-local automated sampling approach, partitioned modeling, and random forest-based resolution-adaptive techniques, demonstrates compelling validation outcomes. Validation against Global Ecosystem Dynamics Investigation (GEDI) reference data yields an R2of 0.49 and an RMSE of 20.8%. Further validation using national forest inventory (NFI) data reinforces this accuracy, achieving an R2of 0.82 and an RMSE of 14.1%. Beyond enhancing comprehension of China's terrestrial ecosystem, the resulting tree cover dataset emerges as a pivotal resource, facilitating informed decision-making in environmental conservation and land-use planning.

Cellular automata (CA) have been prevalently used for the simulation of urban land change. However, how to effectively learn the spatial-temporal dynamics of urban development from time-series data remain an important challenge for CA-based models. To address this issue, we propose a new model for the simulation of urban development based on convolutional long short-term memory (ConvLSTM) neural networks. The core of the proposed model is a sequence of vanilla ConvLSTM cells integrated with the modules of channel attention and contextual embedding. Compared with conventional CA-based models, the proposed ConvLSTM model is more advanced in that it can better leverage the open access annual urban land maps to capture simultaneously the spatial structure and the temporal dependency of historical urban development, and further predict multiple maps of annual development for subsequent years (i.e., Maps-to-Maps). The performance of the ConvLSTM model is evaluated through the case studies in China’s three mega-urban regions, and ConvLSTM outperforms other state-of-the-art deep learning architectures at both the pixel level and the coarser grid level. The results also suggest the satisfactory transferability of ConvLSTM in that the model trained in one mega-urban region can be successfully re-used in others without fine tuning.

Quantifying woody vegetation cover is essential for understanding vegetation evolution and formulating land use management policies. However, most attention has focused on the retrieval of forest attributes, and it is imperative to enhance the capability to monitor the cover of woody vegetation. This study proposed an adaptive stacking regressor with a self-iterative optimization module to improve the quantification of the fractional woody cover. The adaptive stacking model was trained using synthetic mixing techniques and the 2021 Landsat composite to estimate the proportion of woody cover in Shaanxi Province, China. Results show that: 1) the proposed model exhibits a stable relationship ($R^{2}$= 0.92 and RMSE = 0.1020) between very high-resolution (VHR)-derived reference data and the Landsat-based wood cover fraction estimates; 2) the proposed model outperforms other classifiers in heterogeneous region mapping of woody vegetation, followed by random forest (RF) ($R^{2}$= 0.82 and RMSE = 0.1492), gradient-boosting decision tree (GBDT) ($R^{2}$= 0.79 and RMSE = 0.2550), and classification and regression tree (CART) ($R^{2}$= 0.69 and RMSE = 0.2042) models; and 3) uncertainty analysis based on the Monte Carlo method revealed that the adaptive stacking model improved the overestimations of low cover fractions and underestimations of high cover fractions. The adaptive stacking model is expected to advance the systematic monitoring of biophysical parameters of the land surface.

Sub-pixel mapping (SPM) has been widely adopted to alleviate the mixed pixel problem in hyperspectral image, as an extension of spectral unmixing (SU), providing a way to observe the spatial location of the endmember within mixed pixel. However, most of the SPM methods are unmixing-then-mapping (UTM), i.e., SPM process relies on the abundance images generated from SU, in which process uncertainty inherently exists and would be propagated to SPM. Furthermore, the prior knowledge toward the sub-pixel scale distribution is mainly model-driven/handcrafted, which has limitation for geographical-realistic distribution representation. In this letter, we proposed spectral–spatial fusion SPM based on deep neural network (SSNET), to realize the integrative modeling of SU and SPM problem in a unified network fashion to avoid uncertainty accumulation in UTM process, and it can simultaneously generate SU result and SPM result. Besides, SSNET provides a supervised manner to learn prior knowledge with external exemplar pairs of low- and high-resolution images for a geographical-realistic distribution representation. The experiment with two hyperspectral images validated the superiority of the proposed SSNET.

Change detection (CD) aims to distinguish surface changes based on bitemporal images. Since high-resolution (HR) images cannot be typically acquired continuously over time, bitemporal images with different resolutions are often adopted for CD in practical applications. Traditional subpixel-based methods for CD using images with different resolutions may lead to substantial error accumulation when the HR images are employed, which is because of intraclass heterogeneity and interclass similarity. Therefore, it is necessary to develop a novel method for CD using images with different resolutions that are more suitable for the HR images. To this end, we propose a super-resolution-based change detection network (SRCDNet) with a stacked attention module (SAM). The SRCDNet employs a super-resolution (SR) module containing a generator and a discriminator to directly learn the SR images through adversarial learning and overcome the resolution difference between the bitemporal images. To enhance the useful information in multiscale features, a SAM consisting of five convolutional block attention modules (CBAMs) is integrated to the feature extractor. The final change map is obtained through a metric learning-based change decision module, wherein a distance map between bitemporal features is calculated. Ablation study and comparative experiments on two large datasets, building change detection dataset (BCDD) and season-varying change detection dataset (CDD), and a real-image experiment on the Google dataset fully demonstrate the superiority of the proposed method. The source code of SRCDNet is available athttps://github.com/liumency/SRCDNet.

Change detection (CD) aims to identify surface changes from bitemporal images. In recent years, deep learning (DL)-based methods have made substantial breakthroughs in the field of CD. However, CD results can be easily affected by external factors, including illumination, noise, and scale, which leads to pseudo-changes and noise in the detection map. To deal with these problems and achieve more accurate results, a deeply supervised (DS) attention metric-based network (DSAMNet) is proposed in this article. A metric module is employed in DSAMNet to learn change maps by means of deep metric learning, in which convolutional block attention modules (CBAM) are integrated to provide more discriminative features. As an auxiliary, a DS module is introduced to enhance the feature extractor’s learning ability and generate more useful features. Moreover, another challenge encountered by data-driven DL algorithms is posed by the limitations in change detection datasets (CDDs). Therefore, we create a CD dataset, Sun Yat-Sen University (SYSU)-CD, for bitemporal image CD, which contains a total of 20 000 aerial image pairs of size$256\times256$. Experiments are conducted on both the CDD and the SYSU-CD dataset. Compared to other state-of-the-art methods, our network achieves the highest accuracy on both datasets, with an F1 of 93.69% on the CDD dataset and 78.18% on the SYSU-CD dataset.

Current global snow retrieval algorithms based on spaceborne microwave measurements inherit noticeable biases and uncertainties regarding spatial distribution and temporal variations. In this article, we present an improved spatiotemporally dynamic global snow depth retrieval algorithm to account for the heterogeneity of snowpacks in different seasons worldwide. The proposed model adopts nonlinear machine learning to retrieve snow depths from passive microwave measurements and other auxiliary information. We indirectly characterized the variation in snow grain size using the daily profiles of the temperature gradient within the snowpack. In addition, a zoning and multitemporal modeling strategy was employed to reduce the bias and uncertainty caused by snow heterogeneity across different ecoregions and seasons. The proposed model was implemented to retrieve the global daily snow depth from 2001 to 2010. The results were validated byin situobservations and compared with the NASA Advanced Microwave Scanning Radiometer for EOS (AMSR-E) snow water equivalent product (AE_DySno). Satisfactory accuracy was achieved for different ecoregions with regard to daily, monthly, and yearly validations (the root-mean-square error (RMSE) varied from ~7.5 to ~12 cm; the Pearson correlation coefficient$R$ranged from 0.75 to 0.85). The results of ten trials indicated the promising stability of the proposed model in different ecoregions with small variations in RMSE and$R$values. Compared with the AE_DySno products, the estimation results did not exhibit the overestimation problem and provided snow depth patterns with greater spatial heterogeneity, showing RMSEs ~5 cm lower and$R$values ~0.3 higher than those of the AE_DySno products.

Deep sub-pixel mapping network (DSMNet) is a state-of-the-art approach in the field of sub-pixel mapping (SPM, also called super resolution mapping), combining deep learning theory, to solve the mixed pixel problem, which is ubiquitous in remote sensing images due to the spatial-resolving limitation. However, traditional DSMNet usually do not consider the multi-scale distribution characteristics of the real geographical distribution exposed in urban landscape. Furthermore, the heterogeneous distribution characteristics (high-frequency components) are the most important for SPM, but are difficult to learn and usually ignored in the tradition network models. In this paper, the high-frequency component aware (HFCA) module was proposed, based on the hierarchical supervised deep sub-pixel mapping network (HiDSMNet). HiDSMNet establishes a hierarchical supervised architecture for explicit multi-scale supervision to prompt the network to learn a multi-scale representation. Besides, HFCA module is integrated to prompt the network to intensify the learning of the high-frequency representation. The experimental results with three public datasets validated the superiority of the proposed HiDSMNet.

Mixed pixel problem is omnipresent in remote sensing images for urban land use interpretation due to the hardware limitations. Subpixel mapping (SPM) is a usual way to solve this problem by improving the observation scale and realizing a finer spatial resolution land cover mapping. Recently, deep learning-based subpixel mapping network (DLSMNet) was proposed, benefited from its strong representation and learning ability, to restore a visually pleasing finer mapping. However, the spatial context features of artifacts are usually aggregated and progressively lost during the forward pass of the network without sufficient representation, which make it difficult to be learned and restored. In this article, a semantic information modulated (SIM) deep subpixel mapping network (SIMNet) is proposed, which uses low-resolution semantic images as prior, to reinforce the representation of spatial context features. In SIMNet, SIM module is proposed to parametrically incorporate the semantic prior into the state-of-the-art (SOTA) feed forward network architecture in an end-to-end training fashion. Furthermore, stacked SIM module with residual blocks (SIM_ResBlock) is adopted to pass the representation of spatial context feature to the deep layers, to get it fully learned during backpropagation. Experiments have been implemented on three public urban scenario data sets, and the SIMNet generates a clearer outline of artificial facilities with sufficient spatial context, and is distinctive for even individual building, which is challenging for other SOTA DLSMNet. The results demonstrate that the proposed SIMNet is a promising way for high-resolution urban land use mapping from easily available lower resolution remote sensing images.Mixed pixel problem is omnipresent in remote sensing images for urban land use interpretation due to the hardware limitations. Subpixel mapping (SPM) is a usual way to solve this problem by improving the observation scale and realizing a finer spatial resolution land cover mapping. Recently, deep learning-based subpixel mapping network (DLSMNet) was proposed, benefited from its strong representation and learning ability, to restore a visually pleasing finer mapping. However, the spatial context features of artifacts are usually aggregated and progressively lost during the forward pass of the network without sufficient representation, which make it difficult to be learned and restored. In this article, a semantic information modulated (SIM) deep subpixel mapping network (SIMNet) is proposed, which uses low-resolution semantic images as prior, to reinforce the representation of spatial context features. In SIMNet, SIM module is proposed to parametrically incorporate the semantic prior into the state-of-the-art (SOTA) feed forward network architecture in an end-to-end training fashion. Furthermore, stacked SIM module with residual blocks (SIM_ResBlock) is adopted to pass the representation of spatial context feature to the deep layers, to get it fully learned during backpropagation. Experiments have been implemented on three public urban scenario data sets, and the SIMNet generates a clearer outline of artificial facilities with sufficient spatial context, and is distinctive for even individual building, which is challenging for other SOTA DLSMNet. The results demonstrate that the proposed SIMNet is a promising way for high-resolution urban land use mapping from easily available lower resolution remote sensing images.

Modeling urban growth in Economic development zones (EDZs) can help planners determine appropriate land policies for these regions. However, sometimes EDZs are established in remote areas outside of central cities that have no historical urban areas. Existing models are unable to simulate the emergence of urban areas without historical urban land in EDZs. In this study, a cellular automaton (CA) model based on fuzzy clustering is developed to address this issue. This model is implemented by coupling an unsupervised classification method and a modified CA model with an urban emergence mechanism based on local maxima. Through an analysis of the planning policies and existing infrastructure, the proposed model can detect the potential start zones and simulate the trajectory of urban growth independent of the historical urban land use. The method is validated in the urban emergence simulation of the Taiping Bay development zone in Dalian, China from 2013 to 2019. The proposed model is applied to future simulation in 2019–2030. The results demonstrate that the proposed model can be used to predict urban emergence and generate the possible future urban form, which will assist planners in determining the urban layout and controlling urban growth in EDZs.

Urban villages (UVs) are distinctive products formed in the process of rapid urbanization. The fine-grained mapping of UVs from satellite images has always been a considerable challenge because of the complex urban structures and the insufficiency of labeled samples. In this letter, we propose using the domain adaptation strategy to tackle the domain shift problem by employing adversarial learning to tune the semantic segmentation network so as to adaptively obtain similar outputs for input images from different domains. The proposed method was coupled with several segmentation networks, including U-Net, RefineNet, and DeepLab v3+, and the results show that domain adaptation can significantly improve the pixel-level mapping of UVs.

Urban growth boundaries (UGBs) have been applied in many rapid urbanizing areas to alleviate the problems of urban sprawl. Although empirical research has stressed the importance of ecological protection in UGB delineation, existing UGB models lack a component for the assessment of ecologically sensitive areas. To address this problem, we develop an innovative method that is capable of simulating UGB alternatives with economic and ecological constraints. Our method employs a patch-based cellular automaton (i.e. SA-Patch-CA) for simulating future urban growth, constrained by the ecological protection areas produced by an agent-based land allocation optimization model (AgentLA). The delineation of UGBs is also based on the estimated future urban land demand derived from support vector regression (SVR). The proposed method is applied in the Pearl River Delta (PRD), China. Three scenarios are designed to represent different objectives of future industrial transitions. The results indicate that increasing the shares of low energy consumption industries and tertiary industries can effectively reduce urban land demand. By overlapping the simulations, we found that the areal agreement of the simulated UGBs among the three scenarios accounts for approximately 88% of the total area. These areas can then be considered as the primary locations for establishing the UGBs.

The scale effects of the spatial autocorrelation (SA) measurement has been explored for decades. However, the effects of the data aggregation levels and spatial resolution on the SA measurement are often confused. Whether the two types of scale effects are the same is still unclear and requires further investigation. We retrieved the land surface temperature (LST) from Landsat 8 images in 30 capital cities of China. By aggregating the LST images, we observed a decrease in the SA of the LST as the data aggregation level increased; this relationship can be fitted well with a negative logarithmic function. We derived an indicator to measure the scale effects intensity of SA, which was negatively correlated with the spatial complexity of LST. Both aggregating images and the increasing spatial resolution induce weaker SA, but the effect of the former was stronger. The aggregating images negatively affected the SA degree regardless of the spatial resolutions of the original images. The SA degrees of the aggregated images were far below those of the real-life images. This study suggests that the scale effects caused by aggregation levels and spatial resolutions are different, and cautions should be taken when applying relevant conclusions derived from aggregating images.

This study presents a transfer learning method for addressing the insufficient sample problem in hyperspectral image classification. In order to find common feature representation for both the source domain and target domain, we introduce a regularisation based on Bregman divergence into the objective function of the subspace learning algorithm, which can minimise the Bregman divergence between the distribution of training samples in the source domain and the test samples in the target domain. Hyperspectral image with biased sampling is used to evaluate the effectiveness of the proposed method. The results show that the proposed method can achieve a higher classification accuracy than traditional subspace learning methods under the condition of biased sampling.

Urban land-use change is affected by urban planning and government decision-making. Previous urban simulation methods focused only on planning constraints that prevent urban growth from developing in specific regions. However, regional planning produces planning policies that drive urban development, such as traffic planning and development zones, which have rarely been considered in previous studies. This study aims to design two mechanisms based on a cellular automata-based future land-use simulation model to integrate different planning drivers into simulations. The first update mechanism considers the influence of traffic planning, while the second mechanism can model the guiding effect of planning development zones. The proposed mechanisms are applied to the Pearl River Delta region, which is one of the fastest growing areas in China. The first mechanism is validated using simulations from 2000–2013 and demonstrates that simulation accuracy is improved by the consideration of traffic planning. In the simulation from 2013–2052, the two mechanisms are implemented and yield more realistic urban spatial patterns. The simulation outcomes can be employed to identify potential urban expansion inside the master plan. The proposed methods can serve as a useful tool that assists planners in their evaluation of urban evolvement under the impact of different planning policies.

Cellular automata (CA) have been efficiently used to express the complexity and dynamics of cities at different scales. However, those large-scale simulation models typically use only binary values to represent urbanization states without considering mixed types within a cell. They also ignore differences among the cells in terms of their temporal evolution characteristics at different urbanization stages. This study establishes a gradient CA for solving such problems while considering development differences among the cells. The impervious surface area data was used to detect the urbanization states and temporal evolution trends of the grid cells. Transition rules were determined with the incorporation of urban development theory expressed as an S-shaped curve. China was selected as the case study area to validate the performance of the gradient CA for a national simulation. A comparison was also made to a traditional binary logistic-CA. The results demonstrated that the gradient CA achieved higher accuracies in terms of both spatial patterns and quantitative assessment indices. The simulation pattern derived from the gradient CA can better reflect the local disparity and temporal characteristics of urban dynamics. A national urban expansion for 2050 was also simulated, and is expected to provide important data for ecological assessments.

ABSTRACTTo-date few research has successfully integrated big data from multiple sources to characterize urban mixed-use buildings. In this paper, we introduce a probabilistic model to integrate multi-source and geospatial big data (social network data, taxi trajectories, Points of Interest and remote sensing images) to characterize urban mixed-use buildings. The usefulness of our model is demonstrated with a case study of the Tianhe District in megacity Guangzhou, China. The model predicted building functions at 85% accuracy based on ground truth data from field surveys. We further explored the spatial patterns of the identified building functions. Most mixed-use buildings are located along major streets. Our proposed model can identify mixed-use buildings in a city; information is useful for planning evaluation and urban policymaking.

Classification is an important technique for remote sensing data interpretation. In order to enhance the performance of a supervised classifier and ensure the lowest possible cost of the training samples used in the process, active learning (AL) can be used to optimize the training sample set. At the same time, integrating spatial information can help to enhance the separability between similar classes, which can in turn reduce the need for training samples in AL. To effectively integrate spatial information into the AL framework, this paper proposes a new active relearning (ARL) model for remote sensing image classification. In particular, our model is used to relearn the spatial features on the classification map, which contributes significantly to enhancing the performance of the classifier. We integrate the relearning model into the AL framework, with the aim to accelerate the convergence of AL and further reduce the labeling cost. Under the newly developed ARL framework, we propose two spatial–spectral uncertainty criteria to optimize the procedure for selecting new training samples. Furthermore, an adaptive multiwindow ARL model is also introduced in this paper. Our experiments with two hyperspectral images and two very high resolution images indicate that the ARL model exhibits faster convergence speed with fewer samples than traditional AL methods. Our results also suggest that the proposed spatial–spectral uncertainty criteria and the multiwindow version can further improve the performance when implementing ARL.

The reliability of raster cellular automaton (CA) models for fine-scale land change simulations has been increasingly questioned, because regular pixels/grids cannot precisely represent irregular geographical entities and their interactions. Vector CA models can address these deficiencies due to the ability of the vector data structure to represent realistic urban entities. This study presents a new land parcel cellular automaton (LP-CA) model for simulating urban land changes. The innovation of this model is the use of ensemble learning method for automatic calibration. The proposed model is applied in Shenzhen, China. The experimental results indicate that bagging-Naïve Bayes yields the highest calibration accuracy among a set of selected classifiers. The assessment of neighborhood sensitivity suggests that the LP-CA model achieves the highest simulation accuracy with neighbor radius r = 2. The calibrated LP-CA is used to project future urban land use changes in Shenzhen, and the results are found to be consistent with those specified in the official city plan.

Cellular automata (CA) models are widely used to simulate land-use changes because of their simplicity, flexibility, intuitiveness and ability to incorporate the spatial and temporal dimensions of processes. A small number of CA-based models have been developed to simulate changes in multiple land uses, most of which use the hierarchical allocation strategy and/or inertia factors to enable these CA models to do so accurately. However, only some of these models allow explicit determination of the allocation sequence for active land uses according to the hierarchical allocation strategy and the objective calculation of inertia factors. In this paper, we proposed a CA-based model, i.e. the LAND System Cellular Automata model for Potential Effects (LANDSCAPE), with a hierarchical allocation strategy and resistances, to simulate changes in multiple land uses. Furthermore, we introduced effective ways to objectively determine the allocation sequence for active land uses and calculate resistances for individual land uses. The results show that the LANDSCAPE model, with a calibrated allocation sequence and resistances, is reliable and accurate for simulating multiple land-use changes.

Since the late 1990s, there are growing studies on the development of cellular automata (CA) as a simulation tool for assisting urban and regional planning in China. Rapid urban development is the main reason that this country has become one of the best places to test the methodology of CA and analyze the effectiveness of using these models. This paper attempts to summarize the experiences and issues of using CA to solve various environmental and planning problems in China. The analysis is based on the literature review using the search engines of ISI Web of Science and Google Scholar. These experiences could be important for those who want to apply CA in developing countries. For example, which environmental and ecological problems can be solved by using this bottom-up approach? What are the data inputs to these models and how can they be calibrated? Our analyses indicate that CA have the great potential to support land-use planning and policy analysis for fast-growing regions. Some specific features of using CA in China are also identified in the literature review, including delineation of urban growth boundary, prevention of illegal development and formulating zoning schemes. The CA studies in this fast-growing country provided valuable experiences for other developing countries to solve a series of simulation and planning problems by using this bottom-up approach.

Urban cellular automata (CA) models propagate and accumulate errors during the modeling process due to the model structure or stochastic processes involved. It is feasible to assimilate real-time observations into an urban CA model to reduce model uncertainties. However, the assimilation performance is sensitive to the spatio-temporal units in the assimilation algorithm, that is, spatial block size and window length (temporal interval). In this study, we coupled an assimilation model, an ensemble Kalman filter (EnKF) and a Logistic-CA model to simulate the urban dynamic in Beijing over a period of two decades. Our results indicate that the coupled EnKF-CA model outperforms the CA-alone counterpart by about 10% in terms of the figure of merit, which reflects the agreement of modeled pixels. We also find that the assimilation performance using a finer block (1 km) is better than that using a coarser block (5 km and 10 km) because of the better depiction of spatial heterogeneity using a finer block. Moreover, the improvement of intermediate outputs using the coupled EnKF-CA model is effective for a certain period (e.g. 5 years). This implies that a high-frequency assimilation may not significantly improve the model performance. The sensitivity analyses of spatio-temporal assimilation in the EnKF-CA model provide a better understanding of the assimilation mechanism that couples with land-use change models.

Urban land use information plays an important role in urban management, government policy-making, and population activity monitoring. However, the accurate classification of urban functional zones is challenging due to the complexity of urban systems. Many studies have focused on urban land use classification by considering features that are extracted from either high spatial resolution (HSR) remote sensing images or social media data, but few studies consider both features due to the lack of available models. In our study, we propose a novel scene classification framework to identify dominant urban land use type at the level of traffic analysis zone by integrating probabilistic topic models and support vector machine. A land use word dictionary inside the framework was built by fusing natural–physical features from HSR images and socioeconomic semantic features from multisource social media data. In addition to comparing with manual interpretation data, we designed several experiments to test the land use classification accuracy of our proposed model with different combinations of previously acquired semantic features. The classification results (overall accuracy = 0.865, Kappa = 0.828) demonstrate the effectiveness of our strategy that blends features extracted from multisource geospatial data as semantic features to train the classification model. This method can be applied to help urban planners analyze fine urban structures and monitor urban land use changes, and additional data from multiple sources will be blended into this proposed framework in the future.

Information about the functions of urban buildings is helpful not only for developing a better understanding of how cities work, but also for establishing a basis for policy makers to evaluate and improve the effectiveness of urban planning. Despite these advantages, however, and perhaps simply due to a lack of available data, few academic studies to date have succeeded in integrating multi-source ‘big data’ to examine urban land use at the building level. Responding to this deficiency, this study integrated multi-source big data (WeChat users’ real-time location records, taxi GPS trajectories data, Points of Interest (POI) data, and building footprint data from high-resolution Quickbird images), and applied the proposed density-based method to infer the functions of urban buildings in Tianhe District, Guangzhou, China. The results of the study conformed to an overall detection rate of 72.22%. When results were verified against ground-truth investigation data, the accuracy rate remained above 65%. Two important conclusions can be drawn from our analysis: 1.The use of WeChat data delivers better inference results than those obtained using taxi data when used to identify residential buildings, offices, and urban villages. Conversely, shopping centers, hotels, and hospitals, were more easily identified using taxi data. 2. The use of integrated multi-source big data is more effective than single-source big data in revealing the relation between human dynamics and urban complexes at the building scale.

Cellular automata (CA) have been widely used to simulate complex urban development processes. Previous studies indicated that vector-based cellular automata (VCA) could be applied to simulate urban land-use changes at a realistic land parcel level. Because of the complexity of VCA, these studies were conducted at small scales or did not adequately consider the highly fragmented processes of urban development. This study aims to build an effective framework called dynamic land parcel subdivision (DLPS)-VCA to accurately simulate urban land-use change processes at the land parcel level. We introduce this model in urban land-use change simulations to reasonably divide land parcels and introduce a random forest algorithm (RFA) model to explore the transition rules of urban land-use changes. Finally, we simulate the land-use changes in Shenzhen between 2009 and 2014 via the proposed DLPS-VCA model. Compared to the advanced Patch-CA and RFA-VCA models, the DLPS-VCA model achieves the highest simulation accuracy (Figure-of-Merit = 0.232), which is 32.57% and 18.97% higher respectively, and is most similar to the actual land-use scenario (similarity = 94.73%) at the pattern level. These results indicate that the DLPS-VCA model can both accurately split the land during urban land-use changes and significantly simulate urban expansion and urban land-use changes at a fine scale. Furthermore, the land-use change rules that are based on DPLS-VCA mining and the simulation results of several future urban development scenarios can act as guides for future urban planning policy formulation.

Urban land use information plays an essential role in a wide variety of urban planning and environmental monitoring processes. During the past few decades, with the rapid technological development of remote sensing (RS), geographic information systems (GIS) and geospatial big data, numerous methods have been developed to identify urban land use at a fine scale. Points-of-interest (POIs) have been widely used to extract information pertaining to urban land use types and functional zones. However, it is difficult to quantify the relationship between spatial distributions of POIs and regional land use types due to a lack of reliable models. Previous methods may ignore abundant spatial features that can be extracted from POIs. In this study, we establish an innovative framework that detects urban land use distributions at the scale of traffic analysis zones (TAZs) by integrating Baidu POIs and a Word2Vec model. This framework was implemented using a Google open-source model of a deep-learning language in 2013. First, data for the Pearl River Delta (PRD) are transformed into a TAZ-POI corpus using a greedy algorithm by considering the spatial distributions of TAZs and inner POIs. Then, high-dimensional characteristic vectors of POIs and TAZs are extracted using the Word2Vec model. Finally, to validate the reliability of the POI/TAZ vectors, we implement a K-Means-based clustering model to analyze correlations between the POI/TAZ vectors and deploy TAZ vectors to identify urban land use types using a random forest algorithm (RFA) model. Compared with some state-of-the-art probabilistic topic models (PTMs), the proposed method can efficiently obtain the highest accuracy (OA = 0.8728, kappa = 0.8399). Moreover, the results can be used to help urban planners to monitor dynamic urban land use and evaluate the impact of urban planning schemes.

Fine-scale population distribution data at the building level play an essential role in numerous fields, for example urban planning and disaster prevention. The rapid technological development of remote sensing (RS) and geographical information system (GIS) in recent decades has benefited numerous population distribution mapping studies. However, most of these studies focused on global population and environmental changes; few considered fine-scale population mapping at the local scale, largely because of a lack of reliable data and models. As geospatial big data booms, Internet-collected volunteered geographic information (VGI) can now be used to solve this problem. This article establishes a novel framework to map urban population distributions at the building scale by integrating multisource geospatial big data, which is essential for the fine-scale mapping of population distributions. First, Baidu points-of-interest (POIs) and real-time Tencent user densities (RTUD) are analyzed by using a random forest algorithm to down-scale the street-level population distribution to the grid level. Then, we design an effective iterative building-population gravity model to map population distributions at the building level. Meanwhile, we introduce a densely inhabited index (DII), generated by the proposed gravity model, which can be used to estimate the degree of residential crowding. According to a comparison with official community-level census data and the results of previous population mapping methods, our method exhibits the best accuracy (Pearson R = .8615, RMSE = 663.3250, p < .0001). The produced fine-scale population map can offer a more thorough understanding of inner city population distributions, which can thus help policy makers optimize the allocation of resources.

Transition rules are the core of urban cellular automata (CA) models. Although the logistic cellular automata (Logistic-CA) is commonly used for rules extraction, it cannot always achieve satisfactory performance because of the spatial heterogeneity and the inherent complexity of urban expansion. This article presents an ensemble-urban cellular automata (Ensemble-CA) model to achieve better transition rules. First, an uncertainty map that assesses the performance of transition rules spatially was achieved. Then, two auxiliary models (i.e. classification and regression tree, CART; and artificial neural network, ANN), both of which have been stabilized with a Bagging algorithm, were prepared for integration using a proposed self-adaptive -nearest neighbors (-NN) combination algorithm. Thereafter, those unconfident sites were replaced with the ensemble output. This model was applied to Guangzhou, China, for an urban growth simulation from 2003 to 2008. Static validation confirmed that this ensemble framework (i.e. without substitution of uncertain sites) can achieve better performance (0.87) in terms of receiver operating characteristic (ROC) statistics (area under the curve, AUC), and outperformed the best single model (ANN, 0.82) and other common strategies (e.g. weighted average, 0.83). After the substitution of unconfident sites, the AUC of Logistic-CA was elevated from 0.78 to 0.81. Subsequently, two urban growth mechanisms (i.e. pixel- and patch-based) were implemented separately based on the integrated transition rules. Experimental results revealed that the accuracy obtained from simulation of the Ensemble-CA increased considerably. The obtained kappa outperformed the single model, with improvements of 1.74% and 2.76% for pixel- and patch-based approaches, respectively. Correspondingly, landscape similarity index (LSI) improvements of these two mechanisms were 4.24% and 1.82%.

There are many different methods to calibrate cellular automata (CA) models for better simulation results of urban land-use changes. However, few studies have been reported on combination of parameter update and error control using local data in CA calibration procedures. This paper presents a self-modifying CA model (SM-CA) that uses the dual ensemble Kalman filter (dual EnKF), which enables the CA model to simultaneously update model parameters and simulation results by merging observation data (local data). We applied the proposed model to simulate urban land-use changes in a 13-year period (1993–2005) in Dongguan City, a rapidly urbanizing region in south China. Simulation results indicate that this model yields better simulation results than the conventional logistic-regression CA and decision-tree CA models. For example, the validation is carried out using cross-tabulation matrix. The simulation results of SM-CA have allocation disagreement of 10.18%, 19.64%, and 30.03% in 1997, 2001, and 2005, respectively, which are 2.12%, 2.47%, and 6% lower than conventional logistic-regression CA models.

Cellular automata (CA) have been used to understand the complexity and dynamics of cities. The logistic cellular automaton (Logistic-CA) is a popular urban CA model for simulating urban growth based on logistic regression. However, this model usually employs a cell-based simulation strategy without considering the spatial evolution of land-use patches. This drawback largely constrains the Logistic-CA for simulating realistic urban development. We proposed a Patch-Logistic-CA to deal with this problem by incorporating a patch-based simulation strategy into the conventional cell-based Logistic-CA. The Patch-Logistic-CA differentiates new developments into spontaneous growth and organic growth, and uses a moving-window approach to simulate the evolution of urban patches. The Patch-Logistic-CA is tested through the simulation of urban growth in Guangzhou, China, during 2005–2012. The cell-based Logistic-CA was also implemented using the same set of data to make a comparison. The simulation results reflect that the Patch-Logistic-CA has slightly lower cell-level agreement than the cell-based Logistic-CA. However, visual inspection of the results reveals that the cell-based Logistic-CA fails to reflect the actual patterns of urban growth, because this model can only simulate urbanized cells around the edges of initial urban patches. Actually, the pattern-level similarities of the Patch-Logistic-CA are over 18% higher than those of the cell-based Logistic-CA. This indicates that the Patch-Logistic-CA has much better performance of simulating actual development patterns than the cell-based Logistic-CA. In addition, the Patch-Logistic-CA can correctly simulate the fractal structure of actual urban development patterns. By varying the control parameters, the Patch-Logistic-CA can also be used to assist urban planning through the exploration of different development alternatives.

Cellular automata (CA) have emerged as a primary tool for urban growth modeling due to its simplicity, transparency, and ease of implementation. Sensitivity analysis is an important component in CA modeling for a better understanding of errors or uncertainties and their propagation. Most studies on sensitivity analyses in urban CA modeling focus on specific component such as neighborhood configuration or stochastic perturbation. However, sensitivity analysis of transition rules, which is one of the core components in CA models, has not been systematically done. This article proposes a systematic sensitivity analysis of major operational components in urban CA modeling using a stepwise comparison approach. After obtaining transition rules, three stages (i.e. static calibration of transition rules, dynamic evolution with varied time steps, and incorporation with stochastic perturbation) are designed to facilitate a comprehensive analysis. This scheme implemented with a case study in Guangzhou City (China) reveals that gaps in performance from static calibration with different transition rules can be reduced when dynamic evolution is considered. Moreover, the degree of stochastic perturbation is closely related to obtain urban morphology. However, a more realistic (i.e. fragmented) urban landscape is achieved at the cost of decreasing pixel-based accuracy in this study. Thus, a trade-off between pixel-based and pattern-based comparisons should be balanced in practical urban modeling. Finally, experimental results illustrate that models for transition rules extraction with good quality can do an assistance for urban modeling through reducing errors and uncertainty range. Additionally, ensemble methods can feasibly improve the performance of CA models when coupled with nonparametric models (i.e. classification and regression tree).

Traditional urban cellular automata (CA) model can effectively simulate infilling and edge-expansion growth patterns. However, most of these models are incapable of simulating the outlying growth. This paper proposed a novel model called LEI-CA which incorporates landscape expansion index (LEI) with CA to simulate urban growth. Urban growth type is identified by calculating the LEI index of each cell. Case-based reasoning technique is used to discover different transition rules for the adjacent growth type and the outlying growth type, respectively. We applied the LEI-CA model to the simulation of urban growth in Dongguan in southern China. The comparison between logistic-based CA and LEI-CA indicates that the latter can yield a better performance. The LEI-CA model can improve urban simulation accuracy over logistic-based CA by 13.8%, 10.8% and 6.9% in 1993, 1999 and 2005, respectively. Moreover, the outlying growth type hardly exists in the simulation by logistic-based CA, while the proposed LEI-CA model performs well in simulating different urban growth patterns. Our experiments illustrate that the LEI-CA model not only overcomes the deficiencies of traditional CA but might also better understand urban evolution process.

A new image-registration method is presented by integrating the area-based and feature-based methods. The integrated method is characterized by a novel similarity metric based on spatial and mutual information (SMI), the ant colony optimization for continuous domain$(ACO_{\BBR})$, and a two-phase searching strategy. The SMI-based metric takes into account both spatial relations of detected features [spatial information (SI)] and the mutual information (MI) between the reference and sensed images. The spatial relation is to derive a fast transformation of the near global optimum without specifying the initial searching range. The MI is to obtain an optimal transformation with high accuracy.$ACO_{ \BBR}$is adopted to optimize SMI for the first time in this paper, as the function of SMI is generally non-convex and irregular. In addition, a two-phase searching strategy is designed to improve the performance of$ACO_{\BBR}$. Phase-1 only considers the SI and finds some low-accurate solutions. Phase-2 considers both SI and MI so it is to search for a more accurate solution. These two phases are switched according to the diversity of the solutions. The proposed integrated method has been tested using the remote-sensing images acquired from different sensors, including TM, SPOT, and SAR. The experimental results indicate that the SMI-based metric is more robust than the conventional metrics which consider SI or MI alone. This method is able to achieve a highly accurate automatic registration of multisensor images.

The Pareto front can provide valuable information on land-use planning decision by revealing the possible trade-offs among multiple, conflicting objectives. However, seeking the Pareto front of land-use allocation is much more difficult than finding a unique optimal solution, especially when dealing with large-area regions. This article proposes an improved artificial immune system for multi-objective land-use allocation (AIS-MOLA) to tackle this challenging task. The proposed AIS is equipped with three modified operators, namely (1) a heuristic hypermutation based on compromise programming, (2) a non-dominated neighbour-based proportional cloning and (3) a novel crossover operator that preserves connected patches. To validate the proposed algorithm, it was applied in a hypothetical land-use allocation problem. Compared with the Pareto Simulated Annealing (PSA) method, AIS-MOLA can generate solutions more approximate to the Pareto front, with computation time amounting to only 5.1% of PSA. In addition, AIS-MOLA was also applied in the case study of Panyu, Guangdong, PR China, a large area with cells. Experimental results indicate that this algorithm, even dealing with large-area land-use allocation problems, is capable of generating optimal alternative solutions approximate to the true Pareto front. Moreover, the distribution of these solutions can quantitatively demonstrate the complex trade-offs between the spatial suitability and the compactness in the study area. Software and supplementary materials are available at http://www.geosimulation.cn/AIS-MOLA/.

Landscape metrics have been widely used to characterize geographical patterns which are important for many geographical and ecological analyses. Cellular automata (CA) are attractive for simulating settlement development, landscape evolution, urban dynamics, and land-use changes. Although various methods have been developed to calibrate CA, landscape metrics have not been explicitly used to ensure the simulated pattern best fitted to the actual one. This article presents a pattern-calibrated method which is based on a number of landscape metrics for implementing CA by using genetic algorithms (GAs). A Pattern-calibrated GA–CA is proposed by incorporating percentage of landscape (PLAND), patch metric (LPI), and landscape division (D) into the fitness function of GA. The sensitivity analysis can allow the users to explore various combinations of weights and examine their effects. The comparison between Logistic- CA, Cell-calibrated GA–CA, and Pattern-calibrated GA–CA indicates that the last method can yield the best results for calibrating CA, according to both the training and validation data. For example, Logistic-CA has the average simulation error of 27.7%, but Pattern-calibrated GA–CA (the proposed method) can reduce this error to only 7.2% by using the training data set in 2003. The validation is further carried out by using new validation data in 2008 and additional landscape metrics (e.g., Landscape shape index, edge density, and aggregation index) which have not been incorporated for calibrating CA models. The comparison shows that this pattern-calibrated CA has better performance than the other two conventional models.

Few studies have been conducted into the use of knowledge transfer for tackling geo-simulation problems. Cellular automata (CA) have proven to be an effective and convenient means of simulating urban dynamics and land-use changes. Gathering the knowledge required to build the CA may be difficult when these models are applied to large areas or long periods. In this paper, we will explore the possibility that the knowledge from previously collected data can be transferred spatially (a different region) and/or temporally (a different period) for implementing urban CA. The domain adaptation of CA is demonstrated by integrating logistic-CA with a knowledge-transfer technique, the TrAdaBoost algorithm. A modification has been made to the TrAdaBoost algorithm by incorporating a dynamicweight-trimming technique. This proposed model, CAtrans, is tested by choosing different periods and study areas in the Pearl River Delta. The ‘Figure of Merit’ measurements in the experiments indicate that CAtrans can yield better simulation results. The variance of traditional logistic-CA is about 2–5 times the variance of CAtrans until the number of new data reaches 30. The experiments have demonstrated that the proposed method can alleviate the sparse data problem using knowledge transfer.

Spatial population dynamics affects resource allocation in urban planning. Simulation of population dynamics can provide useful information to urban planning for rapidly developing manufacturing metropolises. In such a metropolis with a concentration of immigrant labor forces, individual employment choices could have a significant effect on their residential decisions. There remains a need for an efficient method, which can simulate spatial population dynamics by considering the interactions between employment and residential choices. This article proposes an agent-based model for simulation of spatial population dynamics by addressing the influence of labor market on individual residential decisions. Labor economics theory is incorporated into a multi-agent system in this model. The long-term equilibrium process of labor market is established to define the interactions between labor supply and labor demand. An agent-based approach is adopted to simulate the economic behaviors and residential decisions of population individuals. The residential decisions of individuals would eventually have consequences on spatial population dynamics. The proposed model has been verified by the spatial dynamics simulation (2007 to 2010) of Dongguan, an emerging and renowned manufacturing metropolis in the Pearl River Delta, China. The results indicate that the simulated population size and spatial distribution of each town in Dongguan are close to those obtained from census data. The proposed model is also applied to predict spatial population dynamics based on two economic planning scenarios in Dongguan from 2010 to2015. The predicted results provide insights into the population dynamics of this fast-growing region.

In recent years, agent-based models (ABMs) have become a prevalent approach for modelling complex urban systems. As a class of bottom-up method, ABMs are capable of simulating the decision-making as well as the multiple interactions of autonomous agents and between agents and the environment. The definition of agents' behaviour is a vital issue in implementing ABMs to simulate urban dynamics. Urban economic theory has provided effective ways to cope with this problem. This theory argues that the formation of urban spatial structure is an endogenous process resulting from the interactions among individual actors that are spatially distributed. However, this theory is used to explain urban phenomena regardless of spatial heterogeneity in most cases. This study combines GIS, ABM and urban economic models to simulate complex urban residential dynamics. The time-extended model is incorporated into an ABM so as to define agents' behaviour on a solid theoretical basis. A spatial variable is defined to address the neighbourhood effect by considering spatial heterogeneity. The proposed model is first verified by the simulation of three scenarios using hypothetical data: (1) single dominated preference; (2) varying preferences on the basis of income level; and (3) spatially heterogeneous environment. Then the model is implemented by simulating the residential dynamics in Guangzhou, China.

This article presents a new method of assimilating process context information into change detection for monitoring land use changes. The accurate information about land use changes is important for implementing many global and regional environmental models. Two types of models have been independently developed to obtain such information, including change detection models (e.g. pixel-to-pixel comparison, post-classification comparison and object-based change analysis) and simulation models (e.g. cellular automata (CA) and agent-based modelling). These models may have limitations in capturing land use dynamics when used alone. In this study, the ensemble Kalman filter is used to obtain the best estimate of land use changes by combining remote-sensing observations with urban simulation. Urban simulation is able to provide process context information such as diffusion and coalescence of urban development. This type of complementary information is useful for improving the performance of change detection. Compared with traditional change detection models, this integrated model has the potential to improve the performance of change detection in terms of accuracies and landscape metrics. For example, the assimilating (MLC + CA) method can show improvement of the total accuracy and the kappa coefficient by 2.5–5.2% and 3.6–7.4%, respectively, in this study.

Optimizing land use allocation is a challenging task, as it involves multiple stakeholders with conflicting objectives. In addition, the solution space of the optimization grows exponentially as the size of the region and the resolution increase. This article presents a new ant colony optimization algorithm by incorporating multiple types of ants for solving complex multiple land use allocation problems. A spatial exchange mechanism is used to deal with competition between different types of land use allocation. This multi-type ant colony optimization optimal multiple land allocation (MACO-MLA) model was successfully applied to a case study in Panyu, Guangdong, China, a large region with an area of 1,454,285 cells. The proposed model took only about 25 minutes to find near-optimal solution in terms of overall suitability, compactness, and cost. Comparison indicates that MACO-MLA can yield better performances than the simulated annealing (SA) and the genetic algorithm (GA) methods. It is found that MACO-MLA has an improvement of the total utility value over SA and GA methods by 4.5% and 1.3%, respectively. The computation time of this proposed model amounts to only 2.6% and 12.3%, respectively, of that of the SA and GA methods. The experiments have demonstrated that the proposed model was an efficient and effective optimization technique for generating optimal land use patterns.

Although being increasingly powerful in handling spatial data, geographical information systems (GIS) still lack the powerful functionality for process modeling in terms of simulation and optimization. This article discusses the concepts and methodologies of a geographical simulation and optimization system (GeoSOS). GeoSOS integrates cellular automata (CA), agent-based models (ABMs), and swarm intelligence models (SIMs) for solving process simulation and optimization problems. A general form of the so-called interaction rules is proposed for implementing this integrated system. The GeoSOS software is developed to provide these complementary functions that are unavailable in the current GIS. Experiments have demonstrated that GeoSOS is able to model the reciprocal relationships between urban simulation and spatial optimization (e.g., facility sitting, transport development, and natural protection) in fast-growing regions. Better modeling performances have been achieved using the coupling strategies of GeoSOS.

Optimal zoning of protected natural areas is important for conserving ecosystems. It is an NP-hard problem which is difficult to solve by using common geographic information system (GIS) functions. Another problem is that existing optimization methods ignore potential land-use dynamics in formulating optimal patterns. This article has developed a new method for solving complicated zoning problems by using ant colony optimization (ACO) techniques. Significant modifications have been made, so that traditional ACO can be extended to the solution of area optimization problems. Two strategies, the single-year coupling strategy and the merging-year coupling strategy, have been proposed to couple urban cellular automata with ACO for zoning protected natural areas under a changing landscape. This proposed method has been tested in the metropolitan region of Guangzhou, China, by using Geographical Simulation and Optimization System (GeoSOS) software. The experiments indicate that the modified ACO can effectively solve this optimization problem without getting stuck in local optima. This method has better performances compared to other traditional methods, such as simulated annealing (SA), iterative relaxation (IR), and density slicing (DS). The use of the best coupling strategy can improve the accumulative utility value of the zoning by 4.3%. Moreover, it is also found that the adoption of the best protection pattern could significantly promote the compactness of future urban forms in the study area.

Currently, with rapid expanding of urban area, the rate of conversion of agricultural land to nonagricultural uses in China is increasing. Zoning farmland protection is an important measure to protect limited land resource. This article presented an innovative approach based on the integrated use of remote sensing, GIS, and artificial immune systems (AIS) for generating farmland protection areas. Some modifications have been made for conventional AIS so that it can be further extended to the solution of zoning problems. The optimal objective is to generate farmland protection areas that minimize development potential and maximize agricultural suitability and spatial compactness. First, utility function by addressing the criteria of farmland protection is incorporated into AIS algorithm. Second, encoding and mutation of antibodies is modified so that it can be suited to the solution of spatial optimization problems. The AIS-based zoning model was then applied to a case study in Guangzhou, Guangdong, China. The experiments have demonstrated that the proposed method was an efficient and effective spatial optimization technique, which took only about 194 seconds to generate satisfied farmland protection patterns. Furthermore, the AIS-based zoning model can explore various alternatives conveniently, and it can yield better performances than nonprotection scenario in the utility efficiency of land resources and the site condition for farmland.

Spatial optimization is complex because it usually involves numerous spatial factors and constraints. The optimization becomes more challenging if a large set of spatial data with fine resolutions are used. This article presents an agent-based model for optimal land allocation (AgentLA) by maximizing the total amount of land-use suitability and the compactness of patterns. The essence of the optimization is based on the collective efforts of agents for formulating the optimal patterns. A local and global search strategy is proposed to inform the agents to select the sites properly. Three sets of hypothetical data were first used to verify the optimization effects. AgentLA was then applied to the solution of the actual land allocation optimization problems in Panyu city in the Pearl River Delta. The study has demonstrated that the proposed method has better performance than the simulated annealing method for solving complex spatial optimization problems. Experiments also indicate that the proposed model can produce patterns that are very close to the global optimums.

Cellular automata (CA), which are a kind of bottom-up approaches, can be used to simulate urban dynamics and land use changes effectively. Urban simulation usually involves a large set of GIS data in terms of the extent of the study area and the number of spatial factors. The computation capability becomes a bottleneck of implementing CA for simulating large regions. Parallel computing techniques can be applied to CA for solving this kind of hard computation problem. This paper demonstrates that the performance of large-scale urban simulation can be significantly improved by using parallel computation techniques. The proposed urban CA is implemented in a parallel framework that runs on a cluster of PCs. A large region usually consists of heterogeneous or polarized development patterns. This study proposes a line-scanning method of load balance to reduce waiting time between parallel processors. This proposed method has been tested in a fast-growing region, the Pearl River Delta. The experiments indicate that parallel computation techniques with load balance can significantly improve the applicability of CA for simulating the urban development in this large complex region.

Cellular automata (CA) have been increasingly used in simulating urban expansion and land-use dynamics. However, most urban CA models rely on empirical data for deriving transition rules, assuming that the historical trend will continue into the future. Such inertia CA models do not take into account possible external interventions, particularly planning policies, and thus have rarely been used in urban and land-use planning. This paper proposes to use artificial immune systems (AIS) as a technique for incorporating external interventions and generating alternatives in urban simulation. Inspired by biological immune systems, the primary process of AIS is the evolution of a set of ‘antibodies’ that are capable of learning through interactions with a set of sample ‘antigens’. These ‘antibodies’ finally get ‘matured’ and can be used to identify/classify other ‘antigens’. An AIS-based CA model incorporates planning policies by altering the evolution mechanism of the ‘antibodies’. Such a model is capable of generating different scenarios of urban development under different land-use policies, with which the planners will be able to answer ‘what if’ questions and to evaluate different options. We applied an AIS-based CA model to the simulation of urban agglomeration development in the Pearl River Delta in southern China. Our experiments demonstrate that the proposed model can be very useful in exploring various planning scenarios of urban development.

Land-cover classification is perhaps one of the most important applications of remote-sensing data. There are limitations with conventional (hard) classification methods because mixed pixels are often abundant in remote-sensing images, and they cannot be appropriately or accurately classified by these methods. This paper presents a new approach in improving the classification performance of remote-sensing applications based on mixed-label analysis (MLA). This MLA model can determine class proportions within a pixel in producing soft classification from remote-sensing data. Simulated images and real data sets are used to illustrate the simplicity and effectiveness of this proposed approach. Classification accuracy achieved by MLA is compared with other conventional methods such as linear spectral mixture models, maximum likelihood, minimum distance, and artificial neural networks. Experiments have demonstrated that this new method can generate more accurate land-cover maps, even in the presence of uncertainties in the form of mixed pixels.

Rapid land development in rapidly growing countries has created a series of land-use problems. The implementation of sustainable land use can alleviate some of these problems. It needs a set of tools for the exploration, design, modification, illustration, and evaluation of alternative planning scenarios. This paper demonstrates that the integration of cellular automata and agent-based modelling can provide a spatial exploratory tool for generating alternative development patterns. Sustainable development strategies are embedded in the modelling to regulate agents' behaviours. The use of agents can help to represent human-environment interactions in solving complex land-use problems. It is able to examine the effects of different stakeholders in influencing the process of land development. The proposed model has been applied to the simulation of planning scenarios for residential development in a rapidly expanding city in the Pearl River Delta.

This paper presents a new method to discover transition rules of geographical cellular automata (CA) based on a bottom‐up approach, ant colony optimization (ACO). CA are capable of simulating the evolution of complex geographical phenomena. The core of a CA model is how to define transition rules so that realistic patterns can be simulated using empirical data. Transition rules are often defined by using mathematical equations, which do not provide easily understandable explicit forms. Furthermore, it is very difficult, if not impossible, to specify equation‐based transition rules for reflecting complex geographical processes. This paper presents a method of using ant intelligence to discover explicit transition rules of urban CA to overcome these limitations. This ‘bottom‐up’ ACO approach for achieving complex task through cooperation and interaction of ants is effective for capturing complex relationships between spatial variables and urban dynamics. A discretization technique is proposed to deal with continuous spatial variables for discovering transition rules hidden in large datasets. The ACO–CA model has been used to simulate rural–urban land conversions in Guangzhou, Guangdong, China. Preliminary results suggest that this ACO–CA method can have a better performance than the decision‐tree CA method.

This paper presents a new method to improve the classification performance for remote-sensing applications based on swarm intelligence. Traditional statistical classifiers have limitations in solving complex classification problems because of their strict assumptions. For example, data correlation between bands of remote-sensing imagery has caused problems in generating satisfactory classification using statistical methods. In this paper, ant colony optimization (ACO), based upon swarm intelligence, is used to improve the classification performance. Due to the positive feedback mechanism, ACO takes into account the correlation between attribute variables, thus avoiding issues related to band correlation. A discretization technique is incorporated in this ACO method so that classification rules can be induced from large data sets of remote-sensing images. Experiments of this ACO algorithm in the Guangzhou area reveal that it yields simpler rule sets and better accuracy than the See 5.0 decision tree method.