Hyperspectral images (HSI) are three-dimensional (3D) data cubes with a unique joint dependency structure characterized by spectral continuity and spatial similarity. While Singular Spectrum Analysis (SSA) has proven to be an effective method for feature extraction, existing SSA-like methods do not adequately capture 3D features of HSI. To overcome this limitation, we rethink the 3D representation of HSI through dimensionality extension and propose a novel 3DSSA framework. In 3DSSA, we introduce patch embedding and adaptive embedding techniques to construct the trajectory tensor corresponding to the HSI, allowing us to explore its low-rank intrinsic features through tensor decomposition. Experimental results demonstrate the superior performance of the proposed method in capturing spatial detail information and improving feature classification accuracy. The code of this work is available at https://github.com/RsAI-lab.

The ecological service functions of urban green vegetation are closely related to its three-dimensional spatial structure. Regression method is widely used in extraction of vegetation three-dimensional information, such inversion of living vegetation volume (LVV), owing to its simplicity and efficiency. This kind of inversion models commonly built based on characteristics of vegetations, which always varies with the seasons. Thus, establish inversion models for different seasons separately is desirable. In this study, optical and SAR satellite data are combined to develop a seasonal LVV inversion method in Tangdao Bay coastline, Qingdao, China. In this method, the Sentinel-2 and Sentinel-1 satellite images are firstly used to classify the vegetation in summer and winter. Then, a variety of spectral, polarized and structural features are extracted from Sentinel-2 and Sentinel-1 data to build a LVV candidate inversion feature set. Lastly, two linear regression models based on different inversion features are developed. Accordingly, the seasonal LVV of urban area in Tangdao Bay coastline are obtained. The results show that the seasonal model can effectively improve the accuracy of LVV inversion.

Vegetation indices (VIs) are valuable in numerous remote sensing fields. Nevertheless, it is difficult to accurately discern vegetation using VIs, causing by the effects of the shaded vegetation, saturation vegetation, synthetic turf stadiums, and color steel tile. Measuring the spectral curves of shadows, synthetic turf stadiums, and color steel tiles, we found that the interfering features can be suppressed by a formula composition, named the Original Multiple Spectrum-based Vegetation Index (OMSVI). In addition, we integrated the anti-saturation function into OMSVI to solve the band saturation issue, resulting in the Improved Multiple Spectrum-based Vegetation Index (IMSVI). To address the structure saturation issue, the blue and red edge 1 bands could be used to alter the denominator structure of IMSVI, thus obtaining the final Modified Multiple Spectrum-based Vegetation Index (MMSVI). To demonstrate the generalization of the MMSVI, nine common VIs were compared in three large cities with diverse environments. The results of MMSVI showed that the vegetation extraction accuracy was stabilized at more than 90%, and the saturated LAI position continued to exceed 5.0, both of which were superior to the current VIs. In combination with a conventional optical sensor, it provided an innovative solution for monitoring vegetation in high-dynamic spaces, particularly in subtropical cities where saturation issues are more prevalent.

Spatial information is crucial in deep spectral–spatial hyperspectral image (HSI) classification methods. Spatial features can be divided into central features and surrounding features, which have different significance in high-precision HSI classification tasks. However, the existing algorithms ignore the discrepancy between the two attributes, which weaken decisive characteristics and lose spatial structures. To resolve the above issues, a patch-based anti-hierarchical encoder–decoder network is proposed, namely expansion and stripping network (ESNet). First, a surrounding relation module (SRM) is proposed to change the patch scale. Unlike existing scale change modules, the SRM increases the scale in the encoder and decreases the scale in the decoder by controlling the expansion and stripping of peripheral pixels of the patch, which enables ESNet to amply exploit the critical central features while utilizing the surrounding information in the anti-hierarchical structure. Second, we design a spectral–spatial attention module (SSAM) to extract features. To better represent land cover attributes, SSAM employs large- and small-kernel convolutions to generate spatial, spectral, and spectral–spatial weighted features. Finally, a novel skip connection [deep shallow feature fusion module (DSFM)] is designed to promote the model to fuse deep and shallow features while preserving spectral sequences of patches. Combining DSFM with the encoder–decoder structure can more scientifically fuse deep semantics and shallow details. We conduct extensive experiments in six commonly used HSI datasets to demonstrate the superiority of the proposed model. The codes will be available from the website.

Weakly supervised semantic segmentation methods can effectively alleviate the problem of high cost and difficult access to annotation in traditional methods. Among these approaches, point annotated semantic label not only offers a more affordable option but also provides accurate location and category information, playing an indispensable role in current research. However, point annotation labeling encounters challenges such as missing global and texture information, and limiting segmentation accuracy and efficiency while being susceptible to noise interference. For the above problems, a weakly supervised remote sensing image classification framework based on point annotated semantic label is proposed, which consists of three components: data augmentation, Pixel-Net, and iterative superpixel-based sample expansion (ISSE). First, the data augmentation method is used to generate a sufficient number of training samples. Subsequently, the weakly supervised network Pixel-Net is trained using point annotated semantic labels. Pixel-Net incorporates traditional image processing techniques such as edge detection and blurring into deep learning, enabling effective learning of edge and spectral semantic details while reducing the impact of noise on classification results. Finally, ISSE leverages contextual information from superpixels and pseudo-labels to enrich the valuable information in weakly supervised labels, thereby improving the model’s classification performance. In the experiments, existing semantic segmentation methods and Pixel-Net are evaluated on the Vaihingen and Zurich Summer datasets, and the effectiveness of ISSE is verified. The results show that Pixel-Net achieves the best segmentation accuracy on both datasets, while ISSE can effectively utilize the existing point annotation labels to mitigate the effect of noise and thus improve the accuracy of weakly supervised semantic segmentation.

Due to the cubic structure of a hyperspectral image (HSI), how to characterize its spectral and spatial properties in three dimensions is challenging. Conventional spectral-spatial methods usually extract spectral and spatial information separately, ignoring their intrinsic correlations. Recently, some 3D feature extraction methods are developed for the extraction of spectral and spatial features simultaneously, although they rely on local spatial-spectral regions and thus ignore the global spectral similarity and spatial consistency. Meanwhile, some of these methods contain huge model parameters which require a large number of training samples. In this paper, a novel Tensor Singular Spectral Analysis (TensorSSA) method is proposed to extract global and low-rank features of HSI. In TensorSSA, an adaptive embedding operation is first proposed to construct a trajectory tensor corresponding to the entire HSI, which takes full advantage of the spatial similarity and improves the adequate representation of the global low-rank properties of the HSI. Moreover, the obtained trajectory tensor, which contains the global and local spatial and spectral information of the HSI, is decomposed by the Tensor singular value decomposition (t-SVD) to explore its low-rank intrinsic features. Finally, the efficacy of the extracted features is evaluated using the accuracy of image classification with a support vector machine (SVM) classifier. Experimental results on three publicly available datasets have fully demonstrated the superiority of the proposed TensorSSA over a few state-of-the-art 2D/3D feature extraction and deep learning algorithms, even with a limited number of training samples.

Currently, long-range spectral and spatial dependencies have been widely demonstrated to be essential for hyperspectral image (HSI) classification. Due to the transformer superior ability to exploit long-range representations, the transformer-based methods have exhibited enormous potential. However, existing transformer-based approaches still face two crucial issues that hinder the further performance promotion of HSI classification: 1) treating HSI as 1D sequences neglects spatial properties of HSI, 2) the dependence between spectral and spatial information is not fully considered. To tackle the above problems, a large kernel spectral-spatial attention network (LKSSAN) is proposed to capture the long-range 3D properties of HSI, which is inspired by the visual attention network (VAN). Specifically, a spectral-spatial attention module is first proposed to effectively exploit discriminative 3D spectral-spatial features while keeping the 3D structure of HSI. This module introduces the large kernel attention (LKA) and convolution feed-forward (CFF) to flexibly emphasize, model, and exploit the long-range 3D feature dependencies with lower computational pressure. Finally, the features from the spectral-spatial attention module are fed into the classification module for the optimization of 3D spectral-spatial representation. To verify the effectiveness of the proposed classification method, experiments are executed on four widely used HSI data sets. The experiments demonstrate that LKSSAN is indeed an effective way for long-range 3D feature extraction of HSI.

Extracting buildings from remote sensing images plays an important role in urban development planning, disaster assessment and mapping. Convolutional neural network (CNN) has been widely applied to building extraction because of its powerful deep semantic feature extraction ability. However, existing CNN-based building extraction methods are difficult to accurately extract multiscale buildings with accurate edges because of the limitation of feature receptive fields and the loss of spatial detail information. For the above problems, this paper proposes a multiscale receptive field network (MSRF-Net) to accurately extract multiscale buildings from remote sensing images. MSRF-Net includes multiscale receptive field feature encoder (MRFF-Encoder) and multipath decoder. In the MRFF-Encoder, a multiscale attentional down (MSAD) module and asymmetric residual inception (ARI) module are proposed to capture multiscale receptive field features. In the multipath decoder, convolutions with different kernel size and dilation are used in three parallel paths to learn localization-preserved multiscale features with multiscale receptive field. What’s more, the features of different branches and MRFF-Encoder are fused by the proposed feature combination module, which contribute to capture context information of multiscale receptive field while recovering the resolution of feature space. The experimental results show that compared with the latest MAP-Net, MSRF-Net has achieved F1 score growth of 1.14%, 0.42%, 1.11% and IoU score growth of 1.68%, 0.76% and 1.64% respectively on Massachusetts data set, WHU data set and the Typical Cities Building data set.

Impervious surface (IS) and urban heat island (UHI) effects are always the research hotspots. However, the existing researches either ignore the impacts of IS material on UHI or fail to monitor the seasonal temporal variations of UHI. To this end, we explore the impacts of impervious surface materials on land surface temperature (LST) by analyzing their correlation and seasonal temporal variations. The results show that the mean LST for different impervious surface materials is statistically different from each other. Additionally, the contribution of IS to LST is affected by the material. Finally, the effect of impervious surface materials on LST has seasonal differences. These findings may help decision-makers develop more effective strategies to alleviate the urban heat island phenomenon.

Singular spectral analysis (SSA) has recently been successfully applied to feature extraction in hyperspectral image (HSI), including conventional (1-D) SSA in spectral domain and 2-D SSA in spatial domain. However, there are some drawbacks, such as sensitivity to the window size, high computational complexity under a large window, and failing to extract joint spectral-spatial features. To tackle these issues, in this article, we propose superpixelwise adaptive SSA (SpaSSA), that is superpixelwise adaptive SSA for exploiting local spatial information of HSI. The extraction of local (instead of global) features, particularly in HSI, can be more effective for characterizing the objects within an image. In SpaSSA, conventional SSA and 2-D SSA are combined and adaptively applied to each superpixel derived from an oversegmented HSI. According to the size of the derived superpixels, either SSA or 2-D singular spectrum analysis (2D-SSA) is adaptively applied for feature extraction, where the embedding window in 2D-SSA is also adaptive to the size of the superpixel. Experimental results on the three datasets have shown that the proposed SpaSSA outperforms both SSA and 2D-SSA in terms of classification accuracy and computational complexity. By combining SpaSSA with the principal component analysis (SpaSSA-PCA), the accuracy of land-cover analysis can be further improved, outperforming several state-of-the-art approaches.

As hyperspectral imagery (HSI) contains rich spectral and spatial information, a novel principal component analysis (PCA) and segmented-PCA (SPCA)-based multiscale 2-D-singular spectrum analysis (2-D-SSA) fusion method is proposed for joint spectral–spatial HSI feature extraction and classification. Considering the overall spectra and adjacent band correlations of objects, the PCA and SPCA methods are utilized first for spectral dimension reduction, respectively. Then, multiscale 2-D-SSA is applied onto the SPCA dimension-reduced images to extract abundant spatial features at different scales, where PCA is applied again for dimensionality reduction. The obtained multiscale spatial features are then fused with the global spectral features derived from PCA to form multiscale spectral–spatial features (MSF-PCs). The performance of the extracted MSF-PCs is evaluated using the support vector machine (SVM) classifier. Experiments on four benchmark HSI data sets have shown that the proposed method outperforms other state-of-the-art feature extraction methods, including several deep learning approaches, when only a small number of training samples are available.

As an essential reprocessing method, dimensionality reduction (DR) can reduce the data redundancy and improve the performance of hyperspectral image (HSI) classification. A novel unsupervised DR framework with feature interpretability, which integrates both band selection (BS) and spatial noise reduction method, is proposed to extract low-dimensional spectral-spatial features of HSI. We proposed a new Neighboring band Grouping and Normalized Matching Filter (NGNMF) for BS, which can reduce the data dimension whilst preserve the corresponding spectral information. An enhanced 2-D singular spectrum analysis (E2DSSA) method is also proposed to extract the spatial context and structural information from each selected band, aiming to decrease the intra-class variability and reduce the effect of noise in the spatial domain. The support vector machine (SVM) classifier is used to evaluate the effectiveness of the extracted spectral-spatial low-dimensional features. Experimental results on three publicly available HSI datasets have fully demonstrated the efficacy of the proposed NGNMF-E2DSSA method, which has surpassed a number of state-of-the-art DR methods.

This study presents a spectral–spatial self-attention network (SSSAN) for classification of hyperspectral images (HSIs), which can adaptively integrate local features with long-range dependencies related to the pixel to be classified. Specifically, it has two subnetworks. The spatial subnetwork introduces the proposed spatial self-attention module to exploit rich patch-based contextual information related to the center pixel. The spectral subnetwork introduces the proposed spectral self-attention module to exploit the long-range spectral correlation over local spectral features. The extracted spectral and spatial features are then adaptively fused for HSI classification. Experiments conducted on four HSI datasets demonstrate that the proposed network outperforms several state-of-the-art methods.

Integration of spectral and spatial information is extremely important for the classification of high-resolution hyperspectral images (HSIs). Gravitation describes interaction among celestial bodies which can be applied to measure similarity between data for image classification. However, gravitation is hard to combine with spatial information and rarely been applied in HSI classification. This paper proposes a Bayesian Gravitation based Classification (BGC) to integrate the spectral and spatial information of local neighbors and training samples. In the BGC method, each testing pixel is first assumed as a massive object with unit volume and a particular density, where the density is taken as the data mass in BGC. Specifically, the data mass is formulated as an exponential function of the spectral distribution of its neighbors and the spatial prior distribution of its surrounding training samples based on the Bayesian theorem. Then, a joint data gravitation model is developed as the classification measure, in which the data mass is taken to weigh the contribution of different neighbors in a local region. Four benchmark HSI datasets, i.e. the Indian Pines, Pavia University, Salinas, and Grss_dfc_2014, are tested to verify the BGC method. The experimental results are compared with that of several well-known HSI classification methods, including the support vector machines, sparse representation, and other eight state-of-the-art HSI classification methods. The BGC shows apparent superiority in the classification of high-resolution HSIs and also flexibility for HSIs with limited samples.

Synergetic use of both Synthetic Aperture Radar (SAR) and optical data has been applied for mapping impervious surface (IS) in recent years. However, researchers use only descending or ascending SAR to cooperate with optical data, which cannot avoid the problems caused by side looking of SAR, including layover and SAR shadow. This research explored and analyzed the impact of mapping IS by cooperating descending, ascending SAR and both of them with optical data respectively. To obtain a credible result, support vector machine (SVM) is employed to conduct the classification. The results indicated that the combined use of both descending and ascending SAR with optical data achieved the highest accuracy on IS mapping.

As the source of urban ecology, urban green space (UGS) has always been the focus of urban ecological research. The complex urban surface structure causes great interference to UGS extraction. In areas with high vegetation density, the vegetation index becomes rapidly saturated. Existing vegetation indices are not effective for the two problems due to that these indices do not make full use of the rich spectral information contained in hyperspectral image. To remedy these issues, a hyperspectral image based vegetation index (HSVI) is proposed. In the formulation of the HSVI numerator, we chose four new bands combination sensitive to vegetation to improve the identification of vegetation. We opt the sum of the red edge and green bands as the HSVI denominator and reconstruct the easily saturable band (760nm) in the form of exponential function to weaken the saturation problem. We use the hyperspectral image from Shanghai Theatre Academy and University of Houston with different geomorphological features to verify the effect of HSVI. The performance of HSVI is compared with three widely adopted vegetation indices, i.e., the normalized difference vegetation index (NDVI), the optimized soil-adjusted vegetation index (OSVAI) and the wide-dynamic-range vegetation index (WDRVI). The results show that the UGS extraction accuracy of HSVI is more than 90%, which is significantly better than the other indices. Meanwhile, HSVI can also solve the problem of vegetation index saturation. It can be proved that HSVI can fulfill the requirements of urban ecological research on a fine scale.

Accurate and timely landcover information plays an important role in land resources management and urban planning. In this article, a lightweight and multi-scale convolutional neural network model is proposed for high-resolution remote sensing images classification. Its inputs are the multi -scale patches to capture the multi -scale variability of spatial features in high-resolution remote sensing images. In this model, some standard convolution is replaced by depth separable convolution, which has fewer parameters and requires less calculation cost. The experiments were conducted on high-resolution images of two different regions (i.e., Beijing and Qingdao, China). The promising performance verified the proposed method is very efficient for the classification of high-resolution remote sensing imagery.

Singular spectrum analysis (SSA) and its 2-D variation (2D-SSA) have been successfully applied for effective feature extraction in hyperspectral imaging (HSI). However, they both cannot effectively use the spectral-spatial information, leading to a limited accuracy in classification. To tackle this problem, a novel 2D-SSA based multiscale feature fusion method, combining with segmented principal component analysis (SPCA), is proposed in this paper. The SPCA method is used for dimension reduction and spectral feature extraction, while multiscale 2D-SSA can extract abundant spatial features at different scales. In addition, a postprocessing via SPCA is applied on fused features to enhance the spectral discriminability. Experiments on two widely used datasets show that the proposed method outperforms two conventional SSA methods and other spectral-spatial classification methods in terms of the classification accuracy and computational cost.

Monitoring crop phenology is essential for analyzing the impacts of climate change and agronomic management on agricultural production. In this study, we proposed a new framework to monitor the phonological response of winter wheat using time series of Sentinel-1A as well as rainfall and temperature data, and ground phenological observations. The results showed that the radar backscatter coefficient ratio of VH to VV polarization showed the obvious seasonality that was related to vegetation phenology. Furthermore, the results showed that we can extract the phenology of winter wheat from the backscatter. Overall, this study clearly shows the strong response of Sentinel-1A data to winter wheat phenology and the potential for monitoring winter wheat phenology.

At present, remote sensing images are mutually restricted in temporal and spatial resolution. A single satellite sensor cannot obtain remote sensing images with both high spatial resolution and high temporal resolution. Spatiotemporal fusion of remote sensing images is a promising method to solve this issue. The spatial and temporal adaptive reflectance fusion model (STARFM) is a widely-accepted method for spatiotemporal fusion. However, STARFM selects similar pixels in a regular rectangular window. This neighborhood window has many different land cover types, which leads to wrong selection of similar pixels. Therefore, we develop a novel spatial and temporal adaptive reflectance fusion model based on superpixel, denote by S-STARFM. In the proposed method, the target pixels to be predicted are divided into two categories, including changed pixels and unchanged pixels. Then the superpixels are used to improve the selection of similar pixels. To verify the effectiveness of S-STARFM, the moderate resolution imaging spectrometer (MODIS) and Landsat Enhanced Thematic Mapper Plus (ETM+) data are used to generate high spatiotemporal resolution images. The prediction image accuracy shows that the proposed method outperforms the STARFM.

Feature extraction is of significance for hyperspectral image (HSI) classification. Compared with conventional handcrafted feature extraction methods, convolutional neural network (CNN) can automatically learn hierarchical features with discriminative information. However, two issues exist in applying CNN to HSI classification. One issue is how to represent the land covers at multiscale, the other is how to solve the “salt and pepper” noises caused by pixel-based CNN classification. To solve these issues, in this paper, a multiscale CNN is proposed to extract multiscale features for HSI classification, and then a region-based max voting scheme is applied to the classification map to solve the “salt and pepper” noises. Experiments on two classical data sets demonstrate that the proposed method is effective for HSI classification, especially for images with large scale changes.

The spatial-spectral classification for hyperspectral image (HSI) has been widely concerned. However, the traditional spatial-spectral classification methods are often easily affected by the noisy and heterogeneous pixel in the local region, leading to misclassification. Joint data gravitation classification (JDGC) shows that gravitation can suppress the interference of noisy pixel in the local region, but its flexibility is limited by high local heterogeneity. In this paper, a novel HSI classification method based on the local correlation of pixels and data gravitation (LC-DGC) is proposed. In LC-DGC, each pixel is assigned as an object with local mass, which is defined according to the correlation between each pixel and the central pixel in the local region. The pixels with small correlation contribute less to the local mass, which can decrease the interference of noisy and heterogeneous pixels effectively. Experimental results on two HSI datasets verify the superior classification performance of LC-DGC.

Fine extraction of buildings in very high-resolution (VHR) images plays an important role in urban planning and management. However, the large-variety in appearances and scales makes it challenge to extract buildings with accuracy. Several literatures demonstrate that convolutional neural networks (CNN) is effective in extracting complex buildings, owing to its superiority in high-level features learning. However, traditional CNN always shows poor performance ix extracting multiscale buildings and building boundary, due to its fixed receptive fields and repeated sub-sampling operations, respectively. Therefore, in this paper, we proposed a novel algorithm combining multiscale CNN (MCNN) model and superpixels to meet these two issues. This algorithm firstly designed a MCNN model by constructing a multiscale training samples database and inputs, to produce the preliminary classification building maps. Then, the boundary information provided by superpixels was combined with the CNN classification map using region-based max voting algorithm to produce the final building result. The effectiveness of this algorithm was tested in two well-known VHR datasets. Experimental results demonstrate that our proposed algorithm is outperformed comparison algorithms in extracting complex building in VHR images.

Extracting spectral-spatial information via sparse representation is a hotspot for hyperspectral image (HSI) classification. However, the spectral-spatial information extracted by the traditional joint sparse representation classification (JSRC) method is affected by heterogeneous and noisy pixels, which leads to some misclassifications. In this paper, we proposed a spectral-spatial classification framework based on joint superpixel-constrained and weighted sparse representation for HSI classification. Superpixel constraint is firstly used to remove the effects of heterogeneous pixels, which are located in a fixed sized blocks adopted by JSRC. The weighting scheme is then conducted to suppress the effects of noise. Finally, JSRC is performed on the superpixel-constrained and weighted regions obtained from the first two steps. The experimental results on Indian Pines dataset indicate that the proposed method outperforms compared methods, with more effective and robust performance.

Band selection is an important data dimensionality reduction tool in hyperspectral images (HSIs). To identify the most informative subset band from the hundreds of highly corrected bands in HSIs, a novel hyperspectral band selection method using a crossover‐based gravitational search algorithm (CGSA) is presented in this study. In this method, the discriminative capability of each band subset is evaluated by a combined optimisation criterion, which is constructed based on the overall classification accuracy and the size of the band subset. As the evolution of the criterion, the subset is updated using the V ‐shaped transfer function‐based CGSA. Ultimately, the band subset with the best fitness value is selected. Experiments on two public hyperspectral datasets, i.e. the Indian Pines dataset and the Pavia University dataset, have been conducted to test the performance of the proposed method. Comparing experimental results against the basic GSA and the PSOGSA (hybrid PSO and GSA) revealed that all of the three GSA variants can considerably reduce the band dimensionality of HSIs without damaging their classification accuracy. Moreover, the CGSA shows superiority on both the effectiveness and efficiency compared to the other two GSA variants.

Accurate detection of topographic shadows is of great importance, since topographic shadowing is an inevitable hamper for the interpretation of remotely sensed images covered mountainous areas. In this paper, a novel method is proposed for effective and efficient topographic shadow detection for the images obtained from Sentinel-2A multispectral imager (MSI) by combining both the spectral and spatial information. In this method, four feature indices were firstly extracted from the original Sentinel-2A spectral bands to capture the essential spectral characteristics. Specifically, we constructed a topographic shadow index (TSI) to enhance topographic shadows, focusing on the spectral signatures of shadows in Sentinel-2A MSI imagery. To further enhance the difference between shadows and other objects, the TSI is combined with the first component of the principal component analysis (FPC), the soil-adjusted vegetation index (SAVI) and the normalized water index (NDWI), representing topographic shadows, rocks, vegetation and water, respectively. Finally, a convolutional neural network (CNN) was used by operating directly on indices input due to its remarkable classification performance, which exploits the spatial contextual information and spectral features for effective topographic extraction. Our experiments with one Sentinel-2A image show that the proposed approach has led to satisfactory performances, with few errors in shadow maps and insignificant confusion with spectral-similar land covers.

Balancing exploration and exploitation according to evolutionary states is crucial to meta-heuristic search (M-HS) algorithms. Owing to its simplicity in theory and effectiveness in global optimization, gravitational search algorithm (GSA) has attracted increasing attention in recent years. However, the tradeoff between exploration and exploitation in GSA is achieved mainly by adjusting the size of an archive, named , which stores those superior agents after fitness sorting in each iteration. Since the global property of remains unchanged in the whole evolutionary process, GSA emphasizes exploitation over exploration and suffers from rapid loss of diversity and premature convergence. To address these problems, in this paper, we propose a dynamic neighborhood learning (DNL) strategy to replace the model and thereby present a DNL-based GSA (DNLGSA). The method incorporates the local and global neighborhood topologies for enhancing the exploration and obtaining adaptive balance between exploration and exploitation. The local neighborhoods are dynamically formed based on evolutionary states. To delineate the evolutionary states, two convergence criteria named limit value and population diversity, are introduced. Moreover, a mutation operator is designed for escaping from the local optima on the basis of evolutionary states. The proposed algorithm was evaluated on 27 benchmark problems with different characteristic and various difficulties. The results reveal that DNLGSA exhibits competitive performances when compared with a variety of state-of-the-art M-HS algorithms. Moreover, the incorporation of local neighborhood topology reduces the numbers of calculations of gravitational force and thus alleviates the high computational cost of GSA.