Road crack detection in complex scenarios is challenged by road surface backgrounds such as water stains, oil stains, and printed road signs. To better address these issues, we propose a Multimodal Road Crack Segmentation Network–TMENet, based on complex scenes. In this article, we introduced text guidance in road crack segmentation tasks to supplement more detailed crack features. Firstly, we designed the Textual Crack Enhancer module(TCE Module) to fine-tune the Text Transformer in the Clipseg model to enhance its ability to parse specific crack descriptions. Secondly, we designed a multiscale Mamba module(Ms-Mamba Module) that integrates multiscale composite convolution and residual connections to improve the sensitivity of the model to local and multiscale details and enhance the network’s ability to recover lost features. Then, we designed a Laplacian edge detection module (LED module) that combines deep visual features and high-frequency edge information to enhance the crack edge detection capability in complex scenes through a refined feature adjustment mechanism. Finally, TMENet achieved the best results in comparative testing with seven state-of-the-art segmentation networks on the CampCrack800, AsphaltCrack300, Crackseg9K, and CRKWH100 datasets. In addition, TMENet has achieved excellent results in actual road testing, and the test results show that TMENet is an effective auxiliary method to improve the accuracy of road safety detection.

In the field of early automatic detection of smoke from forest fires, there is the issue of the small size and interference of smoke detection by clouds. The conventional NMS (Non-Maximum Suppression) requires manual adjustment of the threshold, which may result in missed or erroneous detection. This paper proposes a high-accuracy anti-interference forest fire smoke detection network for small objects. Firstly, a window feature extractor based on singular value decomposition (SVD-STR) is designed. This extractor is capable of extracting more representative features, of capturing small and inconspicuous features in the image, and of reducing the complexity and computation of the model. Secondly, a SinThreshold Screening Attention Mechanism (SinAttention) is proposed, which can filter interference information and enhance the discriminative power of the features, thereby facilitating the accurate recognition and distinction of smoke and clouds. Subsequently, a variational particle swarm soft suppression optimization (PGS) is proposed as a means of further enhancing the optimization effect. This is achieved by adjusting the suppression strategy and incorporating a Gaussian variational particle swarm algorithm. In conclusion, an IoT forest fire detection system based on PSSNet has been constructed. The experimental results demonstrate that the mAP50 value of the method is 98.2%, the value of mAP50-95 is 80.4%, and the FPS value is 35.7. These values are superior to those of current forest fire smoke detection methods and can be utilized for the precise detection of forest fire smoke, thereby providing technical support for forest ecological protection.

As a key tropical economic crop, rubber trees play a vital role in both the global rubber industry and the health of ecological systems. Fine-grained segmentation of rubber tree point clouds is essential for accurately extracting structural parameters and achieving effective monitoring and management. However, existing unsupervised segmentation methods are often affected by ground noise and overlapping tree crowns, leading to suboptimal segmentation results and posing significant challenges for individual rubber tree segmentation. To address these issues, this study proposes a fine-grained segmentation network for rubber trees based on UAV LiDAR point clouds, termed RTreeNet. First, we designed a Multi-Scale Feature Aggregation (MSFA) module to tackle the issue of leaf overlap by capturing geometric features at the edges of tree crowns. Secondly, we proposed a Cosine-Space Cross Attention (CSCA) module, which calculates the cosine similarity of vertical and horizontal features for each point, effectively eliminating interference from ground noise. Additionally, an Adaptive Coati Particle Optimization Algorithm (ACPA) was proposed to determine the optimal learning rate for the network, further enhancing segmentation accuracy. Experimental evaluation demonstrates that the proposed RTreeNet outperforms seven state-of-the-art point cloud segmentation architectures and four conventional segmentation algorithms on our custom dataset, achieving a mean Intersection over Union (mIoU) of 86.3% and an F-score of 92.5%. In the generalization experiment, RTreeNet showed high accuracy and stability on three public datasets. The method also measured the specific structural parameters (tree height, crown diameter, and breast diameter) of rubber trees in the two regions, providing strong technical support for the refined management of rubber trees, agricultural planning, pest control, and rubber yield prediction.

As an important economic crop in tropical regions, rubber trees require precise individual-level identification and structural parameter extraction to support productivity assessment and fine-scale plantation management. However, existing point cloud segmentation methods face significant challenges in handling species diversity, overlapping canopies, and clustering parameter sensitivity. Therefore, we propose an incremental learning-driven network, termed RtSegNet, for UAV LiDAR-based rubber tree point cloud segmentation and clustering optimization. Firstly, we design a Tree-Structure Incremental Learning Sampling (TSILS) strategy, which significantly improves model adaptability across diverse forest structures and enhances training efficiency. Secondly, a Fourier Spectrum-enhanced Spatial Graph Clustering (FSGC) method is proposed to improve feature discrimination in overlapping canopy regions. Finally, a Gaussian Random Duck Swarm Algorithm (GR-DSA) is proposed to adaptively optimize clustering parameters, enabling robust segmentation across trees of varying scales. Experimental results on a self-constructed rubber tree dataset show that RtSegNet outperforms ten state-of-the-art methods, achieving a mean Intersection over Union (mIoU) of 76.37%, an F-score of 87.11%, and reducing training time to 25.04 hours. Generalization tests on the FOR-instance and NIBIO MLS public datasets further confirm the method’s cross-environment robustness.Based on the segmentation outputs, key structural parameters such as tree height, canopy diameter, and canopy volume are extracted. The estimated tree height achieves anR2of 0.94 with a root mean square error (RMSE) of 0.52, providing strong technical support for accurate monitoring, yield prediction, health assessment, and the modernization of tropical agriculture. The source code in this work is available at https://github.com/aaaaasleep/RtSegNet.

Modern human productivity and daily life rely on identifying ground objects using remote sensing images (RSIs). Traditional remote sensing object detection (RSOD) techniques lack timeliness and accuracy and fail to meet practical demands. Existing deep learning algorithms face continued challenges when processing RSIs because of the diverse shapes and extensive scale variations of objects, of which a significant proportion are small scale. To address these challenges, we propose the PSWP-DETR, a transformer-based network that leverages adaptive deformation learning and multiscale integration for enhanced object detection in remote sensing. First, we propose PradatorConv (PdConv) to address the significant shape changes of objects because it adaptively learns the horizontal and vertical deformations to perceive the complex geometric features of RSIs. Second, we propose scale-wise differential modules (SDMs), which comprise multiscale convolution (MSC) and edge captor convolution (ECC). SDM integrates features across various scales and captures edge characteristics and local textures. This is advantageous for detecting multiscale objects, tiny objects with limited feature information. Finally, we propose the whale particle optimization (WPO) algorithm for learning rate optimization, which improves convergence speed and accuracy. Experiments using the VisDrone2019-DET, DIOR, and AI-TOD datasets demonstrated that PSWP-DETR achieves the best accuracy benefits, offering significant insights for future RSOD efforts. The source code will be available athttps://github.com/Get1star/PSWP-DETR.git.

Deep reinforcement learning-based object detection approaches center around a pivotal concept: hierarchically scaling image segments that harbor more intricate details. Compared with the traditional object detection approaches, this approach significantly curbs the quantity of region proposals. This reduction holds paramount significance in curtailing the computational overhead. However, common deep reinforcement learning-based approaches suffer from a significant defect in terms of precision. This issue arises from inadequacies in representing image states appropriately and the unstable learning ability exhibited by the agent. To address these issues, we present the LHAR-RLD. First, we design the Low-dimensional RepVGG(LDR) feature extractor to reduce memory consumption and to reduce the difficulty of fitting downstream networks. Second, we propose the Hybrid DQN(HDQN) to enhance the agent's ability to determine the state-action of images in complex environments. Then, the Adaptive Dynamic Reward Function(ADR) is crafted to dynamically adjust the reward based on shifts within the agent's exploration environment. Finally, the ROI Align-based bounding box regression network (RABRNet) is proposed, which aims at further regressing the localization results of reinforcement learning to improve the detection precision. Our method accomplishes 74.4% mAP on the VOC2007, 76.2% mAP on the COCO2017, 75.2% Precision on the SF dataset, with 1.43G FLOPs. The precision outperforms the advanced deep reinforcement learning approaches and the computational cost is far lower than theirs and mainstream object detection methods. This method facilitates highly accurate object localization with minimal computational demands, which means it has notable applications on resource-constrained devices.

Remote sensing (RS) technology plays a critical role in monitoring our constantly changing world, with building change detection (BCD) being a pivotal application that contributes to urban planning, disaster management, and environmental monitoring. In tasks of BCD in high-resolution RS images, it is faced with challenges such as complex backgrounds, redundancy in feature fusion information, and imbalance of positive and negative samples. Utilizing high-resolution RS images for BCD tasks remains challenging. Therefore, a new BCD network named Meta-SGNet with Siamese architecture is proposed. First, a self-calibrating cross module (SCCM) algorithm is proposed to extract the morphological characteristics of buildings in RS images effectively. Subsequently, the gated feature fusion module (GFFM) is proposed to fuse the features of bitemporal buildings dynamically. Finally, a self-learning meta-encoder (Meta-E) is proposed, which uses a meta-learning algorithm to guide the encoder to encode the bitemporal RS image to better pay attention to the learning of positive samples of building changes to improve the accuracy of BCD. Experimental results show that Meta-SGNet outperforms ten state-of-the-art (SOTA) BCD methods on three datasets (Google-CD, WHU-CD, and LEVIR-CD). In the practical application, we acquired 40 pairs of high-resolution image pairs via Google Earth API for a real BCD task. The application results show that Meta-SGNet can accurately capture the range of building changes and shows high adaptability and the ability to quickly detect building changes in different scenarios.

Road crack detection in complex scenarios is challenged by vehicles, traffic facilities, road printed signs and fine cracks. In order to better solve these problems, a novel dense nested depth U-shaped structure for crack image segmentation network named DUCTNet is proposed. Firstly, a depth dense nested structure is designed by combining the superior performance of the Unet$++$dense nested structure and the deep nested structure of U2Net. This structure improves the ability of the model to extract crack features in depth. Second, a novel deep competitive fusion feature extraction block is proposed. It improves the feature dissimilarity between the cracks and the background by competitive fusion. Then, a novel high-density feature fusion attention mechanism is proposed. This method enhances the contextual and sensitive information of cracks both horizontally and vertically by increasing the feature density. Finally, DUCTNet achieves the best results in comparison tests with eight state-of-the-art specialized crack segmentation networks in both self-built datasets and four public datasets. In addition, DUCTNet achieves excellent results in real road tests, which proves that DUCTNet can provide engineers and technicians with a better means of detecting road cracks.

There is a problem in the field of early automatic detection of forest fire smoke that due to low concentration or tiny size, some smoke is difficult to capture. This article proposes a multiscale convergence coordinated pyramid network (MCCPN) with mixed attention and Fast-robust NMS (MMFNet) for the fast detection of forest fire smoke. First, an MCCPN is designed, which combines a dual-attention feature pyramid network and a coordinated convergence module. It improves the detection rate of targets of different sizes. Second, a mixed attention module is designed to focus more on the smoke in the image and enhance the extraction of horizontal and vertical features of smoke. Then, a Fast-robust nonmaximum suppression is proposed to accelerate the convergence of bounding boxes and increase the accuracy of the prediction box. Finally, a forest fire detection system of the Internet of Things using MMFNet is built. The experimental results show that our method achieves 80.72% AP, 88.52% AP50, 83.45% AP75, 46.88% AR, and 154 FPS, which is superior to the state-of-art forest fire smoke detection methods.

Remote sensing image (RSIs) segmentation is widely used in urban planning, natural disaster detection and many other fields. Compared with natural scene images, RSIs have higher resolution, complex imaging, and diverse object shapes and sizes, while semantic segmentation methods based on deep learning often require many data labels. In this paper, we propose a semi-supervised RSIs segmentation network with multi-scale deformable threshold feature extraction module and mixed attention (MDMANet). First, a pyramid ensemble structure is used, which incorporates deformable convolution and bole convolution, to extract features of objects with different shapes and sizes and reduce the influence of redundant features. Meanwhile, a mixed attention (MA) is proposed to aggregate long-range contextual relationships and fuse low-level features with high-level features. Second, an FCN-based full convolution discriminator task network is designed to help evaluate the feasibility of unlabeled image prediction results. We performed experimental validation on three datasets, and the results show that MDMANet segmentation provides more significant improvement in accuracy and better generalization than existing segmentation networks.

Currently, the use of rich spectral and spatial information of hyperspectral images (HSIs) to classify ground objects is a research hotspot. However, the classification ability of existing models is significantly affected by its high data dimensionality and massive information redundancy. Therefore, we focus on the elimination of redundant information and the mining of promising features and propose a novel Bole convolution (BC) neural network with a tandem three-direction attention (TDA) mechanism (BTA-Net) for the classification of HSI. A new BC is proposed for the first time in this algorithm, whose core idea is to enhance effective features and eliminate redundant features through feature punishment and reward strategies. Considering that traditional attention mechanisms often assign weights in a one-direction manner, leading to a loss of the relationship between the spectra, a novel three-direction (horizontal, vertical, and spatial directions) attention mechanism is proposed, and an addition strategy and a maximization strategy are used to jointly assign weights to improve the context sensitivity of spatial–spectral features. In addition, we also designed a tandem TDA mechanism module and combined it with a multiscale BC output to improve classification accuracy and stability even when training samples are small and unbalanced. We conducted scene classification experiments on four commonly used hyperspectral datasets to demonstrate the superiority of the proposed model. The proposed algorithm achieves competitive performance on small samples and unbalanced data, according to the results of comparison and ablation experiments. The source code for BTA-Net can be found athttps://github.com/vivitsai/BTA-Net.

The classification of ground objects from hyperspectral images (HSIs) is of great importance for human perception of information about the terrain and landscape. HSIs have numerous dimensions, and obtaining the data is difficult. The issue of slow convergence of neural network training is brought on by high dimensional data, and the neural network’s performance is impacted by the challenging data acquisition process. In order to achieve the effects of low data dependence and rapid convergence, we propose a redundancy elimination network architecture with decoupled-gaze attention mechanism and phantom fractal modules (DGPF-RENet) for HSIs classification. First, we propose the decoupled-gaze attention mechanism (DGA) to make full use of correlation between adjacent bands and the continuity of neighboring pixels in HSIs. Then, a redundancy elimination module (REM) is proposed to reduce the number of feature points and eliminate redundant information while preserving the contextual information and relationships between pixels. Finally, the phantom fractal module (PFM) is proposed, which improves the scale of feature learning by fractalising convolutions at multiple scales. Four publicly available HSIs datasets, including Indian Pines, Salinas, DFC2018, and WHUHi-HongHu, were used in our experiments. According to experimental findings, when compared to other state-of-the-art methods, our method performs best with a small number of training samples and few iterations. We have released our code and models at https://github.com/yuhua666/DGPF-RENet.

Forest fires are a huge ecological hazard, and smoke is an early characteristic of forest fires. Smoke is present only in a tiny region in images that are captured in the early stages of smoke occurrence or when the smoke is far from the camera. Furthermore, smoke dispersal is uneven, and the background environment is complicated and changing, thereby leading to inconspicuous pixel-based features that complicate smoke detection. In this paper, we propose a detection method called multioriented detection based on a value conversion-attention mechanism module and Mixed-NMS (MVMNet). First, a multioriented detection method is proposed. In contrast to traditional detection techniques, this method includes an angle parameter in the data loading process and calculates the target’s rotation angle using the classification prediction method, which has reference significance for determining the direction of the fire source. Then, to address the issue of inconsistent image input size while preserving more feature information, Softpool-spatial pyramid pooling (Soft-SPP) is proposed. Next, we construct a value conversion-attention mechanism module (VAM) based on the joint weighting strategy in the horizontal and vertical directions, which can specifically extract the colour and texture of the smoke. Ultimately, the DIoU-NMS and Skew-NMS hybrid nonmaximum suppression methods are employed to address the issues of smoke false detection and missed detection. Experiments are conducted using the homemade forest fire multioriented detection dataset, and the results demonstrate that compared to the traditional detection method, our model’s mAP reaches 78.92%, mAP 50 reaches 88.05%, and FPS reaches 122.

In hatching process, temperature, relative humidity and oxygen concentration are three main parameters to the success of the incubator. Incubation system is a multi-variable multi-interference and time delay complex dynamic system, a hybrid intelligent control method is proposed. Based on characteristics of complexity and uncertainty for its mathematical models in the hatching process, this paper will divide the system into two structures, considering of the effect of throttle opening which causes temperature and relative humidity big interference and strong coupling, feed-forward compensation decoupling control algorithm is employed for before-class structure to realize the decoupling of main parameters. After decupling, according to different conditions fuzzy immune PID control algorithm or fuzzy control algorithm is selected for back-class structure to realize incubator intelligent control. Finally, the run of system verifies the effectiveness of the control algorithm.