Fenglei Chen

As a crucial approach for comprehending land surface changes, remote sensing images (RSIs) change detection methods based on deep learning have been extensively studied in recent years. Among these methods, the Siamese network-based method has demonstrated remarkable performance. However, existing approaches are susceptible to interference from irrelevant factors such as shadows, noise, and unrelated changes in a single temporal, which limits the model's ability to accurately perceive bitemporal crucial semantic information. To address the above problems, this paper proposes a frequency-domain information assisted semantic interaction network (FASI-Net) to ensure the effective extraction of crucial semantic information in the original bitemporal features. We attempt to employ frequency-domain information to explicitly interact bitemporal low-frequency semantic information during the encoding process, subsequently preserving and enhancing the high-frequency information of each temporal feature. This enables model comprehensive perception of bitemporal features, effectively distinguishing the semantic similar information from the semantic difference information, and accurately identifying changed regions versus unchanged regions. Extensive experiments on three typical RSIs change detection datasets demonstrate a significant improvement in the performance of our proposed method (with 4.53%/7.57%, 6.66%/10.69%, and 5.11%/7.24% improvements over the baseline in terms of F1/IoU metrics for WHU-CD, LEVIR-CD, and SYSU-CD datasets). Moreover, our FASI-Net comprehensively achieves state-of-the-art results with F1/IoU reaching 92.86%/86.67%, 91.73%/84.71%, and 83.72%/72.00% on WHU-CD, LEVIR-CD, and SYSU-CD datasets respectively. In addition, the proposed bitemporal low-frequency semantic interaction module (BiLIM) can be seamlessly inserted into existing change detection models to achieve effective crucial semantic information extraction. The code is available at https://github.com/FlyC235/FASI-Net.

Salient object detection (SOD) aims to identify the most prominent regions in images. However, the large model sizes , high computational costs , and slow inference speeds of existing RGB-D SOD models have hindered their deployment on real-world embedded devices. To address this issue, we propose a novel method named AirSOD, which is committed to lightweight RGB-D SOD. Specifically, we first design a hybrid feature extraction network, which includes the first three stages of MobileNetV2 and our Parallel Attention-Shift convolution (PAS) module. Using the novel PAS module enables capturing both long-range dependencies and local information to enhance the representation learning while significantly reducing the number of parameters and computational complexity. Secondly, we propose a Multi-level and Multi-modal feature Fusion (MMF) module to facilitate feature fusion, and a Multi-path enhancement for Feature Refinement (MFR) decoder for feature integration. The proposed method significantly reduces the model size by 63%, decreases the computational complexity by 43%, and improves the inference speed by 43% compared with the cutting-edge model (MobileSal). We test our AirSOD on six widely-used RGB-D SOD datasets. Extensive experimental results demonstrate that our method obtains satisfactory performance. The source codes will be made available.

Multi-modal feature fusion and effectively exploiting high-level semantic information are critical in salient object detection (SOD). However, the depth maps complementing RGB image fusion strategies cannot supply effective semantic information when the object is not salient in the depth maps. Furthermore, most existing (UNet-based) methods cannot fully exploit high-level abstract features to guide low-level features in a coarse-to-fine fashion. In this paper, we propose a compensated attention feature fusion and hierarchical multiplication decoder network (CAF-HMNet) for RGB-D SOD. Specifically, we first propose a compensated attention feature fusion module to fuse multi-modal features based on the complementarity between depth and RGB features. Then, we propose a hierarchical multiplication decoder to refine the multi-level features from top down. Additionally, a contour-aware module is applied to enhance object contour. Experimental results show that our model achieves satisfactory performance on five challenging SOD datasets, including NJU2K, NLPR, STERE, DES, and SIP, which verifies the effectiveness of the proposed CAF-HMNet.

This letter introduces a novel Fourier-based data augmentation strategy for visible-thermal person re-identification (VT-ReID). Different from some existing methods which are proposed from the perspective of network structure and loss functions, our method aims to fully consider the semantic information from the perspective of data preprocessing. The main hypothesis is that the phase component in the Fourier domain contains high-level semantic information and the amplitude component contains low-level modality awareness information. In order to make the model pay more attention to semantic information learning, we design a simple but effective Fourier-based semantic augmentation (FSA) module, which can be inserted seamlessly into any existing models. Extensive experiments on RegDB and SYSU-MM01 datasets have shown that our proposed method can improve the VT-ReID performance significantly and achieve state-of-the-art performance.

Multi-spectral semantic segmentation has shown great advantages under poor illumination conditions, especially for remote scene understanding of autonomous vehicles, since the thermal image can provide complementary information for RGB image. However, methods to fuse the information from RGB image and thermal image are still under-explored. In this paper, we propose a simple but effective module, add–multiply fusion (AMFuse) for RGB and thermal information fusion, consisting of two simple math operations—addition and multiplication. The addition operation focuses on extracting cross-modal complementary features, while the multiplication operation concentrates on the cross-modal common features. Moreover, the attention module and atrous spatial pyramid pooling (ASPP) modules are also incorporated into our proposed AMFuse modules, to enhance the multi-scale context information. Finally, in the UNet-style encoder–decoder framework, the ResNet model is adopted as the encoder. As for the decoder part, the multi-scale information obtained from our proposed AMFuse modules is hierarchically merged layer-by-layer to restore the feature map resolution for semantic segmentation. The experiments of RGBT multi-spectral semantic segmentation and salient object detection demonstrate the effectiveness of our proposed AMFuse module for fusing the RGB and thermal information.

This paper focuses on the high-resolution (HR) remote sensing images semantic segmentation task, whose goal is to predict semantic labels in a pixel-wise manner. Due to the rich complexity and heterogeneity of information in HR remote sensing images, the ability to extract spatial details (boundary information) and semantic context information dominates the performance in segmentation. In this paper, based on the frequently used fully convolutional network framework, we propose a boundary enhancing semantic context network (BES-Net) to explicitly use the boundary to enhance semantic context extraction. BES-Net mainly consists of three modules: (1) a boundary extraction module for extracting the semantic boundary information, (2) a multi-scale semantic context fusion module for fusing semantic features containing objects with multiple scales, and (3) a boundary enhancing semantic context module for explicitly enhancing the fused semantic features with the extracted boundary information to improve the intra-class semantic consistency, especially in those pixels containing boundaries. Extensive experimental evaluations and comprehensive ablation studies on the ISPRS Vaihingen and Potsdam datasets demonstrate the effectiveness of BES-Net, yielding an overall improvement of 1.28/2.36/0.72 percent in mF1/mIoU/OA over FCN_8s when the BE and MSF modules are combined by the BES module. In particular, our BES-Net achieves a state-of-the-art performance of 91.4% OA on the ISPRS Vaihingen dataset and 92.9%/91.5% mF1/OA on the ISPRS Potsdam dataset.

The working environment of coal mine roadway has become more severe with the extension of the depth of coal mine roadway. Therefore, there is a higher demand for the size and performance of trackless rubber tire vehicles. The rubber wheel drilling rig developed in combination with the actual needs of coal mine roadway. Based on the detailed calculation and comparison of the two schemes, the applicable drive hydraulic system is proposed. On this basis, the advantages in use are analyzed. Through theoretical analysis, the key parameters affecting the differential and differential problems are proposed. The applicable solutions for the two coal mines are studied, and a more suitable scheme is selected through comparison. The static and dynamic test results show that the system has good operation performance and high efficiency under the condition of limited volume of the rubber drilling rig, which makes it have high passability and stability.