MSA-YOLO: A Remote Sensing Object Detection Model Based on Multi-Scale Strip Attention

Abstract

Remote sensing image object detection holds significant research value in resources and the environment. Nevertheless, complex background information and considerable size differences between objects in remote sensing images make it challenging. This paper proposes an efficient remote sensing image object detection model (MSA-YOLO) to improve detection performance. First, we propose a Multi-Scale Strip Convolution Attention Mechanism (MSCAM), which can reduce the introduction of background noise and fuse multi-scale features to enhance the focus of the model on foreground objects of various sizes. Second, we introduce the lightweight convolution module GSConv and propose an improved feature fusion layer, which makes the model more lightweight while improving detection accuracy. Finally, we propose the Wise-Focal CIoU loss function, which can reweight different samples to balance the contribution of different samples to the loss function, thereby improving the regression effect. Experimental results show that on the remote sensing image public datasets DIOR and HRRSD, the performance of our proposed MSA-YOLO model is significantly better than other existing methods.

A New Lunar Lineament Extraction Method Based on Improved UNet++ and YOLOv5

“…It has been proved that convolutional neural networks have satisfied result on crater detection based on DEM and CCD remote sensing data. For example, the FasterRCNN [21] has already been applied in object detection with higher efficiency compared with previous work,MSA-YOLO [22] is desgined to object detection in DIOR dataset, RSI-YOLOv5 [23] is applied in remote sensing dataset such as DOTA and NWPU-VHR with outperform results, and a Graph Neural Network (GNN) [24] is desgined to execute the semantic segmentation in satellite image data. It can be concluded that deep learning method has great potential in remote sensing data processing.…”

Real-Time Detection of an Undercarriage Based on Receptive Field Blocks and Coordinate Attention

“…Considering that precision and recall have an interactive relationship and cannot be directly involved in the evaluation, the concepts of Average Precision (AP) and Mean Average Precision (mAP) are introduced, and the higher the two indicators are, the better the detection is; see Equations ( 26) and (27).…”

“…For remote sensing monitoring, Li et al [ 26 ] improved the detection precision of small-resolution images by incorporating an attention mechanism into YOLOv5, a bidirectional feature pyramid, and a small-targets detection layer. In contrast, Su et al [ 27 ] lightened the YOLOv5 module and optimized the loss function to enhance the network’s detection performance for remote sensing images.…”