Remote sensing image object detection based on improved SSD
Shuchao Liu, Huajun Shi, Zhan Qi Guo · 2022 3rd International Conference on Computer Vision, Image and Deep Learning & International Conference on Computer Engineering and Applications (CVIDL & ICCEA) · 2022
Aiming at the problem that the detection accuracy of the SSD object detection model is insufficient when detecting remote sensing images, an improved SSD object detection model is proposed. This method uses Mobilenet v3 to replace the backbone network of the ssd algorithm, and adds the FPN network for feature fusion, and the shallow feature map with high resolution is fused with the deep feature map with rich semantic informationAt the same time, a feature pyramid is built between feature maps to enhance the target feature. Experimental results show that the detection accuracy of the improved method is significantly improved compared with the original algorithm, and the effect on the detection of remote sensing image objects is better.