Abstract:The proportion of small objects in remote sensing images is relatively small, which is prone to feature loss or interference from the surrounding complex background in the detection process. To solve this problem, a remote sensing small object detection network (CD-YOLOX) based on enhanceds' feature pyramid network (CS-PANet) and global local combine module (DSAN) based on YOLOX is proposed. Firstly, to solve the problem of small object feature loss and surrounding complex background interference caused by multiple convolution and feature stacking operations in PANet, CS-PANet is proposed. This method improves the network's focus on effective feature channels for small objects by using channel attention at the input of PANet, meanwhile, the input that passes the channel attention is connected to the output of PANet across layers, which makes the PANet retain the richer original information of small objects. Secondly, to further reduce the interference of the surrounding complex background on the small objects, a DSAN module consisting of self-attention mechanism, dilated convolution and residual connection is proposed before network prediction. This module combines self-attention mechanism with dilated convolution so that the network focuses on the global feature region of the small object in the feature map while effectively complementing the local context of this region, and preserving the original information through residual connection. Finally, the effectiveness of the method is verified using the remote sensing dataset NWPU VHR-10 and the general datasets KITTI and PASCAL VOC. The experiment shows that the method improves the detection accuracy by 5.56% in the NWPU VHR-10 dataset and by 1.93% and 2.51% in the KITTI and PASCAL VOC datasets respectively compared to the original network, which fully verifies the effectiveness of the method for detecting the small objects of remote sensing and the ability of the method to detect general purpose objects.

Deconv R-Cnn For Small Object Detection On Remote Sensing Images

$\mathcal{R}^2$ -CNN: Fast Tiny Object Detection in Large-Scale Remote Sensing Images

A Deep Lightweight Convolutional Neural Network Method for Real-Time Small Object Detection in Optical Remote Sensing Images

Sig-NMS-Based Faster R-CNN Combining Transfer Learning for Small Target Detection in VHR Optical Remote Sensing Imagery

An Improved Faster R-CNN for Small Object Detection

R$^2$-CNN: Fast Tiny Object Detection in Large-Scale Remote Sensing Images

Deformable ConvNet with Aspect Ratio Constrained NMS for Object Detection in Remote Sensing Imagery

Remote Sensing Object Detection Based on Strong Feature Extraction and Prescreening Network

High-Resolution Network with Transformer Embedding Parallel Detection for Small Object Detection in Optical Remote Sensing Images

ℱ3-Net: Feature Fusion and Filtration Network for Object Detection in Optical Remote Sensing Images

Cross-Layer Attention Network for Small Object Detection in Remote Sensing Imagery

ERF-RTMDet: An Improved Small Object Detection Method in Remote Sensing Images

A Deep CNN-Based Detection Method for Multi-Scale Fine-Grained Objects in Remote Sensing Images

Parallel Cascade R-CNN for Object Detection in Remote Sensing Imagery

IoU-Adaptive Deformable R-CNN: Make Full Use of IoU for Multi-Class Object Detection in Remote Sensing Imagery

Deep Adaptive Proposal Network for Object Detection in Optical Remote Sensing Images

A small object detection network for remote sensing based on CS-PANet and DSAN

Application of local fully Convolutional Neural Network combined with YOLO v5 algorithm in small target detection of remote sensing image

CABDet: context-and-attention-based detector for small object detection in remote sensing images

Deep Convolutional Feature Enhancement for Remote Sensing Object Detection

AeroDetectNet: A Lightweight, High-Precision Network for Enhanced Detection of Small Objects in Aerial Remote Sensing Imagery.