YOLO-DSD: A YOLO-Based Detector Optimized for Better Balance between Accuracy, Deployability and Inference Time in Optical Remote Sensing Object Detection
Open Access
- 28 July 2022
- journal article
- research article
- Published by MDPI AG in Applied Sciences
- Vol. 12 (15) , 7622
Abstract
Many deep learning (DL)-based detectors have been developed for optical remote sensing object detection in recent years. However, most of the recent detectors are developed toward the pursuit of a higher accuracy, but little toward a balance between accuracy, deployability and inference time, which hinders the practical application for these detectors, especially in embedded devices. In order to achieve a higher detection accuracy and reduce the computational consumption and inference time simultaneously, a novel convolutional network named YOLO-DSD was developed based on YOLOv4. Firstly, a new feature extraction module, a dense residual (DenseRes) block, was proposed in a backbone network by utilizing a series-connected residual structure with the same topology for improving feature extraction while reducing the computational consumption and inference time. Secondly, convolution layer–batch normalization layer–leaky ReLu (CBL) ×5 modules in the neck, named S-CBL×5, were improved with a short-cut connection in order to mitigate feature loss. Finally, a low-cost novel attention mechanism called a dual channel attention (DCA) block was introduced to each S-CBL×5 for a better representation of features. The experimental results in the DIOR dataset indicate that YOLO-DSD outperforms YOLOv4 by increasing mAP0.5 from 71.3% to 73.0%, with a 23.9% and 29.7% reduction in Params and Flops, respectively, but a 50.2% improvement in FPS. In the RSOD dataset, the mAP0.5 of YOLO-DSD is increased from 90.0~94.0% to 92.6~95.5% under different input sizes. Compared with the SOTA detectors, YOLO-DSD achieves a better balance between the accuracy, deployability and inference time.Keywords
Funding Information
- Natural Science Foundation of Guangdong, China (2021A1515012395, CARS-17)
- China Agriculture Research System (2021A1515012395, CARS-17)
This publication has 47 references indexed in Scilit:
- Densely Connected Convolutional Networks2017
- YOLO9000: Better, Faster, Stronger2017
- Aggregated Residual Transformations for Deep Neural Networks2017
- Accurate Object Localization in Remote Sensing Images Based on Convolutional Neural NetworksIEEE Transactions on Geoscience and Remote Sensing, 2017
- SSD: Single Shot MultiBox Detector2016
- A survey on object detection in optical remote sensing imagesISPRS Journal of Photogrammetry and Remote Sensing, 2016
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- Deep Residual Learning for Image Recognition2016
- Fast R-CNN2015
- The Pascal Visual Object Classes (VOC) ChallengeInternational Journal of Computer Vision, 2009