ST-CenterNet: Small Target Detection Algorithm with Adaptive Data Enhancement
Affiliation:
1. College of Computer Science, Guangdong Polytechnic Normal University, Guangzhou 510665, China
Abstract
General target detection with deep learning has made tremendous strides in the past few years. However, small target detection sometimes is associated with insufficient sample size and difficulty in extracting complete feature information. For safety during autonomous driving, remote signs and pedestrians need to be detected from driving scenes photographed by car cameras. In the early period of a medical lesion, because of the small area of the lesion, target detection is of great significance to detect masses and tumors for accurate diagnosis and treatment. To deal with these problems, we propose a novel deep learning model, named CenterNet for small targets (ST-CenterNet). First of all, due to the lack of visual information on small targets in the dataset, we extracted less discriminative features. To overcome this shortcoming, the proposed selective small target replication algorithm (SSTRA) was used to realize increasing numbers of small targets by selectively oversampling them. In addition, the difficulty of extracting shallow semantic information for small targets results in incomplete target feature information. Consequently, we developed a target adaptation feature extraction module (TAFEM), which was used to conduct bottom-up and top-down bidirectional feature extraction by combining ResNet with the adaptive feature pyramid network (AFPN). The improved new network model, AFPN, was added to solve the problem of the original feature extraction module, which can only extract the last layer of the feature information. The experimental results demonstrate that the proposed method can accurately detect the small-scale image of distributed targets and simultaneously, at the pixel level, classify whether a subject is wearing a safety helmet. Compared with the detection effect of the original algorithm on the safety helmet wearing dataset (SHWD), we achieved mean average precision (mAP) of 89.06% and frames per second (FPS) of 28.96, an improvement of 18.08% mAP over the previous method.
Funder
National Natural Science Foundation of China Scientific and Technological Planning Project of Guangzhou Key Project of Guangdong Province Basic Research Foundation Guangdong Province Universities and Colleges Pearl River Scholar Funded Scheme
Subject
General Physics and Astronomy
Reference66 articles.
1. Jiang, Q., Tan, D., Li, Y., Ji, S., Cai, C., and Zheng, Q. (2020). Object detection and classification of metal polishing shaft surface defects based on convolutional neural network deep learning. Appl. Sci., 10. 2. Vaidya, B., and Paunwala, C. (2019). Smart Techniques for a Smarter Planet, Springer. 3. Peng, S., Jiang, W., Pi, H., Li, X., Bao, H., and Zhou, X. (2020, January 13–19). Deep snake for real-time instance segmentation. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA. 4. Akyol, G., Kantarcı, A., Çelik, A.E., and Ak, A.C. (2020, January 5–7). Deep learning based, real-time object detection for autonomous driving. Proceedings of the 2020 28th Signal Processing and Communications Applications Conference (SIU), Gaziantep, Turkey. 5. Luan, B., Sun, Y., Tong, C., Liu, Y., and Liu, H. (2019, January 14–15). R-FCN based laryngeal lesion detection. Proceedings of the 2019 12th International Symposium on Computational Intelligence and Design (ISCID), Hangzhou, China.
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. Metal surface defect detection based on improved YOLOv5;Scientific Reports;2023-11-27 2. Improved YOLOV5 based industrial surface defect detection method;2023 International Conference on Computers, Information Processing and Advanced Education (CIPAE);2023-08-26
|
|