Progressive structure network-based multiscale feature fusion for object detection in real-time application. (November 2021)

• Record Type: Journal Article
• Authors: Wang, Haifeng; Jiang, Lvjiyuan; Zhao, Qian; Li, Hao; Yan, Kai; Yang, Yang; Li, Songlin; Zhang, Yungang; Qiao, Lianliu; Fu, Cuilian; Yin, Hong; Hu, Yun; Yu, Haibin
• Is Part Of: Engineering applications of artificial intelligence. Volume 106(2021)
• Publication Date: 2021-11
• Subjects: Feature fusion -- Deep learning -- Object detection -- Machine learning; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2021.104486
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: Deep learning-based target detection techniques have already made a wide-range impact on our daily life. Currently, a feature pyramid is a widely utilized technique for multiscale target detection, the effectiveness of the technique has already been proved. Nevertheless, in the pyramid structure, problems, such as multiscale feature alignment, model turmoil after fusion, feature redundancy, and no-local feature fusion, exist. In this paper, we propose a novel progressive structure network to solve the aforementioned problems. The proposed structure contains three modules: multiscale feature alignment fusion, different scale channels & spatial location adaptive weighted fusion, and multiscale global and local feature fusion. The proposed structure is capable of fusing information from different feature layers more effectively. Subsequently, the semantic gaps among different scales can be reduced. Furthermore, the proposed structure can maintain the stability of the detection network and its performance has been proved by comparing with other state-of-art feature fusion method. The proposed progressive network structure has also been applied to actual target detection tasks and the practical application effectiveness of our method has been verified.