Multi-scale traffic vehicle detection based on faster R-CNN with NAS optimization and feature enrichment

• Published in: 防务技术 Vol. 17; no. 4; pp. 1542 - 1554
• Main Authors: Ji-qing Luo, Hu-sheng Fang, Fa-ming Shao, Yue Zhong, Xia Hua
• Format: Journal Article
• Language: English
• Published: Department of Mechanical Engineering, College of Field Engineering, Army Engineering University of PLA, Nanjing, 210007, China 2021
• Subjects: K-means; Feature enrichment; Retinex-based image adaptive correction; Neural architecture search; Faster R-CNN; algorithm; UN-DETRAC
• ISSN: 2214-9147; 2214-9147
• DOI: 10.3969/j.issn.2214-9147.2021.04.038

It well known that vehicle detection is an important component of the field of object detection. However, the environment of vehicle detection is particularly sophisticated in practical processes. It is compara-tively difficult to detect vehicles of various scales in traffic scene images, because the vehicles partially obscured by green belts, roadblocks or other vehicles, as well as influence of some low illumination weather. In this paper, we present a model based on Faster R-CNN with NAS optimization and feature enrichment to realize the effective detection of multi-scale vehicle targets in traffic scenes. First, we proposed a Retinex-based image adaptive correction algorithm (RIAC) to enhance the traffic images in the dataset to reduce the influence of shadow and illumination, and improve the image quality. Second, in order to improve the feature expression of the backbone network, we conducted Neural Architecture Search (NAS) on the backbone network used for feature extraction of Faster R-CNN to generate the optimal cross-layer connection to extract multi-layer features more effectively. Third, we used the object Feature Enrichment that combines the multi-layer feature information and the context information of the last layer after cross-layer connection to enrich the information of vehicle targets, and improve the robustness of the model for challenging targets such as small scale and severe occlusion. In the imple-mentation of the model, K-means clustering algorithm was used to select the suitable anchor size for our dataset to improve the convergence speed of the model. Our model has been trained and tested on the UN-DETRAC dataset, and the obtained results indicate that our method has art-of-state detection performance.