期刊
DEFENCE TECHNOLOGY
卷 17, 期 4, 页码 1542-1554出版社
KEAI PUBLISHING LTD
DOI: 10.1016/j.dt.2020.10.006
关键词
Neural architecture search; Feature enrichment; Faster R-CNN; Retinex-based image adaptive correction algorithm; K-means; UN-DETRAC
资金
- National Natural Science Foundation of China [61671470]
- Key Research and Development Program of China [2016YFC0802900]
In this paper, a model based on Faster R-CNN with NAS optimization and feature enrichment is proposed to effectively detect multi-scale vehicle targets in traffic scenes. The model includes an image adaptive correction algorithm to enhance image quality, Neural Architecture Search to optimize backbone network, and object feature enrichment for robust detection of challenging targets. The model demonstrates state-of-the-art detection performance on the UN-DETRAC dataset.
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 comparatively 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 implementation 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. (C) 2021 China Ordnance Society. Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据