期刊
REMOTE SENSING
卷 13, 期 13, 页码 -出版社
MDPI
DOI: 10.3390/rs13132459
关键词
remote sensing image; arbitrary-oriented object detection; lightweight network; knowledge distillation
类别
资金
- National Natural Science Foundation of China [61772399, U1701267, 61773304, 61672405, 61772400]
- Key Research and Development Program in Shaanxi Province of China [2019ZDLGY0905]
- Program for Cheung Kong Scholars and Innovative Research Team in University [IRT_15R53]
- Technology Foundation for Selected Overseas Chinese Scholar in Shaanxi [2017021, 2018021]
A lightweight keypoint-based oriented object detector for remote sensing images is proposed in this paper, which improves detection performance by introducing a semantic transfer block and an adaptive Gaussian kernel, and obtains a lightweight student network using distillation loss associated with object detection. Experimental results show that the method adapts to objects of different scales, obtains accurate bounding boxes, and reduces the influence of complex backgrounds. The comparison with mainstream methods demonstrates comparable performance under lightweight conditions.
Object detection in remote sensing images has been widely used in military and civilian fields and is a challenging task due to the complex background, large-scale variation, and dense arrangement in arbitrary orientations of objects. In addition, existing object detection methods rely on the increasingly deeper network, which increases a lot of computational overhead and parameters, and is unfavorable to deployment on the edge devices. In this paper, we proposed a lightweight keypoint-based oriented object detector for remote sensing images. First, we propose a semantic transfer block (STB) when merging shallow and deep features, which reduces noise and restores the semantic information. Then, the proposed adaptive Gaussian kernel (AGK) is adapted to objects of different scales, and further improves detection performance. Finally, we propose the distillation loss associated with object detection to obtain a lightweight student network. Experiments on the HRSC2016 and UCAS-AOD datasets show that the proposed method adapts to different scale objects, obtains accurate bounding boxes, and reduces the influence of complex backgrounds. The comparison with mainstream methods proves that our method has comparable performance under lightweight.
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