Depth estimation for small obstacles based on monocular vision

Wang Gao, Changqing Wang, Lei Li

科研成果: 书/报告/会议事项章节会议稿件同行评审

1 引用 (Scopus)

摘要

In order to solve the problem of small obstacles avoidance in ultra-low altitude flying of unmanned aerial vehicles (UAVs), a depth estimation algorithm based on monocular visual for small tower obstacles is proposed by combining deep learning and traditional methods. With the advantage of convolutional neural networks to extract the salient features of the images, a lightweight and multi-level residual convolution neural network for obstacle segmentation is proposed. The cross-entropy loss function is optimized by adding the weight of the background and obstacles area. A method based on pairs of coplanar points is introduced to complete the depth estimation of a single frame image. Moreover, the recursive formula of sequence images depth results is derived, and a multi-frame voting optimization algorithm is proposed. Finally, the simulation results show that the proposed algorithm can effectively realize the fast segmentation and accurate depth estimation of the small tower obstacles in the sequence images.

源语言英语
主期刊名Proceedings of 2020 3rd International Conference on Unmanned Systems, ICUS 2020
出版商Institute of Electrical and Electronics Engineers Inc.
407-411
页数5
ISBN(电子版)9781728180250
DOI
出版状态已出版 - 27 11月 2020
已对外发布
活动3rd International Conference on Unmanned Systems, ICUS 2020 - Harbin, 中国
期限: 27 11月 202028 11月 2020

出版系列

姓名Proceedings of 2020 3rd International Conference on Unmanned Systems, ICUS 2020

会议

会议3rd International Conference on Unmanned Systems, ICUS 2020
国家/地区中国
Harbin
时期27/11/2028/11/20

指纹

探究 'Depth estimation for small obstacles based on monocular vision' 的科研主题。它们共同构成独一无二的指纹。

引用此