Optimal view selection method based on object detection. Issue 3 (October 2020)

• Record Type: Journal Article
• Authors: Meng, Fanwu; Xu, Yichen; Huang, Zhipeng; Wang, Lizhong
• Is Part Of: Journal of physics. Volume 1650:Issue 3(2020)
• Publication Date: 2020-10
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1650/3/032141
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: Industrial parts detection are often studied in three-dimensional space, because of subparts occlusion, the choice of view has a great impact on the detection results. In this paper, the method of object detection combined with the 3D CAD model is proposed, the object detection model based on convolutional neural network was used to screen the visual angle, so as to eliminate the influence caused by the occlusion of parts, which provides a general method for the selection of the optimal view.