Open AccessLeft and Right Consistent Stereo Image Detection and Classification Based on Deep Learning
Author(s) -
Lijuan Tang,
Qing Wang,
Guannan Chen
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1575/1/012149
Subject(s) - artificial intelligence , computer vision , monocular , computer science , image (mathematics) , matching (statistics) , binocular disparity , binocular vision , mathematics , statistics
With the development of stereo camera technology, the increasing of the data volume of binocular image makes the target detection and classification of binocular image become one of the research directions in the field of computer vision. Compared with monocular image, binocular image contains the depth information of the image. For using the depth information of the image to improve the target detection and classification accuracy of image, this text puts forward a method of detecting and classifying left and right consistent objects in stereo images based on deep learning, which is based on the Mask R-CNN network and adds the binocular matching algorithm to improve the accuracy of image target detection and classification. The experiment compares the evaluation indexes of monocular and binocular.