Target Detection Based on Binocular Stereo Vision
| Autor: | Xiaopeng Cui, Mingjiang Wang, Binwen Fan, Jingyu Yi |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
business.industry
Computer science Distortion (optics) ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Stereo matching Approximation algorithm Binocular stereo Depth map Calibration RGB color model Computer vision Artificial intelligence business ComputingMethodologies_COMPUTERGRAPHICS Monocular camera |
| Zdroj: | 2017 International Conference on Computer Technology, Electronics and Communication (ICCTEC). |
| DOI: | 10.1109/icctec.2017.00239 |
| Popis: | This paper mainly studies the target detection algorithm based on binocular camera. In this paper, the binocular RGB camera is used to detect the foreground of the human body in the indoor scene. Firstly, the right and left images are corrected by the binocular calibration. Then, the stereo matching is carried out by using the corrected image, and the most suitable stereo matching algorithm will be selected. After that, the depth map obtained by stereo matching algorithm is proposed. Finally, combining with the color information of monocular camera, the separation of the target to be detected from the background is realized. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|