http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
현종길(Jongkil Hyun),이인규(Ingyu Lee),문병인(Byungin Moon) 한국정보기술학회 2017 한국정보기술학회논문지 Vol.15 No.2
Stereo vision is a technique of extracting disparities from a stereo image pair and calculating three-dimensional distance information. Stereo vision is an active research field because it can be used in various applications. Most studies of stereo vision have been focusing on the matching process and paying little attention to the post-processing process which plays a significant role in improving disparity accuracy. Thus, this paper divides the post-processing algorithms into three groups and evaluates and analyzes their performance. According to the experimental results, the left-right consistency check shows the best performance compared with other seed pixel detection methods, the 2-way interpolation method is more efficient than the 1-way interpolation method, and the weight median filter shows outstanding performance to improve disparity accuracy while preserving edge information of disparity maps.