CAJ | 학술논문

为了提高立体匹配算法的精确度，提出了一种结合图割与MeanShift图像分割的立体匹配算法。首先用MeanShift算法对参考图像进行图像分割，得到分割后的标记图，然后将分割信息结合到图割算法的能量函数中，最后用改进的能量函数和图割算法得出致密的视差图。实验结果表明，提出的算法具有更高的精确度和更好的边缘特征。
위료제고입체필배산법적정학도，제출료일충결합도할여MeanShift도상분할적입체필배산법。수선용MeanShift산법대삼고도상진행도상분할，득도분할후적표기도，연후장분할신식결합도도할산법적능량함수중，최후용개진적능량함수화도할산법득출치밀적시차도。실험결과표명，제출적산법구유경고적정학도화경호적변연특정。
In order to improve the accuracy of the stereo matching algorithm, a stereo matching algorithm based on graph cuts and Mean Shift image segmentation is proposed. Firstly, the reference image is segmented using Mean Shift algorithm and the label image has been got. Then the segmentation information is integrated into energy function of graph cuts algorithm. Finally, a dense disparity map is yielded using the improved energy function and graph cuts algorithm. Experimental results show that the proposed method has better accuracy and better edge feature.