CAJ | 학술논문

针对传统行人跟踪算法计算复杂度高的问题，提出一种基于局部图块匹配（local block-graphs matching ， LBGM ）的行人跟踪算法。利用摄像机采集视频流帧画面中含有行人的多幅视频图像，按照行人所在的图像区域进行分割；对图块进行数据关联匹配，通过匹配结果完成对行人的检测过程，得到检测的行人集合；针对行人集合，利用损失最小化进行最小偏差的路径预测，实现行人的跟踪过程。选取多相机行人视频（multi-camera pedestrians video ）进行对比实验，实验结果表明，该算法可以有效地跟踪行人，较传统算法有更好的行人检测率和跟踪准确率。
침대전통행인근종산법계산복잡도고적문제，제출일충기우국부도괴필배（local block-graphs matching ， LBGM ）적행인근종산법。이용섭상궤채집시빈류정화면중함유행인적다폭시빈도상，안조행인소재적도상구역진행분할；대도괴진행수거관련필배，통과필배결과완성대행인적검측과정，득도검측적행인집합；침대행인집합，이용손실최소화진행최소편차적로경예측，실현행인적근종과정。선취다상궤행인시빈（multi-camera pedestrians video ）진행대비실험，실험결과표명，해산법가이유효지근종행인，교전통산법유경호적행인검측솔화근종준학솔。
The traditional human tracking algorithms have the problem of high computational complexity ,and a novel human tracking method based on the local block-graphs matching was proposed .Firstly ,some images were captured by the video frames ,which containing the human ,and the human block-graphs were gotten .Secondly ,the block-graphs were used to achieve the human matching ,and the detection process was achieved by the results of the human matching .Thereafter ,the human set was gotten .Finally ,the human set was used to achieve the minimum deviation forecasting of all frames ,and the human tracking was achieved .The multi-camera pedestrians video was selected as the experimental data ,and a large number of experiments were done .The results show that the proposed method has not only a great effect on tracking human ,but the better detection rate and tracking accuracy .