基于視頻處理的行人檢測(cè)和跟蹤系統(tǒng)-資料下載頁(yè)
2025-06-05 16:52本頁(yè)面
【正文】 al structures revisited: people detection and articulated pose estimation. Proc. IEEE Conf. on Computer Vision and Pattern Recognition, Miami, Florida, USA , 2021, 10141021. [4]N. Shou, H. Peng, H. Wang, . Meng, . Du, An ROIs based pedestrian detection system for single images. Proc. 5th. Int. Congress on Image and Signal Processing (CISP), chongqing, China, Oct. 2021, 12051208. [5]Q. Zhu, S. Avidan, . Yeh amp。 . Cheng, Fast human detection using a cascade of histograms of oriented gradients. Proc. IEEE Conf. Computer Vision Pattern Recogn., New York, 2021, 2: 14901499. [6]Platt J. C. Probabilistic Output for Support Vector Machine and Comparisons to Regularized Likelihood Methods. Advances in Large Margin Classifiers: MIT Press, 1999. [7]C. Wen, A. Azarbayejani, T. Darrell amp。 A. , Realtime tracking of the human body. IEEE Trans. Pattern Anal. Mach. Intell., 1997, 19(7):780785. [8]M. Andriluka, S. Roth amp。 B. Schiele, Peopletrackingbydetection and peopledetectionby tracking. Proc. IEEE Conf. Proc. Computer Vision and Pattern Recognition, Anchorage, Alaska, USA, 2021, 18. [9]. Hou amp。 . Pang, People counting and human detection in a challenging situation. IEEE Trans. Syst. Man Cybern., 2021, 41(1):2433. ? Select positive and negative samples for training the classifier. We select 800 positive samples and 500 negative samples in the INRIA pedestrian database, which are already normalized to . ? Reduce feature dimension of the sample set. If a sample is positive, trim the pedestrian area and normalize it to。 otherwise, directly normalize it to. ? For each sample, extract HOG features[5]. It first divides a sample into blocks of pixels。 then, each block is divided into pixel units, and the step is 8 pixels. The original HOG feature dimension is . After reducing feature dimension, HOG feature dimension is . In summary, the putational plexity reduced by a factor of 5. ? We chose the efficient LIBSVM classifier developed by the Taiwan University, and use the soft output instead of the {1,1} [4].(把 LIBSVM 的硬判斷輸出結(jié)果映射到 ) ? Train LIBSVM classifier with HOG features as input to get a lowdimensional and soft output pedestrian classifier. Meanwhile save the foreground objects and the frames of without clear classification sequentially in the queue. ? Extract HOG features of dimensions for each object. ? Input HOG features of each object to the trained pedestrian classifier, and the output gives whether the object is a pedestrian. At the same time,
畢業(yè)設(shè)計(jì)相關(guān)推薦
鄂ICP備17016276號(hào)-1