The video-based person re-identification is to recognize a person under different cameras, which is a crucial task applied in visual surveillance system. Most previous methods mainly focused on the feature of full body in the frame. In this paper we propose a novel Spatial and Temporal Features Mixture Model (STFMM) based on convolutional neural network (CNN) and recurrent neural network (RNN), in which the human body is split into N parts in horizontal direction so that we can obtain more specific features. The proposed method skillfully integrates features of each part to achieve more expressive representation of each person. We first split the video sequence into N part sequences which include the information of head, waist, legs and so on. Then the features are extracted by STFMM whose 2N inputs are obtained from the developed Siamese network, and these features are combined into a discriminative representation for one person. Experiments are conducted on the iLIDS-VID and PRID-2011 datasets. The results demonstrate that our approach outperforms existing methods for video-based person re-identification. It achieves a rank-1 CMC accuracy of 74% on the iLIDS-VID dataset, exceeding the the most recently developed method ASTPN by 12%. For the cross-data testing, our method achieves a rank-1 CMC accuracy of 48% exceeding the ASTPN method by 18%, which shows that our model has significant stability.
Abnormalities detecting is one important application in the field of image processing and pattern recognition. It can alleviate human workload and improve productivity that employing computer graphic image theory and image processing technology analyzes and matches images in order to detect the abnormal region in image which has broad application prospects. In this paper, we propose a new abnormality detecting method based on similarity matching to address whether either missing or error abnormalities existing in bound books in industrial situation. First of all, we denoise the image by means of digital image processing and transformation, extract the sub rectangular region containing bound books using contour matching and locate the area exactly matching the template image using template matching. After that, we get a binary denoised image to detect the missing abnormality and the error abnormality using shape matching. In addition, we introduce some thresholds to improve the performance. The experiments show that the method we proposed achieve a better or the same performance comparing with the state-of-the-art methods.
Abstract-Semantic overlay could improve query performance in peer-to-peer (P2P) systems. When peers join or leave frequently, it will lead to network traffic surge, since most semantic search methods maintaining a large number of routing tables. In this paper, we address these problems by proposing Interest Attenuation Search(INS), a novel efficient peer-to-peer semantic search approach based on interest attenuation policy. In INS, the interest attenuation policy is introduced to help peers decide whether to forward messages. Before peer floods the request in semantic overlay network(SON), it will check the history information about the request message, then uses INS to make forwarding decision. Simulation results show that INS significantly improves query performance and reduces the traffic overhead generated by unstable network environment.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.