A Study on Person Re-identification in Multi-view Surveillance Videos
0
조회
0
다운로드
4
인용
0
참고문헌
키워드
초록
Nowadays, surveillance cameras are widely installed at every edge due to their largescale applications ranging from monitoring to smart city and indoor offices for video conferencing. However, to perform several tasks such as detection, tracking, and identification of persons in video data, it is an essential step to develop an algorithm specifically using a distributed network. For this purpose, numerous techniques have been developed in the field of computer vison for single-view images/video analysis. However, overviewing the literature of video data analytics, the techniques for multi-view analysis are missing on extreme level. Therefore, in this paper, we propose a method for person re-identification and tracking in multi-view images through hybrid approach of several techniques that are usually applied for a single-view image. At first, we generate our own dataset and pass the images into a preprocessing step to detect person using YOLO and subtract the background through Mask-RCNN. Next, the color features are extracted from the images which are fed into K-mean clustering algorithm. We introduce a novel distancing formula by adding weights to Bhattacharyya’s distance to measure the similarity between the clusters of two multi-view input images. The conducted experiments verify and confirm the affectiveness of the proposed method by obtaining 86.23 average precision.
Abstract
Nowadays, surveillance cameras are widely installed at every edge due to their largescale applications ranging from monitoring to smart city and indoor offices for video conferencing. However, to perform several tasks such as detection, tracking, and identification of persons in video data, it is an essential step to develop an algorithm specifically using a distributed network. For this purpose, numerous techniques have been developed in the field of computer vison for single-view images/video analysis. However, overviewing the literature of video data analytics, the techniques for multi-view analysis are missing on extreme level. Therefore, in this paper, we propose a method for person re-identification and tracking in multi-view images through hybrid approach of several techniques that are usually applied for a single-view image. At first, we generate our own dataset and pass the images into a preprocessing step to detect person using YOLO and subtract the background through Mask-RCNN. Next, the color features are extracted from the images which are fed into K-mean clustering algorithm. We introduce a novel distancing formula by adding weights to Bhattacharyya’s distance to measure the similarity between the clusters of two multi-view input images. The conducted experiments verify and confirm the affectiveness of the proposed method by obtaining 86.23 average precision.
참고문헌 (20)
- Muhammad, K., "DeepReS : A Deep Learning-Based Video Summarization Strategy for Resource-Constrained Industrial Surveillance Scenarios", IEEE Transactions on Industrial Informatics, 16(9), pp.5938~5947 (2019)
- 튜 유 민 울라, "Violence Recognition using Deep CNN for Smart Surveillance Applications", 한국차세대컴퓨팅학회 논문지, 14(5), pp.53~59 (2018), 한국차세대컴퓨팅학회
- 와셈 울라, "Crowd Behavior Detection using Convolutional Neural Network", 한국차세대컴퓨팅학회 논문지, 15(6), pp.7~14 (2019), 한국차세대컴퓨팅학회[DOI]
- Sajjad, M., "Integrating salient colors with rotational invariant texture features for image representation in retrieval systems", Multimedia Tools and Applications, 77(4), pp.4769~4789 (2018)
- 라힘 나시르, "Robust Processing of Visual Contents for Text Recognition in Surveillance Applications", 한국차세대컴퓨팅학회 논문지, 14(1), pp.65~77 (2018), 한국차세대컴퓨팅학회
- López-Cifuentes, A., "Semantic-aware scene recognition", Pattern Recognition, 102, pp.107256~ (2020)
- 노먼 칸, "Efficient Sport Videos Classification via Convolutional Neural Network", 한국차세대컴퓨팅학회 논문지, 16(6), pp.7~16 (2020), 한국차세대컴퓨팅학회[DOI]
- Khan, N., "SD-Net: Understanding overcrowded scenes in real-time via an efficient dilated convolutional neural network", Journal of Real-Time Image Processing, pp.1~15 (2020)
- Rahim, N., "Privacy-preserving image retrieval for mobile devices with deep features on the cloud", Computer Communications, 127, pp.75~85 (2018)
- Haskins, G., "Deep learning in medical image registration : a survey", Machine Vision and Applications, 31(1), pp.8~ (2020)
- Jain, A. K., "Unsupervised texture segmentation using Gabor filters", Pattern recognition, 24(12), pp.1167~1186 (1991)
- Rublee, E., "ORB: An efficient alternative to SIFT or SURF", 2011 International conference on computer vision (2011), Ieee
- Ahonen, T., "Face recognition with local binary patterns", European conference on computer vision (2004), Springer
- Dalal, N., "Histograms of oriented gradients for human detection", 2005 IEEE computer society conference on computer vision and pattern recognition (CVPR'05) (2005), IEEE
- Wang, L., "An image similarity descriptor for classification tasks", 71, pp.102847~ (2020)
- Shnain, N. A., "A feature-based structural measure: an image similarity measure for face recognition", Lu, A feature-based structural measure : an image similarity measure for face recognition, 7(8), pp.786~ (2017)
- Gao, X., "Sparse online learning of image similarity", ACM Transactions on Intelligent Systems and Technology(TIST), 8(5), pp.1~22 (2017)
- Appalaraju, S., "Image similarity using deep CNN and curriculum learning", arXiv preprint arXiv:1709.08761, ~ (2017)
- Liao, L., "Improving image similarity estimation via global distance distribution information", Neurocomputing, 333, pp.381~394 (2019)
- Kim, M.J., "Object Tracking using Similarity Measurement Algorithm in MultiView Surveillance Videos", The 6th International Conference on Next Generation Computing(ICNGC 2020), pp.168~188 (2020)