Dilated spatial–temporal convolutional auto-encoders for human fall detection in surveillance videos
Although methods based on supervised learning have demonstrated remarkable performance on fall detection, these existing fall detection algorithms require a substantial quantity of manually labeled training data. In this paper, we combine dilated convolution and LSTM based on auto-encoder, which can...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-08-01
|
| Series: | ICT Express |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S240595952200100X |
| _version_ | 1797737333109293056 |
|---|---|
| author | Suyuan Li Xin Song Siyang Xu Haoyang Qi Yanbo Xue |
| author_facet | Suyuan Li Xin Song Siyang Xu Haoyang Qi Yanbo Xue |
| author_sort | Suyuan Li |
| collection | DOAJ |
| description | Although methods based on supervised learning have demonstrated remarkable performance on fall detection, these existing fall detection algorithms require a substantial quantity of manually labeled training data. In this paper, we combine dilated convolution and LSTM based on auto-encoder, which can be trained on unlabeled data, further saving time and resources, and a novel fall score is computed based on the high-quality reconstructed frame to detect falls. Extensive experimental results indicate that the proposed method further boosts the performance, achieving recognition rate of 97.1%, sensitivity rate of 93.9% and precision rate of 95.1% on the UR dataset. |
| first_indexed | 2024-03-12T13:27:09Z |
| format | Article |
| id | doaj.art-1b9260483a92413ab17a4f87750d32dd |
| institution | Directory Open Access Journal |
| issn | 2405-9595 |
| language | English |
| last_indexed | 2024-03-12T13:27:09Z |
| publishDate | 2023-08-01 |
| publisher | Elsevier |
| record_format | Article |
| series | ICT Express |
| spelling | doaj.art-1b9260483a92413ab17a4f87750d32dd2023-08-25T04:24:23ZengElsevierICT Express2405-95952023-08-0194734740Dilated spatial–temporal convolutional auto-encoders for human fall detection in surveillance videosSuyuan Li0Xin Song1Siyang Xu2Haoyang Qi3Yanbo Xue4School of Computer Science and Engineering, Northeastern University, Shenyang 110819, ChinaSchool of Computer Science and Engineering, Northeastern University, Shenyang 110819, China; School of Computer and Communication Engineering, Northeastern University at Qinhuangdao, Qinhuangdao 066004, China; Corresponding author at: School of Computer Science and Engineering, Northeastern University, Shenyang 110819, China.School of Computer Science and Engineering, Northeastern University, Shenyang 110819, China; School of Computer and Communication Engineering, Northeastern University at Qinhuangdao, Qinhuangdao 066004, ChinaSchool of Computer Science and Engineering, Northeastern University, Shenyang 110819, China; School of Computer and Communication Engineering, Northeastern University at Qinhuangdao, Qinhuangdao 066004, ChinaBOSS ZhiPin Career Science Lab (CSL), Beijing 100028, ChinaAlthough methods based on supervised learning have demonstrated remarkable performance on fall detection, these existing fall detection algorithms require a substantial quantity of manually labeled training data. In this paper, we combine dilated convolution and LSTM based on auto-encoder, which can be trained on unlabeled data, further saving time and resources, and a novel fall score is computed based on the high-quality reconstructed frame to detect falls. Extensive experimental results indicate that the proposed method further boosts the performance, achieving recognition rate of 97.1%, sensitivity rate of 93.9% and precision rate of 95.1% on the UR dataset.http://www.sciencedirect.com/science/article/pii/S240595952200100XDilated convolutionAuto-encoderFall detectionLSTM |
| spellingShingle | Suyuan Li Xin Song Siyang Xu Haoyang Qi Yanbo Xue Dilated spatial–temporal convolutional auto-encoders for human fall detection in surveillance videos ICT Express Dilated convolution Auto-encoder Fall detection LSTM |
| title | Dilated spatial–temporal convolutional auto-encoders for human fall detection in surveillance videos |
| title_full | Dilated spatial–temporal convolutional auto-encoders for human fall detection in surveillance videos |
| title_fullStr | Dilated spatial–temporal convolutional auto-encoders for human fall detection in surveillance videos |
| title_full_unstemmed | Dilated spatial–temporal convolutional auto-encoders for human fall detection in surveillance videos |
| title_short | Dilated spatial–temporal convolutional auto-encoders for human fall detection in surveillance videos |
| title_sort | dilated spatial temporal convolutional auto encoders for human fall detection in surveillance videos |
| topic | Dilated convolution Auto-encoder Fall detection LSTM |
| url | http://www.sciencedirect.com/science/article/pii/S240595952200100X |
| work_keys_str_mv | AT suyuanli dilatedspatialtemporalconvolutionalautoencodersforhumanfalldetectioninsurveillancevideos AT xinsong dilatedspatialtemporalconvolutionalautoencodersforhumanfalldetectioninsurveillancevideos AT siyangxu dilatedspatialtemporalconvolutionalautoencodersforhumanfalldetectioninsurveillancevideos AT haoyangqi dilatedspatialtemporalconvolutionalautoencodersforhumanfalldetectioninsurveillancevideos AT yanboxue dilatedspatialtemporalconvolutionalautoencodersforhumanfalldetectioninsurveillancevideos |