Research on personnel re-recognition method in coal mine underground based on improved metric learning
In the traditional personnel re-recognition method in coal mine underground based on metric learning, because metric learning ignores the absolute distance between positive and negative samples, the gradient of the loss function disappears or disperses. This results in low recognition precision of u...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Department of Industry and Mine Automation
2023-09-01
|
| Series: | Gong-kuang zidonghua |
| Subjects: | |
| Online Access: | http://www.gkzdh.cn/article/doi/10.13272/j.issn.1671-251x.18100 |
| _version_ | 1797563562038657024 |
|---|---|
| author | ZHANG Liya WANG Yu HAO Bonan |
| author_facet | ZHANG Liya WANG Yu HAO Bonan |
| author_sort | ZHANG Liya |
| collection | DOAJ |
| description | In the traditional personnel re-recognition method in coal mine underground based on metric learning, because metric learning ignores the absolute distance between positive and negative samples, the gradient of the loss function disappears or disperses. This results in low recognition precision of underground personnel position information. In order to solve this problem, a personnel re-recognition method in coal mine underground based on improved metric learning is proposed. Firstly, a feature extraction method for underground personnel based on manual design features is adopted to manually process and extract features such as color space and texture space, enriching the feature dimensions. Secondly, Euclidean distance is used to calculate the similarity of high-dimensional features of personnel. Finally, an improved triple loss function is proposed. Adding adaptive weights to the traditional triple loss function increases the weight of effective samples. It solves the problem of gradient disappearance or dispersion caused by ignoring the absolute distance between positive and negative samples. The traditional recognition method is compared with the personnel re-recognition method in coal mine underground based on improved metric learning for cumulative matching feature curve verification and recognition rate verification. The results show the following points. ① The personnel re-recognition method in coal mine underground based on improved metric learning has a sample matching probability of 100% when the number of similar samples is around 50. ② The personnel re-recognition method in coal mine underground based on improved metric learning reduces the inference time of two different calibration size images by 44 ms and 68 ms, respectively, compared to traditional re-recognition methods. ③ The personnel re-recognition method in coal mine underground based on improved metric learning performs better after discarding the images of personnel heads and feet. It has a sample matching probability of 100% when the number of similar samples is around 42. |
| first_indexed | 2024-03-10T18:45:20Z |
| format | Article |
| id | doaj.art-3ae7d3b5374f440a9f0b2cbffa081823 |
| institution | Directory Open Access Journal |
| issn | 1671-251X |
| language | zho |
| last_indexed | 2024-03-10T18:45:20Z |
| publishDate | 2023-09-01 |
| publisher | Editorial Department of Industry and Mine Automation |
| record_format | Article |
| series | Gong-kuang zidonghua |
| spelling | doaj.art-3ae7d3b5374f440a9f0b2cbffa0818232023-11-20T05:32:11ZzhoEditorial Department of Industry and Mine AutomationGong-kuang zidonghua1671-251X2023-09-014998489, 16610.13272/j.issn.1671-251x.18100Research on personnel re-recognition method in coal mine underground based on improved metric learningZHANG LiyaWANG Yu0HAO BonanSchool of Information and Communication Engineering, Communication University of China, Beijing 100024, ChinaIn the traditional personnel re-recognition method in coal mine underground based on metric learning, because metric learning ignores the absolute distance between positive and negative samples, the gradient of the loss function disappears or disperses. This results in low recognition precision of underground personnel position information. In order to solve this problem, a personnel re-recognition method in coal mine underground based on improved metric learning is proposed. Firstly, a feature extraction method for underground personnel based on manual design features is adopted to manually process and extract features such as color space and texture space, enriching the feature dimensions. Secondly, Euclidean distance is used to calculate the similarity of high-dimensional features of personnel. Finally, an improved triple loss function is proposed. Adding adaptive weights to the traditional triple loss function increases the weight of effective samples. It solves the problem of gradient disappearance or dispersion caused by ignoring the absolute distance between positive and negative samples. The traditional recognition method is compared with the personnel re-recognition method in coal mine underground based on improved metric learning for cumulative matching feature curve verification and recognition rate verification. The results show the following points. ① The personnel re-recognition method in coal mine underground based on improved metric learning has a sample matching probability of 100% when the number of similar samples is around 50. ② The personnel re-recognition method in coal mine underground based on improved metric learning reduces the inference time of two different calibration size images by 44 ms and 68 ms, respectively, compared to traditional re-recognition methods. ③ The personnel re-recognition method in coal mine underground based on improved metric learning performs better after discarding the images of personnel heads and feet. It has a sample matching probability of 100% when the number of similar samples is around 42.http://www.gkzdh.cn/article/doi/10.13272/j.issn.1671-251x.18100precise positioning of mine personnelpersonnel re-recognitionmetric learningsimilarity measurementadaptive triple loss functioncumulative matching feature |
| spellingShingle | ZHANG Liya WANG Yu HAO Bonan Research on personnel re-recognition method in coal mine underground based on improved metric learning Gong-kuang zidonghua precise positioning of mine personnel personnel re-recognition metric learning similarity measurement adaptive triple loss function cumulative matching feature |
| title | Research on personnel re-recognition method in coal mine underground based on improved metric learning |
| title_full | Research on personnel re-recognition method in coal mine underground based on improved metric learning |
| title_fullStr | Research on personnel re-recognition method in coal mine underground based on improved metric learning |
| title_full_unstemmed | Research on personnel re-recognition method in coal mine underground based on improved metric learning |
| title_short | Research on personnel re-recognition method in coal mine underground based on improved metric learning |
| title_sort | research on personnel re recognition method in coal mine underground based on improved metric learning |
| topic | precise positioning of mine personnel personnel re-recognition metric learning similarity measurement adaptive triple loss function cumulative matching feature |
| url | http://www.gkzdh.cn/article/doi/10.13272/j.issn.1671-251x.18100 |
| work_keys_str_mv | AT zhangliya researchonpersonnelrerecognitionmethodincoalmineundergroundbasedonimprovedmetriclearning AT wangyu researchonpersonnelrerecognitionmethodincoalmineundergroundbasedonimprovedmetriclearning AT haobonan researchonpersonnelrerecognitionmethodincoalmineundergroundbasedonimprovedmetriclearning |