Video image feature extraction method for police drones investigation and evidence collection
In the practical application of police drones in investigation and evidence collection, it is not only necessary to operate stably under unfavorable conditions such as night, haze, and complex terrain, but also to ensure the operation continuity and reliability. The drones needs to have good compati...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Office of Journal of XPU
2024-12-01
|
| Series: | Xi'an Gongcheng Daxue xuebao |
| Subjects: | |
| Online Access: | http://journal.xpu.edu.cn/en/#/digest?ArticleID=1523 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849321867578441728 |
|---|---|
| author | CHEN Sixin HE Fengwei |
| author_facet | CHEN Sixin HE Fengwei |
| author_sort | CHEN Sixin |
| collection | DOAJ |
| description | In the practical application of police drones in investigation and evidence collection, it is not only necessary to operate stably under unfavorable conditions such as night, haze, and complex terrain, but also to ensure the operation continuity and reliability. The drones needs to have good compatibility and be able to seamlessly integrate with other police systems to achieve information sharing and collaborative operations. In response to these requirements, a new video image feature extraction method was proposed in the article. Firstly, we made a detailed analysis of the HSV color of the target object in the video image. Then, based on the results of the HSV color space, attention mechanism was introduced for video image feature extraction, and constructed spectral multi-scale feature extraction network and spatial multi-scale feature extraction network with attention mechanism. Finally, the spectrum and spatial network were merged into a whole, and trained and fused to obtain a spectral spatial joint feature extraction network, which achieved the extraction for police unmanned aerial vehicle investigation and evidence collection. The experimental results show that the feature deviation of the method in the article is relatively small, and the highest feature extraction accuracy is 97%. The overall frame per second (FPS) value is between 50~75, which remains at a high level and has good stability and reliability. This not only meets the demand for efficient and accurate feature extraction of police drones, but also provides technical support for their widespread application in complex investigation environments. |
| format | Article |
| id | doaj-art-64d2b3a12c534301858eb4b4de43f3f6 |
| institution | Kabale University |
| issn | 1674-649X |
| language | zho |
| publishDate | 2024-12-01 |
| publisher | Editorial Office of Journal of XPU |
| record_format | Article |
| series | Xi'an Gongcheng Daxue xuebao |
| spelling | doaj-art-64d2b3a12c534301858eb4b4de43f3f62025-08-20T03:49:37ZzhoEditorial Office of Journal of XPUXi'an Gongcheng Daxue xuebao1674-649X2024-12-013869810410.13338/j.issn.1674-649x.2024.06.013Video image feature extraction method for police drones investigation and evidence collectionCHEN Sixin0HE Fengwei1Police Tactics Department, Chongqing Police College, Chongqing 401331, ChinaPolice Tactics Department, Chongqing Police College, Chongqing 401331, ChinaIn the practical application of police drones in investigation and evidence collection, it is not only necessary to operate stably under unfavorable conditions such as night, haze, and complex terrain, but also to ensure the operation continuity and reliability. The drones needs to have good compatibility and be able to seamlessly integrate with other police systems to achieve information sharing and collaborative operations. In response to these requirements, a new video image feature extraction method was proposed in the article. Firstly, we made a detailed analysis of the HSV color of the target object in the video image. Then, based on the results of the HSV color space, attention mechanism was introduced for video image feature extraction, and constructed spectral multi-scale feature extraction network and spatial multi-scale feature extraction network with attention mechanism. Finally, the spectrum and spatial network were merged into a whole, and trained and fused to obtain a spectral spatial joint feature extraction network, which achieved the extraction for police unmanned aerial vehicle investigation and evidence collection. The experimental results show that the feature deviation of the method in the article is relatively small, and the highest feature extraction accuracy is 97%. The overall frame per second (FPS) value is between 50~75, which remains at a high level and has good stability and reliability. This not only meets the demand for efficient and accurate feature extraction of police drones, but also provides technical support for their widespread application in complex investigation environments.http://journal.xpu.edu.cn/en/#/digest?ArticleID=1523police dronesvideo image feature extractionattention mechanismhsv color space |
| spellingShingle | CHEN Sixin HE Fengwei Video image feature extraction method for police drones investigation and evidence collection Xi'an Gongcheng Daxue xuebao police drones video image feature extraction attention mechanism hsv color space |
| title | Video image feature extraction method for police drones investigation and evidence collection |
| title_full | Video image feature extraction method for police drones investigation and evidence collection |
| title_fullStr | Video image feature extraction method for police drones investigation and evidence collection |
| title_full_unstemmed | Video image feature extraction method for police drones investigation and evidence collection |
| title_short | Video image feature extraction method for police drones investigation and evidence collection |
| title_sort | video image feature extraction method for police drones investigation and evidence collection |
| topic | police drones video image feature extraction attention mechanism hsv color space |
| url | http://journal.xpu.edu.cn/en/#/digest?ArticleID=1523 |
| work_keys_str_mv | AT chensixin videoimagefeatureextractionmethodforpolicedronesinvestigationandevidencecollection AT hefengwei videoimagefeatureextractionmethodforpolicedronesinvestigationandevidencecollection |