An Improved Method for Human Activity Detection with High-Resolution Images by Fusing Pooling Enhancement and Multi-Task Learning
Deep learning has garnered increasing attention in human activity detection due to its advantages, such as not relying on expert knowledge and automatic feature extraction. However, the existing deep learning-based approaches are primarily confined to recognizing specific types of human activities,...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/1/159 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841549009095753728 |
|---|---|
| author | Haoji Li Shilong Ren Lei Fang Jinyue Chen Xinfeng Wang Guoqiang Wang Qingzhu Zhang Qiao Wang |
| author_facet | Haoji Li Shilong Ren Lei Fang Jinyue Chen Xinfeng Wang Guoqiang Wang Qingzhu Zhang Qiao Wang |
| author_sort | Haoji Li |
| collection | DOAJ |
| description | Deep learning has garnered increasing attention in human activity detection due to its advantages, such as not relying on expert knowledge and automatic feature extraction. However, the existing deep learning-based approaches are primarily confined to recognizing specific types of human activities, which hinders scientific decision-making and comprehensive environmental protection. Therefore, there is an urgent need to develop a deep learning model to address multiple-type human activity detection with finer-resolution images. In this study, we proposed a new multi-task learning model (named PE-MLNet) to simultaneously achieve change detection and land use classification in GF-6 bitemporal images. Meanwhile, we also designed a pooling enhancement module (PEM) to accurately capture multi-scale change details from the bitemporal feature maps through combining differencing and concatenating branches. An independent annotated dataset at Yellow River Delta was taken to examine the effectiveness of PE-MLNet. The results showed that PE-MLNet exhibited obvious improvements in both detection accuracy and detail handling compared with other existing methods. Further analysis uncovered that the areas of buildings, roads, and oil depots has obviously increased, while the farmland and wetland areas largely decreased over the five years, indicating an expansion of human activities and their increased impacts on natural environments. |
| format | Article |
| id | doaj-art-97dd460170eb4b12a03999b2b82c96f4 |
| institution | Kabale University |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-97dd460170eb4b12a03999b2b82c96f42025-01-10T13:20:26ZengMDPI AGRemote Sensing2072-42922025-01-0117115910.3390/rs17010159An Improved Method for Human Activity Detection with High-Resolution Images by Fusing Pooling Enhancement and Multi-Task LearningHaoji Li0Shilong Ren1Lei Fang2Jinyue Chen3Xinfeng Wang4Guoqiang Wang5Qingzhu Zhang6Qiao Wang7Academician Workstation for Big Data in Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, ChinaAcademician Workstation for Big Data in Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, ChinaAcademician Workstation for Big Data in Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, ChinaAcademician Workstation for Big Data in Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, ChinaAcademician Workstation for Big Data in Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, ChinaAcademician Workstation for Big Data in Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, ChinaAcademician Workstation for Big Data in Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, ChinaAcademician Workstation for Big Data in Ecology and Environment, Environment Research Institute, Shandong University, Qingdao 266003, ChinaDeep learning has garnered increasing attention in human activity detection due to its advantages, such as not relying on expert knowledge and automatic feature extraction. However, the existing deep learning-based approaches are primarily confined to recognizing specific types of human activities, which hinders scientific decision-making and comprehensive environmental protection. Therefore, there is an urgent need to develop a deep learning model to address multiple-type human activity detection with finer-resolution images. In this study, we proposed a new multi-task learning model (named PE-MLNet) to simultaneously achieve change detection and land use classification in GF-6 bitemporal images. Meanwhile, we also designed a pooling enhancement module (PEM) to accurately capture multi-scale change details from the bitemporal feature maps through combining differencing and concatenating branches. An independent annotated dataset at Yellow River Delta was taken to examine the effectiveness of PE-MLNet. The results showed that PE-MLNet exhibited obvious improvements in both detection accuracy and detail handling compared with other existing methods. Further analysis uncovered that the areas of buildings, roads, and oil depots has obviously increased, while the farmland and wetland areas largely decreased over the five years, indicating an expansion of human activities and their increased impacts on natural environments.https://www.mdpi.com/2072-4292/17/1/159human activitychange detectionsemantic segmentationmulti-task learningremote sensing |
| spellingShingle | Haoji Li Shilong Ren Lei Fang Jinyue Chen Xinfeng Wang Guoqiang Wang Qingzhu Zhang Qiao Wang An Improved Method for Human Activity Detection with High-Resolution Images by Fusing Pooling Enhancement and Multi-Task Learning Remote Sensing human activity change detection semantic segmentation multi-task learning remote sensing |
| title | An Improved Method for Human Activity Detection with High-Resolution Images by Fusing Pooling Enhancement and Multi-Task Learning |
| title_full | An Improved Method for Human Activity Detection with High-Resolution Images by Fusing Pooling Enhancement and Multi-Task Learning |
| title_fullStr | An Improved Method for Human Activity Detection with High-Resolution Images by Fusing Pooling Enhancement and Multi-Task Learning |
| title_full_unstemmed | An Improved Method for Human Activity Detection with High-Resolution Images by Fusing Pooling Enhancement and Multi-Task Learning |
| title_short | An Improved Method for Human Activity Detection with High-Resolution Images by Fusing Pooling Enhancement and Multi-Task Learning |
| title_sort | improved method for human activity detection with high resolution images by fusing pooling enhancement and multi task learning |
| topic | human activity change detection semantic segmentation multi-task learning remote sensing |
| url | https://www.mdpi.com/2072-4292/17/1/159 |
| work_keys_str_mv | AT haojili animprovedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning AT shilongren animprovedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning AT leifang animprovedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning AT jinyuechen animprovedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning AT xinfengwang animprovedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning AT guoqiangwang animprovedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning AT qingzhuzhang animprovedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning AT qiaowang animprovedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning AT haojili improvedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning AT shilongren improvedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning AT leifang improvedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning AT jinyuechen improvedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning AT xinfengwang improvedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning AT guoqiangwang improvedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning AT qingzhuzhang improvedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning AT qiaowang improvedmethodforhumanactivitydetectionwithhighresolutionimagesbyfusingpoolingenhancementandmultitasklearning |