A Full-Scale Shadow Detection Network Based on Multiple Attention Mechanisms for Remote-Sensing Images
Shadows degrade image quality and complicate interpretation, underscoring the importance of accurate shadow detection for many image analysis tasks. However, due to the complex backgrounds and variable shadow characteristics of remote sensing images (RSIs), existing methods often struggle with accur...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/16/24/4789 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846102897982439424 |
|---|---|
| author | Lei Zhang Qing Zhang Yu Wu Yanfeng Zhang Shan Xiang Donghai Xie Zeyu Wang |
| author_facet | Lei Zhang Qing Zhang Yu Wu Yanfeng Zhang Shan Xiang Donghai Xie Zeyu Wang |
| author_sort | Lei Zhang |
| collection | DOAJ |
| description | Shadows degrade image quality and complicate interpretation, underscoring the importance of accurate shadow detection for many image analysis tasks. However, due to the complex backgrounds and variable shadow characteristics of remote sensing images (RSIs), existing methods often struggle with accurately detecting shadows of various scales and misclassifying dark, non-shaded areas as shadows. To address these issues, we proposed a comprehensive shadow detection network called MAMNet. Firstly, we proposed a multi-scale spatial channel attention fusion module, which extracted multi-scale features incorporating both spatial and channel information, allowing the model to flexibly adapt to shadows of different scales. Secondly, to address the issue of false detection in non-shadow areas, we introduced a criss-cross attention module, enabling non-shadow pixels to be compared with other shadow and non-shadow pixels in the same row and column, learning similar features of pixels in the same category, which improved the classification accuracy of non-shadow pixels. Finally, to address the issue of important information from the other two modules being lost due to continuous upsampling during the decoding phase, we proposed an auxiliary branch module to assist the main branch in decision-making, ensuring that the final output retained the key information from all stages. The experimental results demonstrated that the model outperformed the current state-of-the-art RSI shadow detection method on the aerial imagery dataset for shadow detection (AISD). The model achieved an overall accuracy (OA) of 97.50%, an <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>F</mi><mn>1</mn></msub></semantics></math></inline-formula> score of 94.07%, an intersection over union (IOU) of 88.87%, a precision of 95.06%, and a BER of 4.05%, respectively. Additionally, visualization results indicated that our model could effectively detect shadows of various scales while avoiding false detection in non-shadow areas. Therefore, this model offers an efficient solution for shadow detection in aerial imagery. |
| format | Article |
| id | doaj-art-f5d08b400602456cb7fe76ef4fce5a10 |
| institution | Kabale University |
| issn | 2072-4292 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-f5d08b400602456cb7fe76ef4fce5a102024-12-27T14:51:13ZengMDPI AGRemote Sensing2072-42922024-12-011624478910.3390/rs16244789A Full-Scale Shadow Detection Network Based on Multiple Attention Mechanisms for Remote-Sensing ImagesLei Zhang0Qing Zhang1Yu Wu2Yanfeng Zhang3Shan Xiang4Donghai Xie5Zeyu Wang6Henan Key Laboratory of Big Data Analysis and Processing, Henan University, Kaifeng 475004, ChinaHenan Key Laboratory of Big Data Analysis and Processing, Henan University, Kaifeng 475004, ChinaInstitute of Surface-Earth System Science, School of Earth System Science, Tianjin University, Tianjin 300072, ChinaHenan Key Laboratory of Big Data Analysis and Processing, Henan University, Kaifeng 475004, ChinaCollege of Mathematics and Information Science, Henan Normal University, Xinxiang 453007, ChinaInstitute of Resource and Environment, Capital Normal University, Beijing 100048, ChinaInstitute of Surface-Earth System Science, School of Earth System Science, Tianjin University, Tianjin 300072, ChinaShadows degrade image quality and complicate interpretation, underscoring the importance of accurate shadow detection for many image analysis tasks. However, due to the complex backgrounds and variable shadow characteristics of remote sensing images (RSIs), existing methods often struggle with accurately detecting shadows of various scales and misclassifying dark, non-shaded areas as shadows. To address these issues, we proposed a comprehensive shadow detection network called MAMNet. Firstly, we proposed a multi-scale spatial channel attention fusion module, which extracted multi-scale features incorporating both spatial and channel information, allowing the model to flexibly adapt to shadows of different scales. Secondly, to address the issue of false detection in non-shadow areas, we introduced a criss-cross attention module, enabling non-shadow pixels to be compared with other shadow and non-shadow pixels in the same row and column, learning similar features of pixels in the same category, which improved the classification accuracy of non-shadow pixels. Finally, to address the issue of important information from the other two modules being lost due to continuous upsampling during the decoding phase, we proposed an auxiliary branch module to assist the main branch in decision-making, ensuring that the final output retained the key information from all stages. The experimental results demonstrated that the model outperformed the current state-of-the-art RSI shadow detection method on the aerial imagery dataset for shadow detection (AISD). The model achieved an overall accuracy (OA) of 97.50%, an <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>F</mi><mn>1</mn></msub></semantics></math></inline-formula> score of 94.07%, an intersection over union (IOU) of 88.87%, a precision of 95.06%, and a BER of 4.05%, respectively. Additionally, visualization results indicated that our model could effectively detect shadows of various scales while avoiding false detection in non-shadow areas. Therefore, this model offers an efficient solution for shadow detection in aerial imagery.https://www.mdpi.com/2072-4292/16/24/4789shadow detectionremote sensingattention mechanismmulti-scale feature extraction |
| spellingShingle | Lei Zhang Qing Zhang Yu Wu Yanfeng Zhang Shan Xiang Donghai Xie Zeyu Wang A Full-Scale Shadow Detection Network Based on Multiple Attention Mechanisms for Remote-Sensing Images Remote Sensing shadow detection remote sensing attention mechanism multi-scale feature extraction |
| title | A Full-Scale Shadow Detection Network Based on Multiple Attention Mechanisms for Remote-Sensing Images |
| title_full | A Full-Scale Shadow Detection Network Based on Multiple Attention Mechanisms for Remote-Sensing Images |
| title_fullStr | A Full-Scale Shadow Detection Network Based on Multiple Attention Mechanisms for Remote-Sensing Images |
| title_full_unstemmed | A Full-Scale Shadow Detection Network Based on Multiple Attention Mechanisms for Remote-Sensing Images |
| title_short | A Full-Scale Shadow Detection Network Based on Multiple Attention Mechanisms for Remote-Sensing Images |
| title_sort | full scale shadow detection network based on multiple attention mechanisms for remote sensing images |
| topic | shadow detection remote sensing attention mechanism multi-scale feature extraction |
| url | https://www.mdpi.com/2072-4292/16/24/4789 |
| work_keys_str_mv | AT leizhang afullscaleshadowdetectionnetworkbasedonmultipleattentionmechanismsforremotesensingimages AT qingzhang afullscaleshadowdetectionnetworkbasedonmultipleattentionmechanismsforremotesensingimages AT yuwu afullscaleshadowdetectionnetworkbasedonmultipleattentionmechanismsforremotesensingimages AT yanfengzhang afullscaleshadowdetectionnetworkbasedonmultipleattentionmechanismsforremotesensingimages AT shanxiang afullscaleshadowdetectionnetworkbasedonmultipleattentionmechanismsforremotesensingimages AT donghaixie afullscaleshadowdetectionnetworkbasedonmultipleattentionmechanismsforremotesensingimages AT zeyuwang afullscaleshadowdetectionnetworkbasedonmultipleattentionmechanismsforremotesensingimages AT leizhang fullscaleshadowdetectionnetworkbasedonmultipleattentionmechanismsforremotesensingimages AT qingzhang fullscaleshadowdetectionnetworkbasedonmultipleattentionmechanismsforremotesensingimages AT yuwu fullscaleshadowdetectionnetworkbasedonmultipleattentionmechanismsforremotesensingimages AT yanfengzhang fullscaleshadowdetectionnetworkbasedonmultipleattentionmechanismsforremotesensingimages AT shanxiang fullscaleshadowdetectionnetworkbasedonmultipleattentionmechanismsforremotesensingimages AT donghaixie fullscaleshadowdetectionnetworkbasedonmultipleattentionmechanismsforremotesensingimages AT zeyuwang fullscaleshadowdetectionnetworkbasedonmultipleattentionmechanismsforremotesensingimages |