Color-Dense Illumination Adjustment Network for Removing Haze and Smoke from Fire Scenario Images
The atmospheric particles and aerosols from burning usually cause visual artifacts in single images captured from fire scenarios. Most existing haze removal methods exploit the atmospheric scattering model (ASM) for visual enhancement, which inevitably leads to inaccurate estimation of the atmospher...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/3/911 |
| _version_ | 1797484754271993856 |
|---|---|
| author | Chuansheng Wang Jinxing Hu Xiaowei Luo Mei-Po Kwan Weihua Chen Hao Wang |
| author_facet | Chuansheng Wang Jinxing Hu Xiaowei Luo Mei-Po Kwan Weihua Chen Hao Wang |
| author_sort | Chuansheng Wang |
| collection | DOAJ |
| description | The atmospheric particles and aerosols from burning usually cause visual artifacts in single images captured from fire scenarios. Most existing haze removal methods exploit the atmospheric scattering model (ASM) for visual enhancement, which inevitably leads to inaccurate estimation of the atmosphere light and transmission matrix of the smoky and hazy inputs. To solve these problems, we present a novel color-dense illumination adjustment network (CIANet) for joint recovery of transmission matrix, illumination intensity, and the dominant color of aerosols from a single image. Meanwhile, to improve the visual effects of the recovered images, the proposed CIANet jointly optimizes the transmission map, atmospheric optical value, the color of aerosol, and a preliminary recovered scene. Furthermore, we designed a reformulated ASM, called the aerosol scattering model (ESM), to smooth out the enhancement results while keeping the visual effects and the semantic information of different objects. Experimental results on both the proposed RFSIE and NTIRE’20 demonstrate our superior performance favorably against state-of-the-art dehazing methods regarding PSNR, SSIM and subjective visual quality. Furthermore, when concatenating CIANet with Faster R-CNN, we witness an improvement of the objection performance with a large margin. |
| first_indexed | 2024-03-09T23:09:58Z |
| format | Article |
| id | doaj.art-984f5e116536477cbdb78f8aa4cef7e3 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T23:09:58Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-984f5e116536477cbdb78f8aa4cef7e32023-11-23T17:47:28ZengMDPI AGSensors1424-82202022-01-0122391110.3390/s22030911Color-Dense Illumination Adjustment Network for Removing Haze and Smoke from Fire Scenario ImagesChuansheng Wang0Jinxing Hu1Xiaowei Luo2Mei-Po Kwan3Weihua Chen4Hao Wang5State Key Laboratory of Nuclear Power Safety Monitoring Technology and Equipment, China Nuclear Power Engineering Co., Ltd., Shenzhen 518172, ChinaShenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, ChinaDepartment of Architecture and Civil Engineering, City University of Hong Kong, Kowloon, Hong Kong 999077, ChinaInstitute of Space and Earth Information Science, The Chinese University of Hong Kong, Shatin, Hong Kong 999077, ChinaState Key Laboratory of Nuclear Power Safety Monitoring Technology and Equipment, China Nuclear Power Engineering Co., Ltd., Shenzhen 518172, ChinaState Key Laboratory of Nuclear Power Safety Monitoring Technology and Equipment, China Nuclear Power Engineering Co., Ltd., Shenzhen 518172, ChinaThe atmospheric particles and aerosols from burning usually cause visual artifacts in single images captured from fire scenarios. Most existing haze removal methods exploit the atmospheric scattering model (ASM) for visual enhancement, which inevitably leads to inaccurate estimation of the atmosphere light and transmission matrix of the smoky and hazy inputs. To solve these problems, we present a novel color-dense illumination adjustment network (CIANet) for joint recovery of transmission matrix, illumination intensity, and the dominant color of aerosols from a single image. Meanwhile, to improve the visual effects of the recovered images, the proposed CIANet jointly optimizes the transmission map, atmospheric optical value, the color of aerosol, and a preliminary recovered scene. Furthermore, we designed a reformulated ASM, called the aerosol scattering model (ESM), to smooth out the enhancement results while keeping the visual effects and the semantic information of different objects. Experimental results on both the proposed RFSIE and NTIRE’20 demonstrate our superior performance favorably against state-of-the-art dehazing methods regarding PSNR, SSIM and subjective visual quality. Furthermore, when concatenating CIANet with Faster R-CNN, we witness an improvement of the objection performance with a large margin.https://www.mdpi.com/1424-8220/22/3/911haze removalvisual enhancementaerosol scattering model |
| spellingShingle | Chuansheng Wang Jinxing Hu Xiaowei Luo Mei-Po Kwan Weihua Chen Hao Wang Color-Dense Illumination Adjustment Network for Removing Haze and Smoke from Fire Scenario Images Sensors haze removal visual enhancement aerosol scattering model |
| title | Color-Dense Illumination Adjustment Network for Removing Haze and Smoke from Fire Scenario Images |
| title_full | Color-Dense Illumination Adjustment Network for Removing Haze and Smoke from Fire Scenario Images |
| title_fullStr | Color-Dense Illumination Adjustment Network for Removing Haze and Smoke from Fire Scenario Images |
| title_full_unstemmed | Color-Dense Illumination Adjustment Network for Removing Haze and Smoke from Fire Scenario Images |
| title_short | Color-Dense Illumination Adjustment Network for Removing Haze and Smoke from Fire Scenario Images |
| title_sort | color dense illumination adjustment network for removing haze and smoke from fire scenario images |
| topic | haze removal visual enhancement aerosol scattering model |
| url | https://www.mdpi.com/1424-8220/22/3/911 |
| work_keys_str_mv | AT chuanshengwang colordenseilluminationadjustmentnetworkforremovinghazeandsmokefromfirescenarioimages AT jinxinghu colordenseilluminationadjustmentnetworkforremovinghazeandsmokefromfirescenarioimages AT xiaoweiluo colordenseilluminationadjustmentnetworkforremovinghazeandsmokefromfirescenarioimages AT meipokwan colordenseilluminationadjustmentnetworkforremovinghazeandsmokefromfirescenarioimages AT weihuachen colordenseilluminationadjustmentnetworkforremovinghazeandsmokefromfirescenarioimages AT haowang colordenseilluminationadjustmentnetworkforremovinghazeandsmokefromfirescenarioimages |