Image Enhancement in the LIP Framework and Noise Reduction with Deep Convolutional Neural Networks to Produce High Quality Images from Low Light Acquisitions
The LIP (Logarithmic Image Processing) model is recognized as an efficient framework to process images acquired in transmitted and reflected light, and to take into account the human visual system. Several image enhancement algorithms have been developed in the LIP framework. Some of them exploit an...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IFSA Publishing, S.L.
2021-02-01
|
| Series: | Sensors & Transducers |
| Subjects: | |
| Online Access: | https://sensorsportal.com/HTML/DIGEST/february_2021/Vol_249/P_3205.pdf |
| _version_ | 1797752189176774656 |
|---|---|
| author | Maxime CARRE Michel JOURLIN |
| author_facet | Maxime CARRE Michel JOURLIN |
| author_sort | Maxime CARRE |
| collection | DOAJ |
| description | The LIP (Logarithmic Image Processing) model is recognized as an efficient framework to process images acquired in transmitted and reflected light, and to take into account the human visual system. Several image enhancement algorithms have been developed in the LIP framework. Some of them exploit an important property of the LIP model consisting in simulating exposure time variations. Applied to very low light images, our LIP algorithms enhance not only the signal, but also the noise and lead to quantized grey levels. Noise reduction based on Deep Convolutional Neural Networks is performed on these enhanced noisy images. The combination of LIP enhancement and DCNN denoising transforms low light images into well-balanced images with low noise level. In addition to classical PSNR evaluation, we propose an estimation of image noise based on LIP contrasts computed at a local scale. The use of LIP contrast results in a noise evaluation consistent with human vision. |
| first_indexed | 2024-03-12T16:59:35Z |
| format | Article |
| id | doaj.art-40ddf83670274606a09362e2748ab610 |
| institution | Directory Open Access Journal |
| issn | 2306-8515 1726-5479 |
| language | English |
| last_indexed | 2024-03-12T16:59:35Z |
| publishDate | 2021-02-01 |
| publisher | IFSA Publishing, S.L. |
| record_format | Article |
| series | Sensors & Transducers |
| spelling | doaj.art-40ddf83670274606a09362e2748ab6102023-08-07T15:27:13ZengIFSA Publishing, S.L.Sensors & Transducers2306-85151726-54792021-02-0124923644Image Enhancement in the LIP Framework and Noise Reduction with Deep Convolutional Neural Networks to Produce High Quality Images from Low Light AcquisitionsMaxime CARRE0Michel JOURLIN1NT2I CompanyHubert Curien LaboratoryThe LIP (Logarithmic Image Processing) model is recognized as an efficient framework to process images acquired in transmitted and reflected light, and to take into account the human visual system. Several image enhancement algorithms have been developed in the LIP framework. Some of them exploit an important property of the LIP model consisting in simulating exposure time variations. Applied to very low light images, our LIP algorithms enhance not only the signal, but also the noise and lead to quantized grey levels. Noise reduction based on Deep Convolutional Neural Networks is performed on these enhanced noisy images. The combination of LIP enhancement and DCNN denoising transforms low light images into well-balanced images with low noise level. In addition to classical PSNR evaluation, we propose an estimation of image noise based on LIP contrasts computed at a local scale. The use of LIP contrast results in a noise evaluation consistent with human vision.https://sensorsportal.com/HTML/DIGEST/february_2021/Vol_249/P_3205.pdflip modelexposure time simulationimage enhancementlow light imagesnoise reductiondeep convolutional neural networks |
| spellingShingle | Maxime CARRE Michel JOURLIN Image Enhancement in the LIP Framework and Noise Reduction with Deep Convolutional Neural Networks to Produce High Quality Images from Low Light Acquisitions Sensors & Transducers lip model exposure time simulation image enhancement low light images noise reduction deep convolutional neural networks |
| title | Image Enhancement in the LIP Framework and Noise Reduction with Deep Convolutional Neural Networks to Produce High Quality Images from Low Light Acquisitions |
| title_full | Image Enhancement in the LIP Framework and Noise Reduction with Deep Convolutional Neural Networks to Produce High Quality Images from Low Light Acquisitions |
| title_fullStr | Image Enhancement in the LIP Framework and Noise Reduction with Deep Convolutional Neural Networks to Produce High Quality Images from Low Light Acquisitions |
| title_full_unstemmed | Image Enhancement in the LIP Framework and Noise Reduction with Deep Convolutional Neural Networks to Produce High Quality Images from Low Light Acquisitions |
| title_short | Image Enhancement in the LIP Framework and Noise Reduction with Deep Convolutional Neural Networks to Produce High Quality Images from Low Light Acquisitions |
| title_sort | image enhancement in the lip framework and noise reduction with deep convolutional neural networks to produce high quality images from low light acquisitions |
| topic | lip model exposure time simulation image enhancement low light images noise reduction deep convolutional neural networks |
| url | https://sensorsportal.com/HTML/DIGEST/february_2021/Vol_249/P_3205.pdf |
| work_keys_str_mv | AT maximecarre imageenhancementinthelipframeworkandnoisereductionwithdeepconvolutionalneuralnetworkstoproducehighqualityimagesfromlowlightacquisitions AT micheljourlin imageenhancementinthelipframeworkandnoisereductionwithdeepconvolutionalneuralnetworkstoproducehighqualityimagesfromlowlightacquisitions |