Low-Light Homomorphic Filtering Network for integrating image enhancement and classification
作者：
Highlights：
• Introduces Low-Light Image (LLI) Enhancement model for Object Classification task.
• Introduces LLHFNet deep learning model for image-to-frequency filter learning.
• Introduces LLI enhancer–classifier to integrate LLHFNet in object classification.
• Performs joint training to optimize both enhancement and classification models.
• Evaluates impact of LLI enhancement on object detecting and classification quality.
• Produces 5.5% improvement in classification accuracy on Pascal VOC & ExDark LLIs.
摘要
•Introduces Low-Light Image (LLI) Enhancement model for Object Classification task.•Introduces LLHFNet deep learning model for image-to-frequency filter learning.•Introduces LLI enhancer–classifier to integrate LLHFNet in object classification.•Performs joint training to optimize both enhancement and classification models.•Evaluates impact of LLI enhancement on object detecting and classification quality.•Produces 5.5% improvement in classification accuracy on Pascal VOC & ExDark LLIs.
论文关键词：Image enhancement,Low-light conditions,Deep learning,Object classification,Homomorphic filtering
论文评审过程：Received 3 August 2021, Revised 23 September 2021, Accepted 30 September 2021, Available online 12 October 2021, Version of Record 21 October 2021.
论文官网地址：https://doi.org/10.1016/j.image.2021.116527
