Low Light Image Enhancement Based on Attention Map

The quality of images captured in low light will often exhibit characteristics like low brightness, low contrast, reduced visibility, high noise levels, and loss of crucial details.For this problem, we present a sophisticated approach for low light image enhancement with Convolution Neural Network (CNN).CNNs have gained prominence in various image processing techniques due to their ability to generate realistic images.To improve the performance of enhancing images compared to other methods, we propose a method on attention map based on M-Net architecture.In this, we place this attention block in the M-Net architecture to extract the details from the image.