A Python Programmed Resnet-50 CNN Designed for Low Light Image Enhancement

Enhancing the quality of photos taken in low light is the goal of the computer vision problem known as Low-Light Image Enhancement. Increasing the brightness, clarity, and visual attractiveness of images while minimizing noise and distortion is the aim of low-light image enhancement. Most CNN-based methods have complex network structures and the network consists of several sub-networks. The challenges posed by insufficient illumination in image enhancement with their visual quality low light condition often lead to reduced visibility and loss of important details, impacting various applications in autonomous system. In contrast to these problems we utilize ResNet-50 for image segmentation up to 50 layers to reduce noises in image, scattering of images and more focusing on higher brightness places in an image. Python programming is used to restoring edges and textures and improving overall image visibility. According to our experimental results, the proposed method achieves improvised image in clarity and environmental issues such as fog, pollution.