Low Light Image Enhancement using Machine Learning

Abstract

Deep Learning is a recent and very powerful machine learning approach which uses neural networks to mimic activities in layers of neurons. Deep learning algorithms have achieved remarkable performances in various image processing and computer vision tasks. low-light image enhancement is very challenging due to the difficulty in handling various factors simultaneously including brightness, contrast, artifacts and noise. So, we are going to propose a deep learning-based method for low-light image enhancement. The model here uses a Convolutional Neural Network (CNN) which makes the use of a dataset of raw short-exposure night-time images, with corresponding long-exposure reference images. This makes results from extreme scenarios like night photography very easy and efficient as compared to traditional denoising and deblurring techniques. The low-light image enhancement is of high importance for several computer vision and computational photography tasks. Low-light and image enhancement is important for video surveillance. In addition, low-light image enhancement leads to increasing the scope of many computer vision algorithms designed to deal with normal light images. However, a high quality low-light image enhancement is a challenging task and developing fast and reliable methods for low-light image enhancement still a topic for intensive research.