Low illumination image enhancement technology based on iterative dense residual network

The images collected by imaging equipment in low-light environments often have low brightness, low contrast, loss of image details, etc., which will cause huge interference in certain application scenarios such as video surveillance, automatic driving, face recognition, etc., according to this article A low-illuminance image enhancement algorithm based on an iterative dense residual network is used to solve the above problems. The network uses the image brightness component V as input, uses a new type of dense residual block to extract image features, adds an attention mechanism to enhance feature learning, and finally obtains an enhanced image through image reconstruction. Experimental results show that compared with mainstream algorithms, the algorithm in this paper performs well in subjective and objective evaluation, and meets the requirements of natural color, contrast, and robustness for image enhancement.