FPGA Implementation of Dehazing Model Based Low-Light Image Enhancement Algorithm.

Low lighting conditions degrade the quality of the captured image by darkening some of its parts. This may lead to the loss of important information contained in the captured image in low-light. It is always desirable to improve the visual quality of low-light images. In this paper, a low-light image enhancement model based on the dehazing principle is proposed to address this issue. Further, its hardware architecture is proposed to meet the real-time system requirement. The proposed pixel-based non-linear transmission estimation technique prevent images from being over-enhanced. Additionally, it suppresses the artifacts around the edges. This curtails the need for an edge-preserving filter. The field-programmable gate array (FPGA) implementation of the proposed architecture requires only 694 LUTs, and it is capable of processing more than 60 images of resolution $1920\times 1080$ per second.