An improved video dehazing algorithm based on ZYNQ architecture

To address the issues of contrast degradation and colour distortion in video images captured by the device in haze and other environments, the contrast-limited adaptive histogram equalization (CLAHE) algorithm is improved, and a joint bilateral filtering algorithm for image enhancement is proposed. The CLAHE algorithm and bilateral filtering algorithm are applied to the bright channel Y in the YUV colour space. The parallel pipeline architecture of the ZYNQ platform is utilized to combine RAM and FPGA to realize hardware and software synergy. The experimental results demonstrate that the algorithm has improved the peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), Underwater Image Quality Measure (UIQM), and Average Gradient (AG) and information entropy (IE) of the image. Consuming only 2.278 W on the FPGA, this hardware circuit falls into the low-power category.