Prioritized air light and transmittance extraction (PATE) using dual weighted deep channel and spatial attention based model for image dehazing

The image dehazing is a complicated dilemma to resolve the haze density influence on the object depth. Though many pixel-based or color space-based algorithms are used by many researchers for resolving this issue, due to lack of utilizing the prominent information makes the objective not achieved properly. The main objective if this work is to dehaze an image with fine-tuned parameters. The proposed research work prioritized air light and transmittance extraction (PATE) using a novel dual weighted deep channel and spatial attention (DWDCA)-based model helps to give proper weightage for the color information to restore the prominent color. The loss information calculation framework is proposed in this model to further enhance the output. The performance analysis is done with the benchmark datasets such as i-haze, o-haze and SOTS datasets where the proposed algorithm gives significant improvements in the metrics such as PSNR, SSIM and CIEDE2000 than the state-of-the-art algorithms. The proposed dehazing model with dual weighted channel and spatial attention block effectively preserves the data and improves the vision of the image.