De-noising the image using DBST-LCM-CLAHE: A deep learning approach

Abstract Histogram Equalization (HE) is one of the most popular techniques for this purpose. Most histogram equalization techniques, including Contrast Limited Adaptive (CLAHE) and Local Modification CLAHE (LCM CLAHE), use a fixed block size technique feature enhancement. Due to this, all these state art are used give poor denoising performance after In paper, deep learning based new approach, namely Dynamic Block Size Technique (DBST), improve image denoising. we Categorical Subjective Image Quality (CSIQ) set, an database generally preprocessing images. The results obtained from experiments show better different important parameters (used by techniques). work novel in images because work, classify depending upon features selecting appropriate sizes dynamically during preprocessing. Proposed outperforms terms PSNR, MSE, NRMSE, SSIM SYNTROPY. average respective values 18.92, 863.86, 0.25, 0.81 19.35 comparison LCM CLAHE.