MUMBAI, India, Aug. 31 -- Intellectual Property India has published a patent application (202641101013 A) filed by Easwari Engineering College; and Srm Institute Of Science And Technology,ramapuram Campus on August 21, 2026, for Joint Noise-Semantic Adaptive Diffusion Scheduling For Low-Light Image Enhancement.
Inventors include Sharmila M; Senthil Kumar M; and Dr. G. Jeyaram.
The application for the patent was published on August 28, 2026, under issue no. 35/2026.
Abstract: The present invention relates to a computer-implemented system and method for low-light image enhancement using joint noise- semantic adaptive diffusion scheduling. The system receives a low-light image and performs preprocessing to obtain an image suitable for diffusion-based enhancement. A noise estimation module determines spatial noise severity information, while a semantic segmentation module generates semantic region information from the input image. The noise severity information and semantic region information are jointly analyzed to determine a regional difficulty measure for different portions of the image. Based on the regional difficulty measure, an adaptive step-budget predictor assigns different denoising step budgets to respective image regions. A conditional diffusion enhancement moduleperforms region-aware reverse diffusion using the assigned step budgets, thereby allocating greater denoising computation to highly degraded or semantically complex regions and reducing unnecessary computation for relatively simple regions. The system generates an enhanced low-light image with improved visibility, noise suppression and detail preservation while potentially reducing computational requirements. Total Number of Words in Abstract: 150
Disclaimer: Curated by HT Syndication.
