MUMBAI, India, Sept. 7 -- Intellectual Property India has published a patent application (202541044519 A) filed by Srm Institute Of Science And Technology Tiruchirappalli on May 08, 2025, for Detecting Toxic Dyes Using Artificial Intelligence.
Inventors include Roshni Bhubaneswari Swain; Subhojit Ram; Vikram Singh; and Dr. J. Jagdeesh Kannan.
The application for the patent was published on September 04, 2026, under issue no. 36/2026.
Abstract: The present invention discloses an AI-driven, real-time system for the non-destructive detection and toxicity classification of harmful synthetic dyes in textile fabrics. The system integrates high-resolution RGB and hyperspectral/multispectral imaging with a digital image preprocessing pipeline and a Convolutional Neural Network (CNN) deep learning classifier trained on over 200 labelled dyed samples. Feature extraction encompasses RGB-HSV colour space analysis, GLCM texture features, and PCA-compressed spectral signatures. The CNN model achieves 94.7% overall classification accuracy with 95.8% recall for toxic dyes, operating with inference times under 1.2 seconds on standard hardware and under 2.5 seconds on edge devices. A novel Toxicity Index, correlated at 89% with expert-labelled hazard assessments, provides quantitative severity scoring. The system is deployable across industrial, regulatory, mobile, and cloud platforms without chemical reagents, enabling sustainable compliance with REACH, EPA, and BIS standards. Applications include textile mill quality control, environmental monitoring, and consumer fabric safety verification.
Disclaimer: Curated by HT Syndication.
