MUMBAI, India, Oct. 26 -- Intellectual Property India has published a patent application (202631123730 A) filed by C. V. Raman Global University on October 15, 2026, for Efficient Time-Distributed Cnn–lstm Model For Driver Drowsiness Detection Using Low-Resolution Eye Image Sequences.
Inventors include Arpita Gouda; Dusari Sandeep; and Dr. Pushpendra Kumar Gupta.
The application for the patent was published on October 23, 2026, under issue no. 43/2026.
Abstract: A method and system for detecting driver drowsiness using low-resolution eye image sequences. Driver eye images are converted to grayscale, resized to 32 × 32 pixels and normalised, and are organised into short temporal sequences using a sliding window. Each sequence is processed by a time-distributed convolutional neural network that extracts spatial features from each frame while preserving temporal order. The ordered features are supplied to a long short-term memory network that captures temporal dependencies such as blinking and prolonged eye closure. A sigmoid output layer produces a probability that the driver is drowsy, and the driver is classified as drowsy or non-drowsy by comparing the probability with a threshold. The model, trained with binary cross-entropy loss and the Adam optimiser, contains approximately 629 thousand trainable parameters and is suitable for real-time execution on embedded and in-vehicle processors. The invention achieves high accuracy on publicly available datasets while reducing computational cost and hardware requirements. It provides a practical and cost-effective driver monitoring solution that can be integrated into advanced driver assistance systems and fleet monitoring platforms to help reduce fatiguerelated road accidents.
Disclaimer: Curated by HT Syndication.
