MUMBAI, India, Jan. 23 -- Intellectual Property India has published a patent application (202641001765 A) filed by Karpagam Academy Of Higher Education; Karpagam Institute Of Technology; C Sasthi Kumar; and L Kavi Bharathi, Coimbatore, Tamil Nadu, on Jan. 7, for 'traffic sign detection and recongnition based on convolutional neural network.'
Inventor(s) include C Sasthi Kumar; and L Kavi Bharathi.
The application for the patent was published on Jan. 23, under issue no. 04/2026.
According to the abstract released by the Intellectual Property India: "Road sign detection and recognition play a crucial role in improving road safety and assisting drivers in adhering to traffic regulations. Misinterpretation or negligence of traffic signboards remains a primary cause of accidents, making automated sign recognition systems essential in modern intelligent transportation solutions. This paper presents a real-time Road Sign Detection and Recognition System designed to automatically capture, analyze, classify, and notify drivers of detected traffic signs using advanced deep learning techniques. The proposed system integrates an onboard vehicle camera that continuously streams video frames containing road scenes. These frames undergo preprocessing operations such as resizing, noise reduction, normalization, and region-of-interest extraction to ensure optimal input quality for the recognition pipeline. A Convolutional Neural Network (CNN) model, trained on a structured dataset of diverse traffic sign categories, is employed for classification. The CNN architecture enables robust learning of spatial features and improves classification accuracy under varying lighting, occlusion, and environmental conditions. The system further incorporates a decision-making module that evaluates confidence levels and filters out false detections, enhancing reliability during real-time operation. Upon successful sign recognition, the system generates immediate alerts to the driver through voice-based notifications delivered via the vehicle's speaker system. This ensures timely awareness of important regulatory notices, warnings, and driving instructions, enabling the driver to respond appropriately. The complete pipeline operates efficiently in real time, making it suitable for integration with Advanced Driver Assistance Systems (ADAS) and autonomous vehicle frameworks. The experimental evaluation demonstrates high accuracy in traffic sign identification and significantly reduces driver dependence on visual interpretation alone. The proposed system contributes to safer driving environments by minimizing human error, improving situational awareness, and providing an automated mechanism for continuous road sign monitoring."
Disclaimer: Curated by HT Syndication.
