MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202641085381 A) filed by Sri Eshwar College Of Engineering on July 11, 2026, for Road Surface Classification Through Inertial Sensors And Deep Learning.
Inventors include Mr. S. Aravind; Mrs. K. Gowthami; Ms. J. Yashwandra; and Dr. Anandakumar Haldorai.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: Road recognition in intelligent driving requires fast real-time accurate high performance, which conventional image recognition technology is currently unable to deliver. Deep learning models are a promising new approach for achieving this performance Classifying Road conditions accurately is crucial for many applications like autonomous driving, road safety, and maintenance. We present a deep learning model to classify the road surface into asphalt, cobalt, concrete, or cobblestone. The results of the project show that a Neural Network model performs the best with an accuracy of 98% in classifying the road surface type. A real-time visualization is added which shows the predicted results on the map. This helps the drivers to improve situational awareness and gives them feedback on conditions. The results help to prevent accidents and can be used for further applications to provide a safe driving experience. It would also help to improve the predictions of autonomous driving vehicles. The road surface detection algorithm trained using the CycleGAN-augmented dataset had a better IoU than the method using imbalanced basic datasets. This result shows that CycleGAN-generated images can be used as datasets for road surface detection to improve the performance of DNN, and this method can help make the data acquisition process easy.
Disclaimer: Curated by HT Syndication.
