MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541122798 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on Dec. 5, 2025, for 'ensemble deep learning system for automated pothole detection.'
Inventor(s) include Dr. E. S. Madhan; Mr. Divyansh Bajpai; and Mr. Bhavishya Rajput.
The application for the patent was published on Jan. 2, under issue no. 01/2026.
According to the abstract released by the Intellectual Property India: "The present disclosure provides a system for automated pothole detection that includes a data acquisition module configured to receive road surface images from at least one input source, an ensemble neural network architecture having at least two different pre-trained convolutional neural network architectures with frozen layers, a feature fusion module configured to combine feature representations extracted from the at least two different pre-trained convolutional neural network architectures, and a classification module configured to process the combined feature representations to generate a binary classification result indicating presence or absence of potholes. The ensemble neural network architecture combines InceptionV3 and VGG16 architectures to leverage complementary feature extraction capabilities. The feature fusion module flattens and concatenates output features into a merged feature vector. The classification module includes dense layers with activation functions, batch normalization, dropout layers for regularization, and a final activation layer for binary classification."
Disclaimer: Curated by HT Syndication.
