MUMBAI, India, Feb. 6 -- Intellectual Property India has published a patent application (202511134324 A) filed by Maharishi Markandeshwar, Ambala, Haryana, on Dec. 31, 2025, for 'hybridization image classification segmentation through dcnn.'
Inventor(s) include Ritu Aggarwal; and Eshaan Aggarwal.
The application for the patent was published on Feb. 6, under issue no. 06/2026.
According to the abstract released by the Intellectual Property India: "The invention discloses a hybridization image classification and segmentation system utilizing deep convolutional neural networks (DCNNs). The system integrates U-Net and DenseNet architectures with transfer learning, feature fusion, and bilinear interpolation upsampling to achieve high-resolution semantic segmentation. Cascaded dense blocks are employed to generate denser feature maps, while skip connections improve gradient propagation and prevent vanishing gradients. The hybridization process fuses outputs from U-Net and DenseNet, producing a unified segmentation map that balances accuracy and computational efficiency. The system computes performance metrics including pixel accuracy, mean pixel accuracy, mean intersection-over-union, precision, recall, and F1-score to validate its effectiveness. Comparative analysis demonstrates superior results compared to conventional models such as SegNet, PSPNet, and ResNet. The invention is scalable across domains including agriculture, defense, environmental monitoring, and medical imaging, providing a robust, efficient, and accurate solution for complex image classification and segmentation tasks."
Disclaimer: Curated by HT Syndication.
