MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641048412 A) filed by Srinivasa Ramanujan Institute Of Technology; N. Zakeeya; K. Archana; M. Kavya Sree; and B. Krishna Teja, Anantapur, Andhra Pradesh, on April 16, for 'gan-based image restoration for imaging logging applications.'
Inventor(s) include Srinivasa Ramanujan Institute Technology; N. Zakeeya; K. Archana; M. Kavya Sree; and B. Krishna Teja.
The application for the patent was published on May 1, under issue no. 18/2026.
According to the abstract released by the Intellectual Property India: "The present invention discloses an intelligent deep learning framework for restoring degraded imaging logging images using an Improved Generative Adversarial Network (GAN) architecture. With the increasing use of imaging logging systems in petroleum, geophysical, medical, and industrial inspection applications, image degradation caused by noise, blur, discontinuities, and missing regions poses significant challenges for accurate interpretation. The proposed invention integrates a fully convolutional generator network with depth- separable residual blocks, Inception modules, multi-scale feature extraction, and channel attention mechanisms to enhance both local texture reconstruction and global structural consistency. A dual-discriminator mechanism using global and local discriminators is employed to ensure semantic realism and detailed restoration. Performance evaluation using Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) demonstrates superior restoration quality compared to traditional interpolation and patch-based methods. The system is applicable in industrial imaging logging, medical imaging, remote sensing, historical image restoration, and other computer vision applications requiring accurate and high-quality image reconstruction."
Disclaimer: Curated by HT Syndication.
