MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202621042886 A) filed by Sage University, Indore, Madhya Pradesh, on April 3, for 'system and method for agricultural object detection.'
Inventor(s) include Dr. Prashant Jain; Poonam Rijhwani; Prof. Vijay Malviya; and Lokesh Vishwakarma.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "The present disclosure relates to agricultural object detection. The system (102) includes an imaging sensor (104) for capturing image data (112) of agricultural scenes, an edge computing platform (110) with a processing circuit (106) and local storage (114), enabling on-device data processing. The processing circuit (106) receives image data (112) from the imaging sensor (104) and executes an artificial neural network model (108) stored in the local storage (114). The artificial neural network model (108) applies differentiated activation intensities to object categories based on per-category training loss values, where categories with higher training loss values receive higher activation intensities. The artificial neural network model (108) undergoes compression through combined layer pruning and channel pruning, removing convolutional layers and filter channels with scaling factor magnitudes below threshold value. The processing circuit (106) generates detection output data (116) identifying agricultural objects, storing results in the local storage (114) for agricultural automation applications."
Disclaimer: Curated by HT Syndication.
