MUMBAI, India, Feb. 27 -- Intellectual Property India has published a patent application (202641017449 A) filed by Vallurupalli Nageswara Rao Vignana Jyothi Institute Of Engineering And Technology, Hyderabad, Telangana, on Feb. 17, for 'embedded raspberry pi based resnet-18 convolutional neural network system for real-time fruit spoilage detection classification.'
Inventor(s) include Sireesha Amaraneni.
The application for the patent was published on Feb. 27, under issue no. 09/2026.
According to the abstract released by the Intellectual Property India: "A real-time fruit spoilage detection system is disclosed wherein an image acquisition unit (101) captures fruit images using a camera module and supplies them to an embedded edge computing unit (102) comprising a Raspberry Pi 4. A preprocessing module (103) resizes the images to a fixed resolution and normalizes pixel values for inference compatibility. A trained convolutional neural network inference module (104) comprising a ResNet-18 architecture executes locally on the embedded edge computing unit (102) to classify fruits into a freshness status comprising fresh or spoiled based on learned spoilage indicators such as discoloration, texture degradation, and surface decay. The classification result is provided through an output interface (105). A prediction validation and re-capture control module (106) computes a confidence score and triggers re-capture and aggregation when confidence falls below a threshold, thereby improving robustness under real-world imaging variations. The system enables accurate low-latency spoilage detection without cloud computation, suitable for household, retail, storage, and transportation applications."
Disclaimer: Curated by HT Syndication.
