Intellectual Property India Publishes Patent Application for Distributed Solar-Powered Iot And Vision System For Machine Learning-Based Pesticide Necessity Prediction

MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202611087147 A) filed by Uttaranchal University on July 16, 2026, for Distributed Solar-Powered Iot And Vision System For Machine Learning-Based Pesticide Necessity Prediction.

Inventors include Arpita Singh; Rahul Mahala; Yashika; Yashita; Neelam Pharswan; and Aisha Ahmed.

The application for the patent was published on September 11, 2026, under issue no. 37/2026.

Abstract: The present invention provides a distributed solar-powered IoT and vision system for machine learning-based pesticide necessity prediction, comprising an allocated network of smart sensory data collector devices (20) deployed up to an arbitrary integer number across a farmland, an allocated network of smart vision data collector devices (40) deployed up to an arbitrary integer number across the farmland, a central cloud infrastructure server (60), and a machine learning execution platform (90) running a pre-trained multimodal machine learning model. Each sensory data collector device (20) includes a localized computing unit (21) operably coupled to an acoustic sensor (29), a temperature and humidity sensor (30), a soil temperature sensor (31), a GPS module (32), a soil moisture sensor (33), a pH sensor (34), a wind speed and direction sensor (35), and a rain sensor (36). The localized computing unit (21) establishes geographic and temporal correlation via a real-time clock (26) and the GPS module (32), archives data locally on an SD card (24), and transmits structured data packets via a Wi-Fi modem (25) to a data set formation module in the cloud server (60). Grid-independent power is sustained via dedicated solar panels (27), rechargeable batteries (28), and a power supply unit (23) managed by an integrated power management circuit. The vision data collector devices (40) are established on a Raspberry Pi computing architecture (41) coupled to an RGB camera (45), an onboard Wi-Fi modem (50), a local display (49), and a data storage drive (48), powered by independent solar panel (44) and battery (43) assemblies. The system automatically synchronizes the environmental parameters and preprocessed crop imagery, inputs a unified feature set into the pre-trained multimodal model, and outputs a binary decision recommendation of "Pesticide Required" or "Pesticide Not Required" to support sustainable farming.

Disclaimer: Curated by HT Syndication.