Chandigarh University researchers use Artificial Intelligence to develop Model for predicting Accurate Crop Yield; Innovation to benefit Indian Farmers
Researchers have leveraged Local weather Knowledge and Satellite tv for pc Know-how for growing the predictive mannequin
CHANDIGARH, India, June 27, 2026 /PRNewswire/ — In a major development for good agriculture and precision farming, researchers at Chandigarh College have developed an Synthetic Intelligence (AI)-powered Transformer Mannequin that’s able to precisely predicting crop yields utilizing satellite tv for pc imagery, local weather knowledge and historic agricultural information.
The innovation could be instrumental in empowering farmers, policymakers and agricultural businesses to make knowledgeable selections whereas strengthening meals safety and advancing resilient farm administration.
The analysis, led by Kusum Lata, Assistant Professor Division of Laptop Science Engineering, Chandigarh College, Dr Navneet Kaur Professor Division of CSE and Dr Simrandeep Singh Professor from College Centre of Analysis and Improvement at Chandigarh College that focuses on bettering crop yield forecasting in Punjab’s agricultural heartland. The examine, just lately offered on the 2026 Worldwide Convention on Sign Processing and Electronics Design (ICSPED) at Chandigarh School of Engineering and Know-how, Chandigarh that introduces a light-weight transformer-based system that leverages multi-source knowledge to estimate crop manufacturing earlier than harvest with higher accuracy and decrease computational prices.
Notably, the correct crop yield prediction has develop into more and more necessary as farmers face rising challenges from local weather variability, altering climate patterns and rising meals demand. Conventional subject surveys are sometimes time-consuming, labour-intensive and restricted in scale. The Chandigarh College researchers sought to beat these limitations by integrating superior AI methods with real-time Earth statement knowledge.
Kusum Lata, Assistant Professor Division of Laptop Science Engineering at CU stated, “The transformer mannequin makes use of knowledge from Sentinel-1 and Sentinel-2 satellites that are superior Earth statement satellites operated by the European House Company (ESA) to constantly monitor agricultural fields and supply info on crop development, vegetation well being, soil moisture and subject situations. The satellite tv for pc observations are mixed with climatic variables comparable to rainfall, temperature and soil moisture, together with historic crop manufacturing information, making a complete image of crop efficiency all through the rising season.”
Kusum added, “Not like typical machine studying fashions, the newly developed light-weight transformer can establish essential crop development phases and be taught complicated temporal patterns that affect ultimate yields. We’ve got designed the mannequin to ship excessive predictive efficiency that require fewer computational assets, making it appropriate for sensible deployment in large-scale agricultural monitoring methods.”
“The mannequin was evaluated on 4 main crops cultivated in Ludhiana district, specifically paddy, maize, moong and sugarcane, utilizing knowledge collected between 2019 and 2023. Experimental outcomes demonstrated that the transformer mannequin outperformed extensively used approaches comparable to Random Forest and Lengthy Brief-Time period Reminiscence (LSTM) fashions, indicating stronger settlement between predicted and precise yields. The framework additionally recorded decrease prediction errors and improved computational effectivity.
Kusum has labored as a Junior Analysis Fellow within the Agriculture and Crop Monitoring Division at Punjab Distant Sensing Centre (PRSC), PAU Ludhiana, contributed to geospatial analysis tasks, together with crop residue burning evaluation utilizing distant sensing and geospatial mapping methods.
The examine additional revealed that the light-weight structure requires practically 40 p.c fewer parameters than typical transformer fashions whereas delivering quicker and correct predictions. This stage of effectivity makes this computerized framework appropriate for close to real-time agricultural purposes, together with regional crop monitoring, manufacturing forecasting and early warning methods.
In keeping with the researchers, the power to precisely forecast crop yields earlier than harvest can have vital implications for farmers and governments alike. Dependable forecasts can assist agricultural planning, optimize useful resource allocation, strengthen crop insurance coverage mechanisms and enhance market administration methods. In a state like Punjab, the place agriculture performs a central function within the economic system, such applied sciences can contribute to extra resilient and sustainable farming methods.
The researchers additionally shared that one of many key strengths of the mannequin lies in its skill to mix a number of sources of knowledge right into a single predictive mannequin. By integrating satellite-derived observations with climatic and historic datasets, the system captures the complicated interactions that affect crop productiveness and gives a extra strong understanding of agricultural outcomes.
The long run developments can even give attention to enabling close to real-time forecasting via cloud-based platforms, paving the best way for broader adoption of AI-driven resolution assist methods in agriculture, added the Chandigarh College researchers.
About Chandigarh College
Chandigarh College is a NAAC A+ Grade College and QS World Ranked College. This autonomous academic establishment is accredited by UGC and is positioned close to Chandigarh within the state of Punjab. It’s the youngest college in India and the one personal college in Punjab to be honoured with A+ Grade by NAAC (Nationwide Evaluation and Accreditation Council). CU presents greater than 109 UG and PG packages within the subject of engineering, administration, pharmacy, regulation, structure, journalism, animation, resort administration, commerce, and others. It has been awarded as The College with Finest Placements by WCRC.
Web site tackle: https://www.cuchd.in/
Picture: https://web3wire.org/wp-content/uploads/2026/06/Chandigarh_University_AI_Crop_Yield.jpg
View authentic content material to obtain multimedia:https://www.prnewswire.com/in/news-releases/chandigarh-university-researchers-use-artificial-intelligence-to-develop-model-for-predicting-accurate-crop-yield-innovation-to-benefit-indian-farmers-302812564.html
