Praveen Pankajakshan VP Data Science & AI
Cropin
India

BiographyPraveen is the VP Data Science & AI at Cropin and Visiting faculty at AMMACHI Labs, Amrita Vishwa Vidyapeetam. Prior to his current appointment, he was with Corteva Agriscience (formerly Dow-DuPont Agriculture division) leading the Data Science and Machine Learning team in Hyderabad. He served as a Data Analytic Scientist with Shell Technology Center Bangalore, and as a Researcher and Technical Manager at Samsung Advanced Institute of Technology (SAIT), Bangalore, India. He was also working as a postdoc and researcher at the Pasteur Institute, Paris. Praveen graduated with his Ph.D. (Summa cum Laude) in Control, Signal, and Image Processing from Ariana, a joint research team of INRIA, CNRS, and UNS INRIA, his Master's in Electrical Engineering from Texas A&M University, College Station, and his undergraduate degree in Electrical Engineering at Indian Institute of Technology (IIT) Roorkee. His research works, in the past two decades, have spanned to the theory and application of statistical signal and image processing, pattern recognition, and machine learning to Energy, Utility, Healthcare, Life Sciences, and Agriculture. He has mentored many students and advisor for Masters and Doctoral students. He is passionate about biodiversity and adopts natural and organic practices on his farm. He is currently a Senior Member of IEEE and is also a certified Heartfulness Meditation trainer.AbstractEvaluating the Effects of Anomalous Dry Spell Periods on Crop Growth and Production for Small Landholding Farmers in IndiaThis study focuses on the effect of dry spell periods on crop growth and production during a particular growing season in India. The Spatio-Temporal analysis of dry and wet spells of rainfall is important for improving water management strategies and reducing socio-economic losses. In general, crops may experience a variety of mild to severe intraseasonal dry spells due to uneven distribution of rainfall or ground water usage, which may subsequently affect the yield unfavorably. Using proprietary machine learning algorithms the crops growing in a season can be identified and their potential yields estimated using a hybrid approach. The machine learning algorithm for crop identification is trained on ground collected data, across India, from farmer plots whos typical land size is about 1 hectare or more. By comparing the current data against the historical normals we can map the stressed regions and identify the anomalous dry spell periods. We then study the impact of this dry spell anomaly on crop at different phenological stages of growth and production. This information on the estimation of dry-spell lengths is particularly useful for a small landholding farmer as it can be used in deciding adaptation strategies like supplementary irrigation and field operations in agriculture.