Drones can spot gaps in sugarcane rows so farmers know where to
replant, but most software assumes large fields with straight rows.
Small Indian farms are a patchwork of plots, with rows running in
different directions and curving around paths.
Working with DronaMaps and Plaksha University, Eknoor joined field
surveys in Muzaffarnagar, Uttar Pradesh, and worked on the analysis
for a method that follows rows using ordinary drone photographs
and no training data.
Research result:
Across 11 fields, the method found 79% of the rows, compared with
9% for a standard method. The additional rows it detected were real.
Far more of each field can now be checked for gaps.
Eknoor is a co-author of the paper, presented at the International
Conference on Agriculture-Centric Computation (ICA 2026), Ropar,
and now under peer review.