Agricultural Landscape Understanding At Country-Scale

TL;DR

This paper presents a novel approach to digitize and analyze agricultural landscapes at a country scale in India using high-resolution imagery and advanced segmentation models, enabling detailed mapping of fields and features.

Contribution

It introduces the first national-scale multi-class panoptic segmentation dataset for Indian agricultural landscapes, advancing automated landscape understanding.

Findings

Identified individual fields across 151.7 million hectares.

Generated detailed maps of water resources and vegetation.

Validated segmentation outputs with external users and case studies.

Abstract

Agricultural landscapes are quite complex, especially in the Global South where fields are smaller, and agricultural practices are more varied. In this paper we report on our progress in digitizing the agricultural landscape (natural and man-made) in our study region of India. We use high resolution imagery and a UNet style segmentation model to generate the first of its kind national-scale multi-class panoptic segmentation output. Through this work we have been able to identify individual fields across 151.7M hectares, and delineating key features such as water resources and vegetation. We share how this output was validated by our team and externally by downstream users, including some sample use cases that can lead to targeted data driven decision making. We believe this dataset will contribute towards digitizing agriculture by generating the foundational baselayer.