Key Skills

Software

Other

Top Subject Matter

No subject matter listed

Top Data Types

Geospatial Tiled Imagery

Image

Top Task Types

Bounding Box

Classification

Land Cover Classification

Mapping

Polygon

Freelancer Overview

I have been working extensively on data annotation and labeling for agricultural machine learning projects, mainly focused on plant disease detection and satellite-based crop monitoring. My work involves preparing high-quality datasets from field images, drone captures, and satellite imagery (such as Sentinel-2, Landsat, and NISAR) to train and evaluate deep learning models for crop analysis. In my plant disease projects, I annotated thousands of images of wheat, corn, and grapevine, identifying symptoms like Fusarium Head Blight (FHB), leaf blight, and nutrient stress. I use tools such as Labeling, CVAT, and my own Python/OpenCV scripts to ensure precise, consistent labeling and to maintain metadata for model training. On the satellite side, I focus on integrating vegetation indices (NDVI, NDWI, NDRE, GNDVI) with field-level data such as soil electrical conductivity (EC), moisture, and yield measurements. I annotate and preprocess these layers to support tasks like crop classification, drought and stress mapping, and variable rate fertilizer management. I also apply quality control and semi-automated labeling techniques to improve efficiency and reduce bias in datasets. These annotated datasets form a key part of my broader research on precision agriculture, combining ground truth with remote sensing and AI to improve crop health monitoring and sustainable resource use.

Labeling Experience

Education

Work History