Hire this AI Trainer
Sign in or create an account to invite AI Trainers to your job.
Roboflow
Image
3D Sensor
Bounding Box
Computer Programming/CodingThrough my work in deep learning-based object detection, I've developed strong hands-on experience with data labeling and AI training data pipelines. For my YOLOv8 projects, one on real-time okra stage identification and another on rice leaf disease classification, I was directly involved in collecting, annotating, and curating image datasets to train and validate the models. I also co-authored a Data-In-Brief publication where we built and published an RGB image dataset specifically designed for okra maturity classification, which required careful labeling standards, quality checks, and ensuring the data was structured for reproducibility and broader research use. What sets me apart is that I don't just label data in isolation, I understand the full loop from raw data collection to model training and deployment. Having trained YOLOv8 models end-to-end, I know firsthand how annotation quality, class balance, and edge cases directly impact model performance. I've also presented this work at IEEE conferences, so I'm comfortable documenting methodologies and maintaining the kind of rigor that high-quality training data demands.
ImageBounding BoxIt was a research paper published in IEEE on maturity classification of Okra. We had three classes namely mature, under mature and over mature. The dataset contained about 900 (approx.) images. We did image acquisition in farmlands, did data preprocessing and annotation, divided the datatset into test, train and valid, trained the model and noted down the observations in the confusion matrix and loss metrics graphs obtained.
Bachelor in Technology, Mechatronics, Robotics and Automation
Junior Robotics Engineer
