Harishankar K.

Key Skills

Software

Other

CVAT

Top Subject Matter

Medical imaging

Agriculture

Computer vision

Top Data Types

Image

Audio

Text

Top Task Types

Diagnosis

Classification

Object Detection

Freelancer Overview

Detection of Viral Pneumonia in Chest X-Ray project (DenseNet/CheXNet). Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other and TensorFlow. Education includes Bachelor of Engineering, University College of Engineering, Kancheepuram (2024) and Bachelor of Engineering, University College of Engineering, Panruti (2024). AI-training focus includes data types such as Medical, DICOM, and Image and labeling workflows including Diagnosis, Classification, and Object Detection.

Labeling Experience

YOLO-Based Traffic Sign Detection System project

OtherImageObject Detection

Developed a real-time object detection system using YOLO for traffic sign recognition, relying on an annotated dataset to train the detector. Deployed the model via a Flask API to support inference use cases on lightweight edge computing devices. Focused on delivering practical detection performance for ongoing traffic sign detection. • Trained and/or tuned YOLO-based traffic sign detector • Used labeled image data for object detection • Integrated inference through a Flask API deployment • Optimized the model for lightweight edge inference

2023 - Present

Plant Disease Detection Using Convolutional Neural Networks (CNN) project

ImageClassification

Built a CNN-based image classification model to distinguish healthy versus diseased plant leaves using labeled image data. Applied image preprocessing and data augmentation techniques to improve classification performance on the training dataset. Trained and refined the model to support accurate plant disease recognition. • Classified plant leaf images into healthy/diseased categories • Performed data preprocessing for model readiness • Applied data augmentation to increase dataset diversity • Optimized CNN-based training for improved accuracy

2025 - 2025

Detection of Viral Pneumonia in Chest X-Ray project (DenseNet/CheXNet)

OtherDiagnosis

Developed a CheXNet-based DenseNet transfer learning model for chest X-ray classification with a focus on viral pneumonia detection. Built and supported a web interface to enable model-assisted screening based on the labeled medical imaging task. Applied deep learning workflows to produce clinically relevant diagnostic outputs. • Used DenseNet/CheXNet architecture for transfer learning • Prepared data and model inputs for X-ray classification • Integrated model outputs into a user-facing detection interface • Worked on evaluation and usability for pneumonia detection

2025 - 2025

Education

Work History