AI Vision Inspection & Video Analytics

Edge Vision AI for inspection, tracking, event detection and real-time video analytics.

Overview

Research and development of a reusable camera-based Edge Vision AI platform that transforms live and recorded visual streams into structured intelligence. The system explores industrial PASS/FAIL inspection, image and video processing, classification, object detection, segmentation, tracking, counting, event analytics, visual evidence capture, and operational monitoring.

Research Focus

Industrial visual inspection and PASS/FAIL decision systems
Real-time object detection, tracking and counting
Image classification and segmentation
Defect and visual anomaly detection
ROI, zone and line-crossing analytics
Safety and event-based video analytics
Edge AI inference and model optimization
Model accuracy and real-time system performance evaluation

Capabilities

Live camera and recorded-video processing
Classical computer vision and image preprocessing
Dataset creation, annotation and model evaluation
Classification, detection and segmentation
Multi-object tracking and direction-aware counting
Industrial defect inspection with PASS/FAIL results
Zone, dwell-time and event detection
Snapshot and event-video evidence recording
Edge inference with FPS and latency measurement
Camera health and event monitoring
Searchable event and inspection history
Vision Operations Center dashboard