Approaches to Object Detection: Edge-Based and AI-Based
summary
Case study detects pests on blue paper using ImageJ + MorphoJ with size and circularity thresholds; raising minimum area removes noise but uneven lighting and background shifts cause many missed or merged detections. Switching to machine learning (examples cited: YOLO / Faster R-CNN and conventional CNNs) uses a labeled dataset with bounding boxes and polygons (polygons give noticeably better segmentation) and produces higher accuracy across variable lighting at the cost of more training data and compute.
AI-written summary · about
chapters (4)
- 0:00Introduction to object detection
- 2:55Edge-based detection approach
- 7:22Transition to machine learning
- 12:47Comparison and conclusion
related talks
13mEnhancing Object Detection & Avoidance Systems for Robots in Challenging Environments Using AI
36mObject detection to Instance segmentation: Learn to apply several algorithms...
01h 15mTrain Object Detection with Small Datasets
25mComputer Vision in Melanoma Detection
12mCascade models in computer vision to boost accuracy and performance
21mThe eyes of industry 4.0 - Vision intelligence and machine learning
23mBuilding high-quality computer vision models using only a few examples
01h 11mObject detection using OpenCV