Traditional quality control in manufacturing involves manual methods that are labor-based prone to human error and, most often, inefficient. The project answers this problem through an automated visual inspection system where it uses AI for real-time detection of defects. The developed system will blend into the process of manufacturing; therefore, defect identification will be prompt and precise enough to preserve product quality at high levels.
Week 1-2: Requirement Analysis and Planning
Activities: Define system requirements, collect data, and plan the project roadmap.
Deliverables: Requirement document and project plan.
Week 3-4: Data Preparation
Activities: Collect and preprocess product images and annotate defects for model training.
Deliverables: Labeled dataset ready for training.
Week 5-6: AI Model Development
Activities: Develop and train machine learning models using TensorFlow.
Deliverables: Trained AI model and performance evaluation.
Week 7-8: System Integration
Activities: Integrate the computer vision pipeline with the manufacturing process.
Deliverables: Integrated system prototype.
Week 9-10: Testing and Optimization
Activities: Test the system for accuracy, performance, and reliability; refine models.
Deliverables: Test results and optimized system.
Week 11-12: Deployment and Documentation
Activities: Deploy the system on the manufacturing line, train operators, and create final documentation.
Deliverables: Deployed system and user manuals.
