Microstructure Analysis ML Model for FYP.pptx

Microstructure Analysis ML Model for FYP.pptx

1. Microstructure Image Analysis and Classification Presented By: Vivek Dutonde (BT22MME028) & Avinash Kumar Ray (BT22MME096) Objectives Automate microstructure analysis to replace manual labor. Enhance accuracy by reducing human bias. Develop a web application for easy access and use. Experimental Methodology Model Development :- 1 CNN Architecture Convolutional neural network (CNN) architecture with convolutional, pooling, and fully connected layers, using activations and ending in classification. 2 Dataset Sourced from public repositories and research publications, preprocessed by resizing, normalization, and augmentation. 3 Training 70% Training, 20% Validation, 10% Testing 4 Web application Upload - Predict - Result

2. Results & Discussion High Accuracy Achieved over % accuracy. Demonstrates strong classification performance. Fast Processing Reduced analysis time significantly. Near real-time predictions are now possible. Web Integration Seamless web application integration. Accessible to a broader audience via web.

3. Thesis Drafting Progress Completed Sections Introduction: Challenges of manual microstructure analysis Methodology: CNN architecture, Dataset Preparation, Training Workflow Web Application: Design and functionality Next Steps Results & Analysis: Model performance metrics, case studies Discussion: Compare with existing literature Conclusion: Impact and future scope Formatting: Align with university guidelines