📸🔥Thermal Camera Diagnosis App Prototype

This repository contains all the source code, scripts, and notebooks used in my bachelor thesis project, "Enhancing Battery Recycling with Thermal Imaging Anomaly Detection." The project explores the integration of thermal imaging and anomaly detection methods to contribute to optimizing battery discharging processes in recycling facilities.

📂Repository Structure

The repository is organized into the following directories:

1. application 🖥️

• Description: Contains all the implementation scripts and deployment files for the Thermal Camera Diagnosis App.
• Highlights:
  • build-amd64.sh: Script for building and deploying the application on AMD64 architecture platforms.
  • build-arm64.sh: Script for building and deploying the application on ARM64 architecture platforms.
  • Integration with Bosch Rexroth's CtrlX CORE system for near-real-time anomaly detection.

2. model-training-and-evaluation 📊📈

• Description: Includes Jupyter notebooks and scripts for training, evaluating, and validating machine learning models for anomaly detection.
• Contents:
  • Datasets used for laboratory and industrial experiments.
  • Implementation of anomaly detection algorithms (e.g., PaDiM, PatchCore, iForest).

3. referenced 📂

• Description: Contains all the referenced scripts and supplementary files used throughout the project.
• Contents:
  • Supporting scripts for data collection.
  • Scripts for preprocessing thermal images and generating synthetic anomalies.
  • Any other files (such as masks) referenced in the main experiments.

🌟Features

• Anomaly Detection Models: Integration of state-of-the-art methods like PaDiM, PatchCore, and Isolation Forest.
• Data Preprocessing Tools: Automated scripts to process thermal images for model training.

🚀Setup & Usage

1. Clone the Repository:

   git clone https://github.com/PetrovEvgeniy/thermal-camera-diagnosis-app.git