PACKINSPECT-ANOMALY-DETECTOR

Detects Defects, Ensures Quality, Accelerates Innovation

Built with the tools and technologies:

---

---

Table of Contents

• Overview
• Features
• Tech Stack
• Project Structure
• Getting Started
  • Clone the Repository
  • Install Requirements
  • Run the Application
• Model
• Analytics
• Export Options
• License
• Author
• Screenshots

Overview

PackInspect is a low-code visual inspection tool built using TensorFlow and Streamlit. It allows users to detect defects in bottle images (or similar packaging components) by uploading images or capturing them via webcam.

It's designed to streamline quality assurance in smart manufacturing, reduce manual errors, and provide real-time feedback using deep learning.

Features

• 📤 Upload or Capture Images
  Choose between uploading images or capturing directly using a webcam.

• 🤖 AI-Based Classification
  Classifies images as Good or Defect using a pre-trained Keras model.

• 📈 Confidence Display
  Shows prediction confidence scores for transparency.

• 🧾 Detection History Log
  View recent predictions with timestamps and export results to CSV.

• 📊 Pie Chart Summary
  Visual overview of detection distribution (Good vs. Defect).

• 🧪 Interactive Sidebar
  Includes collapsible guides, model info, and accuracy stats.

Tech Stack

• Frontend/UI: Streamlit, HTML/CSS (Custom Styling)
• Machine Learning: TensorFlow + Keras
• Image Processing: OpenCV, Pillow
• Data Handling: NumPy, Pandas
• Visualization: Streamlit charts

Project Structure

packinspect-anomaly-detector/
├── app.py                  # Main Streamlit application
├── models/
│   └── keras_model.keras   # Pre-trained ML model
├── data/
│   └── labels.txt          # Contains 'Good' and 'Defect'
├── assets/
│   └── overview_dataset.jpg # Sidebar visual banner
├── requirements.txt        # Python dependencies
├── logs/
│   └── defect_log.csv      # Detection history log
└── README.md               # Project documentation (this file)

Getting Started

1. Clone the Repository

git clone https://github.com/devaldaki3/packinspect-anomaly-detector.git
cd packinspect-anomaly-detector

2. Install Requirements

pip install -r requirements.txt

3. Run the Application

streamlit run app.py

Open your browser at http://localhost:8501

Model

The application uses a binary classification model trained using the included train_model.py script and exported in .keras format.

• Input shape: 224x224 RGB
• Output: Sigmoid (binary classification)
• Classes: Good, Defect

Analytics

• Pie chart showing Good vs. Defect count
• Optional: Extend with bar graphs or trend charts using Streamlit/Pandas

Export Options

• ✅ Download detection history as a CSV report

License

This project is licensed under the MIT License. You can use, modify, and distribute it freely with proper attribution.

Author

Developed with ❤️ by @devaldaki3

Feel free to contribute, raise issues, or suggest improvements.

Screenshots

Prediction Result

⬆️ Back to Top