DataLens Thermal Studio — Streamlit MVP

A Streamlit-based thermal imaging studio that replicates key workflows from FLIR Thermal Studio, built for rapid batch analysis, visualization, and reporting of thermal data.

Developed by DataLens.Tools — empowering researchers and engineers with open, AI-powered data analysis tools.

Overview

This MVP provides a desktop-like web interface for analyzing thermal data (images or temperature matrices).
It supports unified color normalization, quick statistics, video generation, and PDF reporting — all within Streamlit.

Core Features

• Batch Import
  Upload multiple thermal images (PNG/JPG/TIFF) or temperature matrices (CSV/NPY) at once.

• Global Normalization
  Auto or manual scale normalization (°C range) across all frames.

• Quick Analytics
  Calculate min, max, mean, and 95th percentile temperatures per frame.

• Video Creator
  Export animated MP4 or GIF sequences from batches.

• PDF Report Generator
  Build clean, professional thermal reports with title, subtitle, and footer templates.

• Unified Colormap Control
  Supports Inferno, Plasma, Magma, Turbo, Viridis, Jet, and more.

Installation

1️ Clone the Repository

git clone https://github.com/DataLensTools/datalens-thermal-studio.git
cd datalens-thermal-studio

2️ Create a Virtual Environment

python -m venv venv
source venv/bin/activate  # on macOS/Linux
venv\Scripts\activate     # on Windows

3️ Install Dependencies

pip install -r requirements.txt

4️ Run the App

streamlit run app.py

Requirements

streamlit
numpy
pillow
matplotlib
opencv-python
pandas
fpdf2

Note: MP4 export requires an FFmpeg-enabled OpenCV build on your system.
If unavailable, you can export GIFs instead.

File Structure

📦 datalens-thermal-studio/
 ┣ 📜 app.py                ← Main Streamlit app
 ┣ 📜 requirements.txt      ← Python dependencies
 ┣ 📁 sample_data/          ← Example thermal images or CSVs (optional)
 ┣ 📁 docs/                 ← Screenshots or documentation
 ┗ 📜 README.md             ← You’re here

Example Workflow

1. Import & Preview
   Upload your thermal images (PNG/JPG/TIFF) or numerical matrices (CSV/NPY).
   The app converts images to temperature estimates using a linear grayscale → °C mapping.

2. Batch Processing
   Compute per-frame statistics and export normalized PNGs or CSV summaries.

3. Video Creator
   Combine frames into an MP4 or GIF for temporal visualization.

4. PDF Report
   Generate and download a full analysis report (thermal_report.pdf).

Architecture

• Frontend/UI: Streamlit
• Rendering: Matplotlib Colormaps
• Image I/O: Pillow (PIL)
• Video Encoding: OpenCV
• Reporting: FPDF2
• Core Logic: Python Dataclasses for frame management

Example Code Snippet

from dataclasses import dataclass
import numpy as np

@dataclass
class ThermalFrame:
    name: str
    temp: np.ndarray  # temperature matrix in °C

    @property
    def stats(self):
        t = self.temp
        return {
            "min": float(np.nanmin(t)),
            "max": float(np.nanmax(t)),
            "mean": float(np.nanmean(t)),
            "p95": float(np.nanpercentile(t, 95)),
        }

Author

Taufia Hussain (Maintainer) — LinkedIn · GitHub · info@datalens.tools · datalenstools@gmail.com

Developed by: DataLens.Tools
© 2025 DataLens.Tools — Open source for research and educational use.

License

This project is licensed under the MIT License.
You’re free to use, modify, and distribute it with attribution.

MIT License
Copyright (c) 2025 DataLens.Tools

🌐 Links

• 🌍 Website: https://datalens.tools
• 🧑‍💻 GitHub: DataLensTools
• ✉️ Contact: info@datalens.tools

“Analyze, visualize, and report — all in one thermal studio.”