Python • Image Analysis • Research Tool

ExplorationQuantTool

A lightweight quantitative image analysis tool that scores images by measuring non-black pixel density across a configurable grid.

View README GitHub Repo

What it does

  • Splits each image into square cells based on a chosen pixel cell size.
  • Computes non-black pixel percentage in each grid cell.
  • Applies a threshold to classify each cell as active/inactive.
  • Normalises the final score to a 0–100 scale for easy comparison.
  • Exports batch results to CSV for downstream analysis.

Quick start

Install

pip install pillow numpy

Run GUI

python gui.py
# or
python3 gui.py

Run CLI

python run.py --cell-size 50 --threshold 50
# or
python3 run.py --cell-size 50 --threshold 50

Workflow

  1. Select a folder of images.
  2. Choose valid cell size and threshold.
  3. Optionally provide mask.png (white excludes, black keeps).
  4. Process images and export a results_<cell>px_<threshold>.csv file.

Example output

If 30 out of 100 cells pass threshold, the normalised score is 30.0.

CSV columns: Image Name, Normalised Value

Sample heatmaps

Flat heatmap sample
Flat heatmap sample
Triangle heatmap sample
Triangle heatmap sample

Use cases

  • Educational demos of grid-based image quantification.
  • Exploratory visual data scoring for research workflows.
  • Comparing pattern intensity across image sets.