Perception and Grounding

CLAHE_Filter

Enhances local image contrast with clipped adaptive histogram equalization for low-light, shadowed, or low-texture scenes.

Tool Introduction

Core parameters, trigger timing, and visual before/after demo references.

Short Explanation

Use CLAHE as a lightweight visual robustness step when shadows, low dynamic range, or weak texture make perception unreliable.

InputRGB / YUV / LAB image

OutputContrast-enhanced image

Trigger TimingTriggered on demand after the required input files and configuration are prepared.

RuntimeLocal safety preprocessing

BeforeRGB / YUV / LAB image

Prepare the scene, image, video, sensor stream, prompt, or configuration expected by the original project.

AfterContrast-enhanced image

Read the produced visualization, prediction, map, trajectory, mask, grasp pose, or other documented artifact.

Preset Example

A quick-run style example for the documentation page.

Inputtools/clahe-filter/examples/input.png

Promptcolor_space: lab; clip_limit: 2.0; tile_grid_size: 8x8

ExpectedA locally contrast-enhanced image with clearer details and balanced brightness.

Parameters And Output

Readable controls and the meaning of each returned artifact.

Parameter Explanation

input_imagefile

Original RGB, YUV, or LAB image.

color_spaceselectlab

Color space used before applying CLAHE.

clip_limitslider2.0

Upper limit for local histogram clipping.

tile_grid_sizetext8x8

Local grid size used for adaptive equalization.

Output Explanation

enhanced_image

Image after local contrast enhancement.

brightness_distribution

More balanced local intensity distribution.

detail_visibility

Improved visibility of cables, texture, edges, or shadowed objects.

How To Use

Official resources, deployment steps, academic context, citation, and source-reported benchmark numbers.

Resources

OpenCV GitHubhttps://github.com/opencv/opencv Code Downloadhttps://github.com/opencv/opencv/archive/refs/heads/main.zip Original Paper DOIhttps://doi.org/10.1016/S0734-189X(87)80186-X OpenCV CLAHE Docshttps://docs.opencv.org/4.x/d6/db6/classcv_1_1CLAHE.html

Deployment Notes

Use OpenCV or an equivalent image-processing backend with CLAHE support.
Prepare low-light or low-contrast inputs under tools/clahe-filter/examples/.
Run the filter with repository-relative image paths and tuned clipping parameters.
Save enhanced images under tools/clahe-filter/runs/ for downstream perception.

Relative Path Example

python tools/clahe-filter/run.py --input tools/clahe-filter/examples/input.png --color-space lab --clip-limit 2.0 --output tools/clahe-filter/runs/enhanced.png

Expected Result Shape

{
  "tool": "clahe-filter",
  "status": "ok",
  "results": [
    {
      "label": "Local contrast enhancement",
      "score": 0.87,
      "output": "Contrast-enhanced image"
    }
  ],
  "timing": {
    "runtime": "Interactive according to the submitted spreadsheet; exact runtime depends on resolution, tile grid size, color space conversion, and CPU/GPU backend.",
    "device": "documented in source benchmark when available"
  },
  "artifacts": {
    "visualization": "tools/clahe-filter/runs/visualization.png",
    "raw_predictions": "tools/clahe-filter/runs/predictions.json"
  }
}

Paper figure

Academic Info

Paper identity and contribution summary.

TitleAdaptive Histogram Equalization and Its Variations

AuthorsAdd authors

VenueComputer Vision, Graphics, and Image Processing, 1987

ContributionApplies local histogram equalization with clipping to improve contrast while limiting noise amplification.

Citation

@misc{clahefilter1987,
  title={Adaptive Histogram Equalization and Its Variations},
  author={Author},
  year={1987},
  note={Computer Vision, Graphics, and Image Processing, 1987},
  url={https://doi.org/10.1016/S0734-189X(87)80186-X}
}

Benchmark

Only compact, source-reported numbers are shown here.

Dataset	Metric	Value	Runtime	Source
CLAHE is an image-processing primitive rather than a learned model; OpenCV and the original paper do not provide a single canonical cross-dataset benchmark number for this wrapper.	Core result	No universal official numeric benchmark is copied here because contrast improvement is image-dependent and usually evaluated as part of a downstream perception pipeline.	Interactive according to the submitted spreadsheet; exact runtime depends on resolution, tile grid size, color space conversion, and CPU/GPU backend.	Computer Vision, Graphics, and Image Processing, 1987

Artifacts

Tool-sheet description and preprocessing parameters from the submitted spreadsheet.

Demo Images

Visual references from the original tool. Click any image to inspect the original size.