LEGENDFTW/image-filtering-explorer

1---2title: Image Filtering and Denoising Explorer3emoji: 🔍4colorFrom: blue5colorTo: indigo6sdk: streamlit7sdk_version: 1.57.08app_file: app.py9pinned: false10---11 12# 🔍 Image Filtering & Denoising Explorer13 14An interactive Streamlit app for exploring spatial image filters and their denoising behaviour.15Built as a course assignment for an image analysis module.16 17## What it does18 19- Upload your own image **or** choose from 3 built-in noisy samples20- Select one of **5 classic filters**: Gaussian Blur, Median, Bilateral, Box (Mean), Non-local Means21- Tune filter parameters with live sliders22- View **side-by-side input/output** and a **difference image**23- Get **quantitative metrics**: PSNR, SSIM, noise reduction estimate24- Explore **histogram comparison** and a **local noise heatmap**25- Add **synthetic noise** (Gaussian / Salt & Pepper / Speckle) for controlled experiments26- Read concise **theory explanations** for each filter27 28## Run locally29 30\`\`\`bash31git clone <your-repo-url>32cd image-filtering-app33pip install -r requirements.txt34streamlit run app.py35\`\`\`36 37## Hugging Face Space38 39🔗 Live demo: (add your URL here after deployment)40 41## Known limitations42 43- Non-local Means can be slow (5–15s) on images larger than ~512×512px44- Metrics compare input vs output — without a true clean reference, PSNR/SSIM measure *change*, not absolute quality45- Very large uploads (>800px) are auto-resized