Kernels on a real image
The same 3×3 convolution as the cell-by-cell demo, but running on an actual image at full speed. Presets cover blur, sharpen, emboss and the Sobel pair; the nine weights are editable, and you can load your own photo.
Try this
Take the sharpen preset and push the centre weight up one step at a time until the image falls apart. That edge is what "too much gain" looks like.
Kernel
Where this lab comes from
How 2D Convolution Works on Images: Kernels, Filters, and the Math, Interactively2D convolution is the operation behind blur, sharpening, edge detection, and every CNN. Slide a kernel, multiply, sum: see it happen step by step in an interactive demo, apply real kernels to a live image, and get the math plus runnable OpenCV in Python and C++.
More in the image itself
- One colour per photosite — Throw away two thirds of the colour, then guess it back.
- Hitting the colour a meter measured — Drag the gains and watch ΔE against 22 metered patches.
- Sixteen by twelve real pixels — Hover a pixel, read its index and its three numbers.