Edge detection
Find the edges and outlines in an image with Sobel, outline or Laplacian filters, or emboss it, free and in your browser with no upload. Choose the ink and paper colors, or keep the colors the filter produces.
Convolution kernels that respond to change rather than to brightness. Sobel traces outlines; emboss lights them from one side.
In detail
Edge detection finds the outlines in an image by running a 3×3 convolution kernel across it, so what survives is where brightness changes rather than how bright the picture is. Sobel runs two kernels, one measuring change across the image and one down it, and combines them into a single strength per pixel. Outline, Laplacian and emboss are each a single kernel whose nine weights are shown beside the preview.
A threshold drops weak responses, which is how grain and sensor noise go first. You choose the two colors the result is drawn between, or keep the colors the kernels produced. The export runs at full resolution and shows finer detail than the smaller preview.
Edge detection at a glance
- Input
- JPG, PNG, WebP, AVIF, HEIC, GIF (first frame), BMP, TIFF, SVG and other image formats
- Output
- The original format, PNG, WebP or JPEG
- What it does
- Sobel, outline, Laplacian and emboss
- An adjustable threshold
- Monochrome or kept color
- Kernel weights shown
- Batch processing
- Who it is for
- Designers, students of image processing, and anyone turning a photo into line art.
- Limits
- The preview is drawn smaller, so the export shows finer lines.
- Privacy
- Runs entirely in your browser: no file is uploaded, and nothing is kept after you close the tab.
How to find the edges in an image
- Add your images Drop one file or a batch onto the page, or click browse. They are decoded in the tab and nothing is sent anywhere.
- Choose a mode Sobel is the default. Emboss, outline and Laplacian each run one kernel, and the nine weights currently in use are drawn out under the settings — Sobel shows both of its.
- Set the threshold Anything responding more weakly than this is dropped, which is how grain and sensor noise go first. It starts at 24 out of 255; lower it if the result comes out blank.
- Pick the colors Monochrome draws the response between a paper color, where nothing happened, and an ink color, where the response is strongest. Keep color instead keeps what the kernels produced channel by channel — and with Sobel, which answers in one channel, dims the original pixel by the edge strength.
- Apply and download Keep each file in the format it arrived in or switch to PNG, WebP or JPEG, then download one file or the batch as a ZIP.
Edge detection: common questions
Why does the downloaded file show finer detail than the preview?
The preview composes from a source scaled to 1100 pixels on its long edge, and edge detection depends on scale: a 3×3 kernel reads a pixel's immediate neighbors, which cover far more of the picture in a shrunken copy. The export runs the same kernel at full resolution, so it picks up detail the preview cannot show.
What is the difference between Sobel and the other three modes?
Sobel measures how fast brightness changes and in which direction, using two kernels at right angles whose results are combined into one magnitude per pixel. Emboss, outline and Laplacian are each a single 3×3 kernel applied to the color channels directly — outline weighs the center pixel against all eight neighbors, Laplacian uses only the four orthogonal ones for thinner lines, and emboss sits at mid-gray where nothing happened so one side of every edge is raised and the other sunk.
What happens at the edge of the image itself?
The edge pixel is repeated. A kernel that needs a neighbor off the side of the picture reads the nearest real pixel instead, and nothing wraps around to the opposite edge — wrapping would draw a border that is not in the photograph, which is the worst way for an outline finder to be wrong.