Image Edge Detector
Find the outlines in any image. The edge detector measures how quickly brightness changes from pixel to pixel and turns the result into lines – dark on white like a drawing, light on black like neon, in the photo’s own colours, or traced over the original. Smoothing hides texture and noise, the threshold drops weak edges, and strength makes the lines bolder.
- Files stay on your device
- No sign-up
- Free to use
Edges
Uses the Sobel operator, which measures how quickly brightness changes. Smoothing first hides texture and noise so that only real outlines remain.
Output
How to use Image Edge Detector
- Drop one or more images onto the page.
- Choose the line style.
- Adjust strength, threshold and smoothing.
- Download the edge images.
Image Edge Detector features
Sobel operator
Standard gradient-based edge detection.
Four styles
Dark, light, coloured or overlaid lines.
Smoothing
Removes texture before detection.
Threshold
Keeps only strong edges.
Batch
Apply the same settings to up to 50 images.
Private
Processed in your browser; nothing is uploaded.
When to use Image Edge Detector
- Turning photos into line drawings and colouring pages.
- Creating tracing templates for drawing and crafts.
- Visualising structure in technical and scientific images.
- Graphic effects for posters and covers.
Image Edge Detector FAQ
What is edge detection?
A technique that marks places where brightness changes sharply, such as the borders of objects. It is a basic step in computer vision.
What is the Sobel operator?
A pair of small filters that estimate the horizontal and vertical brightness gradient at each pixel. The combined gradient strength becomes the edge value.
Why are there so many small lines?
Texture and noise also create brightness changes. Increase smoothing and the threshold to keep only the main outlines.
How do I make a colouring page?
Choose dark lines on white, smoothing 2–3, threshold 20–30 and strength 60–80. Photos with clear shapes work best.
Which formats can I use?
JPG, PNG, WebP, GIF (first frame), BMP and AVIF in; JPG, PNG or WebP out. PNG and WebP keep transparency; JPG gets a white background.
Are my images uploaded?
No. The images are processed in your browser and never leave your device.
How edges are found
Edges are where an image changes: the border between a building and the sky, the outline of a face, the line between a shadow and a lit surface. Detecting them is one of the oldest problems in image processing and the basis for shape recognition, measurement and many artistic effects.
The Sobel operator estimates the gradient – the direction and speed of brightness change – using two 3×3 filters, one for horizontal and one for vertical changes. Combining both gives the gradient strength at each pixel, which is large on edges and small in uniform areas. The tool normalises the strengths so that the strongest edges reach full intensity regardless of the image’s overall contrast.
Real photos contain texture – grass, fabric, skin pores – and noise, which also produce gradients. Smoothing the brightness first with a small Gaussian blur suppresses these fine details, leaving the larger structures. The threshold then removes the weakest remaining gradients.
The styles turn the edge map into different looks. Dark lines on white resemble a pen drawing; light lines on black look like neon; coloured edges keep the photo’s colours only where there are edges; and the overlay darkens the original along its outlines for a comic-like effect.
All processing runs locally in your browser.