Online Sprite Cutter & Image Splitter

Automatically split sprite sheets and images into transparent PNG cutouts using Canny edge detection and contour extraction — tune parameters in real time and download everything as a ZIP. Completely free, no installation required.

Core Features

All-in-one online image processing — from upload to export in just a few steps

Smart Sprite Splitting

Powered by contour detection algorithms, our tool automatically identifies every individual element in your sprite sheet and precisely splits them into separate transparent PNG files. No manual crop lines needed — dramatically speed up your asset workflow.

Canny Edge Detection Debugger

Built-in parameter panel for Canny edge detection lets you adjust Gaussian blur radius, high/low thresholds, and morphological iteration count in real time. Instantly preview detection results to find the optimal splitting parameters for your image.

One-Click ZIP Download

All split transparent PNG sub-images can be bundled into a single ZIP file for download with one click. Files are numbered sequentially, making it easy to batch-import them into game engines or design tools.

Transparent Background Export

Each extracted sub-image retains its alpha transparency channel with all excess background removed. The exported PNG assets are ready to use directly in game development, UI design, or web production — no post-processing required.

High-Performance Engine

The backend is built with Rust, combined with highly efficient image processing algorithms. Even large sprite sheets or complex assets containing hundreds of sub-images are processed and exported in just seconds.

Privacy & Security

Uploaded images are used only for real-time processing. Temporary files are automatically cleaned up after processing is complete. We do not permanently store your image data — your assets and privacy are fully protected.

Three Simple Steps

No software to download — open your browser and get started

Upload Your Image

Click the upload area or drag and drop an image onto the page. Supports PNG, JPG, WebP, and other common formats. For best results, use sprite sheets with a uniform or transparent background.

Adjust Parameters

Use the parameter panel to fine-tune Canny edge detection thresholds, Gaussian blur radius, and morphological iterations. The page previews detected contours and bounding boxes in real time to help you find the optimal settings.

Preview & Download

Once you're happy with the split results, preview each sub-image's transparent PNG output individually, then click "Download ZIP" to grab all extracted assets in a single archive.

Get Started

Upload your sprite sheet or any image, adjust parameters, then split and download

Supported: .png / .jpeg / .jpg / .webp Limits: max width 16000px, max height 16000px, max contours 1000
Feature overview
One-click sprite/image splitting: auto contour detection → smart segmentation → transparent PNG export → ZIP download.
Use cases
  • Sprite sheet cutting for game assets/UI atlases/stickers with transparency preserved.
  • Split multiple objects from screenshots/product images into individual assets.
  • Contour/edge preview for annotation and pre-matting inspection.
  • Interactive Canny tuning demo with blur/threshold/morphology controls.
Workflow
Upload → tweak (Gaussian blur / Canny thresholds / morphology) → Process & preview → Download cutouts.
See “Tutorial” on the right for examples; your last parameters are saved for next time.

Use Cases

Sprite splitting and image segmentation have broad applications across many fields

Game Development

Split game sprite sheets into individual character frames, item icons, and map elements for easy import into Unity, Godot, Cocos, and other game engines for animation and level design.

UI/UX Design

Quickly extract individual UI elements, buttons, and icons from design comps or icon sets. Export transparent PNGs ready for web frontend development or mobile interface design.

Asset Management

Batch-process art assets by automatically splitting illustrations, stickers, or emoji packs merged onto a single canvas into individual files. Save hours of manual cropping and improve asset organization.

Image Analysis & Research

Use the interactive Canny parameter tuning to visually learn and demonstrate edge detection, contour extraction, and other fundamental computer vision algorithms — perfect for teaching and research.

Frequently Asked Questions

Common questions about using the tool

What image formats are supported?

We currently support PNG, JPG/JPEG, WebP, BMP, and other common image formats. For best splitting results, we recommend using PNG sprite sheets with a transparent background.

Are uploaded images stored on the server?

No. Images are used only for real-time processing. Temporary files are automatically cleaned up after processing completes. We do not permanently store your image data.

What is the maximum image size?

The default limit is 16000×16000 pixels with a maximum file size of 10 MB. For larger images, we recommend resizing before uploading for faster processing.

How should I adjust the Canny parameters?

Start with the default values. If too many or too few contours are detected, adjust the high/low thresholds. A larger Gaussian blur radius reduces detail but improves noise resistance. Increasing morphological iterations merges nearby contours.

How It Works

The core splitting pipeline is based on a classic computer vision algorithm chain. First, the original image undergoes Gaussian blur preprocessing to reduce noise interference. Then, the Canny edge detection algorithm extracts edge information from the image. Next, morphological closing operations (dilation followed by erosion) connect broken edges into closed regions. Finally, a contour extraction algorithm identifies every independent closed contour and computes its minimum bounding rectangle as the crop region.

Every parameter can be adjusted in real time on the page with instant preview: the Gaussian blur radius controls the smoothing level — larger values reduce noise but also blur fine details; the Canny high/low thresholds determine edge detection sensitivity; the morphological iteration count controls how aggressively edges are closed. By combining these parameters, you can adapt the tool to virtually any type of sprite sheet or image asset.

Want to dive deeper into the algorithm? Read our Canny Edge Detection Deep Dive, or check out the Sprite Cutter User Guide for parameter tuning tips.

Ready to Split Your Sprite Sheet?

Completely free. No sign-up. No installation. Just open your browser and go.

Get Started ↑