A real-time edge detection pipeline built using the IoToToy visual programming app for Android.
This project demonstrates real-time Computer Vision on a mobile device. It captures a live video feed and applies a Sobel Operator to highlight edges in the image, displaying both the original source and the processed output simultaneously.
The logic is defined in live-camera-sobel-filter.json and follows this flow:
- Constants (Resolution): A value of
256sets the camera width and height to ensure efficient real-time processing. - Camera: Captures the live video feed based on the input resolution.
- Original Display: One output branch sends the raw frames directly to a display block.
- Sobel Filter: The second branch passes frames through the
image_filter_sobelblock. - Constant (Kernel): A value of
5defines the Sobel kernel size, adjusting the edge detection sensitivity. - Processed Display: Renders the final black-and-white edge map to a larger on-screen display.
- Install IoToToy from the Google Play Store.
- Download the
live-camera-sobel-filter.jsonfile from this repository. - Open IoToToy, then go to FILE -> OPEN and select the JSON file.
- Run the diagram and point your camera at high-contrast objects to see the edge detection in action.
- Android device with camera access.
- IoToToy app (v1.0.5 or newer).
