How to detect shape on a transparent canvas?

Question

I'm looking for a method of detecting a shape in a transparent PNG. For example, I will create a transparent canvas of 940x680, then place a fully opaque object somewhere in that canvas.

I want to be able to detect the size (w, h), and top + left location of that object.

Here is an example of the original image:

Here is an example of what I would like to achieve (Bounding box overlay, with top + left margin data):

I've found a resource that does some transparency detection, but I'm not sure how I scale something like this to what I'm looking for.

var imgData,
    width = 200,
    height = 200;

$('#mask').bind('mousemove', function(ev){
    if(!imgData){ initCanvas(); }
    var imgPos = $(this).offset(),
      mousePos = {x : ev.pageX - imgPos.left, y : ev.pageY - imgPos.top},
      pixelPos = 4*(mousePos.x + height*mousePos.y),
         alpha = imgData.data[pixelPos+3];

    $('#opacity').text('Opacity = ' + ((100*alpha/255) << 0) + '%');
});

function initCanvas(){
    var canvas = $('<canvas width="'+width+'" height="'+height+'" />')[0],
           ctx = canvas.getContext('2d');

    ctx.drawImage($('#mask')[0], 0, 0);
    imgData = ctx.getImageData(0, 0, width, height);
}

Fiddle

Answer 1

What you need to do:

• Get the buffer
• Get a 32-bits reference of that buffer (If your other pixels are transparent then you can use a Uint32Array buffer to iterate).
• Scan 0 - width to find x1 edge
• Scan width - 0 to find x2 edge
• Scan 0 - height to find y1 edge
• Scan height - 0 to find y2 edge

These scans can be combined but for simplicity I'll show each step separately.

Online demo of this can be found here.

Result:

When image is loaded draw it in (if the image is small then the rest of this example would be waste as you would know the coordinates when drawing it - assuming here the image you draw is large with a small image inside it)

(note: this is a non-optimized version for the sake of simplicity)

ctx.drawImage(this, 0, 0, w, h);

var idata = ctx.getImageData(0, 0, w, h),      // get image data for canvas
    buffer = idata.data,                       // get buffer (unnes. step)
    buffer32 = new Uint32Array(buffer.buffer), // get a 32-bit representation
    x, y,                                      // iterators
    x1 = w, y1 = h, x2 = 0, y2 = 0;            // min/max values

Then scan each edge. For left edge you scan from 0 to width for each line (non optimized):

// get left edge
for(y = 0; y < h; y++) {                       // line by line
    for(x = 0; x < w; x++) {                   // 0 to width
        if (buffer32[x + y * w] > 0) {         // non-transparent pixel?
            if (x < x1) x1 = x;                // if less than current min update
        }
    }
}

For the right edge you just reverse x iterator:

// get right edge
for(y = 0; y < h; y++) {                       // line by line
    for(x = w; x >= 0; x--) {                  // from width to 0
        if (buffer32[x + y * w] > 0) {
            if (x > x2) x2 = x;
        }
    }
}

And the same is for top and bottom edges just that the iterators are reversed:

// get top edge
for(x = 0; x < w; x++) {
    for(y = 0; y < h; y++) {
        if (buffer32[x + y * w] > 0) {
            if (y < y1) y1 = y;
        }
    }
}

// get bottom edge
for(x = 0; x < w; x++) {
    for(y = h; y >= 0; y--) {
        if (buffer32[x + y * w] > 0) {
            if (y > y2) y2 = y;
        }
    }
}

The resulting region is then:

ctx.strokeRect(x1, y1, x2-x1, y2-y1);

There are various optimizations you could implement but they depend entirely on the scenario such as if you know approximate placement then you don't have to iterate all lines/columns.

You could do a brute force guess of he placement by skipping x number of pixels and when you found a non-transparent pixel you could make a max search area based on that and so forth, but that is out of scope here.

Hope this helps!

Answer 2

I was in need of something similar to this, just recently. Although the question is answered, I wanted to post my code for a future reference.

In my case, I'm drawing a (font) icon on a blank/transparent canvas, and want to get the bounding box. Even if I know the height of the icon (using font-size, i.e., height), I can't know the width. So I have to calculate it manually.

I'm not sure if there's a clever way to calculate this. First thing that popped into my head was doing it the hard way: manually checking every pixel, and that's what I did.

I think the code is pretty self-explanatory, so I won't do any explanation. I tried to keep the code as clean as possible.

/* Layer 3: The App */

let canvas = document.querySelector("#canvas");
let input  = document.querySelector("#input");
let output = document.querySelector("#output");

canvas.width  = 256;
canvas.height = 256;

let context = canvas.getContext("2d");
context.font = "200px Arial, sans-serif";

let drawnLetter = null;

drawLetter(input.value);

function drawLetter(letter) {
    letter = letter ? letter[0] : null;

    if (!letter) {
        // clear canvas
        context.clearRect(0, 0, canvas.width, canvas.height);
        output.textContent = null;
        return;
    }

    if (letter == drawnLetter) {
        return;
    }

    drawnLetter = letter;

    // clear canvas
    context.clearRect(0, 0, canvas.width, canvas.height);

    // draw letter
    context.fillText(letter, 50, canvas.height - 50);

    // find edges
    let boundingBox = findEdges(context);

    // mark the edges
    context.beginPath();
    context.rect(boundingBox.left, boundingBox.top, boundingBox.width, boundingBox.height);
    context.lineWidth = 2;
    context.strokeStyle = "red";
    context.stroke();

    // output the values
    output.textContent = JSON.stringify(boundingBox, null, "  ");
}


/* Layer 2: Interacting with canvas */

function findEdges(context) {
    let left   = findLeftEdge(context);
    let right  = findRightEdge(context);
    let top    = findTopEdge(context);
    let bottom = findBottomEdge(context);
    // right and bottom are relative to top left (0,0)
    return {
        left,
        top,
        right,
        bottom,
        width  : right - left,
        height : bottom - top,
    };
}

function findLeftEdge(context) {
    let imageData = context.getImageData(0, 0, context.canvas.width, context.canvas.height);
    let emptyPixel = [0, 0, 0, 0].join();
    for (let x = 0; x < context.canvas.width; x++) {
        for (let y = 0; y < context.canvas.height; y++) {
            let pixel = getPixel(imageData, x, y).join();
            if (pixel != emptyPixel) {
                return x;
            }
        }
    }
}

function findRightEdge(context) {
    let imageData = context.getImageData(0, 0, context.canvas.width, context.canvas.height);
    let emptyPixel = [0, 0, 0, 0].join();
    for (let x = context.canvas.width - 1; x >= 0; x--) {
        for (let y = 0; y < context.canvas.height; y++) {
            let pixel = getPixel(imageData, x, y).join();
            if (pixel != emptyPixel) {
                return x;
            }
        }
    }
}

function findTopEdge(context) {
    let imageData = context.getImageData(0, 0, context.canvas.width, context.canvas.height);
    let emptyPixel = [0, 0, 0, 0].join();
    for (let y = 0; y < context.canvas.height; y++) {
        for (let x = 0; x < context.canvas.width; x++) {
            let pixel = getPixel(imageData, x, y).join();
            if (pixel != emptyPixel) {
                return y;
            }
        }
    }
}

function findBottomEdge(context) {
    let imageData = context.getImageData(0, 0, context.canvas.width, context.canvas.height);
    let emptyPixel = [0, 0, 0, 0].join();
    for (let y = context.canvas.height - 1; y >= 0; y--) {
        for (let x = 0; x < context.canvas.width; x++) {
            let pixel = getPixel(imageData, x, y).join();
            if (pixel != emptyPixel) {
                return y;
            }
        }
    }
}


/* Layer 1: Interacting with ImageData */

/**
 * Returns the pixel array at the specified position.
 */
function getPixel(imageData, x, y) {
    return getPixelByIndex(imageData, pos2index(imageData, x, y));
}

/**
 * Returns the RGBA values at the specified index.
 */
function getPixelByIndex(imageData, index) {
    return [
        imageData.data[index + 0],
        imageData.data[index + 1],
        imageData.data[index + 2],
        imageData.data[index + 3],
    ];
}

/**
 * Returns the index of a position.
 */
function pos2index(imageData, x, y) {
    return 4 * (y * imageData.width + x);
}

body {
    background-color: hsl(0, 0%, 95%);
}
canvas {
    background: white;
    image-rendering: pixelated;
    background-image: url(data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQAQMAAAAlPW0iAAAABlBMVEXMzMz////TjRV2AAAAEUlEQVQI12P4z8CAFWEX/Q8Afr8P8erzE9cAAAAASUVORK5CYII=);
    zoom: 0.8; /* this counters the scale up (125%) of my screen; can be removed */
}
input {
    padding: 0.2em;
    margin-top: 0.5em;
}

<canvas id="canvas"></canvas>

<br>

<input type="text" id="input" placeholder="type a letter" value="A" onkeyup="drawLetter(this.value)" />

<pre id="output"></pre>