mirror of
https://github.com/peterbraden/node-opencv.git
synced 2025-12-08 19:45:55 +00:00
495e06a591
As we can pass matrix as template, detecting canny provide more robust match if we only care about shape rather than contrast and color.
400 lines
10 KiB
JavaScript
Executable File
400 lines
10 KiB
JavaScript
Executable File
var fs = require('fs')
|
|
, test = require('tape')
|
|
, cv = null
|
|
|
|
|
|
|
|
test("Smoke tests / Can Import", function(t){
|
|
cv = require('../lib/opencv')
|
|
t.ok(cv, "imported fine")
|
|
t.ok(cv.version, "version is there:" + cv.version)
|
|
t.ok(cv.Point, "point is there")
|
|
t.ok(cv.Matrix, "matrix is there")
|
|
t.end()
|
|
})
|
|
|
|
|
|
test('importing library multiple times is ok', function(t){
|
|
var cv1 = require('../lib/opencv')
|
|
, cv2 = require('../lib/opencv')
|
|
cv1.readImage('./examples/files/mona.png', function(err, im){
|
|
t.error(err)
|
|
cv2.readImage('./examples/files/mona.png', function(err, im){
|
|
t.error(err)
|
|
t.end();
|
|
});
|
|
});
|
|
})
|
|
|
|
|
|
test('Point', function(t){
|
|
|
|
t.ok(new cv.Point(1, 2))
|
|
t.throws(function () { cv.Point(1, 2)}, TypeError, "cannot call without new")
|
|
|
|
t.equal(new cv.Point(1, 2).x, 1)
|
|
t.equal(new cv.Point(1, 2).y, 2)
|
|
t.equal(Math.round(new cv.Point(1.1, 2).x * 100), 110)
|
|
t.equal(Math.round(new cv.Point(1.2, 2.75).y *100), 275)
|
|
|
|
t.throws(function () {new cv.Point(1.1, 2).x = 5}, Error, "Points are immutable")
|
|
t.throws(function () {new cv.Point(1.1, 2).y = 5}, Error, "Points are immutable")
|
|
|
|
var p1 = new cv.Point(3, 6)
|
|
, p2 = new cv.Point(5, 7)
|
|
|
|
t.ok(p1.dot);
|
|
t.equal(p1.dot(p2), 57);
|
|
|
|
t.end()
|
|
})
|
|
|
|
|
|
test('Matrix constructor', function(assert){
|
|
assert.ok(cv.Matrix);
|
|
assert.ok(new cv.Matrix);
|
|
assert.ok(new cv.Matrix(1, 2));
|
|
assert.ok(new cv.Matrix(1, 2, cv.Constants.CV_8U));
|
|
assert.ok(new cv.Matrix(1, 2, cv.Constants.CV_8U, [1]));
|
|
assert.end()
|
|
})
|
|
|
|
test('Matrix accessors', function(assert){
|
|
var mat = new cv.Matrix(1, 2);
|
|
mat.set(0,0,3)
|
|
mat.set(0,1,5000)
|
|
assert.deepEqual(mat.row(0), [3,5000]);
|
|
|
|
mat = new cv.Matrix(1,2);
|
|
assert.equal(mat.set(0,0,3), undefined);
|
|
assert.equal(mat.get(0,0), 3);
|
|
|
|
mat = new cv.Matrix(6,7);
|
|
assert.equal(mat.width(), 7);
|
|
|
|
mat = new cv.Matrix(6,7);
|
|
assert.equal(mat.height(), 6);
|
|
|
|
mat = new cv.Matrix(6,7);
|
|
assert.deepEqual(mat.size(), [6, 7]);
|
|
|
|
mat = new cv.Matrix(6,7);
|
|
assert.equal(mat.width(), 7);
|
|
mat.resize(8,9);
|
|
assert.equal(mat.width(), 8);
|
|
|
|
mat = new cv.Matrix.Eye(4,4)
|
|
assert.deepEqual(mat.row(1), [0,1,0,0])
|
|
assert.deepEqual(mat.row(2), [0,0,1,0])
|
|
|
|
mat = new cv.Matrix.Eye(4,4);
|
|
assert.deepEqual(mat.col(1), [0,1,0,0])
|
|
assert.deepEqual(mat.col(2), [0,0,1,0])
|
|
|
|
assert.equal(new cv.Matrix().empty(), true);
|
|
|
|
mat = new cv.Matrix(1, 2, cv.Constants.CV_8UC3, [1, 2, 3]);
|
|
assert.deepEqual(mat.pixelRow(0), [1, 2, 3, 1, 2, 3]);
|
|
|
|
assert.end()
|
|
})
|
|
|
|
test('Matrix functions', function(assert) {
|
|
// convertTo
|
|
var mat = new cv.Matrix(75, 75, cv.Constants.CV_32F, [2.0]);
|
|
var matNew = new cv.Matrix(75, 75, cv.Constants.CV_8U);
|
|
mat.convertTo(matNew, cv.Constants.CV_8U, 2, 1);
|
|
assert.equal(matNew.pixel(0, 0), 5);
|
|
|
|
// reshape
|
|
mat = new cv.Matrix(75, 75, cv.Constants.CV_8UC1);
|
|
matNew = mat.reshape(1, 1);
|
|
assert.equal(matNew.height(), 1);
|
|
assert.equal(matNew.width(), 5625);
|
|
|
|
// GetRotationMatrix2D
|
|
mat = cv.Matrix.getRotationMatrix2D(0, 0, 90, 1.0);
|
|
assert.deepEqual(mat.size(), [2,3], 'GetRotationMatrix2D');
|
|
|
|
assert.end();
|
|
})
|
|
|
|
test(".norm", function(assert){
|
|
cv.readImage("./examples/files/coin1.jpg", function(err, im) {
|
|
cv.readImage("./examples/files/coin2.jpg", function(err, im2){
|
|
assert.ok(im);
|
|
assert.ok(im2);
|
|
|
|
var errorL2 = im.norm(im2, cv.Constants.NORM_L2);
|
|
assert.equal(errorL2, 7295.591339980605);
|
|
errorL2 = im.norm(im, cv.Constants.NORM_L2);
|
|
assert.equal(errorL2, 0);
|
|
assert.end();
|
|
});
|
|
});
|
|
})
|
|
|
|
test("Matrix toBuffer", function(assert){
|
|
var buf = fs.readFileSync('./examples/files/mona.png')
|
|
|
|
cv.readImage(buf.slice(0), function(err, mat){
|
|
var buf0 = mat.toBuffer()
|
|
assert.ok(buf0);
|
|
assert.end()
|
|
})
|
|
})
|
|
|
|
test("Matrix toBuffer Async", function(assert){
|
|
var buf = fs.readFileSync('./examples/files/mona.png')
|
|
|
|
cv.readImage(buf.slice(0), function(err, mat){
|
|
mat.toBuffer(function(err, buff){
|
|
assert.error(err)
|
|
assert.ok(buf)
|
|
assert.ok(buf.length > 100)
|
|
|
|
assert.end()
|
|
})
|
|
})
|
|
})
|
|
|
|
|
|
test("detectObject", function(assert){
|
|
cv.readImage("./examples/files/mona.png", function(err, im){
|
|
im.detectObject(cv.FACE_CASCADE, {}, function(err, faces){
|
|
assert.error(err)
|
|
assert.ok(faces)
|
|
assert.equal(faces.length, 1)
|
|
assert.end()
|
|
})
|
|
})
|
|
})
|
|
|
|
test(".absDiff and .countNonZero", function(assert){
|
|
cv.readImage("./examples/files/mona.png", function(err, im) {
|
|
cv.readImage("./examples/files/mona.png", function(err, im2){
|
|
assert.ok(im);
|
|
assert.ok(im2);
|
|
|
|
var diff = new cv.Matrix(im.width(), im.height());
|
|
diff.absDiff(im, im2);
|
|
|
|
diff.convertGrayscale();
|
|
assert.equal(diff.countNonZero(), 0);
|
|
assert.end()
|
|
});
|
|
});
|
|
})
|
|
|
|
|
|
test(".bitwiseXor", function(assert){
|
|
var mat1 = new cv.Matrix(1,1);
|
|
mat1.set(0,0, 1);
|
|
|
|
var mat2 = new cv.Matrix(1,1);
|
|
mat2.set(0,0, 1);
|
|
|
|
var xored = new cv.Matrix(1,1);
|
|
xored.bitwiseXor(mat1, mat2);
|
|
|
|
assert.equal(xored.get(0,0), 0);
|
|
|
|
assert.end()
|
|
})
|
|
|
|
|
|
test("Image read from file", function(assert){
|
|
cv.readImage("./examples/files/opencv.png", function(err, im){
|
|
assert.ok(im);
|
|
assert.equal(im.width(), 82);
|
|
assert.equal(im.height(), 99);
|
|
assert.equal(im.channels(), 4);
|
|
assert.equal(im.empty(), false);
|
|
assert.end();
|
|
})
|
|
})
|
|
|
|
|
|
test("read Image from buffer", function(assert){
|
|
cv.readImage(fs.readFileSync('./examples/files/opencv.png'), function(err, im){
|
|
assert.ok(im);
|
|
assert.equal(im.width(), 82);
|
|
assert.equal(im.height(), 99);
|
|
assert.equal(im.channels(), 4);
|
|
assert.equal(im.empty(), false);
|
|
assert.end();
|
|
})
|
|
})
|
|
|
|
test("Cascade Classifier", function(assert){
|
|
assert.ok(new cv.CascadeClassifier("./data/haarcascade_frontalface_alt.xml"), 'test constructor')
|
|
|
|
cv.readImage("./examples/files/mona.png", function(err, im){
|
|
cascade = new cv.CascadeClassifier("./data/haarcascade_frontalface_alt.xml");
|
|
cascade.detectMultiScale(im, function(err, faces){//, 1.1, 2, [30, 30]);
|
|
assert.error(err);
|
|
assert.equal(typeof faces, typeof []);
|
|
assert.equal(faces.length, 1)
|
|
assert.end()
|
|
})
|
|
})
|
|
})
|
|
|
|
|
|
test("ImageDataStream", function(assert){
|
|
var s = new cv.ImageDataStream()
|
|
s.on('load', function(im){
|
|
assert.ok(im)
|
|
assert.equal(im.empty(), false);
|
|
assert.end()
|
|
})
|
|
|
|
fs.createReadStream('./examples/files/mona.png').pipe(s);
|
|
|
|
})
|
|
|
|
test("ImageStream", function(assert){
|
|
var s = new cv.ImageStream()
|
|
, im = fs.readFileSync('./examples/files/mona.png')
|
|
|
|
s.on('data', function(mat){
|
|
assert.deepEqual(mat.size(), [756,500])
|
|
assert.end()
|
|
})
|
|
s.write(im);
|
|
})
|
|
|
|
|
|
test("CamShift", function(assert){
|
|
cv.readImage('./examples/files/coin1.jpg', function(e, im){
|
|
cv.readImage('./examples/files/coin2.jpg', function(e, im2){
|
|
var tracked = new cv.TrackedObject(im, [420, 110, 490, 170], {channel: 'v'});
|
|
assert.ok(tracked);
|
|
var res = tracked.track(im2)
|
|
assert.ok(res);
|
|
assert.ok(res[0] < 396)
|
|
assert.ok(res[0] > 376)
|
|
assert.ok(res[1] < 122)
|
|
assert.ok(res[1] > 102)
|
|
assert.ok(res[2] < 469)
|
|
assert.ok(res[2] > 449)
|
|
assert.ok(res[3] < 176)
|
|
assert.ok(res[3] > 156)
|
|
assert.end()
|
|
})
|
|
})
|
|
})
|
|
|
|
test("fonts", function(t) {
|
|
|
|
function rnd() {
|
|
return Math.round(Math.random() * 255);
|
|
};
|
|
|
|
cv.readImage("./examples/files/coin1.jpg", function(e, im){
|
|
var y = 0;
|
|
var scale = 1;
|
|
|
|
([
|
|
"HERSEY_SIMPLEX",
|
|
"HERSEY_PLAIN",
|
|
"HERSEY_DUPLEX",
|
|
"HERSEY_COMPLEX",
|
|
"HERSEY_TRIPLEX",
|
|
"HERSEY_COMPLEX_SMALL",
|
|
"HERSEY_SCRIPT_SIMPLEX",
|
|
"HERSEY_SCRIPT_COMPLEX",
|
|
"HERSEY_SCRIPT_SIMPLEX"
|
|
]).forEach(function(font) {
|
|
im.putText("Some text", 0, y += 20, font, [rnd(), rnd(), rnd()], 2, scale++);
|
|
});
|
|
|
|
t.ok(im, "image is ok");
|
|
im.save("./examples/tmp/coin1-with-text.jpg");
|
|
t.end();
|
|
});
|
|
})
|
|
|
|
test('LDA Wrap', function(assert) {
|
|
if (cv.LDA === undefined) {
|
|
console.log('TODO: Please port LDAWrap.cc to OpenCV 3')
|
|
assert.end();
|
|
return;
|
|
}
|
|
|
|
// subspaceProject
|
|
var mat = cv.LDA.subspaceProject(new cv.Matrix(1, 2, cv.Constants.CV_64F), new cv.Matrix(), new cv.Matrix(2, 1, cv.Constants.CV_8UC1));
|
|
assert.deepEqual(mat.size(), [2,2], 'subspaceProject');
|
|
|
|
// subspaceReconstruct
|
|
mat = cv.LDA.subspaceReconstruct(new cv.Matrix(1, 2, cv.Constants.CV_64F), new cv.Matrix(), new cv.Matrix(1, 2, cv.Constants.CV_8UC1));
|
|
assert.deepEqual(mat.size(), [1,1], 'subspaceReconstruct');
|
|
|
|
assert.end();
|
|
})
|
|
|
|
|
|
test('Native Matrix', function(assert) {
|
|
var nativemat = require('../build/Release/test_nativemat.node');
|
|
var mat = new cv.Matrix(42, 8);
|
|
|
|
assert.deepEqual(mat.size(), nativemat.size(mat), 'nativemat');
|
|
assert.end();
|
|
})
|
|
|
|
test('Subtract', function(assert) {
|
|
var a = new cv.Matrix.Zeros(1,1);
|
|
a.set(0, 0, 3);
|
|
var b = new cv.Matrix.Zeros(1,1);
|
|
b.set(0, 0, 1);
|
|
a.subtract(b);
|
|
assert.deepEqual(a.get(0, 0), 2);
|
|
assert.end();
|
|
});
|
|
|
|
test('Mean', function(assert) {
|
|
var a = new cv.Matrix.Zeros(2, 2, cv.Constants.CV_8UC3);
|
|
|
|
// Set [0, 0] element to 1 for all three channels
|
|
a.set(0, 0, 1, 0);
|
|
a.set(0, 0, 1, 1);
|
|
a.set(0, 0, 1, 2);
|
|
|
|
var means = a.mean();
|
|
assert.deepEqual(means, [0.25, 0.25, 0.25, 0]);
|
|
assert.end();
|
|
});
|
|
|
|
test('MatchTemplateByMatrix', function(assert) {
|
|
var cv = require('../lib/opencv');
|
|
var targetFilename = "./examples/files/car1.jpg";
|
|
var templateFilename = "./examples/files/car1_template.jpg";
|
|
cv.readImage(targetFilename, function(err, target){
|
|
cv.readImage(templateFilename, function(err, template){
|
|
var TM_CCORR_NORMED = 3;
|
|
var res = target.matchTemplateByMatrix(template, TM_CCORR_NORMED);
|
|
var minMax = res.minMaxLoc();
|
|
var topLeft = minMax.maxLoc;
|
|
assert.ok(topLeft, "RGB Found Match");
|
|
assert.equal(topLeft.x, 42, "match location x === 42");
|
|
assert.equal(topLeft.y, 263, "match location y === 263");
|
|
target.canny(5,300);
|
|
template.canny(5,300);
|
|
res = target.matchTemplateByMatrix(template, TM_CCORR_NORMED);
|
|
minMax = res.minMaxLoc();
|
|
topLeft = minMax.maxLoc;
|
|
target.save("./target.png");
|
|
template.save("./template.png");
|
|
// res.rectangle([topLeft.x, topLeft.y], [template.width(), template.height()], [0, 255,0], 2);
|
|
res.save("./result.png");
|
|
assert.ok(topLeft, "Canny edge Found Match");
|
|
assert.equal(topLeft.x, 42, "match location x === 42");
|
|
assert.equal(topLeft.y, 263, "match location y === 263");
|
|
assert.end();
|
|
});
|
|
})
|
|
});
|
|
|
|
// Test the examples folder.
|
|
require('./examples')()
|