CV.imread( ) does not get Mat

[cike111]

How to load an image under the assets path to generate mat ,if using CV.imread("assets/ material_wood.jpg ") does not get the correct Mat

[abdelaziz-mahdy]

I think assets are not easy to handle specially on mobile

I recommend copying the file to tmp dir and use the new path

[rainyl]

I think assets are not easy to handle specially on mobile

I recommend copying the file to tmp dir and use the new path

yes,it works. a more elegant way is reading assets to bytes and use imdecode.

but for APIs without the support of loading from bytes it will be the only solution for now.

[abdelaziz-mahdy]

And I don't think including the asset reader will work with native assets since the asset reader is from flutter

So people should manage that in their own app, maybe we can add an example?

[rainyl]

And I don't think including the asset reader will work with native assets since the asset reader is from flutter

agree, it's out of the scope of this package. maybe we can develop another package to extend this package and provide some helper functions but not for now, this package is still in heavy development and APIs are not stable enough.

So people should manage that in their own app, maybe we can add an example?

actually the example under this repo has shown how to load image from assets, but we can make another one to show this clearer and add more solutions, e.g., copy to tmp folder and load models.

[abdelaziz-mahdy]

Ok let's open an issue with it I may check when I have time, no promises since I have been busy for a while now

[rainyl]

Ok let's open an issue with it I may check when I have time, no promises since I have been busy for a while now

Sounds good, just do it when you are free.

[rainyl]

Refer rainyl/awesome-opencv_dart#1 for further development.

As for the answer of this issue, please take a look at the example of this repo.

opencv_dart/example/lib/main.dart

Lines 86 to 95 in 9168144

	final data = await DefaultAssetBundle.of(context).load("images/lenna.png");
	final bytes = data.buffer.asUint8List();
	// heavy computation
	// final (gray, blur) = await heavyTask(bytes);
	// setState(() {
	// images = [bytes, gray, blur];
	// });
	final (gray, blur) = await heavyTaskAsync(cv.imdecode(bytes, cv.IMREAD_COLOR));
	setState(() {
	images = [bytes, cv.imencode(".png", gray), cv.imencode(".png", blur)];