HMS Image Super Resolution Application - Huawei Developers

More information like this, you can visit HUAWEI Developer Forum​
Introduction:
HMS ML Kit features image super-resolution service which provides 1x super-resolution capability. This feature removes the compression noise of images to obtain clear images.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
Precautions:
1. Prior to the image super-resolution service, it is necessary to convert images into bitmaps in ARGB format. After
the service processes, the image output are bitmaps in ARGB format.
2. Maximum size of an input image is 1024 x 768 px or 768 x 1024 px. The minimum size is 64 x 64 px.
Integration:
1. Create a project in android studio and Huawei AGC.
2. Provide the SHA-256 Key in App Information Section.
3. Download the agconnect-services.json from AGCand save into app directory.
4. In root build.gradle
Navigate to allprojects > repositories and buildscript > repositories and add the below line.
Code:
maven { url 'http://developer.huawei.com/repo/' }
5. In app build.gradle
Configure the Maven dependency
Code:
implementation 'com.huawei.hms:ml-computer-vision-imageSuperResolution:2.0.2.300'
implementation 'com.huawei.hms:ml-computer-vision-imageSuperResolution-model:2.0.2.300'
Apply plugin
Code:
apply plugin: 'com.huawei.agconnect'
6. Permissions in Manifest
Code:
<uses-permission android:name="android.permission.CAMERA" />
<uses-permission android:name="android.permission.READ_EXTERNAL_STORAGE" />
<uses-permission android:name="android.permission.WRITE_EXTERNAL_STORAGE" />
Code Implementation:
Here is an image convertor application that uses HMS image super-resolution service. This application can select image from gallery and also can capture image. Follow the steps
1. Create an image super-resolution analyzer.
Code:
private void createAnalyzer() {
MLImageSuperResolutionAnalyzerSetting settings = new MLImageSuperResolutionAnalyzerSetting.Factory()
// Set the scale of image super resolution to 1x.
.setScale(MLImageSuperResolutionAnalyzerSetting.ISR_SCALE_1X)
.create();
analyzer = MLImageSuperResolutionAnalyzerFactory.getInstance().getImageSuperResolutionAnalyzer(settings);
}
2. Create an MLFrame object by using android.graphics.Bitmap.
Code:
MLFrame mlFrame = new MLFrame.Creator().setBitmap(srcBitmap).create();
3. Perform super-resolution processing on the image.
Code:
Task<MLImageSuperResolutionResult> task = analyzer.asyncAnalyseFrame(mlFrame);
task.addOnSuccessListener(new OnSuccessListener<MLImageSuperResolutionResult>() {
public void onSuccess(MLImageSuperResolutionResult result) {
// Recognition success.
Toast.makeText(getApplicationContext(), "Success", Toast.LENGTH_SHORT).show();
setImage(result.getBitmap());
}
}).addOnFailureListener(new OnFailureListener() {
public void onFailure(Exception e) {
// Recognition failure.
Toast.makeText(getApplicationContext(), "Failed：" + e.getMessage(), Toast.LENGTH_SHORT).show();
}
});
4. After the recognition is complete, stop the analyzer to release recognition resources.
Code:
private void release() {
if (analyzer == null) {
return;
}
analyzer.stop();
}
5. Capture picture from camera.
Code:
private void capturePictureFromCamera(){
if (checkSelfPermission(Manifest.permission.CAMERA) != PackageManager.PERMISSION_GRANTED)
{
requestPermissions(new String[]{Manifest.permission.CAMERA}, MY_CAMERA_PERMISSION_CODE);
}
else
{
Intent cameraIntent = new Intent(android.provider.MediaStore.ACTION_IMAGE_CAPTURE);
startActivityForResult(cameraIntent, CAMERA_REQUEST);
}
}
6. Acceessing image from Gallery.
Code:
private void getImageFromGallery(){
Intent intent = new Intent();
intent.setAction(Intent.ACTION_GET_CONTENT);
intent.setType("image/*");
startActivityForResult(intent, GALLERY_REQUEST);
}
Screen Shots:
Conclusion:
Image super-resolution service is widely used in day to day life and supports in common scenarios like improving low-quality images on the network, obtaining clear images while reading news, improving image clarity of Identification Card etc.This service intelligently reduces the image noise and provides clear image without changing resolution.
Reference:
HMSCore-Guides

---

Is there any image size limitations.

---

How to Reset the Mobile app?
I have installed an emulated to play the world best game on my computer. But emulator is not working smoothly.
Please share some good technique to run it well so that I can enjoy the taste of the game.
Thank you.

---

cool thing, I'll definitely have to try it, especially with my old photos.

---

Does this work with every image? I've some old black and white photos

---

JohnMes said:
cool thing, I'll definitely have to try it, especially with my old photos.
Click to expand...
Click to collapse
It will be a good choice.:highfive:

---

Rushikesh787 said:
Does this work with every image? I've some old black and white photos
Click to expand...
Click to collapse
I think it can be better to have a try. Maybe we can gain a surprise.

---

will it convert all format images ?

---

I would definitely want to try this out

---

Awesome
I would definitely try this app on my girlfriend's photo