//Match features from canvas
public IEnumerator MatchFeatures()
{
ImageString image;
GameObject instantiatedWait = Instantiate(wait, screenshotPreview.transform);
//Pick the pixels inside the square
ImagePartToByteArray();
//Get imageList
yield return(imageList.GetTextFromURL());
Debug.Log("Image list count: " + imageList.getImageList().Count);
//Don't match features if no images is on the canvas
if (imageList.getImageList().Count == 0)
{
image = new ImageString(Convert.ToBase64String(imagePart), -1);
}
// Match feature from images on canvas.
else
{
image = featureMatcher.MatchFeatures(Convert.ToBase64String(imagePart), imageList.getImageList());
}
Destroy(instantiatedWait);
UploadToInAppBrowser(image.getImageString(), image.getWinnerIndex());
}
public async Task <FileContentResult> Get(ImageString image)
{
if (hoststring == "")
{
hoststring = HttpContext.Request.Host.Value;
}
Image _image = _imageService.GetById(image.ImageKey, image.Username);
return(File(_image.bytes, "image/jpeg", _image.Name + ".jpg"));
}
public void Post([FromBody] ImageString value)
{
Debug.WriteLine("POST COUNT = " + value.base64.Count);
foreach (var item in value.base64)
{
CurrTasks.Enqueue(item);
//Task.Run(() => handleAsyncCalls(item));
}
if (!IsInProcess)
{
ContinueOrStartProcess();
}
//return Redirect("Photo");
}
public void StartClickHandler(object sender, RoutedEventArgs e)
{
Classes.Clear();
foreach (string pathImage in Directory.GetFiles(path).Where(s => s.EndsWith(".JPEG") || s.EndsWith(".jpg")))
{
ImageString obj = new ImageString()
{
path = pathImage,
ImageBase64 = ImageToBase64(new Avalonia.Media.Imaging.Bitmap(pathImage)),
Probability = 0,
ClassImage = "Default"
};
Thread t = new Thread(new ParameterizedThreadStart(Post));
t.Start(obj);
}
}
public ResultClassification Post([FromBody] ImageString obj)
{
ResultClassification result;
OnnxClassifier.OnnxClassifier onnxModel;
using (var db = new ApplicationContext())
{
result = db.FindInDataBase(obj);
if (result != null)
{
return(result);
}
onnxModel = new OnnxClassifier.OnnxClassifier();
result = onnxModel.PredictModel(obj.path);
obj.path = result._PathImage;
obj.Probability = result._Probability;
obj.ClassImage = result._ClassImage;
db.AddToDataBase(obj);
}
return(result);
}
public ResultClassification FindInDataBase(ImageString obj)
{
byte[] array = Convert.FromBase64String(obj.ImageBase64);
lock (this)
{
var images = this.Images.Where(p => p.Hash == GetHashFromBytes(array)).Select(p => p).ToList();
foreach (var img in images)
{
this.Entry(img).Reference(p => p.ImageBytes).Load();
if (array.SequenceEqual(img.ImageBytes.Bytes))
{
img.Call += 1;
this.SaveChanges();
return(new ResultClassification(img.Path, img.Class, img.Prob));
}
}
}
Console.WriteLine("NOT_FOUND");
return(null);
}
public void AddToDataBase(ImageString obj)
{
lock (this)
{
byte[] BytesImage = Convert.FromBase64String(obj.ImageBase64);
Blob ImageBlob = new Blob {
Bytes = BytesImage
};
this.Blobs.Add(ImageBlob);
this.SaveChanges();
RecognitionImage elem = new RecognitionImage
{
Class = obj.ClassImage,
Prob = obj.Probability,
ImageBytes = ImageBlob,
Call = 0,
Hash = GetHashFromBytes(BytesImage),
Path = obj.path
};
this.Images.Add(elem);
this.SaveChanges();
}
}
