public void Recognize(Mat m)
{
Tensor imageTensor = Emgu.TF.TensorConvert.ReadTensorFromMatBgr(m, 224, 224, 128.0f, 1.0f / 128.0f);
//Uncomment the following code to use a retrained model to recognize followers, downloaded from the internet
//Inception _inceptionGraph = new Inception(null, new string[] {"optimized_graph.pb", "output_labels.txt"}, "https://github.com/emgucv/models/raw/master/inception_flower_retrain/", "Mul", "final_result");
//Tensor imageTensor = ImageIO.ReadTensorFromMatBgr(fileName, 299, 299, 128.0f, 1.0f / 128.0f);
//Uncomment the following code to use a retrained model to recognize followers, if you deployed the models with the application
//For ".pb" and ".txt" bundled with the application, set the url to null
//Inception _inceptionGraph = new Inception(null, new string[] {"optimized_graph.pb", "output_labels.txt"}, null, "Mul", "final_result");
//Tensor imageTensor = ImageIO.ReadTensorFromMatBgr(fileName, 299, 299, 128.0f, 1.0f / 128.0f);
float[] probability;
if (_coldSession)
{
//First run of the recognition graph, here we will compile the graph and initialize the session
//This is expected to take much longer time than consecutive runs.
probability = _inceptionGraph.Recognize(imageTensor);
_coldSession = false;
}
//Here we are trying to time the execution of the graph after it is loaded
Stopwatch sw = Stopwatch.StartNew();
probability = _inceptionGraph.Recognize(imageTensor);
sw.Stop();
String resStr = String.Empty;
if (probability != null)
{
String[] labels = _inceptionGraph.Labels;
float maxVal = 0;
int maxIdx = 0;
for (int i = 0; i < probability.Length; i++)
{
if (probability[i] > maxVal)
{
maxVal = probability[i];
maxIdx = i;
}
}
resStr = String.Format("Object is {0} with {1}% probability. Recognized in {2} milliseconds.", labels[maxIdx], maxVal * 100, sw.ElapsedMilliseconds);
}
if (InvokeRequired)
{
this.Invoke((MethodInvoker)(() =>
{
messageLabel.Text = resStr;
pictureBox.Image = m.Bitmap;
}));
}
else
{
messageLabel.Text = resStr;
pictureBox.Image = m.Bitmap;
}
}
public void Recognize(String fileName)
{
Tensor imageTensor = ImageIO.ReadTensorFromImageFile(fileName, 224, 224, 128.0f, 1.0f / 128.0f);
float[] probability;
if (_coldSession)
{
//First run of the recognition graph, here we will compile the graph and initialize the session
//This is expected to take much longer time than consecutive runs.
probability = inceptionGraph.Recognize(imageTensor);
_coldSession = false;
}
//Here we are trying to time the execution of the graph after it is loaded
//If we are not interest in the performance, we can skip the 3 lines that follows
Stopwatch sw = Stopwatch.StartNew();
probability = inceptionGraph.Recognize(imageTensor);
sw.Stop();
String resStr = String.Empty;
if (probability != null)
{
String[] labels = inceptionGraph.Labels;
float maxVal = 0;
int maxIdx = 0;
for (int i = 0; i < probability.Length; i++)
{
if (probability[i] > maxVal)
{
maxVal = probability[i];
maxIdx = i;
}
}
resStr = String.Format("Object is {0} with {1}% probability. Recognition done in {2} in {3} milliseconds.", labels[maxIdx], maxVal * 100, TfInvoke.IsGoogleCudaEnabled ? "GPU" : "CPU", sw.ElapsedMilliseconds);
}
if (InvokeRequired)
{
this.Invoke((MethodInvoker)(() =>
{
fileNameTextBox.Text = fileName;
pictureBox.ImageLocation = fileName;
messageLabel.Text = resStr;
}));
}
else
{
fileNameTextBox.Text = fileName;
pictureBox.ImageLocation = fileName;
messageLabel.Text = resStr;
}
}
partial void inceptionClicked(NSObject sender)
{
messageLabel.StringValue = "Inception Clicked";
mainImageView.Image = null;
Inception inceptionGraph = new Inception();
String fileName = "space_shuttle.jpg";
Tensor imageTensor = Emgu.TF.Models.ImageIO.ReadTensorFromImageFile(fileName, 224, 224, 128.0f, 1.0f / 128.0f);
//MultiboxGraph.Result detectResult = graph.Detect(imageTensor);
float[] probability = inceptionGraph.Recognize(imageTensor);
String resStr = String.Empty;
if (probability != null)
{
String[] labels = inceptionGraph.Labels;
float maxVal = 0;
int maxIdx = 0;
for (int i = 0; i < probability.Length; i++)
{
if (probability[i] > maxVal)
{
maxVal = probability[i];
maxIdx = i;
}
}
resStr = String.Format("Object is {0} with {1}% probability.", labels[maxIdx], maxVal * 100);
}
messageLabel.StringValue = resStr;
mainImageView.Image = new NSImage(fileName);
}
private void OnDownloadComplete(object sender, AsyncCompletedEventArgs e)
{
try
{
var stringResult = new StringBuilder();
using (var imageTensor = ImageIO.ReadTensorFromImageFile <float>(_fileName, 256, 256))
{
var results = _inceptionGraph.Recognize(imageTensor);
foreach (var recognitionResult in results.OrderByDescending(x => x.Probability).Take(5))
{
if (decimal.TryParse(recognitionResult.Probability.ToString(), out decimal resultResult))
{
stringResult.Append($"Object is {recognitionResult.Label} with {resultResult}% probability.")
.AppendLine();
}
}
}
InformationText = stringResult.ToString();
IsModalVisible = false;
}
catch (Exception ex)
{
_exceptionLogDataAccess.LogException(ex.ToString());
}
IsModalVisible = false;
}
async void inceptionClicked(NSObject sender)
{
SetMessage("Please wait while we download Inception model from internet...");
SetImage(null);
if (_inceptionGraph == null)
{
_inceptionGraph = new Inception();
_inceptionGraph.OnDownloadProgressChanged += OnDownloadProgressChanged;
//_inceptionGraph.OnDownloadCompleted += inceptionGraph_OnDownloadCompleted;
}
await _inceptionGraph.Init();
String fileName = "space_shuttle.jpg";
//Tensor imageTensor = Emgu.TF.Models.ImageIO.ReadTensorFromImageFile<float>(fileName, 224, 224, 128.0f, 1.0f / 128.0f);
Tensor imageTensor = Emgu.TF.Models.ImageIO.ReadTensorFromImageFile <float>(fileName, 224, 224, 128.0f, 1.0f);
var recognitionResults = _inceptionGraph.Recognize(imageTensor);
String resStr = String.Empty;
if (recognitionResults != null && recognitionResults.Length > 0)
{
resStr = String.Format("Object is {0} with {1}% probability.", recognitionResults[0].Label, recognitionResults[0].Probability * 100);
}
SetMessage(resStr);
SetImage(new NSImage(fileName));
}
private async void OnButtonClicked(Object sender, EventArgs args)
{
SetMessage("Please wait while the Inception Model is being downloaded...");
await _inception.Init();
if (!_inception.Imported)
{
SetMessage("Failed to initialize Inception Model.");
return;
}
SetImage();
String[] imageFiles = await LoadImages(new string[] { "tulips.jpg" });
//handle user cancel
if (imageFiles == null || (imageFiles.Length > 0 && imageFiles[0] == null))
{
SetMessage("");
return;
}
Stopwatch watch = Stopwatch.StartNew();
var result = _inception.Recognize(imageFiles[0]);
watch.Stop();
String resStr = String.Format("Object is {0} with {1}% probability. Recognition completed in {2} milliseconds.", result[0].Label, result[0].Probability * 100, watch.ElapsedMilliseconds);
SetImage(imageFiles[0]);
SetMessage(resStr);
}
private void RecognizeAndUpdateText(Texture2D texture)
{
if (_inceptionGraph == null)
{
return;
}
Tensor imageTensor = ImageIO.ReadTensorFromTexture2D(texture, 224, 224, 128.0f, 1.0f, true);
float[] probability = _inceptionGraph.Recognize(imageTensor);
//String resStr = String.Empty;
if (probability != null)
{
float maxVal = 0;
int maxIdx = 0;
for (int i = 0; i < probability.Length; i++)
{
if (probability[i] > maxVal)
{
maxVal = probability[i];
maxIdx = i;
}
}
DisplayText.text = String.Format("Object is {0} with {1}% probability.", _inceptionLabels[maxIdx], maxVal * 100);
}
}
private void onDownloadCompleted(object sender, System.ComponentModel.AsyncCompletedEventArgs e)
{
Stopwatch watch = Stopwatch.StartNew();
var result = _inception.Recognize(_imageFiles[0]);
watch.Stop();
String resStr = String.Format("Object is {0} with {1}% probability. Recognition completed in {2} milliseconds.", result[0].Label, result[0].Probability * 100, watch.ElapsedMilliseconds);
SetImage(_imageFiles[0]);
SetMessage(resStr);
}
private void RecognizeAndUpdateText(Texture2D texture)
{
if (!_inceptionGraph.Imported)
{
return;
}
Tensor imageTensor = ImageIO.ReadTensorFromTexture2D(texture, 224, 224, 128.0f, 1.0f, true);
Inception.RecognitionResult[] results = _inceptionGraph.Recognize(imageTensor);
_displayMessage = String.Format("Object is {0} with {1}% probability.", results[0].Label, results[0].Probability * 100);
}
private async void OnButtonClicked(Object sender, EventArgs args)
{
SetMessage("Please wait while the Inception Model is being downloaded...");
await _inception.Init();
SetImage();
String[] imageFiles = await LoadImages(new string[] { "tulips.jpg" });
Stopwatch watch = Stopwatch.StartNew();
var result = _inception.Recognize(imageFiles[0]);
watch.Stop();
String resStr = String.Format("Object is {0} with {1}% probability. Recognition completed in {2} milliseconds.", result[0].Label, result[0].Probability * 100, watch.ElapsedMilliseconds);
SetImage(imageFiles[0]);
SetMessage(resStr);
}
public async Task TestInceptionBatch()
{
//using (Tensor imageTensor = ImageIO.ReadTensorFromImageFile<float>("grace_hopper.jpg", 224, 224, 128.0f, 1.0f))
using (Tensor imageTensor = ImageIO.ReadTensorFromImageFiles <float>(
new String[] { "grace_hopper.jpg", "grace_hopper.jpg" },
224,
224,
128.0f,
1.0f))
using (Inception inceptionGraph = new Inception())
{
await inceptionGraph.Init();
Inception.RecognitionResult[][] results = inceptionGraph.Recognize(imageTensor);
Trace.WriteLine(String.Format("Object is {0} with {1}% probability", results[0][0].Label, results[0][0].Probability * 100));
}
}
public void Recognize(String fileName)
{
Tensor imageTensor = ImageIO.ReadTensorFromImageFile(fileName, 224, 224, 128.0f, 1.0f / 128.0f);
Stopwatch sw = Stopwatch.StartNew();
float[] probability = inceptionGraph.Recognize(imageTensor);
sw.Stop();
String resStr = String.Empty;
if (probability != null)
{
String[] labels = inceptionGraph.Labels;
float maxVal = 0;
int maxIdx = 0;
for (int i = 0; i < probability.Length; i++)
{
if (probability[i] > maxVal)
{
maxVal = probability[i];
maxIdx = i;
}
}
resStr = String.Format("Object is {0} with {1}% probability. Recognition done in {2} milliseconds.", labels[maxIdx], maxVal * 100, sw.ElapsedMilliseconds);
}
if (InvokeRequired)
{
this.Invoke((MethodInvoker)(() =>
{
fileNameTextBox.Text = fileName;
pictureBox.ImageLocation = fileName;
messageLabel.Text = resStr;
}));
}
else
{
fileNameTextBox.Text = fileName;
pictureBox.ImageLocation = fileName;
messageLabel.Text = resStr;
}
}
private void RecognizeAndUpdateText(Color32[] pixels, int width, int height)
{
if (_inceptionGraph == null)
{
return;
}
if (!_inceptionGraph.Imported)
{
return;
}
Inception.RecognitionResult[][] results;
using (Tensor imageTensor = ImageIO.ReadTensorFromColor32(pixels, width, height, 224, 224, 128.0f, 1.0f, true))
{
results = _inceptionGraph.Recognize(imageTensor);
}
_displayMessage = String.Format(
"Object is {0} with {1}% probability.",
results[0][0].Label,
results[0][0].Probability * 100);
}
public void Recognize(String fileName)
{
fileNameTextBox.Text = fileName;
pictureBox.ImageLocation = fileName;
//Use the following code for the full inception model
Inception inceptionGraph = new Inception();
Tensor imageTensor = ImageIO.ReadTensorFromImageFile(fileName, 224, 224, 128.0f, 1.0f / 128.0f);
//Uncomment the following code to use a retrained model to recognize followers, downloaded from the internet
//Inception inceptionGraph = new Inception(null, new string[] {"optimized_graph.pb", "output_labels.txt"}, "https://github.com/emgucv/models/raw/master/inception_flower_retrain/", "Mul", "final_result");
//Tensor imageTensor = ImageIO.ReadTensorFromImageFile(fileName, 299, 299, 128.0f, 1.0f / 128.0f);
//Uncomment the following code to use a retrained model to recognize followers, if you deployed the models with the application
//For ".pb" and ".txt" bundled with the application, set the url to null
//Inception inceptionGraph = new Inception(null, new string[] {"optimized_graph.pb", "output_labels.txt"}, null, "Mul", "final_result");
//Tensor imageTensor = ImageIO.ReadTensorFromImageFile(fileName, 299, 299, 128.0f, 1.0f / 128.0f);
float[] probability = inceptionGraph.Recognize(imageTensor);
String resStr = String.Empty;
if (probability != null)
{
String[] labels = inceptionGraph.Labels;
float maxVal = 0;
int maxIdx = 0;
for (int i = 0; i < probability.Length; i++)
{
if (probability[i] > maxVal)
{
maxVal = probability[i];
maxIdx = i;
}
}
resStr = String.Format("Object is {0} with {1}% probability.", labels[maxIdx], maxVal * 100);
}
messageLabel.Text = resStr;
}
public void TestInception()
{
Tensor imageTensor = ImageIO.ReadTensorFromImageFile("grace_hopper.jpg", 224, 224, 128.0f, 1.0f);
Inception inceptionGraph = new Inception();
float[] probability = inceptionGraph.Recognize(imageTensor);
if (probability != null)
{
String[] labels = inceptionGraph.Labels;
float maxVal = 0;
int maxIdx = 0;
for (int i = 0; i < probability.Length; i++)
{
if (probability[i] > maxVal)
{
maxVal = probability[i];
maxIdx = i;
}
}
Trace.WriteLine(String.Format("Object is {0} with {1}% probability", labels[maxIdx], maxVal * 100));
}
}
public InceptionPage(Model model)
: base()
{
Title = model == Model.Flower ? "Flower Recognition" : "Object recognition (Inception)";
_model = model;
if (_inceptionGraph == null)
{
SessionOptions so = new SessionOptions();
if (TfInvoke.IsGoogleCudaEnabled)
{
Tensorflow.ConfigProto config = new Tensorflow.ConfigProto();
config.GpuOptions = new Tensorflow.GPUOptions();
config.GpuOptions.AllowGrowth = true;
so.SetConfig(config.ToProtobuf());
}
_inceptionGraph = new Inception(null, so);
_inceptionGraph.OnDownloadProgressChanged += onDownloadProgressChanged;
_inceptionGraph.OnDownloadCompleted += onDownloadCompleted;
_inceptionGraph.OnDownloadCompleted += (sender, e) =>
{
OnButtonClicked(sender, e);
};
}
OnImagesLoaded += (sender, image) =>
{
#if !DEBUG
try
#endif
{
SetMessage("Please wait...");
SetImage();
Tensor imageTensor;
if (_model == Model.Flower)
{
imageTensor = Emgu.TF.Models.ImageIO.ReadTensorFromImageFile <float>(image[0], 299, 299, 0.0f, 1.0f / 255.0f, false, false);
}
else
{
imageTensor =
Emgu.TF.Models.ImageIO.ReadTensorFromImageFile <float>(image[0], 224, 224, 128.0f, 1.0f);
}
Inception.RecognitionResult result;
if (_coldSession)
{
//First run of the recognition graph, here we will compile the graph and initialize the session
//This is expected to take much longer time than consecutive runs.
result = _inceptionGraph.Recognize(imageTensor)[0];
_coldSession = false;
}
//Here we are trying to time the execution of the graph after it is loaded
//If we are not interest in the performance, we can skip the following 3 lines
Stopwatch sw = Stopwatch.StartNew();
result = _inceptionGraph.Recognize(imageTensor)[0];
sw.Stop();
String msg = String.Format("Object is {0} with {1}% probability. Recognized in {2} milliseconds.", result.Label, result.Probability * 100, sw.ElapsedMilliseconds);
SetMessage(msg);
var jpeg = Emgu.Models.NativeImageIO.ImageFileToJpeg(image[0]);
SetImage(jpeg.Raw, jpeg.Width, jpeg.Height);
}
#if !DEBUG
catch (Exception excpt)
{
String msg = excpt.Message.Replace(System.Environment.NewLine, " ");
SetMessage(msg);
}
#endif
};
}
public void TestInception()
{
using (Tensor imageTensor = ImageIO.ReadTensorFromImageFile <float>("grace_hopper.jpg", 224, 224, 128.0f, 1.0f))
using (Inception inceptionGraph = new Inception())
{
bool processCompleted = false;
inceptionGraph.OnDownloadCompleted += (sender, e) =>
{
HashSet <string> opNames = new HashSet <string>();
HashSet <string> couldBeInputs = new HashSet <string>();
HashSet <string> couldBeOutputs = new HashSet <string>();
foreach (Operation op in inceptionGraph.Graph)
{
String name = op.Name;
opNames.Add(name);
if (op.NumInputs == 0 && op.OpType.Equals("Placeholder"))
{
couldBeInputs.Add(op.Name);
AttrMetadata dtypeMeta = op.GetAttrMetadata("dtype");
AttrMetadata shapeMeta = op.GetAttrMetadata("shape");
DataType type = op.GetAttrType("dtype");
Int64[] shape = op.GetAttrShape("shape");
Buffer valueBuffer = op.GetAttrValueProto("shape");
Buffer shapeBuffer = op.GetAttrTensorShapeProto("shape");
Tensorflow.TensorShapeProto shapeProto =
Tensorflow.TensorShapeProto.Parser.ParseFrom(shapeBuffer.Data);
}
if (op.OpType.Equals("Const"))
{
AttrMetadata dtypeMeta = op.GetAttrMetadata("dtype");
AttrMetadata valueMeta = op.GetAttrMetadata("value");
using (Tensor valueTensor = op.GetAttrTensor("value"))
{
var dim = valueTensor.Dim;
}
}
if (op.OpType.Equals("Conv2D"))
{
AttrMetadata stridesMeta = op.GetAttrMetadata("strides");
AttrMetadata paddingMeta = op.GetAttrMetadata("padding");
AttrMetadata boolMeta = op.GetAttrMetadata("use_cudnn_on_gpu");
Int64[] strides = op.GetAttrIntList("strides");
bool useCudnn = op.GetAttrBool("use_cudnn_on_gpu");
String padding = op.GetAttrString("padding");
}
foreach (Output output in op.Outputs)
{
int[] shape = inceptionGraph.Graph.GetTensorShape(output);
if (output.NumConsumers == 0)
{
couldBeOutputs.Add(name);
}
}
Buffer buffer = inceptionGraph.Graph.GetOpDef(op.OpType);
Tensorflow.OpDef opDef = Tensorflow.OpDef.Parser.ParseFrom(buffer.Data);
}
using (Buffer versionDef = inceptionGraph.Graph.Versions())
{
int l = versionDef.Length;
}
Inception.RecognitionResult[] results = inceptionGraph.Recognize(imageTensor);
Trace.WriteLine(String.Format("Object is {0} with {1}% probability", results[0].Label, results[0].Probability * 100));
processCompleted = true;
};
inceptionGraph.Init();
while (!processCompleted)
{
Thread.Sleep(1000);
}
}
}
public void TestInception()
{
using (Tensor imageTensor = ImageIO.ReadTensorFromImageFile("grace_hopper.jpg", 224, 224, 128.0f, 1.0f))
using (Inception inceptionGraph = new Inception())
{
bool processCompleted = false;
inceptionGraph.OnDownloadCompleted += (sender, e) =>
{
HashSet <string> opNames = new HashSet <string>();
HashSet <string> couldBeInputs = new HashSet <string>();
HashSet <string> couldBeOutputs = new HashSet <string>();
foreach (Operation op in inceptionGraph.Graph)
{
String name = op.Name;
opNames.Add(name);
if (op.NumInputs == 0 && op.OpType.Equals("Placeholder"))
{
couldBeInputs.Add(op.Name);
}
foreach (Output output in op.Outputs)
{
int[] shape = inceptionGraph.Graph.GetTensorShape(output);
if (output.NumConsumers == 0)
{
couldBeOutputs.Add(name);
}
}
}
using (Buffer versionDef = inceptionGraph.Graph.Versions())
{
int l = versionDef.Length;
}
float[] probability = inceptionGraph.Recognize(imageTensor);
if (probability != null)
{
String[] labels = inceptionGraph.Labels;
float maxVal = 0;
int maxIdx = 0;
for (int i = 0; i < probability.Length; i++)
{
if (probability[i] > maxVal)
{
maxVal = probability[i];
maxIdx = i;
}
}
Trace.WriteLine(String.Format("Object is {0} with {1}% probability", labels[maxIdx], maxVal * 100));
}
processCompleted = true;
};
inceptionGraph.Init();
while (!processCompleted)
{
Thread.Sleep(1000);
}
}
}
public InceptionPage()
: base()
{
var button = this.GetButton();
button.Text = "Perform Image Recognition";
button.Clicked += OnButtonClicked;
OnImagesLoaded += async(sender, image) =>
{
SetMessage("Please wait...");
SetImage();
Task <Tuple <string, string, long> > t = new Task <Tuple <string, string, long> >(
() =>
{
try
{
SetMessage("Please wait while we download the Inception Model from internet.");
Inception inceptionGraph = new Inception();
SetMessage("Please wait...");
Tensor imageTensor = Emgu.TF.Models.ImageIO.ReadTensorFromImageFile(image[0], 224, 224, 128.0f, 1.0f / 128.0f);
//MultiboxGraph.Result detectResult = graph.Detect(imageTensor);
float[] probability = inceptionGraph.Recognize(imageTensor);
String resStr = String.Empty;
if (probability != null)
{
String[] labels = inceptionGraph.Labels;
float maxVal = 0;
int maxIdx = 0;
for (int i = 0; i < probability.Length; i++)
{
if (probability[i] > maxVal)
{
maxVal = probability[i];
maxIdx = i;
}
}
resStr = String.Format("Object is {0} with {1}% probability.", labels[maxIdx], maxVal * 100);
}
return(new Tuple <string, string, long>(image[0], resStr, 0));
//SetImage(t.Result.Item1);
//GetLabel().Text = String.Format("Detected {0} in {1} milliseconds.", t.Result.Item2, t.Result.Item3);
}
catch (Exception e)
{
String msg = e.Message.Replace(System.Environment.NewLine, " ");
SetMessage(msg);
return(new Tuple <string, string, long>(null, msg, 0));
}
});
t.Start();
#if !(__UNIFIED__)
var result = await t;
SetImage(t.Result.Item1);
GetLabel().Text = t.Result.Item2;
#endif
};
}
