protected void PredictDataUsingModel(string imagesLocation,
string imagesFolder,
PredictionEngine <ImageNetData, ImageNetPrediction> model)
{
Console.WriteLine($"Tags file location: {imagesLocation}");
Console.WriteLine("");
Console.WriteLine("=====Identify the objects in the images=====");
Console.WriteLine("");
var testData = ImageNetData.ReadFromCsv(imagesLocation, imagesFolder);
foreach (var sample in testData)
{
var probs = model.Predict(sample).PredictedLabels;
IList <YoloBoundingBox> boundingBoxes = _parser.ParseOutputs(probs);
var filteredBoxes = _parser.NonMaxSuppress(boundingBoxes, 5, .5F);
Console.WriteLine(".....The objects in the image {0} are detected as below....", sample.Label);
foreach (var box in filteredBoxes)
{
Console.WriteLine(box.Label + " and its Confidence score: " + box.Confidence);
}
Console.WriteLine("");
}
}
static void Main(string[] args)
{
var mL = new MLContext();
try
{
IEnumerable <ImageNetData> images = ImageNetData.ReadFromFile(imagesFolder);
IDataView imageDataView = mL.Data.LoadFromEnumerable(images);
var modelScorer = new OnnxModelScorer(imagesFolder, modelPath, mL);
IEnumerable <float[]> probabilities = modelScorer.Score(imageDataView);
YoloOutputParser parser = new YoloOutputParser();
var boundingBoxes = probabilities
.Select(x => parser.ParseOutputs(x))
.Select(x => parser.FilterBoundingBoxes(x, 5, .5f));
for (var i = 0; i < images.Count(); i++)
{
string imageFileName = images.ElementAt(i).Label;
IList <YoloBoundingBox> detectedObjects = boundingBoxes.ElementAt(i);
DrawBoundingBox(imagesFolder, outputFolder, imageFileName, detectedObjects);
LogDetectedObjects(imageFileName, detectedObjects);
}
}
catch (Exception ex)
{
Console.WriteLine(ex.ToString());
}
}
public static void Main()
{
var assetsRelativePath = @"../../../assets";
var assetsPath = GetAbsolutePath(assetsRelativePath);
var modelFilePath = Path.Combine(assetsPath, "Model", "TinyYolo2_model.onnx");
//var modelFilePath = Path.Combine(assetsPath, "Model", "Yolov3.onnx");
//var modelFilePath = Path.Combine(assetsPath, "Model", "model.onnx");
var imagesFolder = Path.Combine(assetsPath, "people");
var outputFolder = Path.Combine(imagesFolder, "output");
// Initialize MLContext
var mlContext = new MLContext();
try
{
// Load Data
var images = ImageNetData.ReadFromFile(imagesFolder);
var imageDataView = mlContext.Data.LoadFromEnumerable(images);
// Create instance of model scorer
var modelScorer = new Scorer(imagesFolder, modelFilePath, mlContext);
// Use model to score data
var probabilities = modelScorer.Score(imageDataView);
// Post-process model output
var parser = new OutputParser();
var boundingBoxes =
probabilities
.Select(probability => parser.ParseOutputs(probability))
.Select(boxes => parser.FilterBoundingBoxes(boxes, 5, .5F))
.ToList()
;
// Draw bounding boxes for detected objects in each of the images
for (var i = 0; i < images.Count(); i++)
{
var imageFileName = images.ElementAt(i).Label;
var detectedObjects = boundingBoxes.ElementAt(i);
DrawBoundingBox(imagesFolder, outputFolder, imageFileName, detectedObjects);
LogDetectedObjects(imageFileName, detectedObjects);
}
}
catch (Exception ex)
{
Console.WriteLine(ex.ToString());
}
Console.WriteLine("========= End of Process..Hit any Key ========");
Console.ReadLine();
}
private static IEnumerable <ImageNetData> GetImagesData(string folder)
{
List <ImageNetData> imagesList = new List <ImageNetData>();
string[] filePaths = Directory.GetFiles(folder);
foreach (var filePath in filePaths)
{
ImageNetData imagedata = new ImageNetData {
ImagePath = filePath, Label = Path.GetFileName(filePath)
};
imagesList.Add(imagedata);
}
return(imagesList);
}
static void Main(string[] args)
{
var assetsRelativePath = @"../../../assets";
string assetsPath = GetAbsolutePath(assetsRelativePath);
var modelFilePath = Path.Combine(assetsPath, "Model", "TinyYolo2_model.onnx");
var imagesFolder = Path.Combine(assetsPath, "images");
var outputFolder = Path.Combine(assetsPath, "images", "output");
MLContext mlContext = new MLContext();
try
{
IEnumerable <ImageNetData> images = ImageNetData.ReadFromFile(imagesFolder);
IDataView imageDataView = mlContext.Data.LoadFromEnumerable(images);
var modelScorer = new OnnxModelScorer(imagesFolder, modelFilePath, mlContext);
// Use model to score data
IEnumerable <float[]> probabilities = modelScorer.Score(imageDataView);
YoloOutputParser parser = new YoloOutputParser();
var boundingBoxes =
probabilities
.Select(probability => parser.ParseOutputs(probability))
.Select(boxes => parser.FilterBoundingBoxes(boxes, 5, .5F));
for (var i = 0; i < images.Count(); i++)
{
string imageFileName = images.ElementAt(i).Label;
IList <YoloBoundingBox> detectedObjects = boundingBoxes.ElementAt(i);
DrawBoundingBox(imagesFolder, outputFolder, imageFileName, detectedObjects);
LogDetectedObjects(imageFileName, detectedObjects);
}
}
catch (Exception ex)
{
Console.WriteLine(ex.ToString());
}
Console.WriteLine("========= End of Process..Hit any Key ========");
Console.ReadLine();
}
public ITransformer LoadModel()
{
// Create IDataView from empty list to obtain input data schema
ImageNetData[] inMemoryCollection = new ImageNetData[]
{
new ImageNetData
{
InputImage = null,
Label = ""
}
};
var data = mlContext.Data.LoadFromEnumerable <ImageNetData>(inMemoryCollection);
// Define scoring pipeline
var pipeline = mlContext.Transforms.ResizeImages(outputColumnName: "image", imageWidth: ImageNetSettings.imageWidth, imageHeight: ImageNetSettings.imageHeight, inputColumnName: "InputImage")
.Append(mlContext.Transforms.ExtractPixels(outputColumnName: "image"))
.Append(mlContext.Transforms.ApplyOnnxModel(modelFile: modelLocation, outputColumnNames: new[] { TinyYoloModelSettings.ModelOutput }, inputColumnNames: new[] { TinyYoloModelSettings.ModelInput }));
// Fit scoring pipeline
var model = pipeline.Fit(data);
return(model);
}
private void btnStart_Click(object sender, EventArgs e)
{
// Frame image buffer
Mat image = new Mat();
_isRunning = true;
btnStart.Enabled = false;
var mlContext = new MLContext();
// Create instance of model scorer
var modelScorer = new OnnxModelScorer(modelFilePath, mlContext);
// Load model only once
var model = modelScorer.LoadModel();
while (_isRunning)
{
_capture.Read(image); // read frame from webcam
if (image.Empty())
{
break;
}
// Store frame as in-memory source for ML.NET
ImageNetData[] inMemoryCollection = new ImageNetData[]
{
new ImageNetData
{
InputImage = OpenCvSharp.Extensions.BitmapConverter.ToBitmap(image),
Label = "",
}
};
var imageDataView = mlContext.Data.LoadFromEnumerable <ImageNetData>(inMemoryCollection);
// Make another copy of the frame. We will use it to draw bounding boxes on it
_mySharpImage = (Image)OpenCvSharp.Extensions.BitmapConverter.ToBitmap(image);
// Use model to score data
IEnumerable <float[]> probabilities = modelScorer.Score(model, imageDataView);
// Post-process model output
YoloOutputParser parser = new YoloOutputParser();
var boundingBoxes =
probabilities
.Select(probability => parser.ParseOutputs(probability))
.Select(boxes => parser.FilterBoundingBoxes(boxes, 5, .5F));
// Since we only have a single frame, it is OK to have i = 0. Otherwise we would need
// to iterate through images.
var i = 0;
IList <YoloBoundingBox> detectedObjects = boundingBoxes.ElementAt(i);
DrawBoundingBox(ref _mySharpImage, detectedObjects);
pictureBox1.SizeMode = PictureBoxSizeMode.StretchImage;
pictureBox1.Image = _mySharpImage;
Cv2.WaitKey(1);
_mySharpImage.Dispose();
inMemoryCollection[0].InputImage.Dispose();
}
btnStart.Enabled = true;
}
