private void TeachThread(object parameter)
{
TeachThreadParameter input = parameter as TeachThreadParameter;
if (input == null)
{
throw new ArgumentNullException("Неправильный параметр потока.");
}
var delta = 1.0 / filesToTeach.Count;
for (int i = input.From; (i < input.To) && (!isTeachingAborted); ++i)
{
try
{
var image = new SimpleImage(imagesToTeach[i]);
var imageData = image.GetGrayData();
var lbp = new LBPCreator(imageData, true, true, false);
var glcm = new GLCMCreator(imageData, true, true, false);
AddFeatures(glcm.Feature, lbp.Feature, filesToTeach[i]);
}
catch (Exception ex)
{
MessageBox.Show(ex.Message);
continue;
}
finally
{
lock ((knownFiles as ICollection).SyncRoot)
{
teachProgress += delta;
}
}
}
input.CommonBarrier.SignalAndWait();
}
public SplitImageThreadParameter(SimpleImage image, int from, int to, List<Rectangle> regions,
List<TextureSample> samples, ManualResetEvent resetEvent)
{
Image = image;
From = from;
To = to;
ResultRegions = regions;
ResultSamples = samples;
ResetEvent = resetEvent;
}
private void SplitImage(SimpleImage image, List<TextureSample> samples, List<Rectangle> regions)
{
int YCount = image.Height / RecognitionParameters.RecognitionFragmentSize;
int threadCount = Math.Min(YCount, Environment.ProcessorCount * 2);
int threadPart = YCount / threadCount;
var resetEvents = new ManualResetEvent[threadCount + 1];
var threadParameters = new SplitImageThreadParameter[threadCount + 1];
for (int i = 0; i < threadCount; ++i)
{
resetEvents[i] = new ManualResetEvent(false);
threadParameters[i] = new SplitImageThreadParameter(
image, threadPart * i, ((i + 1) < threadCount) ? (threadPart * (i + 1)) : (YCount),
regions, samples, resetEvents[i]);
ThreadPool.QueueUserWorkItem(new WaitCallback(SplitImageThread), threadParameters[i]);
}
resetEvents[threadCount] = new ManualResetEvent(false);
threadParameters[threadCount] = new SplitImageThreadParameter(
image, 0, 0, regions, samples, resetEvents[threadCount]);
ThreadPool.QueueUserWorkItem(new WaitCallback(SplitImageAdditionalThread), threadParameters[threadCount]);
WaitHandle.WaitAll(resetEvents);
}
public List<RecognitionResult> RecognizeImages(List<Bitmap> images)
{
recognitionProgress = 0;
isRecognizing = true;
var answers = new List<RecognitionResult>();
for (int i = 0; i < images.Count; ++i)
{
var imageToRecognize = new SimpleImage(images[i]);
var imageData = imageToRecognize.GetGrayData(0, 0, imageToRecognize.Width, imageToRecognize.Height);
var sample = new TextureSample(imageData, false);
answers.Add(GetBestMatch(sample, new Rectangle(0, 0, imageData.Width, imageData.Height)));
recognitionProgress += 1.0 * i / images.Count;
}
recognitionProgress = 1;
isRecognizing = false;
return answers;
}
public RecognitionResult RecognizeImage(Bitmap image, Rectangle region)
{
recognitionProgress = 0;
isRecognizing = true;
var imageToRecognize = new SimpleImage(image);
recognitionProgress = 0.2;
var imageData = imageToRecognize.GetGrayData(region.X, region.Y,
region.X + region.Width, region.Y + region.Height);
var sample = new TextureSample(imageData, false);
recognitionProgress = 0.8;
var answer = GetBestMatch(sample, region);
recognitionProgress = 1;
isRecognizing = false;
return answer;
}
public List<RecognitionResult> RecognizeImage(Bitmap image)
{
recognitionProgress = 0;
isRecognizing = true;
var imageToRecognize = new SimpleImage(image);
recognitionProgress = 0.05;
var regions = new List<Rectangle>();
var samples = new List<TextureSample>();
SplitImage(imageToRecognize, samples, regions);
var matches = new List<RecognitionResult>();
for (int i = 0; i < samples.Count; ++i)
{
var answer = GetBestMatch(samples[i], regions[i]);
matches.Add(answer);
recognitionProgress = 0.95 + (0.05 * i) / samples.Count;
}
recognitionProgress = 1;
isRecognizing = false;
return matches;
}
