// Generate the feature vector of the given image.
public double[] GenerateFeature(short[,] pixelData)
{
// Initialize the parameters of feature (data).
MWStructArray fea_first = new MWStructArray(1, 1, new string[] { "pixels", "maxsize" });
fea_first.SetField("pixels", new MWNumericArray(pixelData));
fea_first.SetField("maxsize", 75);
// Initialize the parameters of dictionary.
MWStructArray dic_first = new MWStructArray(1, 1, new string[] { "dicsize", "patchsize", "samplenum", "dic" });
dic_first.SetField("dicsize", 200);
dic_first.SetField("patchsize", 16);
dic_first.SetField("samplenum", 100);
MWArray array = new MWNumericArray(this.dict_dimension[0], this.dict_dimension[1], MatrixUtil.FlattenMatrix(this.dic, this.dict_dimension[0], this.dict_dimension[1]));
dic_first.SetField("dic", array);
// Orthogonal matching pursuit encoder
Stopwatch ompTimer = Stopwatch.StartNew();
MWArray rgbdfea = ompNormal.extract_feature_normal(fea_first, dic_first);
ompTimer.Stop();
Array feature = rgbdfea.ToArray();
Console.WriteLine("Feature Extraction Time: {0} ms", ompTimer.ElapsedMilliseconds);
double[] feature_vector = new double[feature.Length];
for (int i = 0; i < feature.Length; i++)
{
feature_vector[i] = System.Convert.ToDouble(feature.GetValue(i, 0));
}
return(feature_vector);
}
