feat: amine content analyzer changes

add filtering with Mahalanobis
fix some typos
also fix it after database reforge

GSS2.Core/Analysis/AmineContent/AnalyzerService.cs

@@ -5,7 +5,10 @@ using Microsoft.ML.Data;
 using OpenCvSharp;
 
 using GSS2.Core.Extensions;
-using GSS2.Core.Analysis.AmineContent.Database.Calibration;
+using GSS2.Core.Analysis.AmineContent.Database;
+
+using MathNet.Numerics.LinearAlgebra;
+using MathNet.Numerics.LinearAlgebra.Double;
 
 namespace GSS2.Core.Analysis.AmineContent;
 
@@ -29,6 +32,8 @@ public partial class AnalyzerService
     public const int SEPARATOR_WIDTH = 2969;
     public const int SEPARATOR_HEIGHT = 2965;
 
+    public const double FILTERING_THRESHOLD = 5.0;
+
     private class ClusterizationData
     {
         [ColumnName("Label")]
@@ -72,7 +77,7 @@ public partial class AnalyzerService
     }
 
     private readonly ILogger<AnalyzerService> _logger;
-    private readonly CalibrationContext _calibrationContext;
+    private readonly Context _context;
 
     private MLContext? _ml = null;
     private PredictionEngine<ClusterizationData, ClusterizationPrediction>? _clusterizationEngine = null;
@@ -80,13 +85,13 @@ public partial class AnalyzerService
 
     public bool Initialized { get; private set; } = false;
 
-    public AnalyzerService(ILogger<AnalyzerService> logger, CalibrationContext calibrationContext)
+    public AnalyzerService(ILogger<AnalyzerService> logger, Context Context)
     {
         _logger = logger;
-        _calibrationContext = calibrationContext;
+        _context = Context;
     }
 
-    public async Task Initialize(string? brand = null, CancellationToken cancellationToken = default)
+    public async Task Initialize(BrandRecord brand, SeparatorRecord separator, CancellationToken cancellationToken = default)
     {
         Initialized = false;
 
@@ -101,7 +106,7 @@ public partial class AnalyzerService
 
         try
         {
-            await InitializeMlClusterizationEngine(cancellationToken);
+            await InitializeMlClusterizationEngine(brand, separator, cancellationToken);
             await InitializeMlRegressionEngine(brand, cancellationToken);
         }
         catch (Exception ex)
@@ -112,83 +117,134 @@ public partial class AnalyzerService
 
         Initialized = true;
     }
-    private async Task InitializeMlClusterizationEngine(CancellationToken cancellationToken)
+    private async Task InitializeMlClusterizationEngine(BrandRecord brand, SeparatorRecord separator, CancellationToken cancellationToken)
     {
-        _logger.LogInformation("Инициализация модели кластеризации");
+        _logger.LogInformation("Инициализация модели кластеризации для марки: \"{} - {}\" и сепаратора: \"{}\"", brand.BrandName, brand.MixtureName, separator.Type);
-
-        var separatorPrefix = "SEPARATOR";
-        var samplePrefix = "SAMPLE";
 
         if (_ml is null)
-            throw new InvalidOperationException();
+            throw new InvalidOperationException("Анализатор не инициализирован");
+        cancellationToken.ThrowIfCancellationRequested();
 
+        _logger.LogInformation("Подготовка данных");
+        var sampleRecordIds = _context.SampleRecords
+            .Where(r => r.BrandId == brand.Id)
+            .Select(r => r.Id);
+        _logger.LogInformation("Найдено калибровочных записей проб для марки: {}", sampleRecordIds.Count());
+        if (sampleRecordIds.Count() == 0)
+            throw new InvalidDataException($"Не найдено калибровочных записей для марки: \"{brand.BrandName} - {brand.MixtureName}\"");
+        cancellationToken.ThrowIfCancellationRequested();
+
+        var separatorDataSet = _context.FeatureRecords
+            .Where(r => r.SeparatorRecordId == separator.Id)
+            .Select(r => new ClusterizationData
+            {
+                Label = "separator",
+                Features = r.Values.ToArray()
+            });
+        _logger.LogInformation("Загружено векторов для сепаратора: {}", separatorDataSet.Count());
+        if (separatorDataSet.Count() == 0)
+            throw new InvalidDataException($"Не найдено калибровочных данных для сепаратора: \"{separator.Type}\"");
+        cancellationToken.ThrowIfCancellationRequested();
+
+        var sampleDataSet = _context.FeatureRecords
+            .Where(r => r.SeparatorRecordId != 0)
+            .Where(r => sampleRecordIds.Contains(r.SampleRecordId))
+            .Select(r => new ClusterizationData
+            {
+                Label = "sample",
+                Features = r.Values.ToArray()
+            })
+            .Shuffle();
+        _logger.LogInformation("Загружено векторов для пробы: {}", sampleDataSet.Count());
+        if (separatorDataSet.Count() == 0)
+            throw new InvalidDataException($"Не найдено калибровочных данных для марки: \"{brand.BrandName} - {brand.MixtureName}\"");
+        cancellationToken.ThrowIfCancellationRequested();
+
+        var trainDataSet = Enumerable.Concat(separatorDataSet, sampleDataSet);
+        _logger.LogInformation("Загружено векторов для обучения: {}", trainDataSet.Count());
+        var trainData = _ml.Data.LoadFromEnumerable(trainDataSet);
         cancellationToken.ThrowIfCancellationRequested();
 
         _logger.LogInformation("Создание конвейера");
         var pipeline = _ml.Transforms.Conversion.MapValueToKey("Label")
             .Append(_ml.MulticlassClassification.Trainers.LbfgsMaximumEntropy())
             .Append(_ml.Transforms.Conversion.MapKeyToValue("PredictedLabel"));
-
-        cancellationToken.ThrowIfCancellationRequested();
-
-        _logger.LogInformation("Подготовка данных");
-        var trainDataSet = _calibrationContext.VectorRecords.Select(r =>
-            new ClusterizationData
-            {
-                Label = r.SeparatorRecord != null ?
-                        $"{separatorPrefix}|{r.SeparatorRecord.Type}" :
-                        r.SampleRecord != null ?
-                        $"{samplePrefix}|{r.SampleRecord.SampleBrand}" :
-                        "",
-                Features = r.Values
-            })
-            .ToList();
-        trainDataSet = trainDataSet.Shuffle().ToList();
-        _logger.LogInformation("Записей для обучения: {}", trainDataSet.Count());
-        var trainData = _ml.Data.LoadFromEnumerable(trainDataSet);
-
         cancellationToken.ThrowIfCancellationRequested();
 
         _logger.LogInformation("Обучение модели");
         var model = pipeline.Fit(trainData);
+        cancellationToken.ThrowIfCancellationRequested();
 
         _clusterizationEngine = _ml.Model.CreatePredictionEngine<ClusterizationData, ClusterizationPrediction>(model);
         model.Dispose();
     }
-    private async Task InitializeMlRegressionEngine(string? brand, CancellationToken cancellationToken)
+    private async Task InitializeMlRegressionEngine(BrandRecord brand, CancellationToken cancellationToken)
     {
-        _logger.LogInformation("Инициализация модели регрессии для марки \"{}\"", brand);
+        _logger.LogInformation("Инициализация модели регрессии для марки \"{} - {}\"", brand.BrandName, brand.MixtureName);
 
         if (_ml is null)
-            throw new InvalidOperationException();
+            throw new InvalidOperationException("Анализатор не инициализирован");
-
-        _logger.LogInformation("Создание конвейера");
-
-        var pipeline = _ml.Regression.Trainers.FastTree();
-
         cancellationToken.ThrowIfCancellationRequested();
 
         _logger.LogInformation("Подготовка данных");
-        var trainDataSet = _calibrationContext.VectorRecords
+        var sampleRecordIds = _context.SampleRecords
-            .Where(r => r.SampleRecord != null)
+            .Where(r => r.BrandId == brand.Id)
-            .Where(r => brand == null || r.SampleRecord!.SampleBrand == brand)
+            .Select(r => r.Id);
+        _logger.LogInformation("Найдено калибровочных записей проб для марки: {}", sampleRecordIds.Count());
+        if (sampleRecordIds.Count() == 0)
+            throw new InvalidDataException($"Не найдено калибровочных записей для марки: \"{brand.BrandName} - {brand.MixtureName}\"");
+        cancellationToken.ThrowIfCancellationRequested();
+
+        _logger.LogInformation("Подготовка данных");
+        var trainDataSet = _context.FeatureRecords
+            .Where(r => r.SeparatorRecordId != 0)
+            .Where(r => sampleRecordIds.Contains(r.SampleRecordId))
+            .ToList()
             .Select(r =>
-                new RegressionData
+            {
+                var sample = _context.SampleRecords.First(s => s.Id == r.SampleRecordId);
+                return new RegressionData
                 {
                     Value = (float)(
-                        r.SampleRecord!.MeasuredContent >= 0 ?
+                        sample.MeasuredContent >= 0 ? sample.MeasuredContent :
-                        r.SampleRecord!.MeasuredContent :
+                        sample.MixtureActualRate
-                        r.SampleRecord!.MixtureActualRate >= 0 ?
-                        r.SampleRecord!.MixtureActualRate :
-                        r.SampleRecord!.MixtureNormalRate
                     ),
-                    Features = r.Values
+                    Features = r.Values.ToArray()
-                })
+                };
-            .ToList();
+            })
-        trainDataSet = trainDataSet.Shuffle().ToList();
+            .Shuffle();
-        _logger.LogInformation("Записей для обучения: {}", trainDataSet.Count());
+        _logger.LogInformation("Загружено векторов для обучения: {}", trainDataSet.Count());
-        var trainData = _ml.Data.LoadFromEnumerable(trainDataSet);
+        if (trainDataSet.Count() == 0)
+            throw new InvalidDataException($"Не найдено калибровочных данных для марки: \"{brand.BrandName} - {brand.MixtureName}\"");
+        cancellationToken.ThrowIfCancellationRequested();
 
+        int rowCount = trainDataSet.Count();
+        double[][] featuresArray = trainDataSet.Select(d => d.Features.Select(f => (double)f).ToArray()).ToArray();
+        Matrix<double> matrix = DenseMatrix.OfRowArrays(featuresArray);
+        for (int i = 0; i < IMAGES_COUNT; i++)
+        {
+            matrix = matrix.RemoveColumn(i * 3);
+            matrix = matrix.RemoveColumn(i * 3);
+            matrix = matrix.RemoveColumn(i * 3);
+        }
+        cancellationToken.ThrowIfCancellationRequested();
+
+        Vector<double> meanVector = matrix.ColumnSums() / matrix.RowCount;
+        Matrix<double> covarianceMatrix = MatrixCovariance(matrix, meanVector);
+        Matrix<double> invCovarianceMatrix = covarianceMatrix.Inverse();
+        trainDataSet = trainDataSet.Where((data, index) =>
+        {
+            var row = matrix.Row(index);
+            double distance = CalculateMahalanobis(row, meanVector, invCovarianceMatrix);
+
+            return distance <= FILTERING_THRESHOLD;
+        }).ToList();
+        _logger.LogInformation("Векторов для обучения после фильтрации: {}", trainDataSet.Count());
+        var trainData = _ml.Data.LoadFromEnumerable(trainDataSet);
+        cancellationToken.ThrowIfCancellationRequested();
+
+        _logger.LogInformation("Создание конвейера");
+        var pipeline = _ml.Regression.Trainers.FastTree();
         cancellationToken.ThrowIfCancellationRequested();
 
         _logger.LogInformation("Обучение модели");
@@ -311,9 +367,9 @@ public partial class AnalyzerService
             // var scale = oldMean / newMean;
             // Cv2.ConvertScaleAbs(result, result, scale);
 
-            // Добавляем эрозию к результатм и маске, удаляющую края, и пересчитываем контура
+            // Добавляем эрозию к результатам и маске, удаляющую края, и пересчитываем контура
-            // поскольку края грунул подсвечиваются близлежайщими гранулами и сепаратором,
+            // поскольку края гранул подсвечиваются близлежащими гранулами и сепаратором,
-            // их нельзя считаль представительными
+            // их нельзя считать представительными
             Cv2.Erode(result, result, Cv2.GetStructuringElement(MorphShapes.Ellipse, new Size(3, 3)));
             Cv2.Erode(mask, mask, Cv2.GetStructuringElement(MorphShapes.Ellipse, new Size(3, 3)));
             Cv2.FindContours(mask, out var newContours, out _, RetrievalModes.List, ContourApproximationModes.ApproxSimple);
@@ -588,4 +644,34 @@ public partial class AnalyzerService
         return CONTOUR_MIN_SIZE < bbox.Size.Width && CONTOUR_MIN_SIZE < bbox.Size.Height && CONTOUR_MIN_AREA < area &&
                CONTOUR_MAX_SIZE > bbox.Size.Width && CONTOUR_MAX_SIZE > bbox.Size.Height && CONTOUR_MAX_AREA > area;
     }
+
+    // Формула расстояния Махаланобиса: d = sqrt((x - mu)^T * Sigma^-1 * (x - mu))
+    private static double CalculateMahalanobis(Vector<double> x, Vector<double> mean, Matrix<double> invCovariance)
+    {
+        var diff = x - mean;
+        var distanceSq = diff * invCovariance * diff;
+        return Math.Sqrt(distanceSq);
+    }
+
+    // Ручной расчет матрицы ковариации (для экономии памяти и контроля)
+    private static Matrix<double> MatrixCovariance(Matrix<double> matrix, Vector<double> meanRows)
+    {
+        int rows = matrix.RowCount;
+        int cols = matrix.ColumnCount;
+        var cov = new DenseMatrix(cols, cols);
+
+        for (int i = 0; i < cols; i++)
+            for (int j = i; j < cols; j++)
+            {
+                double sum = 0;
+                for (int k = 0; k < rows; k++)
+                {
+                    sum += (matrix[k, i] - meanRows[i]) * (matrix[k, j] - meanRows[j]);
+                }
+                double val = sum / (rows - 1);
+                cov[i, j] = val;
+                cov[j, i] = val;
+            }
+        return cov;
+    }
 }

GSS2.Core/GSS2.Core.csproj

@@ -7,6 +7,7 @@
   </PropertyGroup>
 
   <ItemGroup>
+    <PackageReference Include="MathNet.Numerics" Version="5.0.0" />
     <PackageReference Include="Microsoft.Extensions.Hosting" Version="10.0.0" />
     <PackageReference Include="Microsoft.Data.SqlClient" Version="6.1.3" />
     <PackageReference Include="Microsoft.Data.Sqlite" Version="10.0.0" />
@@ -18,16 +19,19 @@
     <PackageReference Include="Microsoft.EntityFrameworkCore.Sqlite" Version="10.0.0" />
     <PackageReference Include="Microsoft.Extensions.DependencyInjection" Version="10.0.0" />
     <PackageReference Include="Microsoft.ML" Version="5.0.0" />
+    <PackageReference Include="Microsoft.ML.FastTree" Version="5.0.0" />
     <PackageReference Include="OpenCvSharp4" Version="4.11.0.20250507" />
     <PackageReference Include="OpenCvSharp4.runtime.linux-arm" Version="4.11.0.20250506" />
     <PackageReference Include="System.Device.Gpio" Version="4.0.1" />
     <PackageReference Include="Iot.Device.Bindings" Version="4.0.1" />
-    <PackageReference Include="LibCameraSharp" Version="0.5.2-20260216-4" />
+    <PackageReference Include="System.Security.Cryptography.Xml" Version="9.0.15" />
+    <PackageReference Include="LibCameraSharp" Version="0.5.2-20260615" />
   </ItemGroup>
 
   <ItemGroup>
     <ProjectReference Include="..\GSS2.LightsControl\CSharpClient\GSS2.LightsControl.CSharpClient.csproj" />
     <ProjectReference Include="..\GSS2.FrameDecoders\GSS2.FrameDecoders.csproj" />
+    <!-- <ProjectReference Include="..\..\LibCameraSharp\LibCameraSharp\LibCameraSharp.csproj" /> -->
   </ItemGroup>
 
 </Project>