move image extraction logic from Octave to Dataset class

apps/sft/include/sft/octave.hpp

@@ -46,17 +46,70 @@
 #include <opencv2/ml/ml.hpp>
 #include <sft/common.hpp>
 
+#include <opencv2/imgproc/imgproc.hpp>
+#include <opencv2/highgui/highgui.hpp>
 namespace sft
 {
 
-class Dataset
+class Preprocessor
 {
 public:
-    Dataset(const sft::string& path, const int octave);
+    Preprocessor() {}
 
-// private:
-    svector pos;
-    svector neg;
+    void apply(const cv::Mat& frame, cv::Mat& integrals) const
+    {
+        CV_Assert(frame.type() == CV_8UC3);
+
+        int h = frame.rows;
+        int w = frame.cols;
+
+        cv::Mat channels, gray;
+
+        channels.create(h * BINS, w, CV_8UC1);
+        channels.setTo(0);
+
+        cvtColor(frame, gray, CV_BGR2GRAY);
+
+        cv::Mat df_dx, df_dy, mag, angle;
+        cv::Sobel(gray, df_dx, CV_32F, 1, 0);
+        cv::Sobel(gray, df_dy, CV_32F, 0, 1);
+
+        cv::cartToPolar(df_dx, df_dy, mag, angle, true);
+        mag *= (1.f / (8 * sqrt(2.f)));
+
+        cv::Mat nmag;
+        mag.convertTo(nmag, CV_8UC1);
+
+        angle *=  6 / 360.f;
+
+        for (int y = 0; y < h; ++y)
+        {
+            uchar* magnitude = nmag.ptr<uchar>(y);
+            float* ang = angle.ptr<float>(y);
+
+            for (int x = 0; x < w; ++x)
+            {
+                channels.ptr<uchar>(y + (h * (int)ang[x]))[x] = magnitude[x];
+            }
+        }
+
+        cv::Mat luv, shrunk;
+        cv::cvtColor(frame, luv, CV_BGR2Luv);
+
+        std::vector<cv::Mat> splited;
+        for (int i = 0; i < 3; ++i)
+            splited.push_back(channels(cv::Rect(0, h * (7 + i), w, h)));
+        split(luv, splited);
+
+        float shrinkage = static_cast<float>(integrals.cols - 1) / channels.cols;
+
+        CV_Assert(shrinkage == 0.25);
+
+        cv::resize(channels, shrunk, cv::Size(), shrinkage, shrinkage, CV_INTER_AREA);
+        cv::integral(shrunk, integrals, cv::noArray(), CV_32S);
+    }
+
+    enum {BINS = 10};
 };
 
 struct ICF
@@ -74,7 +127,7 @@ struct ICF
     }
 
 
-    float operator() (const Mat& integrals, const cv::Size& model) const
+    float operator() (const cv::Mat& integrals, const cv::Size& model) const
     {
         int step = model.width + 1;
 
@@ -95,11 +148,11 @@ private:
     cv::Rect bb;
     int channel;
 
-    friend void write(cv::FileStorage& fs, const string&, const ICF& f);
+    friend void write(cv::FileStorage& fs, const std::string&, const ICF& f);
     friend std::ostream& operator<<(std::ostream& out, const ICF& f);
 };
 
-void write(cv::FileStorage& fs, const string&, const ICF& f);
+void write(cv::FileStorage& fs, const std::string&, const ICF& f);
 std::ostream& operator<<(std::ostream& out, const ICF& m);
 
 class ICFFeaturePool : public cv::FeaturePool
@@ -108,7 +161,8 @@ public:
     ICFFeaturePool(cv::Size model, int nfeatures);
 
     virtual int size() const { return (int)pool.size(); }
-    virtual float apply(int fi, int si, const Mat& integrals) const;
+    virtual float apply(int fi, int si, const cv::Mat& integrals) const;
+    virtual void preprocess(const cv::Mat& frame, cv::Mat& integrals) const;
     virtual void write( cv::FileStorage& fs, int index) const;
 
     virtual ~ICFFeaturePool();
@@ -124,12 +178,30 @@ private:
 
     static const unsigned int seed = 0;
 
+    Preprocessor preprocessor;
+
     enum { N_CHANNELS = 10 };
 };
 
 
 using cv::FeaturePool;
 
+
+
+class Dataset
+{
+public:
+    typedef enum {POSITIVE = 1, NEGATIVE = 2} SampleType;
+    Dataset(const sft::string& path, const int octave);
+
+    cv::Mat get(SampleType type, int idx) const;
+    int available(SampleType type) const;
+
+private:
+    svector pos;
+    svector neg;
+};
+
 // used for traning single octave scale
 class Octave : cv::Boost
 {
@@ -163,7 +235,7 @@ protected:
        const cv::Mat& sampleIdx=cv::Mat(), const cv::Mat& varType=cv::Mat(), const cv::Mat& missingDataMask=cv::Mat());
 
     void processPositives(const Dataset& dataset, const FeaturePool* pool);
-    void generateNegatives(const Dataset& dataset);
+    void generateNegatives(const Dataset& dataset, const FeaturePool* pool);
 
     float predict( const Mat& _sample, const cv::Range range) const;
 private:

apps/sft/octave.cpp

@@ -44,9 +44,6 @@
 #include <sft/random.hpp>
 
 #include <glob.h>
-#include <opencv2/imgproc/imgproc.hpp>
-#include <opencv2/highgui/highgui.hpp>
-
 #include <queue>
 
 // ============ Octave ============ //
@@ -138,85 +135,26 @@ void sft::Octave::setRejectThresholds(cv::Mat& thresholds)
 
 namespace {
 using namespace sft;
-class Preprocessor
-{
-public:
-    Preprocessor(int shr) : shrinkage(shr) {}
-
-    void apply(const Mat& frame, Mat& integrals)
-    {
-        CV_Assert(frame.type() == CV_8UC3);
-
-        int h = frame.rows;
-        int w = frame.cols;
-
-        cv::Mat channels, gray;
-
-        channels.create(h * BINS, w, CV_8UC1);
-        channels.setTo(0);
-
-        cvtColor(frame, gray, CV_BGR2GRAY);
-
-        cv::Mat df_dx, df_dy, mag, angle;
-        cv::Sobel(gray, df_dx, CV_32F, 1, 0);
-        cv::Sobel(gray, df_dy, CV_32F, 0, 1);
-
-        cv::cartToPolar(df_dx, df_dy, mag, angle, true);
-        mag *= (1.f / (8 * sqrt(2.f)));
-
-        cv::Mat nmag;
-        mag.convertTo(nmag, CV_8UC1);
-
-        angle *=  6 / 360.f;
-
-        for (int y = 0; y < h; ++y)
-        {
-            uchar* magnitude = nmag.ptr<uchar>(y);
-            float* ang = angle.ptr<float>(y);
-
-            for (int x = 0; x < w; ++x)
-            {
-                channels.ptr<uchar>(y + (h * (int)ang[x]))[x] = magnitude[x];
-            }
-        }
-
-        cv::Mat luv, shrunk;
-        cv::cvtColor(frame, luv, CV_BGR2Luv);
-
-        std::vector<cv::Mat> splited;
-        for (int i = 0; i < 3; ++i)
-            splited.push_back(channels(cv::Rect(0, h * (7 + i), w, h)));
-        split(luv, splited);
-
-        cv::resize(channels, shrunk, cv::Size(), 1.0 / shrinkage, 1.0 / shrinkage, CV_INTER_AREA);
-        cv::integral(shrunk, integrals, cv::noArray(), CV_32S);
-    }
 
-    int shrinkage;
-    enum {BINS = 10};
-};
 }
 
 void sft::Octave::processPositives(const Dataset& dataset, const FeaturePool* pool)
 {
-    Preprocessor prepocessor(shrinkage);
-
     int w = boundingBox.width;
     int h = boundingBox.height;
 
     integrals.create(pool->size(), (w / shrinkage + 1) * (h / shrinkage * 10 + 1), CV_32SC1);
 
     int total = 0;
-    for (svector::const_iterator it = dataset.pos.begin(); it != dataset.pos.end(); ++it)
+    // for (svector::const_iterator it = dataset.pos.begin(); it != dataset.pos.end(); ++it)
+    for (int curr = 0; curr < dataset.available( Dataset::POSITIVE); ++curr)
     {
-        const string& curr = *it;
-
-        cv::Mat sample = cv::imread(curr);
+        cv::Mat sample = dataset.get( Dataset::POSITIVE, curr);
 
         cv::Mat channels = integrals.row(total).reshape(0, h / shrinkage * 10 + 1);
         sample = sample(boundingBox);
 
-        prepocessor.apply(sample, channels);
+        pool->preprocess(sample, channels);
         responses.ptr<float>(total)[0] = 1.f;
 
         if (++total >= npositives) break;
@@ -228,7 +166,7 @@ void sft::Octave::processPositives(const Dataset& dataset, const FeaturePool* po
     nnegatives = cvRound(nnegatives * total / (double)npositives);
 }
 
-void sft::Octave::generateNegatives(const Dataset& dataset)
+void sft::Octave::generateNegatives(const Dataset& dataset, const FeaturePool* pool)
 {
     // ToDo: set seed, use offsets
     sft::Random::engine eng(65633343L);
@@ -237,9 +175,7 @@ void sft::Octave::generateNegatives(const Dataset& dataset)
     // int w = boundingBox.width;
     int h = boundingBox.height;
 
-    Preprocessor prepocessor(shrinkage);
-
-    int nimages = (int)dataset.neg.size();
+    int nimages = dataset.available(Dataset::NEGATIVE);
     sft::Random::uniform iRand(0, nimages - 1);
 
     int total = 0;
@@ -248,7 +184,7 @@ void sft::Octave::generateNegatives(const Dataset& dataset)
     {
         int curr = iRand(idxEng);
 
-        Mat frame = cv::imread(dataset.neg[curr]);
+        Mat frame = dataset.get(Dataset::NEGATIVE, curr);
 
         int maxW = frame.cols - 2 * boundingBox.x - boundingBox.width;
         int maxH = frame.rows - 2 * boundingBox.y - boundingBox.height;
@@ -262,7 +198,7 @@ void sft::Octave::generateNegatives(const Dataset& dataset)
         frame = frame(cv::Rect(dx, dy, boundingBox.width, boundingBox.height));
 
         cv::Mat channels = integrals.row(i).reshape(0, h / shrinkage * 10 + 1);
-        prepocessor.apply(frame, channels);
+        pool->preprocess(frame, channels);
 
         dprintf("generated %d %d\n", dx, dy);
 
@@ -386,7 +322,7 @@ bool sft::Octave::train(const Dataset& dataset, const FeaturePool* pool, int wea
 
     // 1. fill integrals and classes
     processPositives(dataset, pool);
-    generateNegatives(dataset);
+    generateNegatives(dataset, pool);
 
     // 2. only sumple case (all features used)
     int nfeatures = pool->size();
@@ -455,6 +391,11 @@ sft::ICFFeaturePool::ICFFeaturePool(cv::Size m, int n) : FeaturePool(), model(m)
     fill(nfeatures);
 }
 
+void sft::ICFFeaturePool::preprocess(const Mat& frame, Mat& integrals) const
+{
+    preprocessor.apply(frame, integrals);
+}
+
 float sft::ICFFeaturePool::apply(int fi, int si, const Mat& integrals) const
 {
     return pool[fi](integrals.row(si), model);
@@ -572,3 +513,14 @@ Dataset::Dataset(const string& path, const int oct)
     CV_Assert(pos.size() != size_t(0));
     CV_Assert(neg.size() != size_t(0));
 }
+
+cv::Mat Dataset::get(SampleType type, int idx) const
+{
+    const std::string& src = (type == POSITIVE)? pos[idx]: neg[idx];
+    return cv::imread(src);
+}
+
+int Dataset::available(SampleType type) const
+{
+    return (int)((type == POSITIVE)? pos.size():neg.size());
+}
\ No newline at end of file

modules/ml/include/opencv2/ml/ml.hpp

@@ -2140,6 +2140,8 @@ public:
     virtual float apply(int fi, int si, const Mat& integrals) const = 0;
     virtual void write( cv::FileStorage& fs, int index) const = 0;
 
+    virtual void preprocess(const Mat& frame, Mat& integrals) const = 0;
+
     virtual ~FeaturePool() = 0;
 };