Made the python object detection tools process color images

in color instead of always converting to grayscale.

python_examples/train_object_detector.py

@@ -46,7 +46,7 @@ options.add_left_right_image_flips = True
 # empirically by checking how well the trained detector works on a test set of
 # images you haven't trained on.  Don't just leave the value set at 1.  Try a
 # few different C values and see what works best for your data.
-options.C = 1 
+options.C = 5 
 # Tell the code how many CPU cores your computer has for the fastest training.
 options.num_threads = 4
 options.be_verbose = True 

tools/python/src/object_detection.cpp

@@ -68,10 +68,10 @@ std::vector<rectangle> run_detector (
 )
 {
     pyramid_down<2> pyr;
-    array2d<unsigned char> temp;
 
     if (is_gray_python_image(img))
     {
+        array2d<unsigned char> temp;
         if (upsampling_amount == 0)
         {
             return detector(numpy_gray_image(img));
@@ -94,6 +94,7 @@ std::vector<rectangle> run_detector (
     }
     else if (is_rgb_python_image(img))
     {
+        array2d<rgb_pixel> temp;
         if (upsampling_amount == 0)
         {
             return detector(numpy_rgb_image(img));

tools/python/src/simple_object_detector.h

@@ -127,7 +127,7 @@ namespace dlib
         if (options.C <= 0)
             throw error("Invalid C value given to train_simple_object_detector(), C must be > 0.");
 
-        dlib::array<array2d<unsigned char> > images;
+        dlib::array<array2d<rgb_pixel> > images;
         std::vector<std::vector<rectangle> > boxes, ignore;
         ignore = load_image_dataset(images, boxes, dataset_filename);
 
@@ -225,7 +225,7 @@ namespace dlib
         const std::string& detector_filename
     )
     {
-        dlib::array<array2d<unsigned char> > images;
+        dlib::array<array2d<rgb_pixel> > images;
         std::vector<std::vector<rectangle> > boxes, ignore;
         ignore = load_image_dataset(images, boxes, dataset_filename);