def test_hotspot(self):
# Read in a pickle file with bboxes saved
# Each item in the "all_bboxes" list will contain a
# list of boxes for one of the images shown above
box_list = pickle.load(open("bbox_pickle.p", "rb"))
# Read in image similar to one shown above
image = mpimg.imread(IMG_FILE)
image_with_boxes = draw_boxes(image, box_list)
heat = np.zeros_like(image[:, :, 0]).astype(np.float)
# Add heat to each box in box list
heat = uut.add_heat(heat, box_list)
# Apply threshold to help remove false positives
heat = uut.apply_threshold(heat, 1)
# Visualize the heatmap when displaying
heatmap = np.clip(heat, 0, 255)
# Find final boxes from heatmap using label function
labels = label(heatmap)
draw_img = uut.draw_labeled_bboxes(np.copy(image), labels)
final_image = image_util.arrangeImages(
[image, image_with_boxes, heatmap, draw_img],
["Original", "Boxes that detected car", "Heatmap", "labeld boxes"],
2)
image_util.saveImage(final_image, TEST_OUT_DIR + "/hotspot.jpg")
def draw_sliding_windows(self, filename, img, y_start, y_stop, scale):
xy_window = (int(64*scale),int(64*scale))
boxes = slide_window(img, y_start_stop=[y_start, y_stop], xy_window=xy_window, xy_overlap=(0.75, 0.75))
img = draw_boxes(img,boxes)
img = cv2.rectangle(img,(0,y_start),(xy_window[0],y_start+xy_window[1]), (255,0,0), 6)
print (len(boxes))
print (img.shape)
image_util.saveImage(img, filename)
def feed_image(self,img,isVideo=True):
draw_image1 = np.copy(img)
draw_image2 = np.copy(img)
if self.imgsize==None or not isVideo:
self.imgsize = img.shape[0:2]
self.heatmap = np.zeros(self.imgsize,dtype='float64')
# Detect Cars
hot_windows = []
for search_param in self.search_params:
xstart = search_param["xstart"]
xstop = search_param["xstop"]
ystart = search_param["ystart"]
ystop = search_param["ystop"]
scale = search_param["scale"]
hot_windows += lf.find_cars(img, xstart,xstop,ystart, ystop, scale, self.svc, self.X_scaler, self.orient,
self.pix_per_cell, self.cell_per_block, self.spatial_size, self.hist_bins,
color_space=self.color_space,hog_channel=self.hog_channel,spatial_feat=self.spatial_feat,
hist_feat=self.hist_feat, hog_feat=self.hog_feat)
window_img = lf.draw_boxes(draw_image1, hot_windows, color=(0, 0, 255), thick=6)
# Add to heatmap
if not isVideo:
self.heatmap = lf.add_heat(self.heatmap,hot_windows)
self.heatmap_thresh = lf.apply_threshold(self.heatmap, self.minheat)
else:
self.heatmap = np.clip(self.heatmap-1,0,self.minheat)
self.heatmap = lf.add_heat(self.heatmap,hot_windows)
self.heatmap_thresh = lf.apply_threshold(self.heatmap, self.minheat)
self.heatmap += self.heatmap_thresh
labels = label(self.heatmap_thresh)
label_boxes = lf.get_labeled_bboxes(labels)
heatmap_img = np.asarray(np.dstack([self.heatmap,self.heatmap,self.heatmap])/np.max(self.heatmap)*255,dtype='uint8')
out_img = lf.draw_boxes(draw_image2, label_boxes, color=(0, 255, 0), thick=6)
results_dict = {"window_img":window_img,"out_img":out_img,"heat_map":heatmap_img,"heat_map_thresh":self.heatmap_thresh}
return results_dict
def draw_labeled_bboxes_with_history_verbose(self, img):
heatmap_img = np.zeros_like(img[:,:,0])
boxed_image = np.copy(img)
for bbox_list in self.history:
heatmap = self.add_heat(heatmap_img, bbox_list)
boxed_image = draw_boxes(boxed_image,bbox_list)
heatmap = self.apply_threshold(heatmap, self.history_max_size * 1.5)
labels = label(heatmap)
draw_img = self.draw_labeled_bboxes(img, labels)
#draw_img = self.draw_heatmap(heatmap)
return draw_img, labels, heatmap, boxed_image
def get_cars(self, image):
hot_windows = []
if self.subsample:
scales = [1, 1.5]
for scale in scales:
hot_windows.extend(
lf.find_cars(image,
self.svc,
self.X_scaler,
spatial_size=(self.ssahb, self.ssahb),
hist_bins=self.ssahb,
scale=scale))
else:
hot_windows.extend(
lf.detect_cars_in_image(image,
self.svc,
self.X_scaler,
color_space=self.color,
channel=self.channel,
spatial_size=(self.ssahb, self.ssahb),
hist_bins=self.ssahb,
xy_windows=self.xy_windows))
self.windows.append(hot_windows)
if len(self.windows) > self.history:
self.windows.pop(0)
hw = []
for wind in self.windows:
hw.extend(wind)
heatmap, labels = lf.heat_map(image, hw, self.heat_thres)
draw_image = np.copy(image)
if not self.heat_only:
draw_image = lf.draw_boxes(draw_image,
hw,
color=(0, 0, 255),
thick=6)
draw_img = lf.draw_labeled_bboxes(draw_image, labels)
return draw_img
def process_pipeline(img, car_classifier, heatmapper, windows, debug=False):
norm_img = img.astype(np.float32) / 255.0
box_list = search_windows(
img, windows, car_classifier.clf, car_classifier.feature_scaler,
car_classifier.color_space, car_classifier.spatial_size,
car_classifier.hist_bins, (0, 256), car_classifier.orient,
car_classifier.pix_per_cell, car_classifier.cell_per_block,
car_classifier.hog_channel, car_classifier.spatial_feat,
car_classifier.hist_feat, car_classifier.hog_feat)
if debug:
boxes_img = np.copy(img)
boxes_img = draw_boxes(img, box_list)
# creat heatmap using box_list
heatmap = heatmapper.compute_heatmap(box_list)
labels = label(heatmap)
draw_img = draw_labeled_bboxes(np.copy(img), labels)
if debug:
print("debug process pipeline")
return draw_img, boxes_img
else:
return draw_img
def process_image(imgfn, svc, X_scaler,
ssahb, channel, color, train_big, dc, heat_only, ss):
"""Searches for cars in images while varying features"""
image = mpimg.imread(imgfn)
if ss: # select subsampeling algorithm
heat_thres = 2
scales = [1, 1.5]
windows = []
for scale in scales:
windows.extend(lf.find_cars(image, svc, X_scaler,
spatial_size=(ssahb, ssahb), hist_bins=ssahb,
scale=scale))
heatmap, labels = lf.heat_map(image, windows, heat_thres)
cars = np.copy(image)
if not heat_only:
cars = lf.draw_boxes(cars, windows, color=(0, 0, 255), thick=6)
cars = lf.draw_labeled_bboxes(cars, labels)
else:
cars = dc.get_cars(image)
# hot_windows = lf.detect_cars_in_image(image, svc, X_scaler)
# threshold = 4
# heatmap, labels = lf.heat_map(image, hot_windows, threshold)
return cars
for img_file in test_win_images:
t = time.time()
img = mpimg.imread(img_file)
win_img = np.copy(img)
img = img.astype(np.float32)/255
print(np.min(img), np.max(img))
h = img.shape[0]
y_start_stop = [h//2, h]
print('y_start_stop', y_start_stop)
windows = slide_window(img, x_start_stop=[None, None],
y_start_stop=y_start_stop, xy_window=(96, 96), xy_overlap=(overlap, overlap))
search_wins = search_windows(img, windows, svc, X_scaler, color_space=color_space,
spatial_size=spatial_size, hist_bins=hist_bins, orient=orient,
pix_per_cell=pix_per_cell, cell_per_block=cell_per_block, hog_channel=hog_channel,
spatial_feat=spatial_feat, hist_feat=hist_feat, hog_feat=hog_feat)
window_img = draw_boxes(win_img, search_wins, color=(0, 0, 255), thick=5)
temp_images.append(window_img)
titles.append(img_file)
print(time.time()-t, 'seconds to process one image searching', len(windows), 'windows')
fig = plt.figure(figsize=(12,18), dpi=300)
visualize(fig, 5,2, temp_images, titles)
write_name= 'output_images/search_draw_win/search_draw_win'+'.jpg'
print('car_plot image: ', write_name)
fig.savefig(write_name)
def _lane_car_locate_pipeline(rgb_img):
"""Run both vehicle_zone and lane_zone pipelines."""
car_matches = vehicle_zone.locate_nearby_cars(rgb_img)
lane_img = lane_zone.locate_lane_bounds(rgb_img)
return lesson_functions.draw_boxes(lane_img, car_matches)
def draw_rectangles(self, image, rects, color=(0, 0, 1), thick=6):
return lesson_functions.draw_boxes(image, rects, color, thick)
draw_img = draw_labeled_bboxes(image, labels)
if show_heat:
heatmap = np.clip(heat, 0, 255)
return draw_img, heatmap, hot_windows
else:
return draw_img
if __name__ == '__main__':
image = mpimg.imread(TEST_IMAGE)
t_start = time.time()
draw_img, heatmap, hot_windows = detect_vehicle(image, show_heat=True)
t_end = time.time()
print("Time take to process image:", round(t_end - t_start, 2))
window_img = draw_boxes(image, hot_windows, color=(0, 0, 255), thick=6)
# fig = plt.figure(figsize=(12,6))
# plt.subplot(131)
# plt.imshow(window_img)
# plt.title('All Detected Windows')
# plt.subplot(132)
# plt.imshow(draw_img)
# plt.title('Car Positions')
# plt.subplot(133)
# plt.imshow(heatmap, cmap='hot')
# plt.title('Heat Map')
# fig.tight_layout()
# plt.show()
plt.imshow(window_img)
plt.axis('off')
def videopipe(video_image):
# this is the way to treat each videoframe
video_image_png = video_image.astype(np.float32) / 255
draw_image = np.copy(video_image)
# print(str(np.amax(draw_image)))
'''
# For the output of a single image:
test_file_name = 'test1.jpg'
test_file_folder = '../test_images/'
test_file = test_file_folder+test_file_name
test_image_name = test_file_name+'_LUVonlySpat.jpg'
test_image_png = mpimg.imread(test_file)
test_image = test_image_png.astype(np.float32)/255
draw_image = np.copy(test_image)
'''
# Search for matches in image:
hot_windows_spat = detect.search_windows(video_image_png,
windows,
svc_spat,
X_scaler_spat,
color_space=color_space,
spatial_size=spatial_size,
hist_bins=hist_bins,
orient=orient,
pix_per_cell=pix_per_cell,
cell_per_block=cell_per_block,
hog_channel=hog_channel,
spatial_feat=spatial_feat,
hist_feat=False,
hog_feat=False)
hot_windows_hist = detect.search_windows(video_image_png,
windows,
svc_hist,
X_scaler_hist,
color_space=color_space,
spatial_size=spatial_size,
hist_bins=hist_bins,
orient=orient,
pix_per_cell=pix_per_cell,
cell_per_block=cell_per_block,
hog_channel=hog_channel,
spatial_feat=spatial_feat,
hist_feat=hist_feat,
hog_feat=False)
hot_windows_hog = detect.search_windows(video_image_png,
windows,
svc_hog,
X_scaler_hog,
color_space=color_space,
spatial_size=spatial_size,
hist_bins=hist_bins,
orient=orient,
pix_per_cell=pix_per_cell,
cell_per_block=cell_per_block,
hog_channel=hog_channel,
spatial_feat=False,
hist_feat=False,
hog_feat=hog_feat)
#print("hot_windows_spat: "+str(len(hot_windows_spat)))
#print("hot_windows_hist: "+str(len(hot_windows_hist)))
#print("hot_windows_hog: "+str(len(hot_windows_hog )))
hot_windows = []
hot_windows.extend(hot_windows_spat)
hot_windows.extend(hot_windows_hist)
hot_windows.extend(hot_windows_hog)
burning_windows = len(hot_windows)
font = cv2.FONT_HERSHEY_SIMPLEX
# ''' # for debugging:
# Drawing the identified rectangles on the image
window_img = lesson_functions.draw_boxes(draw_image,
hot_windows_spat,
color=(255, 0, 255),
thick=2)
cv2.putText(window_img, str(nnn), (50, 100), font, 1.5, (255, 255, 255), 2,
cv2.LINE_AA)
cv2.putText(window_img, "hot windows found: " + str(len(hot_windows_spat)),
(150, 100), font, 1.5, (255, 0, 255), 2, cv2.LINE_AA)
cv2.putText(window_img, "spatial bins: " + str(spatial_size), (100, 150),
font, 1, (255, 0, 255), 2, cv2.LINE_AA)
cv2.imwrite(folder + '/spat/spat_' + str(nnn) + '.jpg', window_img)
window_img = lesson_functions.draw_boxes(draw_image,
hot_windows_hist,
color=(0, 255, 255),
thick=2)
cv2.putText(window_img, str(nnn), (50, 100), font, 1.5, (255, 255, 255), 2,
cv2.LINE_AA)
cv2.putText(window_img, "hot windows found: " + str(len(hot_windows_hist)),
(150, 100), font, 1.5, (0, 255, 255), 2, cv2.LINE_AA)
cv2.putText(window_img, "hist_bins: " + str(hist_bins), (100, 150), font,
1, (0, 255, 255), 2, cv2.LINE_AA)
cv2.imwrite(folder + '/hist/hist_' + str(nnn) + '.jpg', window_img)
window_img = lesson_functions.draw_boxes(draw_image,
hot_windows_hog,
color=(255, 255, 0),
thick=2)
cv2.putText(window_img, str(nnn), (50, 100), font, 1.5, (255, 255, 255), 2,
cv2.LINE_AA)
cv2.putText(window_img, "hot windows found: " + str(len(hot_windows_hog)),
(150, 100), font, 1.5, (255, 255, 0), 2, cv2.LINE_AA)
cv2.putText(
window_img, "orient: " + str(orient) + " pix_per_cell: " +
str(pix_per_cell) + " cell_per_block: " + str(cell_per_block) +
" hog_channel: " + str(hog_channel), (100, 150), font, 1,
(255, 255, 0), 2, cv2.LINE_AA)
cv2.imwrite(folder + '/hog/hog_' + str(nnn) + '.jpg', window_img)
window_img = lesson_functions.draw_boxes(draw_image,
hot_windows,
color=(0, 0, 255),
thick=2)
cv2.putText(window_img, str(nnn), (50, 100), font, 1.5, (255, 255, 255), 2,
cv2.LINE_AA)
cv2.putText(window_img, "hot windows found: " + str(len(hot_windows)),
(150, 100), font, 1.5, (0, 0, 255), 2, cv2.LINE_AA)
cv2.putText(
window_img, "abs_threshold: " + str(abs_threshold) + " dyn_lim: " +
str(dyn_lim) + " = " + str(int(dyn_lim * len(hot_windows))),
(100, 150), font, 1, (0, 0, 255), 2, cv2.LINE_AA)
cv2.putText(
window_img, "spat: " + str(len(hot_windows_spat)) + " hist: " +
str(len(hot_windows_hist)) + " hog: " + str(len(hot_windows_hog)),
(100, 200), font, 1, (0, 0, 255), 2, cv2.LINE_AA)
# document single image output
# '''
global nnn
nnn = nnn + 1
# consolidate heatmap
# threshold = 6
# dyn_threshold = int(burning_windows*0.15) # % of all frames must be overlapping object for detection
dyn_threshold = int(
burning_windows *
dyn_lim) # % of all frames must be overlapping object for detection
labels, heat_max = heatmap.heathot(hot_windows, draw_image, dyn_threshold,
abs_threshold, max_threshold)
# print("labels shape after: "+labels.shape)
##############################
# Another filter function added with love to the Udacity reviewers
##############################
# generate list of boundary boxes
bbox_list = heatmap.draw_labeled_bboxes(video_image_png, labels)[1]
# print("before append: "+str(history))
history.append(bbox_list)
# print("after append: "+str(history))
# print(hot_windows_hist)
tmp = []
for n in history:
tmp.extend(n)
# print("tmp = ")
# print(tmp)
# print(history[0].shape)
if nnn > history_length + 1:
labels = heatmap.heathot(tmp, draw_image, history_threshold, -1,
100)[0]
# print(bbox_list)
# print(len(bbox_list))
# print(np.amax(bbox_list))
# print(labels[0].shape)
if (len(hot_windows_spat) >
abs_limit) and (len(hot_windows_hist) > abs_limit) and (
len(hot_windows_hog) > abs_limit
): # there have to be at least some detections in each filter
draw_image = heatmap.draw_labeled_bboxes(draw_image, labels)[0]
# else:
# draw_img = draw_image
# ''' for documentation purposes add results to output image
if (len(hot_windows_spat) >
abs_limit) and (len(hot_windows_hist) > abs_limit) and (
len(hot_windows_hog) > abs_limit
): # there have to be at least some detections in each filter
docu_img = heatmap.draw_labeled_bboxes(window_img, labels)[0]
else:
docu_img = window_img
cv2.putText(docu_img, "heat_max: " + str(heat_max), (100, 250), font, 1,
(255, 255, 255), 2, cv2.LINE_AA)
cv2.imwrite(folder + '/all/all_' + str(nnn) + '.jpg', docu_img)
# '''
# label each frame with a counter for debugging / threshold
cv2.putText(draw_image, str(nnn), (50, 100), font, 1, (255, 255, 255), 2,
cv2.LINE_AA)
cv2.putText(draw_image, "hot windows found: " + str(len(hot_windows)),
(150, 100), font, 1, (0, 0, 255), 2, cv2.LINE_AA)
cv2.imwrite(folder + '/output/out_' + str(nnn) + '.jpg', draw_image)
return draw_image
hist_feat=False,
hog_feat=hog_feat)
#print("hot_windows_spat: "+str(len(hot_windows_spat)))
#print("hot_windows_hist: "+str(len(hot_windows_hist)))
#print("hot_windows_hog: " +str(len(hot_windows_hog )))
hot_windows = []
hot_windows.extend(hot_windows_spat)
hot_windows.extend(hot_windows_hist)
hot_windows.extend(hot_windows_hog)
burning_windows = len(hot_windows)
print("hot_windows found: " + str(len(hot_windows)))
# Drawing the identified rectangles on the image
window_img = lesson_functions.draw_boxes(draw_image,
hot_windows_spat,
color=(255, 0, 255),
thick=2)
cv2.imwrite(folder + '/testpic/' + test_file_name + '_spat.jpg', window_img)
window_img = lesson_functions.draw_boxes(draw_image,
hot_windows_hist,
color=(0, 255, 255),
thick=2)
cv2.imwrite(folder + '/testpic/' + test_file_name + '_hist.jpg', window_img)
window_img = lesson_functions.draw_boxes(draw_image,
hot_windows_hog,
color=(255, 255, 0),
thick=2)
cv2.imwrite(folder + '/testpic/' + test_file_name + '_hog.jpg', window_img)
window_img = lesson_functions.draw_boxes(draw_image,
hot_windows,
color=(0, 0, 255),
print('spatial_feat: ', params['spatial_feat'])
print('hist_feat: ', params['hist_feat'])
print('hog_feat: ', params['hog_feat'])
#img = mpimg.imread('test_images/test1.jpg')
#img = mpimg.imread('test_images/test2.jpg')
#img = mpimg.imread('test_images/test3.jpg')
img = mpimg.imread('test_images/test4.jpg')
#img = mpimg.imread('test_images/test5.jpg')
#img = mpimg.imread('test_images/test6.jpg')
ystart = 400
ystop = 656
scale = 1.5
t = time.time()
draw_img = np.copy(img)
img = img.astype(np.float32) / 255
bboxes = find_cars(img, ystart, ystop, scale, svc, X_scaler, params)
t2 = time.time()
print(round(t2 - t, 5), 'Seconds to detect using scale:', scale)
draw_img = draw_boxes(draw_img, bboxes)
plt.figure(figsize=(8, 8))
plt.xlim(0, 1280)
plt.ylim(720, 0)
plt.imshow(draw_img)
