In this project, we created a vehicle detection and tracking pipeline with OpenCV, histogram of oriented gradients (HOG), and support vector machines (SVM), optimizing and evaluated the model on video data from a automotive camera taken during highway driving (start with the test_video.mp4 and later implement on full project_video.mp4).
The goals / steps of this project are the following:
Author : Tran Ly Vu
Dataset was provided by Udacity: Here are links to the labeled data for vehicle and non-vehicle
These example images come from a combination of the GTI vehicle image database, the KITTI vision benchmark suite, and examples extracted from the project video itself.
There is also recently released Udacity labeled dataset to augment your training data. However, for this project, I did not use this dataset
In [1]:
import matplotlib.image as mpimg
import matplotlib.pyplot as plt
import numpy as np
import cv2
import glob
from skimage.feature import hog
from skimage import color, exposure
import random
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.svm import LinearSVC
import time
from moviepy.editor import VideoFileClip
from scipy.ndimage.measurements import label
from IPython.display import HTML
%matplotlib inline
In [2]:
vehicles_images = glob.glob('../../../vehicles/vehicles/*/*.png')
non_vehicles_images = glob.glob('../../../non-vehicles/non-vehicles/*/*.png')
def load_data(my_list):
new_list = []
for image in my_list:
img = cv2.imread(image)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
new_list.append(img)
return new_list
cars = load_data(vehicles_images)
non_cars = load_data(non_vehicles_images)
print('Number of car images: ', len(cars))
print('Number of non-car images: ', len(non_cars))
In [3]:
"""plotting 10 randome traffic sign images"""
def plot_10_random_images(images):
fig, axes = plt.subplots(2, 5, figsize=(13, 6))
fig.subplots_adjust(left=None, right=None, hspace = .02, wspace=0.1)
for i in range(2):
for j in range(5):
randomindex = random.randint(0, len(images) - 1)
axes[i,j].axis('off')
axes[i,j].imshow(images[randomindex])
plot_10_random_images(cars)
"""Saving 2 sample images"""
plt.imsave('../output_images/sample_car_img.jpg', cars[0])
plt.imsave('../output_images/sample_non_car_img.jpg', non_cars[0])
In [4]:
# Define a function to return HOG features and visualization
def get_hog_features(img, orient, pix_per_cell, cell_per_block, vis=False, feature_vec=True):
if vis == True:
features, hog_image = hog(img, orientations=orient, pixels_per_cell=(pix_per_cell, pix_per_cell),
cells_per_block=(cell_per_block, cell_per_block), transform_sqrt=False,
visualise=True, feature_vector=False)
return features, hog_image
else:
features = hog(img, orientations=orient, pixels_per_cell=(pix_per_cell, pix_per_cell),
cells_per_block=(cell_per_block, cell_per_block), transform_sqrt=False,
visualise=False, feature_vector=feature_vec)
return features
In [5]:
sample_car_gray = cv2.cvtColor(cars[0], cv2.COLOR_RGB2GRAY)
sample_non_car_gray = cv2.cvtColor(non_cars[0], cv2.COLOR_RGB2GRAY)
# Define HOG parameters
orient = 9
pix_per_cell = 8
cell_per_block = 2
# Call our function with vis=True to see an image output
features1, hog_image1 = get_hog_features(sample_car_gray, orient, pix_per_cell, cell_per_block, vis=True, feature_vec=False)
features2, hog_image2 = get_hog_features(sample_non_car_gray, orient, pix_per_cell, cell_per_block, vis=True, feature_vec=False)
# Plot the examples
fig = plt.figure()
plt.subplot(121)
plt.imshow(cars[0], cmap='gray')
plt.title('Example Car Image')
plt.subplot(122)
plt.imshow(hog_image1, cmap='gray')
plt.title('HOG Visualization')
plt.imsave('../output_images/sample_car_hog.jpg', hog_image1)
In [6]:
plt.subplot(121)
plt.imshow(non_cars[0], cmap='gray')
plt.title('Example non-Car Image')
plt.subplot(122)
plt.imshow(hog_image2, cmap='gray')
plt.title('HOG Visualization')
plt.imsave('../output_images/sample_non_car_hog.jpg', hog_image2)
I tried various combinations of parameters with trials and errors, I finaly chose to use spatial binning, color histogram and hog features with the following parameters
color space : 'YCrCb'
the number of orientation bins: 9
pix_per_cell = 8
cell_per_block = 2
hog_channel = "ALL"
spatial_size = (16, 16)
hist_bins = 16
In [7]:
def bin_spatial(img, size=(32, 32)):
color1 = cv2.resize(img[:,:,0], size).ravel()
color2 = cv2.resize(img[:,:,1], size).ravel()
color3 = cv2.resize(img[:,:,2], size).ravel()
return np.hstack((color1, color2, color3))
def color_hist(img, nbins=32): #bins_range=(0, 256)
# Compute the histogram of the color channels separately
channel1_hist = np.histogram(img[:,:,0], bins=nbins)
channel2_hist = np.histogram(img[:,:,1], bins=nbins)
channel3_hist = np.histogram(img[:,:,2], bins=nbins)
# Concatenate the histograms into a single feature vector
hist_features = np.concatenate((channel1_hist[0], channel2_hist[0], channel3_hist[0]))
# Return the individual histograms, bin_centers and feature vector
return hist_features
# Define a function to extract features from a list of images
# Have this function call bin_spatial() and color_hist()
def extract_features(imgs,
color_space='RGB',
spatial_size=(32, 32),
hist_bins=32, orient=9,
pix_per_cell=8,
cell_per_block=2,
hog_channel=0,
spatial_feat=True,
hist_feat=True,
hog_feat=True):
# Create a list to append feature vectors to
features = []
# Iterate through the list of images
for image in imgs:
file_features = []
# Read in each one by one
#image = mpimg.imread(file)
# apply color conversion if other than 'RGB'
if color_space != 'RGB':
if color_space == 'HSV':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2HSV)
elif color_space == 'LUV':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2LUV)
elif color_space == 'HLS':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2HLS)
elif color_space == 'YUV':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2YUV)
elif color_space == 'YCrCb':
feature_image = cv2.cvtColor(image, cv2.COLOR_RGB2YCrCb)
else: feature_image = np.copy(image)
if spatial_feat == True:
spatial_features = bin_spatial(feature_image, size=spatial_size)
file_features.append(spatial_features)
if hist_feat == True:
# Apply color_hist()
hist_features = color_hist(feature_image, nbins=hist_bins)
file_features.append(hist_features)
if hog_feat == True:
# Call get_hog_features() with vis=False, feature_vec=True
if hog_channel == 'ALL':
hog_features = []
for channel in range(feature_image.shape[2]):
hog_features.append(get_hog_features(feature_image[:,:,channel],
orient, pix_per_cell, cell_per_block,
vis=False, feature_vec=True))
hog_features = np.ravel(hog_features)
else:
hog_features = get_hog_features(feature_image[:,:,hog_channel], orient,
pix_per_cell, cell_per_block, vis=False, feature_vec=True)
# Append the new feature vector to the features list
file_features.append(hog_features)
features.append(np.concatenate(file_features))
# Return list of feature vectors
return features
def visualize(fig, rows, cols, imgs, titles):
for i, img in enumerate(imgs):
plt.subplot(rows, cols, i + 1)
plt.title(i + 1)
img_dims = len(img.shape)
if img_dims < 3:
plt.imshow(img, cmap='hot')
plt.title(titles[i])
else:
plt.imshow(img)
I trained a linear SVM using the above features and parameters.
I initially used RGB for color space but 'YCrCb' yielded better result. Number of orientation bins is 9 as it is recommended by original HOG paper. SVM was first recommended by udacity and actually provided good result so i did not tried other models.
I was also normalizing the training data as recommended by Udacyty with sklearn.preprocessing.StandardScaler()
The final test accuracy was 0.99
In [8]:
"""Parameters"""
COLOR_SPACE = 'YCrCb' # Can be RGB, HSV, LUV, HLS, YUV, YCrCb
ORIENTATION = 9 # HOG orientations
PIX_PER_CELL = 8 # HOG pixels per cell
CELL_PER_BLOCK = 2 # HOG cells per block
HOG_CHANNEL = "ALL" # Can be 0, 1, 2, or "ALL"
SPATIAL_SIZE = (16, 16) # Spatial binning dimensions
HIST_BINS = 16 # Number of histogram bins
IS_SPATIAL_FEAT = True # Spatial features on or off
IS_HIST_FEAT = True # Histogram features on or off
IS_HOG_FEAT = True # HOG features on or off
t=time.time()
#n_samples =1000
#random_idxs = np.random.randint(0, len(cars), n_samples)
#test_cars = np.array(cars)[random_idxs]
#test_noncars = np.array(non_cars)[random_idxs]
car_features = extract_features(cars,
color_space = COLOR_SPACE,
spatial_size= SPATIAL_SIZE,
hist_bins = HIST_BINS,
orient = ORIENTATION,
pix_per_cell = PIX_PER_CELL,
cell_per_block = CELL_PER_BLOCK,
hog_channel = HOG_CHANNEL,
spatial_feat = IS_SPATIAL_FEAT ,
hist_feat = IS_HIST_FEAT,
hog_feat = IS_HOG_FEAT)
notcar_features = extract_features(non_cars,
color_space = COLOR_SPACE,
spatial_size= SPATIAL_SIZE,
hist_bins = HIST_BINS,
orient = ORIENTATION,
pix_per_cell = PIX_PER_CELL,
cell_per_block = CELL_PER_BLOCK,
hog_channel = HOG_CHANNEL,
spatial_feat = IS_SPATIAL_FEAT ,
hist_feat = IS_HIST_FEAT,
hog_feat = IS_HOG_FEAT)
print(time.time()-t, 'Seconds to compute features...')
X = np.vstack((car_features, notcar_features)).astype(np.float64)
# Fit a per-column scaler
X_scaler = StandardScaler().fit(X)
# Apply the scaler to X
scaled_X = X_scaler.transform(X)
# Define the labels vector
y = np.hstack((np.ones(len(car_features)), np.zeros(len(notcar_features))))
# Split up data into randomized training and test sets
rand_state = np.random.randint(0, 100)
X_train, X_test, y_train, y_test = train_test_split(scaled_X, y, test_size=0.2, random_state=rand_state)
print('Using:',orient,'orientations',pix_per_cell,'pixels per cell and', cell_per_block,'cells per block')
print('Feature vector length:', len(X_train[0]))
# Use a linear SVC
SVC = LinearSVC()
# Check the training time for the SVC
SVC.fit(X_train, y_train)
t2 = time.time()
print(round(t2-t, 2), 'Seconds to train SVC...')
# Check the score of the SVC
print('Test Accuracy of SVC = ', round(SVC.score(X_test, y_test), 4))
For searching cars in an input image I use sliding window technique that taught by Udacity. It means that I iterate over image area that could contain cars with approximately car sized box and try to classify whether box contain car or not. I use sliding window sizes of 96 pixels side size. While iterating I use 50% window overlapping in horizontal and vertical directions. I also decided to search random window positions at random scales from bottom half of the image. Here is a sample of test images:
In [9]:
# Define a function to extract features from a single image window
# This function is very similar to extract_features()
# just for a single image rather than list of images
def single_img_features(img, color_space='RGB', spatial_size=(32, 32),
hist_bins=32, orient=9,
pix_per_cell=8, cell_per_block=2, hog_channel=0,
spatial_feat=True, hist_feat=True, hog_feat=True):
#1) Define an empty list to receive features
img_features = []
#2) Apply color conversion if other than 'RGB'
if color_space != 'RGB':
if color_space == 'HSV':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2HSV)
elif color_space == 'LUV':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2LUV)
elif color_space == 'HLS':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2HLS)
elif color_space == 'YUV':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2YUV)
elif color_space == 'YCrCb':
feature_image = cv2.cvtColor(img, cv2.COLOR_RGB2YCrCb)
else: feature_image = np.copy(img)
#3) Compute spatial features if flag is set
if spatial_feat == True:
spatial_features = bin_spatial(feature_image, size=spatial_size)
#4) Append features to list
img_features.append(spatial_features)
#5) Compute histogram features if flag is set
if hist_feat == True:
hist_features = color_hist(feature_image, nbins=hist_bins)
#6) Append features to list
img_features.append(hist_features)
#7) Compute HOG features if flag is set
if hog_feat == True:
if hog_channel == 'ALL':
hog_features = []
for channel in range(feature_image.shape[2]):
hog_features.extend(get_hog_features(feature_image[:,:,channel],
orient, pix_per_cell, cell_per_block,
vis=False, feature_vec=True))
else:
hog_features = get_hog_features(feature_image[:,:,hog_channel], orient,
pix_per_cell, cell_per_block, vis=False, feature_vec=True)
#8) Append features to list
img_features.append(hog_features)
#9) Return concatenated array of features
return np.concatenate(img_features)
# Define a function you will pass an image
# and the list of windows to be searched (output of slide_windows())
def search_windows(img, windows, clf, scaler, color_space='RGB',
spatial_size=(32, 32), hist_bins=32,
hist_range=(0, 256), orient=9,
pix_per_cell=8, cell_per_block=2,
hog_channel=0, spatial_feat=True,
hist_feat=True, hog_feat=True):
#1) Create an empty list to receive positive detection windows
on_windows = []
#2) Iterate over all windows in the list
for window in windows:
#3) Extract the test window from original image
test_img = cv2.resize(img[window[0][1]:window[1][1], window[0][0]:window[1][0]], (64, 64))
#4) Extract features for that window using single_img_features()
features = single_img_features(test_img,
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)
#5) Scale extracted features to be fed to classifier
test_features = scaler.transform(np.array(features).reshape(1, -1))
#6) Predict using your classifier
prediction = clf.predict(test_features)
#7) If positive (prediction == 1) then save the window
if prediction == 1:
on_windows.append(window)
#8) Return windows for positive detections
return on_windows
# Define a function that takes an image,
# start and stop positions in both x and y,
# window size (x and y dimensions),
# and overlap fraction (for both x and y)
def slide_window(img, x_start_stop=[None, None], y_start_stop=[None, None],
xy_window=(64, 64), xy_overlap=(0.5, 0.5)):
# If x and/or y start/stop positions not defined, set to image size
if x_start_stop[0] == None:
x_start_stop[0] = 0
if x_start_stop[1] == None:
x_start_stop[1] = img.shape[1]
if y_start_stop[0] == None:
y_start_stop[0] = 0
if y_start_stop[1] == None:
y_start_stop[1] = img.shape[0]
# Compute the span of the region to be searched
xspan = x_start_stop[1] - x_start_stop[0]
yspan = y_start_stop[1] - y_start_stop[0]
# Compute the number of pixels per step in x/y
nx_pix_per_step = np.int(xy_window[0]*(1 - xy_overlap[0]))
ny_pix_per_step = np.int(xy_window[1]*(1 - xy_overlap[1]))
# Compute the number of windows in x/y
nx_buffer = np.int(xy_window[0]*(xy_overlap[0]))
ny_buffer = np.int(xy_window[1]*(xy_overlap[1]))
nx_windows = np.int((xspan-nx_buffer)/nx_pix_per_step)
ny_windows = np.int((yspan-ny_buffer)/ny_pix_per_step)
# Initialize a list to append window positions to
window_list = []
# Loop through finding x and y window positions
# Note: you could vectorize this step, but in practice
# you'll be considering windows one by one with your
# classifier, so looping makes sense
for ys in range(ny_windows):
for xs in range(nx_windows):
# Calculate window position
startx = xs*nx_pix_per_step + x_start_stop[0]
endx = startx + xy_window[0]
starty = ys*ny_pix_per_step + y_start_stop[0]
endy = starty + xy_window[1]
# Append window position to list
window_list.append(((startx, starty), (endx, endy)))
# Return the list of windows
return window_list
# Define a function to draw bounding boxes
def draw_boxes(img, bboxes, color=(0, 0, 255), thick=6):
# Make a copy of the image
imcopy = np.copy(img)
# Iterate through the bounding boxes
for bbox in bboxes:
# Draw a rectangle given bbox coordinates
cv2.rectangle(imcopy, bbox[0], bbox[1], color, thick)
# Return the image copy with boxes drawn
return imcopy
In [23]:
t1 = time.time()
img = mpimg.imread("../test_images/test1.jpg")
y_start_stop = [img.shape[0]//2, img.shape[0]] # Min and max in y to search in slide_window()
draw_image = np.copy(img)
windows = slide_window(img,
x_start_stop = [None, None],
y_start_stop = y_start_stop,
xy_window=(96, 96),
xy_overlap=(0.5, 0.5))
hot_windows = search_windows(img,
windows,
clf = SVC,
scaler = X_scaler,
color_space = COLOR_SPACE ,
spatial_size = SPATIAL_SIZE,
hist_bins = HIST_BINS,
orient = ORIENTATION ,
pix_per_cell = PIX_PER_CELL,
cell_per_block = CELL_PER_BLOCK,
hog_channel = HOG_CHANNEL,
spatial_feat = IS_SPATIAL_FEAT,
hist_feat = IS_HIST_FEAT,
hog_feat = IS_HOG_FEAT )
window_img = draw_boxes(draw_image, hot_windows, color=(0, 0, 255), thick=6)
print(time.time()-t, ' seconds to process one image searching ', len(windows), 'windows')
plt.imshow(window_img)
plt.imsave('../output_images/window_search.jpg', window_img)
In [11]:
example_images = glob.glob("../test_images/*.jpg")
images = []
titles = []
for img_src in example_images:
t1 = time.time()
img = mpimg.imread(img_src)
draw_image = np.copy(img)
y_start_stop = [img.shape[0]//2, img.shape[0]] # Min and max in y to search in slide_window()
windows = slide_window(img,
x_start_stop=[None, None],
y_start_stop=y_start_stop,
xy_window=(96, 96),
xy_overlap=(0.5, 0.5))
hot_windows = search_windows(img,
windows,
clf = SVC,
scaler = X_scaler,
color_space = COLOR_SPACE ,
spatial_size = SPATIAL_SIZE,
hist_bins = HIST_BINS,
orient = ORIENTATION ,
pix_per_cell = PIX_PER_CELL,
cell_per_block = CELL_PER_BLOCK,
hog_channel = HOG_CHANNEL,
spatial_feat = IS_SPATIAL_FEAT,
hist_feat = IS_HIST_FEAT,
hog_feat = IS_HOG_FEAT )
window_img = draw_boxes(draw_image, hot_windows, color=(0, 0, 255), thick=6)
images.append(window_img)
titles.append('')
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, images, titles)
In [12]:
img_boxes = []
def convert_color(img, conv='RGB2YCrCb'):
if conv == 'RGB2YCrCb':
return cv2.cvtColor(img, cv2.COLOR_RGB2YCrCb)
if conv == 'BGR2YCrCb':
return cv2.cvtColor(img, cv2.COLOR_BGR2YCrCb)
if conv == 'RGB2LUV':
return cv2.cvtColor(img, cv2.COLOR_RGB2LUV)
# Define a single function that can extract features using hog sub-sampling and make predictions
def find_cars(img, ystart, ystop, scale, svc, X_scaler, orient, pix_per_cell, cell_per_block, spatial_size, hist_bins):
draw_img = np.copy(img)
#img = img.astype(np.float32)/255
heat_map = np.zeros_like(img[:,:,0]).astype(np.float)
img_tosearch = img[ystart:ystop,:,:]
ctrans_tosearch = convert_color(img_tosearch, conv='RGB2YCrCb')
if scale != 1:
imshape = ctrans_tosearch.shape
ctrans_tosearch = cv2.resize(ctrans_tosearch, (np.int(imshape[1]/scale), np.int(imshape[0]/scale)))
ch1 = ctrans_tosearch[:,:,0]
ch2 = ctrans_tosearch[:,:,1]
ch3 = ctrans_tosearch[:,:,2]
# Define blocks and steps as above
nxblocks = (ch1.shape[1] // pix_per_cell) - cell_per_block + 1
nyblocks = (ch1.shape[0] // pix_per_cell) - cell_per_block + 1
nfeat_per_block = orient*cell_per_block**2
# 64 was the orginal sampling rate, with 8 cells and 8 pix per cell
window = 64
nblocks_per_window = (window // pix_per_cell) - cell_per_block + 1
cells_per_step = 2 # Instead of overlap, define how many cells to step
nxsteps = (nxblocks - nblocks_per_window) // cells_per_step + 1
nysteps = (nyblocks - nblocks_per_window) // cells_per_step + 1
# Compute individual channel HOG features for the entire image
hog1 = get_hog_features(ch1, orient, pix_per_cell, cell_per_block, feature_vec=False)
hog2 = get_hog_features(ch2, orient, pix_per_cell, cell_per_block, feature_vec=False)
hog3 = get_hog_features(ch3, orient, pix_per_cell, cell_per_block, feature_vec=False)
for xb in range(nxsteps):
for yb in range(nysteps):
ypos = yb*cells_per_step
xpos = xb*cells_per_step
# Extract HOG for this patch
hog_feat1 = hog1[ypos:ypos+nblocks_per_window, xpos:xpos+nblocks_per_window].ravel()
hog_feat2 = hog2[ypos:ypos+nblocks_per_window, xpos:xpos+nblocks_per_window].ravel()
hog_feat3 = hog3[ypos:ypos+nblocks_per_window, xpos:xpos+nblocks_per_window].ravel()
hog_features = np.hstack((hog_feat1, hog_feat2, hog_feat3))
xleft = xpos*pix_per_cell
ytop = ypos*pix_per_cell
# Extract the image patch
subimg = cv2.resize(ctrans_tosearch[ytop:ytop+window, xleft:xleft+window], (64,64))
# Get color features
spatial_features = bin_spatial(subimg, size=spatial_size)
hist_features = color_hist(subimg, nbins=hist_bins)
# Scale features and make a prediction
test_features = X_scaler.transform(np.hstack((spatial_features, hist_features, hog_features)).reshape(1, -1))
#test_features = X_scaler.transform(np.hstack((shape_feat, hist_feat)).reshape(1, -1))
test_prediction = svc.predict(test_features)
if test_prediction == 1:
xbox_left = np.int(xleft*scale)
ytop_draw = np.int(ytop*scale)
win_draw = np.int(window*scale)
cv2.rectangle(draw_img,(xbox_left, ytop_draw+ystart),(xbox_left+win_draw,ytop_draw+win_draw+ystart),(0,0,255),6)
img_boxes.append(((xbox_left, ytop_draw+ystart),(xbox_left+win_draw, ytop_draw+win_draw+ystart)))
heat_map[ytop_draw+ystart:ytop_draw+win_draw+ystart, xbox_left:xbox_left+win_draw] +=1
return draw_img, heat_map
# Here is your draw_boxes function from the previous exercise
def draw_boxes(img, bboxes, color=(0, 0, 255), thick=6):
# Make a copy of the image
imcopy = np.copy(img)
# Iterate through the bounding boxes
for bbox in bboxes:
# Draw a rectangle given bbox coordinates
cv2.rectangle(imcopy, bbox[0], bbox[1], color, thick)
# Return the image copy with boxes drawn
return imcopy
In [26]:
"""Parameters"""
COLOR_SPACE = 'YCrCb' # Can be RGB, HSV, LUV, HLS, YUV, YCrCb
ORIENTATION = 9 # HOG orientations
PIX_PER_CELL = 8 # HOG pixels per cell
CELL_PER_BLOCK = 2 # HOG cells per block
HOG_CHANNEL = "ALL" # Can be 0, 1, 2, or "ALL"
SPATIAL_SIZE = (16, 16) # Spatial binning dimensions
HIST_BINS = 16 # Number of histogram bins
IS_SPATIAL_FEAT = True # Spatial features on or off
IS_HIST_FEAT = True # Histogram features on or off
IS_HOG_FEAT = True # HOG features on or off
YSTART= 400
YSTOP = 656
SCALE = 1.5
img = mpimg.imread("../test_images/test4.jpg")
out_img , heatmap = find_cars(img,
ystart=YSTART,
ystop=YSTOP,
scale=SCALE,
svc = SVC,
X_scaler = X_scaler,
orient= ORIENTATION,
pix_per_cell = PIX_PER_CELL,
cell_per_block= CELL_PER_BLOCK,
spatial_size = SPATIAL_SIZE,
hist_bins = HIST_BINS)
plt.imshow(out_img)
plt.imsave('../output_images/sample_heatmap.jpg', heatmap)
In order to eliminate overlapping detection and false positive, I recorded the positions of positive detections in each frame of the video. From the positive detections I created a heatmap and then thresholded that map to identify vehicle positions. I then used scipy.ndimage.measurements.label() to identify individual blobs in the heatmap. I then assumed each blob corresponded to a vehicle. I constructed bounding boxes to cover the area of each blob detected.
In [27]:
def add_heat(heatmap, bbox_list):
# Iterate through list of bboxes
for box in bbox_list:
# Add += 1 for all pixels inside each bbox
# Assuming each "box" takes the form ((x1, y1), (x2, y2))
heatmap[box[0][1]:box[1][1], box[0][0]:box[1][0]] += 1
# Return updated heatmap
return heatmap
def apply_threshold(heatmap, threshold):
# Zero out pixels below the threshold
heatmap[heatmap <= threshold] = 0
# Return thresholded map
return heatmap
def draw_labeled_bboxes(img, labels):
# Iterate through all detected cars
for car_number in range(1, labels[1]+1):
# Find pixels with each car_number label value
nonzero = (labels[0] == car_number).nonzero()
# Identify x and y values of those pixels
nonzeroy = np.array(nonzero[0])
nonzerox = np.array(nonzero[1])
# Define a bounding box based on min/max x and y
bbox = ((np.min(nonzerox), np.min(nonzeroy)), (np.max(nonzerox), np.max(nonzeroy)))
# Draw the box on the image
cv2.rectangle(img, bbox[0], bbox[1], (0,0,255), 6)
# Return the image
return img
In [29]:
labels = label(heatmap)
#draw bouding box on cop of image
draw_img = draw_labeled_bboxes(np.copy(img), labels)
plt.imsave('../output_images/sample_draw_img.jpg', draw_img)
fig = plt.figure()
plt.subplot(121)
plt.imshow(draw_img)
plt.title('Car Positions')
plt.subplot(122)
plt.imshow(heatmap, cmap='hot')
plt.title('Heat Map')
fig.tight_layout()
In [25]:
out_images = []
out_titles = []
for i, img_src in enumerate(example_images):
img = mpimg.imread(img_src)
out_img, heat_map = find_cars(img,
ystart=YSTART,
ystop=YSTOP,
scale=SCALE,
svc = SVC,
X_scaler = X_scaler,
orient= ORIENTATION,
pix_per_cell = PIX_PER_CELL,
cell_per_block= CELL_PER_BLOCK,
spatial_size = SPATIAL_SIZE,
hist_bins = HIST_BINS)
labels = label(heat_map)
#draw bouding box on cop of image
draw_img = draw_labeled_bboxes(np.copy(img), labels)
out_images.append(draw_img)
out_titles.append('')
out_images.append(heatmap)
out_titles.append('')
fig = plt.figure(figsize=(12,24))
visualize(fig, 8, 2, out_images, out_titles)
In [20]:
def process_image(img):
# Find final boxes from heatmap using label function
out_img, heatmap = find_cars(img,
ystart=YSTART,
ystop=YSTOP,
scale=SCALE,
svc = SVC,
X_scaler = X_scaler,
orient= ORIENTATION,
pix_per_cell = PIX_PER_CELL,
cell_per_block= CELL_PER_BLOCK,
spatial_size = SPATIAL_SIZE,
hist_bins = HIST_BINS)
labels = label(heatmap)
draw_img = draw_labeled_bboxes(np.copy(img), labels)
return draw_img
In [21]:
clip1 = VideoFileClip('../project_video.mp4')
video_clip = clip1.fl_image(process_image) #NOTE: this function expects color images!!
video_output = '../output_videos/project_video.mp4'
%time video_clip.write_videofile(video_output, audio=False)
In [22]:
HTML("""
<video width="960" height="540" controls>
<source src="{0}">
</videot
""".format(video_output))
Out[22]:
From my video, there are some redundant boxes although cars have been detected. Here are a few things I could consider doing if i were to have more times:
- Try new classifier, i.e decision tree
- Try to use more data, i.e [Udacity labeled dataset](https://github.com/udacity/self-driving-car/tree/master/annotations)
- Doing more parameters tuning ,etc
In [ ]: