def processing(img, object_points, img_points, M, Minv, left_line, right_line):
#camera calibration, image distortion correction
undist = utils.cal_undistort(img, object_points, img_points)
#get the thresholded binary image
thresholded = thresholding(undist)
#perform perspective transform
thresholded_wraped = cv2.warpPerspective(thresholded,
M,
img.shape[1::-1],
flags=cv2.INTER_LINEAR)
#perform detection
if left_line.detected and right_line.detected:
left_fit, right_fit, left_lane_inds, right_lane_inds = utils.find_line_by_previous(
thresholded_wraped, left_line.current_fit, right_line.current_fit)
else:
left_fit, right_fit, left_lane_inds, right_lane_inds = utils.find_line(
thresholded_wraped)
left_line.update(left_fit)
right_line.update(right_fit)
#draw the detected laneline and the information
area_img = utils.draw_area(undist, thresholded_wraped, Minv, left_fit,
right_fit)
curvature, pos_from_center = utils.calculate_curv_and_pos(
thresholded_wraped, left_fit, right_fit)
result = utils.draw_values(area_img, curvature, pos_from_center)
return result
def pipline_img(img,M,Minv):
'''img 为已经二值化的经过滤波的图像'''
thresholded = utils.thresholding(img)
thresed_warped = Mtrans.warper(thresholded,M)
# find line points
left_fit, right_fit, left_lane_inds, right_lane_inds,leftx_base,rightx_base = find_line(thresed_warped)
cur,dst = utils.calculate_curv_and_pos(thresed_warped,left_fit, right_fit)
# df = np.concatenate([left_fit.reshape(1,-1), right_fit.reshape(1,-1)],axis = 0)
# print(df,left_fit.reshape(1,-1))
res = utils.draw_area(img,thresed_warped,Minv,left_fit,right_fit)
return utils.draw_values(res,cur,dst)
def processing(img, object_points, img_points, M, Minv, left_line, right_line):
undist = utils.cal_undistort(img, object_points, img_points)
thresholded = thresholding(undist)
thresholded_wraped = cv2.warpPerspective(thresholded,
M,
img.shape[1::-1],
flags=cv2.INTER_LINEAR)
if left_line.detected and right_line.detected:
left_fit, right_fit, left_lane_inds, right_lane_inds = utils.find_line_by_previous(
thresholded_wraped, left_line.current_fit, right_line.current_fit)
else:
left_fit, right_fit, left_lane_inds, right_lane_inds = utils.find_line(
thresholded_wraped)
left_line.update(left_fit)
right_line.update(right_fit)
area_img = utils.draw_area(undist, thresholded_wraped, Minv, left_fit,
right_fit)
curvature, pos_from_center = utils.calculate_curv_and_pos(
thresholded_wraped, left_fit, right_fit)
result = utils.draw_values(area_img, curvature, pos_from_center)
return result
def processing(img, M, Minv, left_line, right_line):
prev_time = time.time()
img = Image.fromarray(img)
undist = img
#get the thresholded binary image
img = np.array(img)
blur_img = cv2.GaussianBlur(img, (3, 3), 0)
Sobel_x_thresh = utils.abs_sobel_thresh(blur_img,
orient='x',
thresh_min=90,
thresh_max=255)
kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (5, 5))
dilated = cv2.dilate(Sobel_x_thresh, kernel, iterations=2)
#perform perspective transform
thresholded_wraped = cv2.warpPerspective(dilated,
M,
img.shape[1::-1],
flags=cv2.INTER_LINEAR)
#perform detection
if left_line.detected and right_line.detected:
left_fit, right_fit = utils.find_line_by_previous(
thresholded_wraped, left_line.current_fit, right_line.current_fit)
else:
left_fit, right_fit = utils.find_line(thresholded_wraped)
left_line.update(left_fit)
right_line.update(right_fit)
# #draw the detected laneline and the information
undist = Image.fromarray(img)
area_img = utils.draw_area(undist, thresholded_wraped, Minv, left_fit,
right_fit)
area_img = np.array(area_img)
curr_time = time.time()
exec_time = curr_time - prev_time
info = "time: %.2f ms" % (1000 * exec_time)
print(info)
return area_img, thresholded_wraped
def processing(img, M, Minv, left_line, right_line):
#img = RotateClockWise90(img)
#img = np.rot90(img)
prev_time = time.time()
img = Image.fromarray(img)
undist = img
#get the thresholded binary image
img = np.array(img)
thresholded = thresholding(img)
#perform perspective transform
thresholded_wraped = cv2.warpPerspective(thresholded,
M,
img.shape[1::-1],
flags=cv2.INTER_LINEAR)
#perform detection
if left_line.detected and right_line.detected:
left_fit, right_fit, left_lane_inds, right_lane_inds = utils.find_line_by_previous(
thresholded_wraped, left_line.current_fit, right_line.current_fit)
else:
left_fit, right_fit, left_lane_inds, right_lane_inds = utils.find_line(
thresholded_wraped)
left_line.update(left_fit)
right_line.update(right_fit)
# #draw the detected laneline and the information
undist = Image.fromarray(img)
area_img, gre1 = utils.draw_area(undist, thresholded_wraped, Minv,
left_fit, right_fit)
curvature, pos_from_center = utils.calculate_curv_and_pos(
thresholded_wraped, left_fit, right_fit)
area_img = np.array(area_img)
result = utils.draw_values(area_img, curvature, pos_from_center)
curr_time = time.time()
exec_time = curr_time - prev_time
info = "time: %.2f ms" % (1000 * exec_time)
print(info)
return result, thresholded_wraped
def processing(img, object_points, img_points, M, Minv, left_line, right_line):
# camera calibration, image distortion correction
# undist = utils.cal_undistort(img,object_points,img_points)
cv2.imshow("img", img)
hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
# 设置阈值下限和上限，去除背景颜色
lower_red = np.array([80, 0, 0])
upper_red = np.array([160, 255, 150])
lower_white = np.array([0, 0, 221])
upper_white = np.array([180, 30, 255])
lower_yellow = np.array([100, 43, 46])
upper_yellow = np.array([120, 255, 255])
lower_lane = np.array([0, 0, 50])
upper_lane = np.array([180, 43, 120])
# 创建掩膜
# 将原图像和掩膜做位与运算
white_img = colorMask("white_mask", img, hsv, lower_white, upper_white)
# yellow_img = colorMask("yellow_mask", img, hsv, lower_yellow, upper_yellow)
# lane_img = colorMask("lane_mask", img, hsv, lower_lane, upper_lane)
undist = img
gray = cv2.cvtColor(white_img, cv2.COLOR_RGB2GRAY)
ret, binary = cv2.threshold(gray, 200, 255,
cv2.THRESH_BINARY | cv2.THRESH_TRIANGLE)
ret, binary2 = cv2.threshold(gray, 200, 1,
cv2.THRESH_BINARY | cv2.THRESH_TRIANGLE)
#cv2.imshow("binary", binary)
rgb = cv2.cvtColor(binary, cv2.COLOR_GRAY2RGB)
cv2.imshow("rgb", rgb)
#cv2.imshow("binary2", binary)
# get the thresholded binary image
thresholded = thresholding(rgb)
cv2.imshow("thresholded*255", thresholded * 255)
#perform perspective transform
thresholded_wraped = cv2.warpPerspective(binary2,
M,
img.shape[1::-1],
flags=cv2.INTER_LINEAR)
cv2.imshow("thresholded_wraped*255", thresholded_wraped * 255)
#perform detection
if left_line.detected and right_line.detected:
print("find line")
left_fit, right_fit, left_lane_inds, right_lane_inds = utils.find_line_by_previous(
thresholded_wraped, left_line.current_fit, right_line.current_fit)
else:
print("no find line")
left_fit, right_fit, left_lane_inds, right_lane_inds = utils.find_line(
thresholded_wraped)
left_line.update(left_fit)
right_line.update(right_fit)
#draw the detected laneline and the information
area_img = utils.draw_area(undist, thresholded_wraped, Minv, left_fit,
right_fit)
cv2.imshow("area_img", area_img)
cv2.waitKey(50)
#print(area_img)
#curvature,pos_from_center = utils.calculate_curv_and_pos(thresholded_wraped,left_fit, right_fit)
#result = utils.draw_values(area_img,curvature,pos_from_center)
#result=area_img
result = rgb
return result
plt.savefig('lane fitting.png')
# In[23]:
for binary_warped in perspective_transformed_images:
left_fit, right_fit, left_lane_inds, right_lane_inds = utils.find_line(
binary_warped)
curvature, distance_from_center = utils.calculate_curv_and_pos(
binary_warped, left_fit, right_fit)
# In[24]:
plt.figure(figsize=(20, 40))
for i in range(0, (len(undistorted_test_images))):
left_fit, right_fit, left_lane_inds, right_lane_inds = utils.find_line(
perspective_transformed_images[i])
curvature, distance_from_center = utils.calculate_curv_and_pos(
perspective_transformed_images[i], left_fit, right_fit)
result = utils.draw_area(undistorted_test_images[i],
perspective_transformed_images[i], Minv, left_fit,
right_fit)
img = utils.draw_values(result, curvature, distance_from_center)
plt.subplot(len(undistorted_test_images), 1, i + 1)
plt.title('result')
plt.imshow(img)
plt.savefig('result.png')
# In[ ]: