Loading train1/pre-processing.py +68 −27 Original line number Diff line number Diff line Loading @@ -17,7 +17,7 @@ def transform(input_img, num): M = cv2.getPerspectiveTransform(pts1, pts2) dst = cv2.warpPerspective(input_img, M, (600, 600)) dst_extract = abs_thresh(dst, sobel_kernel=3, mag_thresh=(50, 210), direction='x') dst_extract = abs_thresh(dst, sobel_kernel=3, mag_thresh=(35, 210), direction='x') # dst_extract = np.where(dst_extract>0.5,1,0) # histogram = np.sum(dst_extract[dst_extract.shape[0] // 2:, :], axis=0) Loading Loading @@ -57,27 +57,30 @@ def abs_thresh(img, sobel_kernel=9, mag_thresh=(0, 255), return_grad=False, dire def add_line_mark(input_img): # determind start point input_img[450:599, 0:150] = 0 input_img[450:599, 400:599] = 0 # determined start point # delete triangle mult_matrix = np.tri(600,600,420).T input_img = input_img * mult_matrix * np.flip(np.tri(600,600,440).T,1) # input_img[450:599, 0:150] = 0 # input_img[450:599, 450:599] = 0 # fill blank input_img_hist_shape = input_img[400:599, 0:599] # plt.plot(histogram) # plt.show() # use window to find points (windows size: H30 W50) ##LEFT histogram = np.sum(input_img_hist_shape[input_img_hist_shape.shape[0] // 2:, :], axis=0) peak_L = find_peaks(histogram[0:299], distance=150)[0] def gen_windows(bottom_H, midpoint_W, H, W): window = input_img[bottom_H - 30:bottom_H, midpoint_W - 25:midpoint_W + 25] window = input_img[bottom_H - 30:bottom_H, midpoint_W - 50:midpoint_W + 50] H = 30 W = 50 W = 100 # window_arr = input_img[] histogram_window = np.sum(window[window.shape[0] // 2:, :], axis=0) window_peak = find_peaks(histogram_window, distance=50)[0] window_peak = find_peaks(histogram_window, distance=100)[0] return_mid = 0 if window_peak.size > 0: if window_peak > 5: Loading @@ -86,40 +89,78 @@ def add_line_mark(input_img): return_mid = midpoint_W # plt.plot(histogram_window) # plt.show() # print(return_mid) return bottom_H - H / 2, return_mid lane_list_L = [] # gen_windows(299,peak[0],30,50) plt.imshow(input_img) # iteraton(L) # print(peak_L) midpoint_tmp_L = peak_L[0] for bottom in range(599, 29, -30): window_H, window_WM = gen_windows(bottom, midpoint_tmp_L, 30, 50) midpoint_tmp_L = int(window_WM) plt.scatter(midpoint_tmp_L, bottom) ##RIGHT histogram = np.sum(input_img_hist_shape[input_img_hist_shape.shape[0] // 2:, :], axis=0) peak_R = find_peaks(histogram[300:599], distance=150)[0][0]+299 if peak_R.size > 0 and histogram[peak_R] > 50: pass def find_hist_iteration(baseline=500,start_W=0): input_img_hist_iter = input_img[baseline:baseline+ 99, start_W:start_W+300] find_hist_iteration_iter = np.sum(input_img_hist_iter[input_img_hist_iter.shape[0] // 2:, :], axis=0) peak_hist = find_peaks(find_hist_iteration_iter, distance=180)[0] if find_hist_iteration_iter[peak_hist] < 10 or peak_hist.size==0: # print(find_peaks(find_hist_iteration_iter, distance=180)) # print(baseline) find_hist_iteration(baseline-50,start_W) else: peak_R = 400 # iteraton(L) print(peak_R) midpoint_tmp_R = peak_R pass return (peak_hist[0] + start_W) # find_hist_iteration() ##RIGHT # histogram = np.sum(input_img_hist_shape[input_img_hist_shape.shape[0] // 2:, :], axis=0) # peak_L = find_peaks(histogram[300:599], distance=150)[0] # midpoint_tmp_L = peak_L[0] + offset midpoint_tmp_R = find_hist_iteration(500, 300) print(midpoint_tmp_R) for bottom in range(599, 29, -30): window_H, window_WM = gen_windows(bottom, midpoint_tmp_R, 30, 50) window_H, window_WM = gen_windows(bottom, midpoint_tmp_R, 30, 100) midpoint_tmp_R = int(window_WM) plt.scatter(midpoint_tmp_R, bottom) midpoint_tmp_L = find_hist_iteration(500, 0) print(midpoint_tmp_L) for bottom in range(599, 29, -30): window_H, window_WM = gen_windows(bottom, midpoint_tmp_L, 30, 100) midpoint_tmp_L = int(window_WM) plt.scatter(midpoint_tmp_L, bottom) #np.polyfit plt.show() # ##RIGHT # histogram = np.sum(input_img_hist_shape[input_img_hist_shape.shape[0] // 2:, :], axis=0) # peak_R = find_peaks(histogram[300:599], distance=150)[0][0]+299 # if peak_R.size > 0 and histogram[peak_R] > 50: # pass # else: # peak_R = 400 # # iteraton(L) # print(peak_R) # midpoint_tmp_R = peak_R # for bottom in range(599, 29, -30): # window_H, window_WM = gen_windows(bottom, midpoint_tmp_R, 30, 50) # midpoint_tmp_R = int(window_WM) # plt.scatter(midpoint_tmp_R, bottom) # # plt.show() return 0 for i in range(19,20): for i in range(10, 11): input_img = cv2.imread('1492626805094402903/' + str(i) + '.jpg') input_img = cv2.cvtColor(input_img, cv2.COLOR_BGR2RGB) transform(input_img, i) Loading
train1/pre-processing.py +68 −27 Original line number Diff line number Diff line Loading @@ -17,7 +17,7 @@ def transform(input_img, num): M = cv2.getPerspectiveTransform(pts1, pts2) dst = cv2.warpPerspective(input_img, M, (600, 600)) dst_extract = abs_thresh(dst, sobel_kernel=3, mag_thresh=(50, 210), direction='x') dst_extract = abs_thresh(dst, sobel_kernel=3, mag_thresh=(35, 210), direction='x') # dst_extract = np.where(dst_extract>0.5,1,0) # histogram = np.sum(dst_extract[dst_extract.shape[0] // 2:, :], axis=0) Loading Loading @@ -57,27 +57,30 @@ def abs_thresh(img, sobel_kernel=9, mag_thresh=(0, 255), return_grad=False, dire def add_line_mark(input_img): # determind start point input_img[450:599, 0:150] = 0 input_img[450:599, 400:599] = 0 # determined start point # delete triangle mult_matrix = np.tri(600,600,420).T input_img = input_img * mult_matrix * np.flip(np.tri(600,600,440).T,1) # input_img[450:599, 0:150] = 0 # input_img[450:599, 450:599] = 0 # fill blank input_img_hist_shape = input_img[400:599, 0:599] # plt.plot(histogram) # plt.show() # use window to find points (windows size: H30 W50) ##LEFT histogram = np.sum(input_img_hist_shape[input_img_hist_shape.shape[0] // 2:, :], axis=0) peak_L = find_peaks(histogram[0:299], distance=150)[0] def gen_windows(bottom_H, midpoint_W, H, W): window = input_img[bottom_H - 30:bottom_H, midpoint_W - 25:midpoint_W + 25] window = input_img[bottom_H - 30:bottom_H, midpoint_W - 50:midpoint_W + 50] H = 30 W = 50 W = 100 # window_arr = input_img[] histogram_window = np.sum(window[window.shape[0] // 2:, :], axis=0) window_peak = find_peaks(histogram_window, distance=50)[0] window_peak = find_peaks(histogram_window, distance=100)[0] return_mid = 0 if window_peak.size > 0: if window_peak > 5: Loading @@ -86,40 +89,78 @@ def add_line_mark(input_img): return_mid = midpoint_W # plt.plot(histogram_window) # plt.show() # print(return_mid) return bottom_H - H / 2, return_mid lane_list_L = [] # gen_windows(299,peak[0],30,50) plt.imshow(input_img) # iteraton(L) # print(peak_L) midpoint_tmp_L = peak_L[0] for bottom in range(599, 29, -30): window_H, window_WM = gen_windows(bottom, midpoint_tmp_L, 30, 50) midpoint_tmp_L = int(window_WM) plt.scatter(midpoint_tmp_L, bottom) ##RIGHT histogram = np.sum(input_img_hist_shape[input_img_hist_shape.shape[0] // 2:, :], axis=0) peak_R = find_peaks(histogram[300:599], distance=150)[0][0]+299 if peak_R.size > 0 and histogram[peak_R] > 50: pass def find_hist_iteration(baseline=500,start_W=0): input_img_hist_iter = input_img[baseline:baseline+ 99, start_W:start_W+300] find_hist_iteration_iter = np.sum(input_img_hist_iter[input_img_hist_iter.shape[0] // 2:, :], axis=0) peak_hist = find_peaks(find_hist_iteration_iter, distance=180)[0] if find_hist_iteration_iter[peak_hist] < 10 or peak_hist.size==0: # print(find_peaks(find_hist_iteration_iter, distance=180)) # print(baseline) find_hist_iteration(baseline-50,start_W) else: peak_R = 400 # iteraton(L) print(peak_R) midpoint_tmp_R = peak_R pass return (peak_hist[0] + start_W) # find_hist_iteration() ##RIGHT # histogram = np.sum(input_img_hist_shape[input_img_hist_shape.shape[0] // 2:, :], axis=0) # peak_L = find_peaks(histogram[300:599], distance=150)[0] # midpoint_tmp_L = peak_L[0] + offset midpoint_tmp_R = find_hist_iteration(500, 300) print(midpoint_tmp_R) for bottom in range(599, 29, -30): window_H, window_WM = gen_windows(bottom, midpoint_tmp_R, 30, 50) window_H, window_WM = gen_windows(bottom, midpoint_tmp_R, 30, 100) midpoint_tmp_R = int(window_WM) plt.scatter(midpoint_tmp_R, bottom) midpoint_tmp_L = find_hist_iteration(500, 0) print(midpoint_tmp_L) for bottom in range(599, 29, -30): window_H, window_WM = gen_windows(bottom, midpoint_tmp_L, 30, 100) midpoint_tmp_L = int(window_WM) plt.scatter(midpoint_tmp_L, bottom) #np.polyfit plt.show() # ##RIGHT # histogram = np.sum(input_img_hist_shape[input_img_hist_shape.shape[0] // 2:, :], axis=0) # peak_R = find_peaks(histogram[300:599], distance=150)[0][0]+299 # if peak_R.size > 0 and histogram[peak_R] > 50: # pass # else: # peak_R = 400 # # iteraton(L) # print(peak_R) # midpoint_tmp_R = peak_R # for bottom in range(599, 29, -30): # window_H, window_WM = gen_windows(bottom, midpoint_tmp_R, 30, 50) # midpoint_tmp_R = int(window_WM) # plt.scatter(midpoint_tmp_R, bottom) # # plt.show() return 0 for i in range(19,20): for i in range(10, 11): input_img = cv2.imread('1492626805094402903/' + str(i) + '.jpg') input_img = cv2.cvtColor(input_img, cv2.COLOR_BGR2RGB) transform(input_img, i)
