lane_detection_2.py

from numpy import imag
from Utils.misc import *
from Utils.lane_functions import *
import matplotlib.pyplot as plt

def main():
    video = 'challenge.mp4'
    cap = cv2.VideoCapture(video)
    fourcc = cv2.VideoWriter_fourcc('m', 'p', '4', 'v')
    video = cv2.VideoWriter('lane2.mp4',fourcc, 15, (1280, 720))
    if (cap.isOpened()== False):
        print("Error opening video stream or file")
    while True:
        ret, frame = cap.read()
        if ret == True:
            image = frame.copy()
            hls_img = cv2.cvtColor(image, cv2.COLOR_BGR2HLS)
            # mask for white lane
            white_mask = cv2.inRange(hls_img, (0, 200, 0), (255, 255,255))
            # mask for yellow lane
            yellow_mask = cv2.inRange(hls_img, (20,120,80), (45, 200, 255))
            # final mask
            mask = cv2.bitwise_or(white_mask, yellow_mask)
            # thresholding the L channel of HLS 
            _,L_thresh = cv2.threshold(hls_img[:,:,1], 128, 255, cv2.THRESH_BINARY)
            L_thresh = cv2.GaussianBlur(L_thresh, (3, 3), 0)
            # Perform sobel edge detection on the thresholded channel
            sobel_binary = sobel(L_thresh)            
            combined_binary = cv2.bitwise_or(mask, sobel_binary.astype(np.uint8))
            src_pts = np.array([[240, image.shape[0]-40], [590, image.shape[0]- 260 ], [745, image.shape[0]-260], [1100,image.shape[0]-40]], np.float32)
            # cv2.polylines(image, np.int32([src_pts]), True, (0,0,255), 3)
            dst_pts = np.array([[240, image.shape[0]-40], [240, 0], [1100, 0], [1100, image.shape[0]-40]], np.float32)
            lane_warped, inv_matrix = perspective_transform(combined_binary, src_pts, dst_pts)
            histogram = np.sum(lane_warped[lane_warped.shape[0]//2:,:], axis=0)
            midpoint = int(histogram.shape[0]/2)
            peak1 = np.argmax(histogram[:midpoint])
            peak2 = np.argmax(histogram[midpoint:]) + midpoint
            if histogram[peak1] > histogram[peak2]:
                flag = True
            else:
                flag =False
            # print(histogram[peak1])
            # print(histogram[peak2])
            lane_line_img, ind1, ind2, y_ind, curv, turn= find_lane_pixels(lane_warped, peak1, peak2, image.shape[1], 2)
            img_warped, inv_matrix = perspective_transform(image, src_pts, dst_pts)
            final_lane = overlay_lane_lines(img_warped, image, ind1, ind2, y_ind, inv_matrix, flag)
            text = 'Left curvature: ' + str(curv[0])
            text1 = 'Right curvature: ' + str(curv[1])
            final_lane = cv2.putText(final_lane, text, (50, 50), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 0, 0), 1, cv2.LINE_AA)
            final_lane = cv2.putText(final_lane, text1, (50, 100), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 0, 0), 1, cv2.LINE_AA)
            final_lane = cv2.putText(final_lane, turn, (50, 150), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 0, 0), 1, cv2.LINE_AA)
            # plt.plot(histogram)
            # plt.show()
            cv2.imshow('frame', final_lane)
            # cv2.waitKey(0)
            video.write(final_lane)
            # plt.imshow(image, cmap = 'gray')
            if cv2.waitKey(1) & 0xFF ==ord('q'):
                break
        else:
            break

    cap.release()

if __name__ == '__main__':
    main()