L28_detector_image.py

import cv2, numpy as np
from matplotlib import pyplot as plt


def region_of_interest(img, vertices):
    mask = np.zeros_like(img)  # matrice di 0 con dimensioni come img
    #channel_count = img.shape[2]
    match_mask_color = 255  # maschero con il nero
    cv2.fillPoly(mask, vertices, match_mask_color)
    masked_img = cv2.bitwise_and(img, mask)
    return masked_img

def draw_lines(img, lines):
    new_img = np.copy(img)
    blank_img = np.zeros((new_img.shape[0], new_img.shape[1], new_img.shape[2]), dtype=np.uint8)
    for line in lines:
        for x1, y1, x2, y2 in line:
            cv2.line(blank_img, (x1,y1), (x2,y2), (0,0,255), thickness=3)
    img = cv2.addWeighted(new_img, 0.8, blank_img, 1, 0.0)
    return img

img = cv2.imread('road.png')
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB) # per usare matplotlib

print(img.shape)
height = img.shape[0]
width = img.shape[1]

# define a ROI (Region Of Interest)
ROI_vertices = [
    (0, height),
    (width/2, height/2),
    (width, height)
]

# Canny edge detection before masking
gray_img = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
canny_img = cv2.Canny(gray_img, 100, 200)

# masking the ROI
masked_img = region_of_interest(canny_img, np.array([ROI_vertices], np.int32))

# probabilistic Hough line transform
lines = cv2.HoughLinesP(masked_img, rho=6, theta=np.pi/60, threshold=160, lines=np.array([]), minLineLength=40, maxLineGap=25)
line_img = draw_lines(img, lines)

#plt.imshow(img)
#plt.imshow(masked_img)
#plt.imshow(canny_img)
plt.imshow(line_img)
plt.show()