Writeup.md

Udacity SDCND Project_1 : Finding lane lines on the Road

In this project, Use computer vision technology to find lane lines in the picture

---

Process

From the pictures of the road I extracts the part where the lane is present. Then, the part is clarified in color and synthesized with the original photograph to recognize the lane. And apply the result to video of the car

---

Details (soildWhiteRight soildYellowLeft)

First, include library to deal with mathematical operation and receive image

• mpimg.imread()

And make the image gry_color then use the Gaussian function to smooth the wrong pixels in the image

• cv2.cvtColor(image,cv2.COLOR_RGB2GRAY), cv2.GaussianBlur(gray,(kernel_size, kernel_size),0)

Use a candy effect to select boundaries on the photo.

• cv2.Canny(blur_gray, low_threshold, high_threshold)

The rate of change with respect to the color of the pixel becomes larger at the boundary value of the object of the photograph. Pixels smaller than the minimum threshold of this rate of change are removed and pixels between the highest threshold and the lowest threshold are alive if they are associated with pixels above the highest threshold.

Now that an area where the lane is present in the picture is constant, Remove other pixels except the area.

• cv2.fillPoly(mask, vertices, ignore_mask_color)

Detect lines on the image through the hough transform to fill the lines with color and plot the lines in blank image

• cv2.HoughLinesP(masked_edges, rho, theta, threshold, np.array([]), min_line_length, max_line_gap)

The last process is to combine original image and detected lines

• cv2.addWeighted(line_image, 1, image, 1, 0)