dawn360 / advanced-lane-lines Goto Github PK

Project Goals

The goals / steps of this project are the following:

• Compute the camera calibration matrix and distortion coefficients given a set of chessboard images.
• Apply a distortion correction to raw images.
• Use color transforms, gradients, etc., to create a thresholded binary image.
• Apply a perspective transform to rectify binary image ("birds-eye view").
• Detect lane pixels and fit to find the lane boundary.
• Determine the curvature of the lane and vehicle position with respect to center.
• Warp the detected lane boundaries back onto the original image.
• Output visual display of the lane boundaries and numerical estimation of lane curvature and vehicle position.

The code for this step is contained in the first code cell of the IPython notebook located in "P2.ipynb".

I start by preparing "object points", which will be the (x, y, z) coordinates of the chessboard corners in the world. Here I am assuming the chessboard is fixed on the (x, y) plane at z=0, such that the object points are the same for each calibration image. Thus, objp is just a replicated array of coordinates, and objpoints will be appended with a copy of it every time I successfully detect all chessboard corners in a test image. imgpoints will be appended with the (x, y) pixel position of each of the corners in the image plane with each successful chessboard detection.

I then used the output objpoints and imgpoints to compute the camera calibration and distortion coefficients using the cv2.calibrateCamera() function. I applied this distortion correction to the test image using the cv2.undistort() function and obtained this result:

To demonstrate this step, I applied this distortion correction to the test image using the cv2.undistort() function and obtained this result:

I used a combination of color and gradient thresholds to generate a binary image (thresholding steps at cell 4 through 5 ). After experimenting with a combination of gabs_sobel_thresh (x and y), mag_thresh, dir_threshold and hls_threshold, I settled on a combination of sobel_x,mag_thresh and a new method i call alternative sobel_x( where i used the s channel in hls instead of gray image for the kernel convolution)

combined = np.zeros_like(gradx)
combined[ ((mag_binary == 1) & (gradx == 1))] = 1
combined[ ((combined == 1) | (gradxAlt == 1))] = 1

Here's an example of my output for this step.

The code for my perspective transform includes a function called warp_transform(), which appears in cell 6 of the IPython notebook. The warper() function takes as inputs an image (img) and an (offset). I chose the hardcode the source and destination points in the following manner:

src = np.float32(
    [[(img_size[0] / 2) - 57, img_size[1] / 2 + 97],
    [((img_size[0] / 6) - 10), img_size[1]],
    [(img_size[0] * 5 / 6) + 60, img_size[1]],
    [(img_size[0] / 2 + 60), img_size[1] / 2 + 97]])
    dst = np.float32(
    [[(img_size[0] / 4), 0],
    [(img_size[0] / 4), img_size[1]],
    [(img_size[0] * 3 / 4), img_size[1]],
    [(img_size[0] * 3 / 4), 0]])

This resulted in the following source and destination points:

I verified that my perspective transform was working as expected by drawing the src and dst points onto a test image and its warped counterpart to verify that the lines appear parallel in the warped image.

The code for finding lane pixels uses a function called find_lane_pixels found in cell 9 of the IPython notebook. find_lane_pixels takes in a (binary_warped) image uses a histogram of the sum of pixels on the vertical axis to detect the lanes under the hood in the image. Then with a sliding widow, we trace the groups of pixels along the vertical axis to detect both left and right lane lines. Then we fit the detected lane pixels with a 2nd order polynomial kinda like this:

 # Find our lane pixels first
    leftx, lefty, rightx, righty, out_img = find_lane_pixels(binary_warped)

    # Fit a second order polynomial to each using `np.polyfit`
    left_fit = np.polyfit(lefty, leftx, 2)
    right_fit = np.polyfit(righty, rightx, 2)

I did this in cells 10 line 32 of the IPython notebook. The code for calculating the curvature uses a function called measure_curvature_real which accepts the x and y pixel values and converts from pixels space to meters. The image center and offset from lane center is calculated using the code below.

# Calculate the radius of curvature in meters for both lane lines
    left_curverad, right_curverad = measure_curvature_real(left_fitx,right_fitx,ploty)
    avg_curve = (left_curverad+right_curverad)/2

    # calculate offset from center
    # lane_center = (right_fitx[0]+left_fitx[0])/2
    left_value = left_fit[0]*image.shape[0]**2 + left_fit[1]*image.shape[0] + left_fit[2]
    right_value = right_fit[0]*image.shape[0]**2 + right_fit[1]*image.shape[0] + right_fit[2]
    lane_center = (right_value+left_value)/2
    xm_per_pix = 3.7/700
    image_center = image.shape[1]/2
    offset_center = round((image_center-lane_center)*xm_per_pix,3)

I implemented this step in cell 7 line 82 in the function map_lane(). Here is an example of my result on a test image:

Here's a link to my video result

Here I'll talk about the approach I took, what techniques I used, what worked and why. The major factor i found in detecting the lanes correctly is getting a good binary image that emphasises the lane pixels as much as possible. Another problem was whenever a car drove by in the next lane, close to my lane lines, the sliding window includes the car pixels results in a wrong lane detection. To resolve this i had to reduce the size of the sliding window on the function search_around_poly from 100 to 70 to narrow the boundary for the sliding window. This pipeline will fail in generating reliable binary_images with emphasises on lane lines, other crossing vechiles and repaved road edges could be included in the edge detection phase. I think this could be improved by adding color masking mainly to detect the yellow lane lines, masking of white lines in addition to the combined_binary could help improve highlighting of lane lines for detection.