Scribd
Upload
73 views11 pages

Lane Detection Using Computer Vision

The document describes an experiment to detect lane lines in two images, numbered 8 and 392, using computer vision models. The lane detection process includes pre-processing the images using grayscale and Gaussian blur, applying Canny edge detection, masking the region of interest, applying Hough transform, and adding the extrapolated lines back to the original image. Code snapshots are provided with each step commented with the author's BITS ID and results shown side by side with the time visible. Modifications are suggested to eliminate horizontal line detection, such as increasing the Canny edge detection threshold and adjusting the polygon mask height.

Uploaded by

shubhamforme
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
73 views11 pages

Lane Detection Using Computer Vision

The document describes an experiment to detect lane lines in two images, numbered 8 and 392, using computer vision models. The lane detection process includes pre-processing the images using grayscale and Gaussian blur, applying Canny edge detection, masking the region of interest, applying Hough transform, and adding the extrapolated lines back to the original image. Code snapshots are provided with each step commented with the author's BITS ID and results shown side by side with the time visible. Modifications are suggested to eliminate horizontal line detection, such as increasing the Canny edge detection threshold and adjusting the polygon mask height.

Uploaded by

shubhamforme
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

ADAS Assignment

Lane Detection for Image number ‘8’ & Image number ‘392’

A) Aim of the experiment –


Aim of the experiment is to detect lane line using computer vision
models.
B) Brief description of the experiment-
Detecting Lane lines is a fundamental task for autonomous vehicles
while driving on the road. It is the building block to other path planning
and control actions like breaking and steering. The lane detection follows
these steps:

1. Pre-process image using grayscale and gaussian blur


2. Apply canny edge detection to the image
3. Apply masking region to the image
4. Apply Hough transform to the image
5. Add the extrapolated lines to the input image
1) Pre-process image using grayscale and gaussian blur:

Grayscale Image - 8

Grayscale Image – 392


2) Apply canny edge detection to the image

Canny Edge Image – 8

Canny Edge Image – 392


3) Apply masking region to the image

Masking region Image – 8

Masking region Image – 392


4) Apply Hough transform to the image

Hough transform Image – 8

Hough transform Image – 392


5) Add the extrapolated lines to the input image

Lane Detected Image- 8

Lane Detected Image- 392


C) Program snapshot with each line of the code commented with
your BITS ID –

Image-1 with all the libraries and function definitions

Image-2 with all the function calls and implementation

Note: Same code is applicable for Image 392 the only difference in code would be the image
number in the code. In image 2, original_image = cv2.imread('8.jpg') #2022ht65110 shall be
replaced with original_image = cv2.imread('392.jpg') #2022ht65110 and rest will be kept same.
D) Snapshot with program and results side by side. The time on the
computer should be visible-
For Image number 8
For Image number 392

E) The title of the plots/Images should be your BITS ID and same


should be included in your report

All the above snapshots are with having titles as BITS ID and each line
of code is commented with BITS ID which is 2022ht65110.
Enhanced results: From the above results it is evident that apart from
Lane detection there is a horizontal line detection so in order to eliminate
the horizontal line following modifications can be done

1) canny_edge = cv2.Canny(lane_gs_blur,150,150)
Comment: First threshold value changed from 60 to 150
2) poly_mask =
np.array([[(0,650),(int(width/2),int(height/1.65)),(width,650)]])
Comment: height is changed to 650.

For Image 8

Enhanced Lane Detected Image- 8


For Image 392

Enhanced Lane Detected Image- 392

You might also like