Scribd
Upload
15 views

Presentation Major Project

The document presents a project on road lane line detection using computer vision to enhance traffic safety and prevent accidents caused by lane departures. It reviews existing literature, defines the problem of lane detection, and outlines a methodology involving image processing techniques such as Gaussian blur, Canny edge detection, and Hough transform. The ultimate goal is to implement a system that accurately identifies lane markings for autonomous vehicles to maintain lane discipline.

Uploaded by

tanu giri
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
15 views

Presentation Major Project

The document presents a project on road lane line detection using computer vision to enhance traffic safety and prevent accidents caused by lane departures. It reviews existing literature, defines the problem of lane detection, and outlines a methodology involving image processing techniques such as Gaussian blur, Canny edge detection, and Hough transform. The ultimate goal is to implement a system that accurately identifies lane markings for autonomous vehicles to maintain lane discipline.

Uploaded by

tanu giri
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 19

Road Lane Line Detection

System
BY – TANISHA GIRI
MENTOR- MR. DEVENDRA KUMAR MISRA
FINAL SEM PROJECT
GROUP NO. 31
Flow of the Presentation
• Introduction
• Literature Review
• Problem Definition
• Objectives
• Methodology
1. Introduction
• Traffic safety is becoming increasingly crucial as urban traffic grows.
People exiting lanes without respecting the laws cause the majority of
accidents on the avenues.
• The majority of these are the outcome of the driver's interrupted and
sluggish behavior. Lane discipline is essential for both drivers and
pedestrians on the road.
• Computer vision is a form of technology that enables automobiles to
comprehend their environment. It's an artificial intelligence branch
that helps software to understand picture and video input.
• The system's goal is to find the lane markings. Its goal is to provide a
safer environment and better traffic conditions.
1. Introduction
• In lane recognition and departure warning systems, accurate detection of
lane roads is crucial. When a vehicle breaches a lane boundary, vehicles
equipped with the predicting lane borders system direct the vehicles to
avoid crashes and issue an alarm.
• These intelligent systems always provide safe travel, but it is not always
necessary that lane boundaries are clearly visible, as poor road conditions,
insufficient quantity of paint used for marking the lane boundaries, and
other factors can make it difficult for the system to detect the lanes
accurately.
• Other factors can include environmental effects such as shadows cast by
objects such as trees or other automobiles, or streetlights, day and nighttime
conditions, or fog caused by invariant lightening conditions.
2. Literature Review
Name of the Year Journal/Paper/others Details & Summary
author
Shu-Chung Yi, 2015 A lane detection approach The paper presented a lane detection and lane departure
Yeong-Chin Chen, based on intelligent vision methods. Author experiment showed that proposed
Ching-HaurChang approach can easily locate the corresponding lane
parameter space. The accuracy reached 93.8%. The
proposed algorithm speed is about 0.18 s/frame
for640*480 frame size.
Huaizhong Chen, 2010 Research on Real-time Lane The practical application of intelligent vehicle is
Zheliang JIn Line Detection Technology considered, the image segmentation process is requested
Based on Machine Vision to have better robustness, real-time and practicality. The
inverted perspective principle of the camera is used to
realize two-dimensional reconstruction of the lane lines to
reach lane identification purposes, and it has important
application signification to further develop intelligent
vehicle technology.
Name of the author Year Journal/Paper/others Details & Summary

K. Dinakaran, 2020 Advanced lane detection technique The proposed a lane detection
A.StephenSagayaraj, S.K. forstructural highway based on technique using computervision based
Kabilesh,T. Mani, A. computervision algorithm algorithm which accurately detects the
Anandkumar,Gokul lanesand curve on the road and make
Chandrasekaran prior warning to thedriver. So that
many accidents can be avoided due to
theover speeding of vehicles on the
curved regions of the road.

HG Zhu 2021 An Effective Lane Line Detection According to theoretical and


Method Based on Computer Vision experimental analysis, methodcan only
achieve high detection accuracy in
various harshenvironments, but also
achieve fast, accurate and real-
timedetection in processing high
resolution images, whichcannot be
achieved by many neural network
methods.
3. PROBLEM DEFINATION
• Problem 1 : Lane Line detection is a critical component for self-
driving cars and also for computer vision in general. This concept is
used to describe the path for self-driving cars and to avoid the risk of
getting in another lane.
4. Methodology
Experimental procedure
• Pre-process using Gray scale and gaussian blur
• Apply canny edge detection to the image
• Apply masking region to the image
• Apply Hough transform to the image
• Extrapolate the lines found in the Hough transform to construct the left
and right lane lines
• Add the extrapolated lines to the input image
1. Grey Scale
• We convert our image to Gray scale so that we can abstract away from
a Tensor to a Matrix and deal with raw pixel values.(960,540,3)->
(960,540)
• Treat yellow or white lane lines the same
• Pixel list = color channel
• Average color channel values within pixel list (R+G+B/3)
2. Gaussian Blur
• Apply Gaussian Blur to reduce image noise and
details.
• Smooth out the image before applying Canny
edge detection so we do not detect faint edges
• Blur_gray= cv2.GaussianBlur(src, ksize,
std_dev)
• Src: image to modify
• Ksize = kernel size. Larger kernel size implies
averaging/ smoothing over a larger
area.(Sample size for convolution)
3. Canny Edge Detection
• Used to detect boundaries of an image
• Uses differential of pixel values
• Three regions:
A. Pixel higher than upper threshold = edge (kept)
B. Pixel between high and low is accepted if
connected
C. Pixel below low rejected
• Recommend 1:2 or 1:3 low: high threshold ratio
4. Masked Image
• We want to narrow down our analysis to a
section of the image
• Apply a trapezoidal mask over edges
• Use bitwise AND to return image only where
mask pixels are non- zero.
5. Hough Transform
• Goal: To identify straight lines
• Process: For each pixel at (x, y) the hough transform algorithm determines if there is enough evidence of a straight
line at that pixel
• Line= cv2.HoughLinesP(canny_edges, rho, theta, threshold, min_line_length, max_line_gap)
• Rho: Distance of P(pixels). Minimum value of 1. higher Rho= Increased distance resolution
• Theta: angular resolution. Higher theta = higher angular resolution
• Min_line_length: minimum length of a line in pixels you will accept in the output
• Max_line_gap: max distance (in pixels) between segments that you allow to connected into a single line
6. Overlay Hough Image on Original
• Cv2.addWeighted(initial_img, a, img, ß, A)
• Initial_img: original img to write over
• Img: img to overlay on tp of initial_img
Test images(Data used)
Conclusion
• So, by using computer vision techniques in Python, we will identify
road lane lines in which autonomous cars must run. This will be
acritical part of autonomous cars, as the self driving cars should not
cross its lane and should not go in opposite lane to avoid accidents.
REFERENCES
1. Shu-Chung Yi, Yeong-Chin Chen, Ching-Haur Chang, A lane
detection approach based on intelligent vision, 2015.
2. Huaizhong Chen, Zheliang Jin, Research on Real-time Lane Line
DetectionTechnology Based on Machine Vision, 2010.
3. K. Dinakaran, A.Stephen Sagayaraj, S.K. Kabilesh, T. Mani, A.
Anandkumar, Gokul Chandrasekaran, Advanced lane detection
technique for structural highway based on computer vision
algorithm, 2020.
4. HG Zhu, An Effective Lane Line Detection Method Based on
Computer Vision, 2021.
Thank You

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy