Scribd
Upload
61 views

Edge Based Lane Detection: Seminar On

The document discusses edge-based lane detection techniques. It begins with an overview of edge detection and types of edge detection algorithms. It then describes different edge detection methods like Roberts, Prewitt, Sobel, and Canny edge detectors. It also discusses the steps to implement a lane detection algorithm using edge detection. The document evaluates and compares various edge detection algorithms based on time taken and accuracy. It concludes that Roberts edge detector provides better image quality and speed compared to others for lane detection applications.

Uploaded by

Prem Sagar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
61 views

Edge Based Lane Detection: Seminar On

The document discusses edge-based lane detection techniques. It begins with an overview of edge detection and types of edge detection algorithms. It then describes different edge detection methods like Roberts, Prewitt, Sobel, and Canny edge detectors. It also discusses the steps to implement a lane detection algorithm using edge detection. The document evaluates and compares various edge detection algorithms based on time taken and accuracy. It concludes that Roberts edge detector provides better image quality and speed compared to others for lane detection applications.

Uploaded by

Prem Sagar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 21

JSS ACADEMY OF TECHNICAL EDUCATION

DEPARTMENT OF ELECTRONICS AND COMMUNICATION

Seminar On

EDGE BASED LANE


DETECTION
Presented by,
Prem Sagar P V
( 1JS15EC064 )
OVERVIEW

 Edge detection

 Types of edge detection algorithms

 Implementing lane detection algorithm

 Comparison of edge detection algorithms


EDGES

 Abrupt change in the intensity of pixels.

 Discontinuity in image brightness or contrast.

 Usually edges occur on the boundary of two


regions.
EDGE DETECTION

 Process of identifying edges in an


image to be used as a fundamental
asset in image analysis

 Locating areas with strong intensity


contrasts
EDGE TYPES

 Step Edge

 Ramp Edge

 Ridge

 Roof
EDGE DETECTION STEPS

Steps in edge detection process:


1. Noise Reduction

2. Edge Enhancement

3. Edge Localization
GRADIENT BASED EDGE
DETECTION
 Gradient - Directional change of intensity
 Magnitude of the gradient - strength of the edge

 𝐺 = 𝐺𝑥 2 + 𝐺𝑦 2

 Direction – perpendicular to the direction of edge

−1 𝐺𝑦
 𝜃= tan
𝐺𝑥
TYPES OF GRADIENT BASED EDGE
DETECTION

Roberts Edge Detector

Prewitt Edge Detector

Sobel Edge Detector

Canny Edge Detector


1. ROBERTS EDGE DETECTOR

 2X2 Convolution Mask


 Convolution Mask
 𝐺𝑥 𝐺𝑦
1 0 0 −1
0 −1 1 0
2. PREWITT EDGE DETECTOR

 3X3 Convolution Mask


 Convolution Mask
 𝐺𝑥 𝐺𝑦
−1 0 1 1 1 1
−1 0 1 0 0 0
−1 0 1 −1 −1 −1
3. SOBEL EDGE DETECTOR
 3X3 Convolution Mask
 Convolution Mask
 𝐺𝑥 𝐺𝑦
−1 0 1 1 2 1
−2 0 2 0 0 0
−1 0 1 −1 −2 −1
Canny’s Algorithm

 Step 1

 Smoothing: using a Gaussian filter – reduce noise.

 Step 2

 Gradient magnitude and direction: Sobel or Prewitt

 Step 3

 Non maximum suppression – edge thinning


Cont. Canny’s Algorithm
 Step 4
 Double Threshold
 Step 5
 Edge tracking by hysteresis
APPLICATION OF EDGE DETECTION

 Extract important features of an image

 Corners

 Lines

 Curves

 Recognize objects, boundaries, segmentation.


LANE DETECTION METHODOLOGY

 Step 1
 Conversion of the colour image to grey image.
 Step 2
 Removal of noise
 Step 3
 Locating boundary
 Step 4
 Extracting region of interest (ROI)
 Step 5
 Applying Hough transform
CRITERIA FOR EVALUATING QUALITY
OF EDGE DETECTOR

 Time taken to execute the algorithm

 The probability of a false positive

 The probability of a false negative


Edge detection algorithm based on time
evaluation

Edge First Second Third Average


Operator Elapsed Time Elapsed Time Elapsed Time Time (s)
(s) (s) (s)

Sobel 0.397447 0.400295 0.435820 0.411187

Canny 0.555339 0.658794 0.638951 0.617695

Prewitt 0.432182 0.500475 0.523980 0.485546

Roberts 0.456619 0.448964 0.431709 0.445764


Lane detection algorithm based on time
evaluation

Edge First Second Third Average


Operator Elapsed Elapsed Elapsed time Time (s)
time(s) Time(s) (s)

Sobel 1.129469 1.135940 1.146956 1.137455

Canny 1.147756 1.137728 1.149455 1.144980

Prewitt 1.098794 0.853286 0.992367 0.981482

Roberts 0.854637 0.811514 0.849829 0.838660


CONCLUSION

 Roberts edge detector - simple, better image


quality and better speed

 Canny, Sobel and Prewitt - more sensitive and


time consuming
REFERENCES

 Development of Edge-Based Lane Detection Algorithm Using Image


Processing Moveh Samuel , Maziah Mohamad, Shaharil Mad Saad ,
Mohamed Hussein, INTERNATIONAL JOURNAL ON INFORMATICS
VISUALIZATION, VOL 2 (2018) NO 1
 Shameem A., Rajayalakshmi D. and Syed A. S., A Theoretical Survey
for Edge Detection Techniques and Watershed Transformation,
International Journal of Computer Technology and Electronics
Engineering (IJCTEE) Volume 2, Issue 1, pp: 54-58
 Meghana D. M., Andurkar G. K., 2012. Edge Detection Technique: A
Comparative Approach, World Journal of Science and Technology
 Digital image processing, Rafael C Gonzalez, Richard E Woods
THANK YOU

You might also like