logical+filt1

Views:
 
     
 

Presentation Description

FOR FINDING EDGES OF IMAGE BY USING LOGICAL FILTERS

Comments

Presentation Transcript

Image Edge Detection based on Logical Filtering:

Image Edge Detection based on Logical Filtering By A.OBULESH Asst Prof STUDENT OF Dr.V.V VIJAY KUMAR

Edge detection:

Edge detection How can you tell that a pixel is on an edge?

PowerPoint Presentation:

What is an Edge? Edges: Sudden changes in certain image properties that extend along a contour The perception of edges changes with scale Edge map of an image is very informative From Prof. Al Bovik

What are edges in an image?:

What are edges in an image? Edges are those places in an image that correspond to object boundaries. Edges are pixels where image brightness changes abruptly. Brightness vs. Spatial Coordinates

Types of Edges::

Types of Edges: Discontinuity of intensities in the image Edge models Step Roof Ramp Spike Step Ramp Roof Spike Edges can be used to represent a shape of an object

Edge Detection:

Edge Detection Edge information in an image is found by looking at the relationship a pixel has with its neighborhoods. If a pixel’s gray-level value is similar to those around it, there is probably not an edge at that point. If a pixel’s has neighbors with widely varying gray levels, it may present an edge point.

Edge is Where Change Occurs:

Edge is Where Change Occurs Change is measured by derivative in 1D Biggest change, derivative has maximum magnitude Or 2 nd derivative is zero.

Edge Detectors:

Edge Detectors Gradient operators Prewitt Sobel Robert Laplacian of Gaussian (Marr-Hildreth) Gradient of Gaussian (Canny) Facet Model Based Edge Detector (Haralick)

PowerPoint Presentation:

Motivation Detect sudden changes in image intensity Gradient: sensitive to intensity changes Gradient-Based Methods Gradient operator image Thresholding edge map x(m,n) g(m,n) e(m,n) Gradient: edge pixel threshold non edge pixel

PowerPoint Presentation:

Gradient Operators Gradient-Based Methods Robert: Prewitt: Sobel: g 1 g 2 Local gradient vector: Gradient magnitude: Approximation:

Roberts Operator::

Roberts Operator: It is the oldest edge detector operator It is simple It is used less than the others It can’t find the edges that are multiples of 45 degrees. Primary disadvantage: High sensitivity to noise. Few pixels are used to approximate the gradient The edge’s are not accurate.

The Sobel operator: :

The Sobel operator: -1 0 1 -2 0 2 -1 0 1 1 2 1 0 0 0 -1 -2 -1 The standard defn. of the Sobel operator omits the 1/8 term doesn’t make a difference for edge detection. sobel is the standard one. it is well in finding diagonal edges.

Sobel Edge Detector:

Sobel Edge Detector Image I Threshold Edges

Canny Edge Detector:

Canny Edge Detector It is not effected by noise It is the modern standard one Steps to find canny edges: Smooth by Gaussian Compute x and y derivatives Compute gradient magnitude and orientation

LOG operator::

LOG operator: 1. Smoothing with the Gaussian filter for to avoid noise. The Gaussian filter h(r) = here r 2 =x 2 +y 2 The degree blurring is depending upon the 2.Laplacian of Gaussian function is h(r ) Laplacian operator: 0 1 0 1 -4 1 0 1 0

Proposed Method:

Proposed Method Some operators find edges that are not identified by others. Each operator edges are having some more or less edges. Some edges are find by sobel are not identified by Roberts and vice versa. By good analysis of all edge methods, this new method is proposed for accurate edge. This method well in identifying in miner edge pixels also.

Algorithm:

Algorithm Read given image Find Sobel edge image and Robert edge image then apply logical filtering. Find canny edge image and LOG edge image then apply logical filtering. Apply logical filtering result obtained by step ii and step iii. By post processing we get final edge image.

PowerPoint Presentation:

Original image Sobel operator extract edge (F1) Roberts operator extract edge (F2) Canny operator extract edge (F3) LOG operator extract edge (F4) Logical filtering on F1 & F2 (F5) Logical filtering on F3 & F4 (F6) Logical filtering on F5 & F6 (F7) Edge image by post processing (F7) Fig.1.Flow chart of the logical filtering method

PowerPoint Presentation:

EXPERIMENTAL RESULTS: Fig.2.Original medical image. Fig.2.1.Canny edge image Fig.2.2.Proposed edge image.

PowerPoint Presentation:

Fig.3.Original vehicle image. Fig.3.1.Canny edge image Fig.3.2.Proposed edge image. Fig.4.Original real image. Fig.4.1.Canny edge image Fig.4.2.Proposed edge image.

PowerPoint Presentation:

Conclusion: Compare with canny edge detector, the proposed method is well in case of identifying the minor edge pixels and maintain some continuity. This feature of proposed method is useful for clear diagnosis of medical images, for highlight the vehicle edges in videos and real image edge detection. The proposed approach operates without intervention of high knowledge and by prior information

QUERIES? :

QUERIES?

PowerPoint Presentation:

Thank you

authorStream Live Help