Scribd
Upload
41 views

A Tool For Object Detection and Extraction in

Uploaded by

Ravi Chandra
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
41 views

A Tool For Object Detection and Extraction in

Uploaded by

Ravi Chandra
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 12

SACHIN KANTIPUDI(07BCE264)

CHINNI MAHESH KUMAR (07BCE091)


MAIN OBJECTIVE
Locate the objects present in a given gray scale image

Effective Way of Non-Destructive Testing using


image processing

Objects can also be defects in weld metals


Scope
In Industries it takes lot of time for defect recognition
as
-> Weld Metal is hot to carry
->Manual process is tedious
-> Need Work Labour for it

No existing Automated Tool as of now which counts


number of defects too
WHAT WE HAVE LEARNED
Basic Working of Matlab
Weld Metal Defects
Image processing
->Image conversions
-> Segmentation
->Morphological Operations
->Denoising
HOW WE DIFFER?
Automatic Counting of Objects

Implementation without the need of Neural


Networks
Flow of Steps:
Psuedo Code
Image Preprocessing
Input image
 Convert into gray scale image using rgb2gray
command and display it
 Convert into double data type suitable for
thresholding
 Convert into binary type using im2bw command
using threshold value and display it
 Filter it using average filter
Region of Interest isolation
 Detect edge using sobel edge detector
 Do the following
 Assign [ro,col] to size(x1);
 for j=1 to col
 for i=1 to ro
 if x1(i,j) = 1
 assign i to y(k)
 assign ro to i
 end
 end
 increment k
 end
 find median(y)-10, median(y)+10
 isolate weld bead using above
 use canny edge detector for detecting defects in weld bead
Object counting
 Assign [row,col] to size(im1)
 Do the following
 for i=2 to col
 for j=2 to row
 if im1(j,i)==1
 if im1(j-1,i-1)!=1 && im1(j-1,i+1)!=1 && im1(j-1,i)!=1 && im1(j,i+1)!=1
%&& im1(j,i-1)!=1
 increment k
 end
 end
 end
 end
 Display K/2 as number of defects
Modules:
Denoising and filtering

Image segmentation

Morphological operations
Further Developments
Graphical User Interface

Automatic Threshold Calculation

You might also like