Shape-Based Image Retrieval Using Canny Edge Detection and K-Means Clustering Algorithms for Medical Images

1. CBMIR: Content Based Medical
Image Retrieval System
Supervisor:
Rethwan Faiz
Group member:
Haider, Dollon(15-28614-1)
Khan, Md Akib Shahriar(15-28592-1)
Hasan, Shah Md Nahid(14-27959-3)
Mallick, Rabin(14-26346-1)
Maisha Rowshon Islam(15-28427-1)

2. In this common era accumulation of large collections of digital images has
become very difficult. Considering the Medical Sector it plays a vital role.
To prevent this problem there arrived CBMIR(Content Based Medical Image
Retrieval ) system.

3. CBMIR(Content Based Medical Image Retrieval
) system
In this work,
“CBMIR: Shape-Based Image Retrieval Using Canny Edge Detection and K-
Means Clustering Algorithms for Medical Images”, has been developed to
retrieve the medical images from huge volume of medical databases

4. Efficient medical image retrieval system
Pre
Processing
Feature
Extraction
Classification
Retrieval

5. CBMIR Framework:

6. Image Enhancement
 Preprocessing - Correcting Nonuniform
illumination
 Read Image
 Use Morphological Opening to Estimate the Background
 Subtract the Background Image from the Original Image
 Increase the Image Contrast
 Threshold the Image

7. Preprocessing
Read
Image
Morphological
Opening to
Estimate the
Background
Background
Subtraction
Read Image Subtract the Background

8. Preprocessing
Background
Subtraction
Increase the
Image Contrast
Threshold the
Image
Increase the Image Contrast Threshold the Image

9. Preprocessing
MATLAB code for Preprocessing

10. Edge detection
 Image processing technique to finding the boundaries of objects within
images.
 Why canny edge detector?

11. Canny Edge Detector Steps
 i) Find magnitude and orientation of gradient.
 ii) Apply “Non-maximum Suppression”.
 iii) Edge Threshold.

12. Different threshold value applied
Fig: 0.2 threshold Fig: 0.3 thresholdFig: 0.1 threshold

13. Canny on medical Image

14. MATLAB code for canny edge detector:

15. Image Classification
using
K-Means Clustering

16. Clustering
Clustering:
– Unsupervised learning
– Requires data, but no labels
– Detect patterns e.g. in
• Group emails or search results
• Customer shopping patterns
• Regions of images
– Useful when don’t know what you’re
looking for
– But: can get gibberish

17. Clustering
• Basic idea: group together similar instances
• Example: 2D point patterns

18. K-Means
• An iterative clustering
algorithm
– Initialize: Pick K random
points as cluster centers
– Alternate:
1. Assign data points to
closest cluster center
2. Change the cluster
center to the average
of its assigned points
– Stop when no pointsʼ
assignments change

19. K-Means
• An iterative clustering
algorithm
– Initialize: Pick K random
points as cluster centers
– Alternate:
1. Assign data points to
closest cluster center
2. Change the cluster
center to the average
of its assigned points
– Stop when no pointsʼ
assignments change

20. K-‐meansclustering: Example
• Pick K random
points as cluster
centers (means)
Shown here for K=2
11

21. K-‐meansclustering: Example
Iterative Step 1
• Assign data points to
closest cluster center
12

22. K-‐meansclustering: Example
13
Iterative Step 2
• Change the cluster
center to the average of
the assigned points

23. K-‐meansclustering: Example
• Repeat until
convergence
14

24. K-‐meansclustering: Example
15

25. Matlab Code for K-Means

26. Stored into database

27. Image Retrieval
Similarity Comparison :

28. RESULT ANALYSIS
Retrieval Efficiency:
For retrieval efficiency, traditional measures namely precision and recall were computed
with 1000 real time medical images [2, 3]. Standard formulas have been computed
for determining the precision and recall
measures.
Precision (P) is the ratio of the relevant images to the total number of images retrieved
P=r/n1
Where,
r-number of relevant images retrieved
n1-total number of images retrieved
Recall(R) is the percentage of relevant images among all possible relevant images

29. EXPERIMENTAL SETUP AND RESULT
ANALYSIS
R=r/n2
Where,
r-number of relevant images retrieved
n2-total number of relevant images in the database
By randomly selecting some sample query images from the MATLAB-Image
Processing tool Box-Workspace
Database, the system was tested and the results are shown in the following
Table.

30. EXPERIMENTAL SETUP AND RESULT
ANALYSIS
Table :
PRECISION AND RECALL VALUES IN %

31. CONCLUSION

32. Reference :
 [1] B.Ramamurthy, CBMIR: Shape-Based Image Retrieval using Canny Edge Detection
and K-means Clustering Algorithms for Medical Images, International Journal of
Engineering Science and Technology (IJEST), Vol. 3 No. 3 pp.1870-1877, 2011.
 [2] S. Nandagopalan, Dr. B. S. Adiga, and N. Deepak “A Universal Model for Content-
Based Image Retrieval” World Academy of Science, Engineering and Technology 46 2008
 [3] Mohammed Eisa and Ibrahim Elhenawy and A. E. Elalfi and Hans Burkhardt, “Image
Retrieval based on Invariant Features and Histogram Refinement”, ICGST International
Journal on Graphics, Vision and Image Processing, March 2006, pp. 7-11.