Region based segmentation

about region based segmentation

1. REGION BASED SEGMENTATION PRESENTED BY M.RAMYA M.sc[CS&IT] NADAR SARASWATHI COLLEGE OF ARTS &SCIENCE, VADAPUDUPATTI , THENI.

2. Group similar components (such as, pixels in an image, image frames in a video) to obtain a compact representation. • Applications: Finding tumors, veins, etc. in medical images, finding targets in satellite/aerial images, finding people in surveillance images, summarizing video, etc. • Methods: Thresholding, K-means clustering, etc. INTRODUCTION

3. BASIC FORMULATION n U R i = R i =1 WHERE: R i is a connected region ,i=1,2,….,n R i ᴧ R j = ᴓ for all i and j,i ǂ j P(R i )=TRUE for i=1,2,……,n P(R i U R j ) =FALSE for i=j

4. REGION GROWING Region growing is a procedure that groups pixels or sub regions into larger regions. The simplest of these approaches is pixel aggregation, which starts with a set of “seed” points and from these grows regions by appending to each seed points those neighboring pixels that have similar properties (such as gray level, texture, color, shape). Region growing based techniques are better than the edge-based techniques in noisy images where edges are difficult to detect.

5. Original figure The Seed Points

6. Result of region growing Boundaries of segmented defective welds

7. THE ADVANTAGES AND DISADVANTAGES OF REGION GROWING Advantages Region growing methods can correctly separate the regions that have the same properties we define. Region growing methods can provide the original images which have clear edges with good segmentation results. The concept is simple. We only need a small number of seed points to represent the property we want, then grow the region.

8. Disadvantages Computationally expensive It is a local method with no global view of the problem. Sensitive to noise. Unless the image has had a threshold function applied to it, a continuous path of points related to color may exist which connects any two points in the image.

9. REGION SPLITTING AND MERGING Region Splitting Region growing starts from a set of seed points. An alternative is to start with the whole image as a single region and subdivide the regions that do not satisfy a condition of homogeneity.

10. Region Merging Region merging is the opposite of region splitting. Start with small regions (e.g. 2x2 or 4x4 regions) and merge the regions that have similar characteristics (such as gray level, variance). Typically, splitting and merging approaches are used iteratively CONTU……….

11. Let R represent the entire image region and select a predicate . One approach for segmenting R is to subdivide it successively into smaller and smaller quadrant regions s o that , for R i , P(R i ) = TRUE. If P(R)FALSE divide the image into quadrants . If P is FALSE for any quadrant , subdivide that , quadrants and so on. This particular splitting technique has a convenient representation in the form called quad tree.

12. R R2 R1 R44 R42 R41 R43 R3 R4 Partitioned image Corresponding quad tree

13. Split into four disjoint quadrants any region R i for which P(R i )=FALSE. Merge any adjacent regions R j and R k for which P(R j U R k ) = TRUE. Stop when no further merging or splitting is possible.

14. REGION-ORIENTED SEGMENTATION (a)Original image (b)Result of split and merge procedure (c)Result of thresholding in a

15. P(Ri) =TRUE if at least 80% of the pixels in Ri have the property ǀ zj - mi ǀ ≤ 2 σ i , where zj is the gray level of the jth pixel in Ri , mi is the mean gray level of that region , and σ i is the standard deviation of the gray levels in Ri. If P(Ri) = TRUE under this condition , the values of all the pixels in Ri were set equal to mi.

16. Splitting and merging was done using the algorithm outlined previously. The resulting of applying this technique to the image in (a) is shown in (b) . Note that the image was segmented perfectly. The image (c) was obtained by thresholding (a) , with a threshold placed midway between the two principal peaks of the histogram.

17. CONCLUSION Region and boundary information for the purpose of segmentation. Image segmentation is an essential step in most automatic graphic pattern recognition and scene analysis problems. One segmentation technique over another is dictated mostly by the peculiar characteristics of problem being measured.