|Title:||Approaches using mathematical morphology for finding closed-loop skeletons|
|Subject:||Image processing -- Mathematics|
Image processing -- Digital techniques
Hong Kong Polytechnic University -- Dissertations
|Pages:||iv, 80, 2 leaves : ill. ; 30 cm|
|Abstract:||Obtaining an outline (skeleton) of a closed-loop object is an important procedure in shape analysis and image recognition. There are three major steps, namely segmentation, thinning and skeletal leg removal, in finding the skeleton of a closed-loop object from a raw grey-scale image. We propose approaches based on mathematical morphology to obtain the skeleton in an accurate and fast way. To segment the input grey-scale image accurately, we first estimate the sizes of the noise, the main object, and the varying background based on the concept of grey-scale morphological opening. A morphological image enhancement algorithm is then formulated which makes use of the size information to remove the noise and the background. Finally a threshold value selection algorithm is used and the segmentation is completed by a global thresholding at the selected threshold value. Once a binary image is obtained, an integrated approach which uses mathematical morphology, fuzzy logic and multiresolution techniques is employed to thin the image and remove unwanted skeletal legs. We propose three speed-up algorithms for fast leg removal, and compare their efficiencies in a formal probabilistic analysis. To obtain prior information on the skeletal leg properties to determine the fastest algorithm, a fuzzy logic approach is used. We propose procedures to formulate the fuzzification functions and the fuzzy inference mechanism. Finally, illustrations of applying these approaches to real-scene images are included.|
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item: