(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

Image segmentation algorithm based on integration of k-means clustering, watershed and binary partition tree

Image segmentation algorithm based on integration of k-means clustering, watershed and binary partition tree / Wong Kok Choy
Pada masa kini, teknologi imej komputer sedang berkembang dengan kadar yang sangat cepat. Dengan ini, ahli-ahli sains komputer telah pun memberi tumpuan kepada pengembangan algorima-algorima pemprosesan imej. Untuk merealisasikan algorima yang boleh mengesan ciri dan mengumpul maklumat secara automatik, bidang peruasan imej selalunya menjadi kajian kegemaran penyelidik. Sungguhpun pelbagai teknik peruasan imej diperkenalkan, namun ketepatan dan keberkesanan teknik-teknik ini masih jauh daripada pengesanan yang dibuat oleh manusia, terutamanya apabila imej tersebut mengalamai variasi pencahayaan seperti pencahayaan tidak sekata, pantulan cahaya dan sebagainya. Jadi, objektif utama projek ini adalah untuk meningkatkan keteguhan algorima peruasan imej terhadap pelbagai variasi pencahayaan. Peningkatan algorima peruasan imej yang dibentangkan di sini melibatkan penggunaan algorima Pengelompokan Purata K sebagai pre-peruasan, dan hasilnya akan diproses dengan Transformasi Tadahan Air (Watershed Transform) sebelum gabungan pokok bahagian binari (Binary Partition Tree, BPT) dilaksanakan. Pengelompokan Purata K dijalankan untuk mengurangkan variasi pencahayaan, dan hasilnya akan mengalami Transformasi Tadahan Air. Bahagian-bahagian yang dihasilkan semasa Transformasi Tadahan Air akan dijadikan nod dedaun dalam BPT. Dengan berpandukan kriteria-kriteria gabungan yang telah ditentukan, bahagian-bahagian tersebut akan bergabung dua dengan dua sehingga ia sampai ke punca pokok iaitu seluruh imej tersebut. Untuk menilai prestasi algorima yang dicadangkan tersebut, imej dan hasil peruasan asas di dalam pangkalan data peruasan Sharon Alpert telah digunakan untuk mengkaji kesan pampasan pencahayaan. Berdasarkan keputusan yang diperoleh, kita boleh mengatakan bahawa dengan adanya implimentasi algorima Pengelompokan Purata K, proses peruasan ini akan menjadi lebih teguh ke atas variasi pencahayaan. _______________________________________________________________________________________________________ Nowadays, computer vision technology had grown at a supreme speed. Along with the advancement of computer vision technology, many computer scientists had focused in the development of image processing algorithm. With the aim to construct automated feature detection and information extraction image processing algorithm, the image segmentation is being explored by many researchers. Although many types of image segmentation technique had been proposed, their accuracy and efficiency are still far away from detection done by human, especially when the image experiences some illumination variations (Uneven lighting, light reflection, etc.). Thus, the main objective of this project is to improve the robustness of an image segmentation algorithm against various illumination conditions. The improved image segmentation algorithm presented here is the use of K-Means Clustering algorithm as presegmentation, and its output will undergo Watershed Transform before Binary Partitioning Tree (BPT) merging process takes place. K-Means Clustering is implemented to reduce illumination changes, and its output image undergoes Watershed Transform. The resultant regions obtained from Watershed Transform are then used as the leaves node in BPT. Based on merging criteria, these regions will be merged two by two until it reaches the root of the tree (the entire image). To evaluate the performance of the proposed algorithm, images and ground truth result in Sharon Alpert’s segmentation database are used to evaluate the illumination compensation effect. From the result obtained we can say that by adding K-Means Clustering algorithm, the segmentation process is now more robust against illumination variation.
Contributor(s):
Wong Kok Choy - Author
Primary Item Type:
Final Year Project
Identifiers:
Accession Number : 875004725
Language:
English
Subject Keywords:
computer; vision; technology
First presented to the public:
6/1/2012
Original Publication Date:
7/14/2020
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 95
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2020-07-14 17:02:12.399
Submitter:
Mohd Jasnizam Mohd Salleh

All Versions

Thumbnail Name Version Created Date
Image segmentation algorithm based on integration of k-means clustering, watershed and binary partition tree1 2020-07-14 17:02:12.399