Information and Communication Technology for Development for Africa. First International Conference, ICT4DA 2017, Bahir Dar, Ethiopia, September 25–27, 2017, Proceedings

Research Article

Comparison of Moving Object Segmentation Techniques

  • @INPROCEEDINGS{10.1007/978-3-319-95153-9_26,
        author={Yaecob Gezahegn and Abrham Gebreselasie and Dereje Gebreal and Maarig Hagos},
        title={Comparison of Moving Object Segmentation Techniques},
        proceedings={Information and Communication Technology for Development for Africa. First International Conference, ICT4DA 2017, Bahir Dar, Ethiopia, September 25--27, 2017, Proceedings},
        proceedings_a={ICT4DA},
        year={2018},
        month={7},
        keywords={Clustering Segmentation PCA KM GA GAIK},
        doi={10.1007/978-3-319-95153-9_26}
    }
    
  • Yaecob Gezahegn
    Abrham Gebreselasie
    Dereje Gebreal
    Maarig Hagos
    Year: 2018
    Comparison of Moving Object Segmentation Techniques
    ICT4DA
    Springer
    DOI: 10.1007/978-3-319-95153-9_26
Yaecob Gezahegn1,*, Abrham Gebreselasie2,*, Dereje Gebreal1,*, Maarig Hagos3,*
  • 1: Addis Ababa University
  • 2: Addis Ababa Science and Technology University
  • 3: Mekelle University
*Contact email: yaecob.girmay@gmail.com, kgabrham@gmail.com, dereje.hailemariam@aait.edu.et, maarig2000@gmail.com

Abstract

Moving object segmentation is the extraction of meaningful features from series of images. In this paper, different types of moving object segmentation techniques such as rincipal omponent nalysis (PCA), -eans clustering (KM), enetic lgorithm (GA) and enetic lgorithm nitialized -means clustering (GAIK) have been compared. From our analysis we have observed that PCA reduces dimension or size of data for further processing, which in return reduces the computational time. However, the segmentation quality sometimes becomes unacceptable. On the other hand, due to random initialization of its centroids, KM clustering sometimes converges to local minimum which results in bad segmentation. Another algorithm which has been considered in this study is GA, which searches all the feature space and results in a global optimum clustering. Although the segmentation quality is good, it is computationally expensive. To mitigate these problems, KM and GA are merged to form GAIK, where GA helps to initialize the centroids of the clustering. From our study, it has been found out that GAIK is superior to GA in both the quality of segmentation and computational time. Therefore, in general, the analyses of the four algorithms shows that GAIK is optimal for segmenting a moving object.