TITLE

Segmentation of complex document

AUTHOR(S)
Oudjemia, Souad; Ameur, Zohra; Ouahabi, Abdeldjali
PUB. DATE
January 2014
SOURCE
Carpathian Journal of Electronic & Computer Engineering;2014, Vol. 7 Issue 1, p13
SOURCE TYPE
Academic Journal
DOC. TYPE
Article
ABSTRACT
In this paper we present a method for segmentation of documents image with complex structure. This technique based on GLCM (Grey Level Co-occurrence Matrix) used to segment this type of document in three regions namely, 'graphics', 'background' and 'text'. Very briefly, this method is to divide the document image, in block size chosen after a series of tests and then applying the co-occurrence matrix to each block in order to extract five textural parameters which are energy, entropy, the sum entropy, difference entropy and standard deviation. These parameters are then used to classify the image into three regions using the k-means algorithm; the last step of segmentation is obtained by grouping connected pixels. Two performance measurements are performed for both graphics and text zones; we have obtained a classification rate of 98.3% and a Misclassification rate of 1.79%.
ACCESSION #
100343989

 

Related Articles

  • A modified method for MRF segmentation and bias correction of MR image with intensity inhomogeneity. Xie, Mei; Gao, Jingjing; Zhu, Chongjin; Zhou, Yan // Medical & Biological Engineering & Computing;Jan2015, Vol. 53 Issue 1, p23 

    Markov random field (MRF) model is an effective method for brain tissue classification, which has been applied in MR image segmentation for decades. However, it falls short of the expected classification in MR images with intensity inhomogeneity for the bias field is not considered in the...

  • K-Means Clustering and Affinity Clustering based on Heterogeneous Transfer Learning. Shrivastava, Shailendra Kumar; Rana, J. L.; Jain, R. C. // Journal of Computing;Jan2013, Vol. 5 Issue 1, p31 

    Heterogeneous Transfer Learning aims to extract the knowledge form one or more tasks from same feature space and applies this knowledge to target task of another features space. In this paper two clustering algorithms K-means clustering and Affinity clustering both based on Heterogeneous...

  • Learning what is where from unlabeled images: joint localization and clustering of foreground objects. Chandrashekar, Ashok; Torresani, Lorenzo; Granger, Richard // Machine Learning;Feb2014, Vol. 94 Issue 2, p261 

    'What does it mean, to see? The plain man's answer would be, to know what is where by looking.' This famous quote by David Marr (Vision: A Computational Investigation into the Human Representation and Processing of Visual Information, Freeman, New York, ) sums up the holy grail of vision:...

  • MICROSCOPIC RGB COLOR IMAGES ENHANCEMENT FOR BLOOD CELLS SEGMENTATION IN YCbCr COLOR SPACE FOR K-MEANS CLUSTERING. ABBAS, NAVEED; MOHAMAD, DZULKIFLI // Journal of Theoretical & Applied Information Technology;9/10/2013, Vol. 55 Issue 1, p117 

    Blood is the main component of human body. Blood is composed of Red Blood Cells, White Blood Cells, Platelets and other artifacts. Blood is considered as the main source of identifying various diseases like Malaria, Anemia, and Leukemia etc. In these diseases the main region of interests are the...

  • Semi-Automated Classification of the Physiological Condition of the Carotid Artery in 2D Ultrasound Image Sequences. MUHAMMED, HAMED HAMID; AZAR, JIMMY C. // WSEAS Transactions on Biology & Biomedicine;2014, Vol. 11, p35 

    A novel automated method for the classification of the physiological condition of the carotid artery in 2D ultrasound image sequences is introduced. Unsupervised clustering was applied for the segmentation process in which both spatial and temporal information was utilized. Radial distension is...

  • Customer Segmentation for Telecom with the k-means Clustering Method. Ye, Luo; Qiuru, Cai; Haixu, Xi; Guangping, Zhu // Information Technology Journal;2013, Vol. 12 Issue 3, p409 

    Development of data mining applications is very important for the telecommnnication enterprise which is a typical data-intensive industry. Customer segmentation can help analyze customer composition accurately and promote the quality of service and marketing. Using k-means clustering and the...

  • Vessel Segmentation in Retinal Images Using Multi-scale Line Operator and K-Means Clustering. Saffarzadeh, Vahid Mohammadi; Osareh, Alireza; Shadgar, Bita // Journal of Medical Signals & Sensors;Apr-Jun2014, Vol. 4 Issue 2, p122 

    Detecting blood vessels is a vital task in retinal image analysis. The task is more challenging with the presence of bright and dark lesions in retinal images. Here, a method is proposed to detect vessels in both normal and abnormal retinal fundus images based on their linear features. First,...

  • Parallelization of the Algorithm k-means Applied in Image Segmentation. Lóopez Del Álamo, Cristian José; Fuentes Pérez, Lizeth Joseline; Romero Calla, Luciano Arnaldo // International Journal of Computer Applications;Feb2014, Vol. 88, p1 

    Algorithm k-means is useful for grouping operations; however, when is applied to large amounts of data, its computational cost is high. This research propose an optimization of k-means algorithm by using parallelization techniques and synchronization, which is applied to image segmentation. In...

  • Parallelization of the Algorithm k-means Applied in Image Segmentation. López Del Álamo, Cristian José; Fuentes Pérez, Lizeth Joseline; Romero Calla, Luciano Arnaldo // International Journal of Computer Applications;Feb2014, Vol. 87, p1 

    Algorithm k-means is useful for grouping operations; however, when is applied to large amounts of data, its computational cost is high. This research propose an optimization of k-means algorithm by using parallelization techniques and synchronization, which is applied to image segmentation. In...

Share

Read the Article

Courtesy of THE LIBRARY OF VIRGINIA

Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics