A New Algorithm for Greyscale Objects Representation by Means of the Polar Transform and Vertical and Horizontal Projections

Intelligent computer vision systems can be based on various approaches, methods and algorithms. One of them is the usage of object descriptors, devoted to the representation of objects extracted from digital images (or video sequences) mainly by means of low-level features. There are several features that are applied in computer vision theory and practice, e.g. color, context of the information, luminance, movement, shape, and texture. Amongst them shape, color and texture are especially popular. To the contrary, object representation based on greyscale is less popular. The paper proposes and analyses a new simple and fast algorithm for greyscale object representation. The description method is based on the usage of the polar transform of pixels belonging to an object, and projections – vertical and horizontal. Apart from these operations, some additional steps are also applied in order to improve the efficiency of the developed approach, e.g. median and low-pass filtering. The properties of the proposed greyscale object representation algorithm are analyzed experimentally by means of an exemplary application from the computer vision domain. The ear images (taken from The West Pomeranian University of Technology Ear Database) are applied in the experiment. The obtained results constitute the basis for certain conclusions as well as the proposition of future plans and works on the problem.

[1]  Brian C. Lovell,et al.  Fisher tensors for classifying human epithelial cells , 2014, Pattern Recognit..

[2]  Dariusz Frejlichowski.  Identification of Erythrocyte Types in Greyscale MGG Images for Computer-Assisted Diagnosis , 2011, IbPRIA.

[3]  B. S. Manjunath,et al.  Texture Features for Browsing and Retrieval of Image Data , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[4]  Przemyslaw Klesk,et al.  Boosted Classifiers for Antitank Mine Detection in C-Scans from Ground-Penetrating Radar , 2014, ACS.

[5]  Tat-Jun Chin,et al.  Boosting histograms of descriptor distances for scalable multiclass specific scene recognition , 2011, Image Vis. Comput..

[6]  Dariusz Frejlichowski,et al.  The West Pomeranian University of Technology Ear Database - A Tool for Testing Biometric Algorithms , 2010, ICIAR.

[7]  Balasubramanian Raman,et al.  Center symmetric local binary co-occurrence pattern for texture, face and bio-medical image retrieval , 2015, J. Vis. Commun. Image Represent..

[8]  Michele Nappi,et al.  MEG: Texture operators for multi-expert gender classification , 2017, Comput. Vis. Image Underst..

[9]  Dariusz Frejlichowski.  An Algorithm for Binary Contour Objects Representation and Recognition , 2008, ICIAR.

[10]  David G. Lowe,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004, International Journal of Computer Vision.

[11]  Sébastien Lefèvre,et al.  Chapter 1 - Morphological Texture Description of Grey-Scale and Color Images , 2011 .

[12]  Christine Fernandez-Maloigne,et al.  Evaluation of local and global descriptors for emotional impact recognition , 2016, J. Vis. Commun. Image Represent..

[13]  Loris Nanni,et al.  A simple method for improving local binary patterns by considering non-uniform patterns , 2012, Pattern Recognit..

[14]  Odemir Martinez Bruno,et al.  Three-dimensional connectivity index for texture recognition , 2016, Pattern Recognit. Lett..

[15]  Th. M. Hupkens,et al.  Noise and intensity invariant moments , 1995, Pattern Recognit. Lett..

[16]  Stavros Paschalakis,et al.  Pattern recognition in grey level images using moment based invariant features , 1999 .

[17]  Kostas Delibasis,et al.  Designing texture filters with genetic algorithms: An application to medical images , 1997, Signal Process..

[18]  Loris Nanni,et al.  Combination of projectors, standard texture descriptors and bag of features for classifying images , 2016, Neurocomputing.

[19]  Odemir Martinez Bruno,et al.  Local fractal dimension and binary patterns in texture recognition , 2016, Pattern Recognit. Lett..

[20]  Dariusz Frejlichowski.  Application of the Polar-Fourier Greyscale Descriptor to the Problem of Identification of Persons Based on Ear Images , 2011, IP&C.

[21]  Sasan Mahmoodi,et al.  Gaussian Markov random field based improved texture descriptor for image segmentation , 2014, Image Vis. Comput..