Distance Based Edge Linking (DEL) for Character Recognition

This article proposes an minimum distance based edge linking algorithms for handwritten character images. Improvement of performance for machine recognition is challenging task due to noise and degraded input images. In the proposed system we enhance the recognition rate of object reconstruction for broken edges by using edge linking. Such edges of objects are reconstructed by using novel Distance based Edge Linking (DEL) approach. Developed new benchmark approach is fill the gaps between nearest edge segment of Binary image map (BIM). We obtain state-of-art performance of proposed system on character recognition (CR) using two datasets MNIST and ISI.

[1]  Jack Bresenham,et al.  A linear algorithm for incremental digital display of circular arcs , 1977, CACM.

[2]  Frank Y. Shih,et al.  Adaptive mathematical morphology for edge linking , 2004, Inf. Sci..

[3]  Marco Dorigo,et al.  Ant colony optimization , 2006, IEEE Computational Intelligence Magazine.

[4]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  Thomas Stützle,et al.  Ant colony optimization: artificial ants as a computational intelligence technique , 2006 .

[6]  Peter Wai-Ming Tsang,et al.  A genetic algorithm for affine invariant recognition of object shapes from broken boundaries , 1997, Pattern Recognit. Lett..

[7]  Laurent Wendling,et al.  Character recognition based on non-linear multi-projection profiles measure , 2015, Frontiers of Computer Science.

[8]  Bidyut Baran Chaudhuri,et al.  Indian script character recognition: a survey , 2004, Pattern Recognit..

[9]  Cuneyt Akinlar,et al.  PEL: A Predictive Edge Linking algorithm , 2016, J. Vis. Commun. Image Represent..

[10]  Hong Yan,et al.  Linking broken character borders with variable sized masks to improve recognition , 1996, Pattern Recognit..

[11]  Joe-Air Jiang,et al.  Mathematical-morphology-based edge detectors for detection of thin edges in low-contrast regions , 2007 .

[12]  Bart Lamiroy,et al.  Spatial Similarity Based Stroke Number and Order Free Clustering , 2010, 2010 12th International Conference on Frontiers in Handwriting Recognition.

[13]  Liu Wei,et al.  An improved sequential edge linking model for contour detection in medical images , 2009, 2009 4th IEEE Conference on Industrial Electronics and Applications.

[14]  Parshuram M. Kamble,et al.  Deep neural network for handwritten Marathi character recognition , 2017 .

[15]  H. Liu,et al.  Automated extraction of coastline from satellite imagery by integrating Canny edge detection and locally adaptive thresholding methods , 2004 .

[16]  J. Ramsay.  Kernel smoothing approaches to nonparametric item characteristic curve estimation , 1991 .

[17]  Bart Lamiroy,et al.  Relative Positioning of Stroke-Based Clustering: a New Approach to Online Handwritten Devanagari Character Recognition , 2012, Int. J. Image Graph..

[18]  Chien-Chang Chen,et al.  Edge detection improvement by ant colony optimization , 2008, Pattern Recognit. Lett..

[19]  Ye Zhou,et al.  Segmentation of petrographic images by integrating edge detection and region growing , 2004, Comput. Geosci..