Content Based Image Retrieval Using Edge Based Feature Extraction In Deep Learning Algorithm

Siva Krishna Sajarao, M Pradeep

Abstract


Content-based image retrieval (CBIR) using edge-based feature extraction in deep learning algorithm has been implemented in this paper. CBIR is widely used in securities, entertainment, medicine and many military applications. the retrieval system process is based on the decomposition of the input image and different levels of filtering analysis such as edge, enhancement and different scaling factors of the training images. Deep learning analysis is majorly concentrated on the different convolution layers. The convolutional neural network layer has been implemented on this process. The convolutional neural network is having one of the layers is convolutional. It is used to extract the feature based on the different kernels to form the feature extraction map. This paper is implemented on the COREL database. As compared to existing methods this paper gives better results in terms of recognition efficiency.


Full Text:

PDF

References


J. T. Favata, G. Srikantan, and S. N. Srihari, “Handprinted

character/digit recognition using a multiple feature/resolution philosophy,” in IWFHR-IV, pp. 57–66, Dec 1994.

A. Marzal and E. Vidal, “Computation of normalized edit dis-

tance and applications,” IEEE Transactions on Pattern Analy- sis and Machine Intelligence, vol. 15, pp. 926–932, Sep 1993.

P. A. V. Hall and G. R. Dowling, “Approximate string match-

ing,” ACM Computing Surveys, vol. 12, pp. 381–402, Dec 1980.

CEDAR, “Intelligent character recognition.” http://www.cedar.buffalo.edu/ICR, 1997.

R. A. Wagner and M. J. Fischer, “The string-to-string correc- tion problem,” Journal of the ACM, vol. 6, pp. 168–178, Jan 1974.

G. M. Landau and U. Vishkin, “Introducing efficient paral-

lelism into approximate string matching and a new serial al- gorithm,” in Proceedings of the 18th Annual ACM Symposium on Theory of Computing, pp. 220–230, 1986.

R. R. Bradford and R. B. Bradford, Introduction to Handwrit-

ing Examination and Identification. Nelson-Hall Publishers: Chicago, 1992.

G. Kim and V. Govindaraju, “Efflcient chain code based image

manipulation for handwritten word recognition,” in the SPIE sym. on electronic imaging science and technology (Document Recognition III), vol. 2660, pp. 262–272, Feb 1996.

CEDAR, “http://www.cedar.buffalo.edu/databases/cdrom1/.”

J. J. Hull, “A database for handwriten text recognition re- search,” Transactions on Pattern Analysis and Machine Intel- ligence, pp. 550–554, May 1994.

R. O. Duda and P. E. Hart, Pattern classification and scene

analysis. New York, Wiley, 1st ed., 1973.


Refbacks

  • There are currently no refbacks.