Paper Details: Downloads: 1083
Serial Number: P1151547443
Title: An Enhanced Character Segment Method in Image-based Email Classification
Authors: Mallikka Rajalingam and Valliappan Raman and Putra Sumari
Abstract: As the number of internet users increases, email has been an efficient and popular communication mechanism. The integral of classification of text and image based email by machine learning approaches were used. Image email is one of the most recent tricks introduced by spammers. The unwanted text embedding into images which are sent as email attachments. This paper proposed an enhanced way to segment characters in image based email classification domain. A hybrid approach of Discrete Wavelet Transform (DWT) and Hough Transform based character segmentation is used. This algorithm can enhance the segmentation accuracy. The experimental results show that proposed method produces less number of false negatives when compared with existing machine learning techniques.
Keywords: character segmentation; discrete wavelet transform; email classification; Hough transform;
Journal/Conference: International Journal of Graphics, Vision and Image Processing
Volume: 15
Issue: 2
Submission Date: 11/19/2015 12:00:00 AM
Review Date: 12/20/2015 1:39:02 PM
Publishing Date: 12/28/2015 12:56:00 AM
Article Downloads: 1083
Download:

Facebook