Lecture on “Recognition System for Text Printed on Clear Reflected Material”

On 2 October 2012, the BZU Faculty of Information Technology held a lecture titled: “Recognition System for Text Printed on Clear Reflected Material”, delivered by Dr. Khader Mohammad*.
Abstract of the lecture:
Text embedded in an image contains useful information for applications in medical, industrial, commercial, and research products. While many systems have been designed to correctly identify text in an image, no work addressing the recognition of degraded text on clear plastic has been found. This paper posits novel methods and an apparatus for extracting text from an image with the following qualities: a) poor background contrast, b) white, curved, and/or differing fonts or character width between sets of images, d) dotted text printed on curved reflective material, and/or e) touching characters. Illustrative examples were used for testing the system and verifying the proposed solution. Methods were evaluated using images containing a variety of text captured from water bottles.  These tests averaged a processing time of (~10 seconds using MATLAB8 on HP 8510W with 4G of RAM and 2.3GHz of processor speed), and experimental results yielded an average recognition rate of (90% -to- 93%) using customized systems generated by the proposed development.
* Khader Mohammad worked as a senior design engineer at Intel Corporation, Austin, where he was engaged in designing the next generation of Intel microprocessors (ATOM processer). He has more than 13 years of experience in processers with an emphasis on circuit and physical design. He received his B.Sc. degree in Electrical Engineering from Birzeit University, Palestine in 1996 and his M.Sc. degree from Portland State University, USA.  Currently, Khader works as an assistant professor at the BZU Department of Information Technology. His research interests includes DSP design, processors, computer networks and chip SOC design, and has published more than fourteen conference papers and journal articles.