Performance Analysis of Time in Line and Word Segmentation using Different Fonts

Abstract: Segmentation is a widely used process in current-day image processing. Various segmentation processes exist to segment lines and words. Header lines are detected and converted as straight lines. Text line segmentation is avn important step because inaccurately segmented text lines will cause errors in the recognition stage. Text characteristics can vary in font, size, and orientation. This paper introduces a comparative analysis of time of the segmentation process that is carried out on various style text to find the time taken in the segmentation process. The text is segmented into lines, lines into words. Here we make a comparative study of time consumed during line and word segmentation using different style fonts and find the time in which fonts type, both segmentation take minimum amount of CPU time. Here we proposed a method that supports all segmentation process(line and word segmentation) to retrieve text, make boundary boxes to the fully lines, words and perform segmentation. It consists of recording the start time of the process as well as end time of the process and find the time difference. This process is applied to different fonts of text to make a comparative analysis.

Keywords: Image processing, Printed/Handwritten document analysis, Text font time analysis, segmentation, line segmentation and word segmentation.