Character scaling by contour method

Resolution enhancement of textual images: a survey of single image‐based methods

“…Namane and Sid‐Ahmed [54] proposed an up‐scaling contour method specifically designed for characters captured by a camera. Their method is based on detecting the borders of the character, scaling the contour and then magnifying it by using the cubic spline interpolation.…”

Color image coding using regional correlation of primary colors

“…In [5], the information is approximated as lines and arcs, whereas in [18,19] a basic algorithm for chain coding are presented. This work uses Freeman's chain-code [20] with 8 connected contours.…”

CCIGS: A data compression system for Chinese fonts and binary images using classification techniques

“…From Tables I and 11, we observed that a character generally contains about 7.3 contours, and that the average size of Chinese character descriptors for each character is 16.3 bytes. Then we use equation (1) to calculate the original storage requirement of Boldface style and it is 2,956,844 + 211,900 = 3,168,744 (bytes). Similarly, the original storage requirement of three kinds of style of Chinese patterns is 9,065, 850 + 502,400 = 9,568,250 (bytes).…”

“…Using manual labour to do the classification and compression work has some disadvantages: (1) it needs large amounts of manpower and time, (2) it is very difficult to predict the quality of the patterns generated from these basic radicals, (3) it is very difficult to detect the small difference among these contours. To use a computer to build the basic radical database instead of manual labour, we must overcome the following problems with Chinese characters: line algorithm3 to form the borders of a pattern.…”