-
公开(公告)号:JPH09237338A
公开(公告)日:1997-09-09
申请号:JP4230296
申请日:1996-02-29
Applicant: IBM
Inventor: NEMOTO NAOYUKI , TAKAHASHI HIROYASU
Abstract: PROBLEM TO BE SOLVED: To speed up OCR processing by executing smoothing operation for the n pieces of line conversion by using only shift, NOT, OR, and AND operation. SOLUTION: A bit pattern is inputted to a working array (line conversion) and two upper and lower lines of the array are blanked. A loop variable (i) is increased one by one from 2 up to (n+1) and the preceding line (i-1), current line (i) and succeeding line (i+1) of the i-th line of the working array are respectively substituted for a working variable. Then the working variables are shifted to right and left respectively by one bit and the shifted variables are substituted for respectively different working variables. Working variables not to be shifted on the current lines in four patterns A to D are held as they are, the bits of working variables corresponding to the positions of black points are inverted and the total OR of each bit is found out by eight working variables. Then AND between the OR result and a result working variable is found out in each bit, the AND is substituted for the result working variable, and finally the result working variable is substituted for the i-th line pattern of the working array.
-
2.
公开(公告)号:JPH1173497A
公开(公告)日:1999-03-16
申请号:JP23304897
申请日:1997-08-28
Applicant: IBM
Inventor: NEMOTO NAOYUKI , TAKAHASHI HIROYASU
Abstract: PROBLEM TO BE SOLVED: To increase the processing speed when the outline of a figure is smoothed by performing the smoothing process by using only logical operation without using 'if then else' or 'switch' syntax of a conditional judgement and a branch instruction as an element causing a decrease in the processing speed. SOLUTION: Process units (t bits) are put one over another, a pattern (process object s bits) for writing a process result is loaded in the center part of an (r)-bit regsiter, and contiguous patterns of a source image are loaded to right and left adjacent parts and processed. Here, a process unit (t) consists of 64 bits the process object (s) which is processed and written consists of 48 bits. When a figure and characters drawn in a bit pattern is smoothed, not the if then else' or 'switch' syntax of a conditional judgement and a branch instruction as an element causing a decrease in process speed, but only logical operations for shifting a bit array, NOT, AND, OR, etc., are used to perform a process for smoothing.
-
公开(公告)号:JPH09237319A
公开(公告)日:1997-09-09
申请号:JP3974096
申请日:1996-02-27
Applicant: IBM
Inventor: NEMOTO NAOYUKI , TAKAHASHI HIROYASU
IPC: G06K9/42
Abstract: PROBLEM TO BE SOLVED: To prevent a normalization result from being changed due to different character size or a different position by executing normalization by the adjust ment of parameters without especially changing a processing flow. SOLUTION: The image of a character and parameters for the character are applied to a scheduling program for executing practical enlargement/ reduction and seamless normalization is executed by the adjustment of the parameter without especially changing a processing flow. The parameters are the height of a virtual character frame, the height of the character, the width of the character, a distance from the virtual character frame up to the left end of the character, and a distance from the virtual character frame up to the upper end of the character. When a normalized output pattern is 24×16 (length X width), the aspect ratio of the character is the value of (character height)/(character width) and o- height, o- width, t- pad, and l- pad respectively indicate the practicaly height of a normalized character, the practical width of the normalized character, a distance from the upper frame of the virtual character frame up to the upper end of the normalized character, and a distance from the left frame of the virtual character frame up to the left end of the normalized character.
-
-