Topic
Character interval
About: Character interval is a research topic. Over the lifetime, 88 publications have been published within this topic receiving 246 citations.
Papers published on a yearly basis
Papers
Patent•
26 Aug 2009
TL;DR: Zhang et al. as mentioned in this paper proposed a character extraction method in a digital video based on character segmentation and color cluster, which comprises the following steps: (1) Character segmentation: utilizing the characteristic differences of a character area and a character interval area to carry out vertical projection to segment images in the character area, namely, segmenting each row of area image containing a plurality of characters into sub-area images only containing a single character so as to reduce the post operating and treating difficulties and improve the identifying accuracy rate of OCR.
Abstract: The invention relates to a character extracting method in a digital video based on character segmentation and color cluster, which comprises the following steps: (1) character segmentation: utilizing the characteristic differences of a character area and a character interval area to carry out vertical projection to segment images in the character area, namely, segmenting each row of area image containing a plurality of characters into a plurality of subarea images only containing a single character so as to reduce the post operating and treating difficulties and improve the identifying accuracy rate of OCR; and (2) character extraction: firstly, using the character color characteristic in the image to cluster colors, finding out an image layer containing maximum character information as a target image layer, and deleting the background area; and then, using the communicating characteristics of the characters to analyze a communicating area of the target image layer, and removing non-character areas to obtain such three results as single character images, an integral image of the character area and an integral image spliced by the single character images respectively, wherein all the three results are input to an OCR system to be identified, and the latter two results use the semantic processing function of the OCR and can accurately determine the characters with similar forms according to the context to improve the identifying effect.
30 citations
Patent•
1 Mar 1982TL;DR: In this paper, an image data conversion method for converting the image data signal to the runlength signal through the process for reading once or several times a runlength code from a run length conversion table for a given image signal and a process for adding the pertinent run length code when the same type of code appears continuously.
Abstract: An image data conversion method for converting the image data signal to the runlength signal through the process for reading once or several times a runlength code from a runlength conversion table for a given image data signal and a process for adding the pertinent runlength code when the same type of runlength code appears continuously. A character code/character pattern conversion apparatus which uses said image data conversion method, and provides, moreover, a multiplication device for multiplying character enlarging coefficient and the runlength code and a character interval inserting means which adds a character interval width code to the enlarged runlength code and outputs the input character code after converting it to the enlarged character pattern along with the interval code.
17 citations
Patent•
12 Mar 1990
TL;DR: In this article, the authors propose to reduce the quantity of read designation information given before the character readout to simplify the character reading, by obtaining a position of the next character line in the calculation based on a preliminarily given character interval and storing the pattern of the last line in a pattern memory on a basis of this calculation result.
Abstract: PURPOSE:To reduce the quantity of read designation information given before the character readout to simplify the character reading, by obtaining a position of the next character line in the calculation based on a preliminarily given character interval and storing the pattern of the next character line in a pattern memory on a basis of this calculation result. CONSTITUTION:A business form or the like is scanned by a photoelectric converting part 1, and read information is converted photoelectrically, and a character pattern of one-line components of characters of the video signal from the converting part 1 is stored in a pattern memory part 2. The state of black at the tip of this character pattern is detected by a black detecting part 3, and the detection signal is applied to a control part 7. One character unit of the character pattern stored in the memory part 3 is cut by a detecting and cutting part 4, and the cut character pattern is recognized by a recognizing part 6. The position of a character line is calculated on a basis of the recognition result of the recognizing part 6 by the control part 7, and the position of the next character line is calculated on a basis of this character line position and a preliminarily given character line interval. A pattern of the next character line is stored in the memory part 2 on a basis of the calculation result, thus reducing the quantity of read designation information.
14 citations
Patent•
13 May 2015
TL;DR: In this article, a character cutting and recognizing method was proposed for solving the problems that the recognition capacity of an existing slitting method for characteristics under complex backgrounds is not high, and the smudginess and interference prevention capacity of the existing slining method is poor.
Abstract: The embodiment of the invention discloses a character cutting and recognizing method. The method is used for solving the problems that the recognition capacity of an existing slitting method for characteristics under complex backgrounds is not high, and the smudginess and interference prevention capacity of the existing slitting method is poor. The method includes the steps of collecting image data to obtain an image to be recognized; positioning a character line candidate area on the image to be recognized; obtaining preset character line prior information, wherein the character line prior information includes the character number, the character interval and the character size; obtaining a corresponding slitting point template corresponding to the character line prior information; obtaining reliability of different positions when the slitting point template traverses the character line candidate area; determining the position with the highest credibility as the optimal slitting position; slitting the character line candidate area according to the slitting point template and the optimal slitting position to obtain a plurality of single character areas; conducting character recognition on the single character areas to obtain corresponding recognition results.
14 citations
Patent•
1 Jun 1992TL;DR: In this paper, a method for using the printer unit of a telefax device as a page printer for a PC is described, where the data representing the information are supplied as hexadecimal values from the PC to the buffer memories PSp1 and PSp2 and to the page memory SSp.
Abstract: The invention relates to a method for using the printer unit of a telefax device as a page printer for a PC. It is the object to make this printer unit usable for different PCs and to achieve high printing quality and printing speed. This is achieved by providing a control unit which detects a print requirement and switches the telefax device from facsimile operation to printing operation and vice versa, the information data being transmitted as hexadecimal characters. The data representing the information are supplied as hexadecimal values from the PC to the buffer memories PSp1 and PSp2 and to the page memory SSp. Control signals for actual character shapes, which correspond to the hexadecimal values and take over control of the printing printer section (thermal comb), are stored in the register R of the printer unit. As a result, the character is printed out not in pixel form but in high quality as an actual character in original form. The user can set a character interval after which the control unit STE automatically switches from printing mode to facsimile mode. When copies are made, this switching is automatically extended to begin with the end of the last copy.
10 citations
Related Topics (5)
Performance Metrics
| Year | Papers |
|---|---|
| 2020 | 1 |
| 2018 | 1 |
| 2017 | 1 |
| 2016 | 2 |
| 2015 | 1 |
| 2014 | 3 |