阅读说明：本技术 图形识别修正的自适应校正方法 (Self-adaptive correction method for pattern recognition correction ) 是由 刘海旭 于 2018-06-13 设计创作，主要内容包括：本发明公开了图形识别修正的自适应校正方法,主要包括以下步骤：①基于知识的扩边过程是将视口从小到大逐渐扩大,直到能识别连通成分的可连性为止,在扩边的过程中可以得到视口边框大小、边线出口个数及位置、边框内连通成分的端点数及它的位置信息；②修正过程由扩边过程所得到的有关信息计算出连通数和出口数,再根据连通数和出口数分类合并。实施时,视口为正方形；本发明图形识别修正的自适应校正方法,适用范围广,运用方便。(The invention discloses a self-adaptive correction method for pattern recognition and correction, which mainly comprises the following steps: the method comprises the following steps that firstly, in the edge expanding process based on knowledge, a viewport is gradually expanded from small to large until the connectivity of connected components can be identified, and the size of a viewport frame, the number and the position of side line outlets, the number of endpoints of the connected components in the frame and the position information of the endpoints can be obtained in the edge expanding process; secondly, in the correction process, the connection number and the exit number are calculated according to the related information obtained in the edge expanding process, and then the connection number and the exit number are classified and combined. In practice, the view port is square; the self-adaptive correction method for pattern recognition and correction has the advantages of wide application range and convenient application.)

1. The self-adaptive correction method for pattern recognition correction is characterized by mainly comprising the following steps of:

the method comprises the following steps that firstly, in the edge expanding process based on knowledge, a viewport is gradually expanded from small to large until the connectivity of connected components can be identified, and the size of a viewport frame, the number and the position of side line outlets, the number of endpoints of the connected components in the frame and the position information of the endpoints can be obtained in the edge expanding process;

secondly, in the correction process, the connection number and the exit number are calculated according to the related information obtained in the edge expanding process, and then the connection number and the exit number are classified and combined.

2. The adaptive correction method for pattern recognition modification of claim 1, wherein the view port is square.

Technical Field

The invention relates to a character recognition algorithm, in particular to a self-adaptive correction method for pattern recognition correction.

Background

Because the Chinese characters have various font styles and fonts and the characteristics of the same Chinese character are different, the Chinese characters with different font styles and fonts can be identified in order to conveniently describe and extract the characteristics of the same Chinese character in a unified way, a foundation is laid for the Chinese character identification work, and the size normalization operation of the Chinese character image is needed before the Chinese character characteristics are extracted. The size normalization is to perform scaling operation on the actually extracted characters to finally obtain a character image with a preset size; the purpose of normalizing the stroke width is to make the binary image a skeleton only one pixel wide, a process also referred to as thinning. The essence of refinement is to find the central axis or skeleton of the graph. And replacing the pattern with its skeleton. The pixel width of the thinned graph becomes 1, but structural information of the original graph, such as position, direction, length and the like, can be still maintained. In a modern pattern recognition system, refinement processing becomes one of the most critical preprocessing steps, and the quality of the refinement effect directly influences the recognition speed and the recognition accuracy. It can be said that whether effective refinement can be performed becomes the key for identifying the success or failure of the system. Due to the irregularity of the image boundary and the sensitivity of the framework to noise, a large amount of distortion exists in the extracted framework, so that the results of data fitting and vectorization tracking cannot correctly represent the original information, and the future identification quality of the image is seriously influenced. In this regard, a general correction strategy is to examine the relationship between the distance and a threshold between two connected components of a given skeleton to determine the connectivity of the two connected components

Disclosure of Invention

In order to solve the technical problems in the background art, the invention provides a self-adaptive correction method for pattern recognition correction.

The invention is realized by the following technical scheme:

setting a fixed view port, mainly comprising the following steps:

the method comprises the following steps that firstly, in the edge expanding process based on knowledge, a viewport is gradually expanded from small to large until the connectivity of connected components can be identified, and the size of a viewport frame, the number and the position of side line outlets, the number of endpoints of the connected components in the frame and the position information of the endpoints can be obtained in the edge expanding process;

secondly, in the correction process, the connection number and the exit number are calculated according to the related information obtained in the edge expanding process, and then the connection number and the exit number are classified and combined.

Further, the view port is square.

Compared with the prior art, the invention has the following advantages and beneficial effects:

1. the self-adaptive correction method for pattern recognition and correction has the advantages of wide application range and convenient application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to examples, and the exemplary embodiments and descriptions thereof are only used for explaining the present invention and are not used as limitations of the present invention.