Image characteristics extraction method used for handwritten Chinese character recognition
A technology for image feature extraction and Chinese character recognition, which is applied in character and pattern recognition, instruments, computer parts, etc., can solve the problem that positioning cannot extract effectively adapt to different writing styles and scale invariant features, and achieve the effect of improving recognition performance.
Inactive Publication Date: 2010-12-29
SOUTH CHINA UNIV OF TECH
View PDF2 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] The purpose of the present invention is to overcome the problem that the direct application of SIFT feature point positioning cannot extract effective scale-invariant features suitable for different writing styles, combined with the characteristics of handwritten Chinese character images, using elastic grid technology and SIFT features, and designing a Handwritten Chinese Character Feature Extraction Method Based on SIFT Feature in Local Elastic Region
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0026] The flow chart of the elastic local area SIFT Chinese character feature extraction method of the present invention is as attached figure 1 As shown, the specific method is to first perform two different elastic segmentations on the input Chinese character image to obtain a total of 18 different image regions (blocks), and then linearly normalize these blocks to a uniform size, and then extract them SIFT describes the sub-features, and finally stitches all the descriptor features into the final feature vector in sequence.
[0027] The extraction flow chart of the SIFT descriptor feature applied in the elastic local area SIFT Chinese character feature extraction process of the present invention is as attached figure 2 , and its structure diagram is attached image 3 , specifically, the first step is to use the SOBEL operator in the feature extraction area to calculate the magnitude and direction of the gradient of each pixel in the area, such as image 3 As shown on t...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention provides an image characteristic extraction method for handwritten Chinese character recognition. A handwritten Chinese character image is cut into a series of stable characteristic extraction regions, and then scale invariant characteristic transformation algorithm is used for carrying out characteristic extraction on the regions. The invention overcomes a problem that the direct application of SIFT characteristic points for positioning cannot extract effective scale invariant characteristics adapted to different writing styles; with the combination with the characteristics of the handwritten Chinese character image and the application of the elastic mesh technology and SIFT characteristics, the invention designs the handwritten Chinese character characteristic extraction method based on the SIFT characteristics of a secondary local elastic region.
Description
technical field [0001] The invention belongs to the technical field of pattern recognition and artificial intelligence, in particular to a handwritten Chinese character image recognition processing method. technical background [0002] A handwritten Chinese character recognition system is divided into four modules: preprocessing, feature extraction, classification recognition, and postprocessing. Feature extraction is considered to be one of the key steps in Chinese character recognition and has an important impact on the final performance of the entire system. In recent years, many scholars have done a lot of research work on how to obtain effective features, and achieved many excellent results. The Gabor feature is one of the more effective features among various Chinese character features, and its application has a good biological vision theory support behind it. In fact, pattern recognition has always been closely related to computer vision and biological vision theory....
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00G06K9/46
CPCG06K2209/011G06K9/4671G06V10/462G06V30/287
Inventor 金连文张志毅丁凯
Owner SOUTH CHINA UNIV OF TECH