A Classification Identification Method Based on Krawtchouk Moment and KNN-SMO Classifier

A recognition method and classifier technology, which is applied in character and pattern recognition, instruments, calculations, etc., can solve the problem that classified marks cannot be effectively recognized, and achieve the effect of overcoming the shortcomings of recognition

Active Publication Date: 2018-11-23
深圳元物质科技集团有限公司
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Aiming at the fact that the usual text recognition algorithm cannot effectively identify the classified mark after being attacked, a secret mark recognition method based on Krawtchouk moment and KNN-SMO classifier is proposed, and the characteristics of the image can be well described by using the low-order Krawtchouk moment , and the quantity has good stability under common attacks, and the use of KNN-SMO combined classifier makes the classifier not only have the ability of KNN fast classification but also has the advantage of SMO in overcoming the small sample problem, thus improving the confidentiality identification Recognition accuracy and speed

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Classification Identification Method Based on Krawtchouk Moment and KNN-SMO Classifier
  • A Classification Identification Method Based on Krawtchouk Moment and KNN-SMO Classifier
  • A Classification Identification Method Based on Krawtchouk Moment and KNN-SMO Classifier

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] The scheme of the present invention is specifically described below in conjunction with the accompanying drawings:

[0049] [1] Different attacks (including affine transformation, JPEG compression, brightness reduction, fuzzy processing, median filtering, mean filtering, contrast enhancement, etc.) are carried out on the classified mark to obtain experimental data.

[0050] [2] Divide the experimental data into two parts: training samples and test samples, which do not contain each other.

[0051] [3] Preprocessing training samples, including image grayscale, image inversion and binarization, image denoising, tilt correction, line word segmentation, thinning and normalization and other steps.

[0052] [4] Calculate the low-order Krawtchouk moments of the training samples after preprocessing as training features.

[0053] [5] To construct a KNN-SMO classifier, first use the KNN algorithm to prune the training set, determine the choice according to the similarities and d...

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

No PUM Login to view more

Abstract

A secret-level identification method based on Krawtchouk moment and KNN-SMO classifier, the identification method adopted is: applying the theory based on Krawtchouk moment and KNN-SMO to the identification of secret-level identification in electronic forensics, the method first passes After image preprocessing of the dense-level identification, the feature vector is formed by calculating the low-order Krawtchouk moments of the image, and then the KNN-SMO classifier is used to classify and identify the dense-level identification image. On the one hand, the low-order Krawtchouk moment can be used to describe the features of the image well, and the quantity has good stability under common attacks. On the other hand, the KNN-SMO combined classifier makes the classifier both have KNN fast classification It also has the advantages of SMO in overcoming the problem of small samples, thereby improving the accuracy and speed of the identification of classified labels.

Description

technical field [0001] The invention belongs to the field of electronic forensics, and in particular relates to a secret-level sign recognition method based on Krawtchouk moments and a KNN-SMO classifier. Background technique [0002] In electronic forensics, after completing the retrieval of secret-related documents, it is necessary to further identify the secret-level identification of the secret-related documents, and record the results of the inspection to form a log as evidence for forensics. [0003] Existing recognition methods, common methods include template matching method (Aravinda C V, Prakash HN. Template matching method for Kannada Handwritten recognition based on correlation analysis [C] / / Contemporary Computing and Informatics (IC3I), 2014International Conference on. IEEE, 2014: 857-861.), Character Feature Statistics (Das S, jyotiChoudhury S, Das A K, et al. Selection of Graph-Based Features for Character Recognition Using Similarity Based Feature Dependency ...

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
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/62
CPCG06V10/758G06F18/24147
Inventor 傅德胜经正俊
Owner 深圳元物质科技集团有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products