Image classification and identification method based on regional bicubic interpolation technology

A bicubic interpolation, classification and recognition technology, applied in character and pattern recognition, instruments, computer parts and other directions, can solve the problems of image distortion and image classification recognition rate reduction, achieve distortion suppression, good classification and recognition effect, improve The effect of classification recognition accuracy

Inactive Publication Date: 2019-10-15
HARBIN UNIV OF SCI & TECH
View PDF5 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problem that the existing image processing method has image distortion caused by scale transformation, which reduces the recognition rate of image classification, and proposes an image classification recognition method based on regional bicubic interpolation technology

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
  • Image classification and identification method based on regional bicubic interpolation technology
  • Image classification and identification method based on regional bicubic interpolation technology
  • Image classification and identification method based on regional bicubic interpolation technology

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0014] A kind of image classification and recognition method based on area bicubic interpolation technology of this embodiment, such as figure 1 As shown, the method is realized through the following steps:

[0015] Step 1. After obtaining the training data through the steps of SIFT feature extraction, K-means clustering statistics word frequency and classification using the training samples, train the classifier model;

[0016] Step 2: Perform the same zoom-in and zoom-out changes on the test sample image through regional bicubic interpolation, and then perform retrieval and classification through the BOF algorithm;

[0017] Step 3: Send the test samples processed in Step 2 into the classifier model trained in Step 1 for classification and recognition.

[0018] From the classification results, it can be seen that the method of combining BOF algorithm and regional bicubic interpolation can improve the problem of reducing the recognition rate of image classification due to siz...

specific Embodiment approach 2

[0020] The difference from Embodiment 1 is that in this embodiment, an image classification and recognition method based on area bicubic interpolation technology,

[0021] After using the training samples to extract features through SIFT, K-means clustering to count word frequency, and classify the training data, the process of training the classifier model is as follows:

[0022] Step 11, use Gaussian blur to obtain scale space, (Gaussian convolution kernel is the only linear change kernel to realize scale change, which is a proven conclusion. Scale space is used to simulate the multi-scale characteristics of image data), use Gaussian The difference (Difference of Gaussian, DOG) pyramid represents that the scale space L(x, y, σ) of an image is defined as the difference between a Gaussian function G(x, y, σ) of varying scales and its original image I(x, y). The convolution between, as shown in formula (1):

[0023] L(x,y,σ)=G(x,y,σ)*I(x,y) (1)

[0024] in

[0025]

[002...

specific Embodiment approach 3

[0053] The difference from the specific embodiment 1 or 2 is that in this embodiment, an image classification and recognition method based on the area bicubic interpolation technology,

[0054] In the described step 2, first calculate the variance Var of the pixel values ​​of the four neighborhood points of the corresponding point of the point to be interpolated in the source image, if the variance is less than the set threshold T (T is 20), then directly take these 4 pixels The mean value E of the value is used as the pixel value of the point to be interpolated; otherwise, it is still calculated by bicubic interpolation; the calculation formula of the variance Var is shown in formula (14):

[0055] Var=(E-f 11 ) 2 +(E-f 12 ) 2 +(E-f 21 ) 2 +(E-f 22 ) 2 (14)

[0056] Among them, f 11 , f 12 , f 21 , f 22 They are the pixel values ​​of the 4 points in the neighborhood of the corresponding point of the source image at the current point to be interpolated; E is the a...

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

The invention discloses an image classification and identification method based on the regional bicubic interpolation technology, and belongs to the field of image processing. An existing image processing method has the problem that the image classification identification rate is reduced due to image distortion caused by scale transformation. According to the image classification and identification method based on the regional bicubic interpolation technology, the method comprises the steps of extracting the features and K-of a training sample through SIFT; performing means clustering statistics on the word frequency, and training a classifier model after training data is obtained in the classification step; subjecting the images of a test sample to identical zooming-in and zooming-out changes through regional bicubic interpolation, and then carrying out retrieval and classification through the BOF algorithm. According to the invention, the classification and identification accuracy ofdistorted images caused by zooming can be improved.

Description

technical field [0001] The invention relates to an image classification and recognition method based on area bicubic interpolation technology. Background technique [0002] Computer vision processes the images or videos acquired by the computer, so that it has the ability to recognize, understand and adapt to the environment autonomously. As an important part of image processing and machine vision, feature extraction and classification technology can efficiently process visual information, obtain the information people need, and bring convenience to life and industrial production. Commonly used image processing methods are: deep convolutional neural network (Deep Convolutional Neural Network, DCNN), word bag model (Bag of Features, BOF), K-means clustering (K-Means) and support vector machine (Support VectorMachine). , SVM) coupling algorithm (KM-SVM) and other related improved algorithms. At present, the following methods are proposed: the weather state of the image is id...

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 Applications(China)
IPC IPC(8): G06K9/62G06K9/46
CPCG06V10/50G06V10/462G06F18/23213G06F18/2411
Inventor 刘明珠鲍雪蒋燚铭
Owner HARBIN UNIV OF SCI & TECH
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap