Unlock instant, AI-driven research and patent intelligence for your innovation.

Multi-packet image classification method based on two-dimensional empirical mode decomposition

A technology of empirical mode decomposition and classification method, which is applied in the direction of character and pattern recognition, instruments, computer components, etc., can solve the problems of insufficient utilization of essential features of images and low classification accuracy, and is conducive to popularization and application. The effect of improving the classification accuracy

Active Publication Date: 2010-10-06
HARBIN INST OF TECH
View PDF5 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the problem of insufficient utilization of image essential features and low classification accuracy by traditional classification methods, and provides a multi-group image classification method based on two-dimensional empirical mode decomposition

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
  • Multi-packet image classification method based on two-dimensional empirical mode decomposition
  • Multi-packet image classification method based on two-dimensional empirical mode decomposition
  • Multi-packet image classification method based on two-dimensional empirical mode decomposition

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0015] Specific implementation mode one: the following combination figure 1 and figure 2 To describe this embodiment,

[0016] In 1998, Dr. Huang E of NASA (National Aeronautics and Space Administration, NASA) proposed Hilbert-Huang Transform (HHT) according to the mathematical theory design of modern mathematician Hilbert. As a powerful tool for analyzing nonlinear and non-stationary signals, HHT is divided into two steps. First use Empirical Mode Decomposition (EMD) to obtain a limited number of Intrinsic Mode Functions (IMF), and then use Hilbert Transform (HilbertTransform, HT) and instantaneous frequency method to obtain the instantaneous frequency and amplitude of IMF, Finally, the time-spectrum of the signal is obtained.

[0017] The core of HHT is EMD. EMD is completely driven by data, and its essence is to screen the signal according to the characteristic time scale of the signal. This process is represented by a scale band-pass filter to filter the signal, so th...

specific Embodiment approach 2

[0061] Specific implementation mode two: the following combination Figure 3 to Figure 14 This embodiment is described, and a specific example is given, using multi-group images of 92AV3C hyperspectral images.

[0062] Hyperspectral images are typical multi-group images. The selected 92AV3C hyperspectral image comes from the remote sensing image of an agricultural area in northwestern Indiana, USA collected by the AVIRIS (Airborne Visible / Infrared Imaging Spectrometer) sensor. The data set contains 224 continuous bands, from 0.40 μm to 2.45 μm approximately every 10 nm. Remove 4 zero-value bands and 20 bands affected by the absorption of water vapor in the earth's atmosphere, and the number of bands used in the experiment is 200. The ground objects with the largest number of 7 types of pixels (ie corn-notill, corn-mintill, grass / trees, soybeans-notill, soybean-mintill, soybean-cleantill and woods) were selected as experimental samples. The total number of pixels of these 7 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

No PUM Login to View More

Abstract

The invention relates to a multi-packet image classification method based on two-dimensional empirical mode decomposition, which belongs to the field of image processing. In order to solve the problems that traditional classification methods do not make full use of the essential characteristics of images and have low classification precision, the method comprises the following steps that: 1. two-dimensional empirical mode decomposition is respectively carried out to all wave bands in a multi-packet image to obtain the first K two-dimensional intrinsic mode function component IMFs with the frequency gradually decreasing from high to low and one residual error; 2. the first K two-dimensional component IMFs are summated as characteristic values; 3: characteristic values of a plurality of multi-packet images are randomly selected by the ratio as training samples and testing samples for supporting a vector machine so as to carry out parameter training to the training samples on supporting the vector machine, judge attribution and form a plurality of sub-classifiers for supporting the vector machine; and 4: a plurality of sub-classifiers for supporting the vector machine are constructed into a multiple classifier based on a one-to-one strategy, decisions are made on the attribution and the category of the testing samples based on a decision making function so as to complete the classification of the multi-packet images.

Description

technical field [0001] The invention relates to a multi-group image classification method based on two-dimensional empirical mode decomposition, which belongs to the field of image processing. Background technique [0002] Multi-group images are a group of multi-band images with high correlation. There are a large number of physical prototypes in the fields of earth observation, medical diagnosis, and radar detection, such as hyperspectral images, medical ultrasound images, and sea level fluctuation images. They are generally continuous observations or multi-spectral spectroscopic observations of the same area, and often contain images of hundreds or thousands of bands, and the images of each band generally have a high correlation. Distinguish information, and there is a large amount of redundant information. [0003] Multi-group image classification can obtain more internal rules of the image, which is convenient for subsequent image processing, so multi-group image classi...

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): G06K9/66
Inventor 沈毅贺智张淼
Owner HARBIN INST OF TECH