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

Spectral Image Lossless Compression Method Based on Support Vector Regression

A support vector regression and spectral image technology, applied in the field of spectral remote sensing, can solve the problems of spectral image transmission and storage difficulties, large amount of spectral data, etc., and achieve the effects of good technical support, removal of inter-band redundancy, and high compression ratio

Active Publication Date: 2018-07-31
XIDIAN UNIV +1
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a lossless compression method for spectral images based on support vector regression in view of the large amount of spectral data generated by imaging spectrometers, difficulties in transmission and storage of spectral images, and methods for effectively removing redundant spectral data.

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
  • Spectral Image Lossless Compression Method Based on Support Vector Regression
  • Spectral Image Lossless Compression Method Based on Support Vector Regression
  • Spectral Image Lossless Compression Method Based on Support Vector Regression

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0041] In order to overcome the large amount of spectral data generated by current imaging spectrometers, the difficulties in transmission and storage of spectral images, and the shortcomings of methods for effectively removing spectral data redundancy, the present invention provides a method such as figure 1 , figure 2 , image 3 and Figure 4 The shown spectral image lossless compression method based on support vector regression includes the following steps:

[0042] (1) Import the spectral image, that is, the original image;

[0043] (2) Select a clustering algorithm, classify and preprocess the spectral image, and obtain the corresponding clustering index;

[0044] (3) Select the prediction algorithm, carry out the design of the prediction model, according to the clustering index that step (2) obtains and the prediction model that this step produces, predict each pixel of the whole spectrum image, obtain the prediction image;

[0045] (4) Perform a difference between ...

Embodiment 2

[0080] On the basis of Embodiment 1, the effects of the present invention are further illustrated through the following simulation experiments.

[0081] The simulation condition of the present invention: computer configuration environment is Intel (R) Core (TM) 34.00Ghz, internal memory 2G, system windows7, computer simulation software adopts the MATLABR2010a of integrated Libsvm. The experimental database uses the corrected Yellowstone hyperspectral dataset (Scene0, Scene3, Scene10, Scene11, Scene18) obtained by the American AVIRIS scanner in 2006.

[0082] Simulation content of the present invention: select one group of data in the experimental database, such as Scene0; In the simulation of this embodiment, the number of clusters selected is 16, the number of training samples of support vector regression is 300, and the prediction order is 10. In the process of training the predictive model by support vector regression, the parameters are selected as g=1, c=0.00001-1 (the st...

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 belongs to the technical field of spectral remote sensing, and provides a method for lossless compression of spectral images based on support vector regression, comprising the following steps: (1) importing the original image; (2) selecting a clustering algorithm, performing classification preprocessing on the spectral image, and obtaining the corresponding (3) select the prediction algorithm, design the prediction model, and predict each pixel of the entire spectral image according to the obtained cluster index and prediction model, and obtain the prediction image; (4) the original image (5) Arithmetic coding is used for the residual image, and at the same time, the prediction coefficient obtained by the prediction model and the cluster index obtained by the clustering algorithm are encoded to obtain the code stream file. The invention implements spectral lossless compression based on support vector regression by clustering the read-in images, establishing a prediction model for each category, and encoding residual images and side information, achieving a better lossless compression effect. The prediction accuracy is high and the residual error is small.

Description

technical field [0001] The invention belongs to the technical field of spectral remote sensing, and relates to a lossless compression and coding technology of spectral images, in particular to a lossless compression method of spectral images based on support vector regression. Background technique [0002] Spectral remote sensing technology is a kind of earth observation technology that emerged in the 1980s. It is widely used in ocean remote sensing, geological exploration, atmospheric and environmental remote sensing, and military reconnaissance. With the improvement of spatial and interspectral resolution of imaging spectrometers, spectral data has increased dramatically. However, the expansion of data volume and data dimension will bring great difficulties to the transmission and storage of spectral images. Compression of spectral images is necessary. [0003] Since the imaging spectrum is imaged in a narrow spectral range, this makes the spectral image have a strong int...

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): H04N19/50H04N19/59
CPCG06F18/23G06F18/2411
Inventor 吴家骥白静任改玲张敏焦李成张向荣王爽熊涛刘红英
Owner XIDIAN UNIV