### HyperSpectral Toolbox ###
For better viewing, visit <http://davidkun.github.io/HyperSpectralToolbox/>
Originally created by [Isaac Gerg](http://www.gergltd.com/home/) and maintained by him [here](https://github.com/isaacgerg/matlabHyperspectralToolbox).
# Note #
**This repository is no longer being maintained or updated.**
Isaac (the original creator of 99% of this work) has [his own GitHub repository](https://github.com/isaacgerg/matlabHyperspectralToolbox) with this work and intends to actively maintain / update it there.
_Please use that repository and see his BibTeX citation for correctly referencing the software in your research._
**Dependencies**
FastICA -- [from Aalto University](http://research.ics.aalto.fi/ica/fastica/code/dlcode.shtml))
## Setup ##
In the terminal type:
cd ~/path-to-directory
git clone https://github.com/davidkun/HyperSpectralToolbox.git
git clone https://github.com/davidkun/FastICA.git
Open Matlab. The default directory should contain a file `startup.m`. If not, create it:
% in Matlab command window
uPath = userpath;
cd(uPath(1:end-1)); % removes trailing colon
edit startup.m % may ask if you'd like to create it; click Yes
Add the following code to it (make sure to modify `path-to-directory` so it matches the actual path):
addtopath('~/path-to-directory/FastICA', ...
'~/path-to-directory/HyperSpectralToolbox/functions', ...
'~/path-to-directory/HyperSpectralToolbox/newFunctions');
You're ready to go now! Check out the demo files `hyperDemo.m` in `functions/` and `hyperDemo2.m` in `newFunctions/` to learn how to use the toolbox, or see the examples further down this page.
***
[Back to top](https://github.com/davidkun/HyperSpectralToolbox#welcome-to-my-project)
***
## Description ##
The open source Matlab Hyperspectral Toolbox is a Matlab toolbox containing various hyperspectral exploitation algorithms. The toolbox is meant to be a concise repository of current state-of-the-art exploitation algorithms for learning and research purposes. The toolbox includes functions for:
* **Target detection**
* Constrained Energy Minimization (CEM)
* Orthogonal Subspace Projection (OSP)
* Generalized Likelihood Ratio Test (GLRT)
* Adaptive Cosine/Coherent Estimator (ACE)
* Adaptive Matched Subspace Detector (AMSD)
* **Endmember Finders**
* Automatic Target Generation Procedure (ATGP)
* Independent component analysis - endmember extraction algorithm (ICA-EEA)
* **Material abundance map (MAM) generation**
* **Spectral Comparison**
* Spectral angle mapper (SAM)
* Spectral information divergence (SID)
* Normalize cross correlation
* **Anomaly Detectors**
* Reed-Xiaoli Detector (RX)
* **Least Square Solvers** (for abundance map estimation)
* Fully-constrained least squares (FCLS)
* Non negative least squares (NNLS)
* **Material Count Estimation**
* HFC virtual dimensionality (VD) for material count estimate
* **Automated processing**
* **Change detection**
* **Visualization**
* **Reading / writing files** (.rfl, .asd, ect)
***
[Back to top](https://github.com/davidkun/HyperSpectralToolbox#welcome-to-my-project)
***
## Examples ##
Download the Cuprite, Nevada hyperspectral image (HSI) from [here](http://aviris.jpl.nasa.gov/data/free_data.html). This will contain reflectance data and a .spc file with the spectral bands. The following samples of code are from `hyperDemo2.m`.
Show a 'slice' of the HSI:
slice = hyperReadAvirisRfl(rflFile, [1 512], [1 614], [bndnum bndnum]);
figure; imagesc(slice); axis image; colormap(gray);
![Figure 1: 1997 AVIRIS flight over Cuprite, NV](https://raw.githubusercontent.com/davidkun/HyperSpectralToolbox/master/figures/sampleSlice.png)
_Figure 1: 1997 AVIRIS flight over Cuprite, NV_
View an enhanced truecolor composite of the HSI:
tColor = hyperTruecolor(rflFile, 512, 614, 224, rgbBands, 'stretchlim');
figure; imagesc(tColor); axis image
![Figure 2: Truecolor composite from RGB bands](https://raw.githubusercontent.com/davidkun/HyperSpectralToolbox/master/figures/truecolor.png)
_Figure 2: Truecolor composite from RGB bands_
Plot the spectral signatures of 20 random pixels in order to determine which bands are greatly affected by water absorption and/or have a low signal-to-noise ratio (SNR):
![Figure 3: Pre-processing: removal of poor spectral bands from original HSI](https://raw.githubusercontent.com/davidkun/HyperSpectralToolbox/master/figures/targets_spectra.png)
_Figure 3: Pre-processing: removal of poor spectral bands from original HSI_
Using the resampled HSI cube, perform an endmember extraction algorithm, for example, the N-FINDR algorithm:
Unfindr = hyperNfindr(M2d, q);
figure; plot(lambdasNm, Unfindr, '.'); grid on;
![Figure 4: Endmember signatures estimated by PPI](https://raw.githubusercontent.com/davidkun/HyperSpectralToolbox/master/figures/endmmbrs-nfindr.png)
_Figure 4: Endmember signatures estimated by PPI_
Generate abundance maps using the non-negative constrained least squares method for each extracted endmember signature, for example:
abundanceMaps = hyperNnls(M2d, Uppi);
abundanceMaps = hyperConvert3d(abundanceMaps, h, w, q);
figure; imagesc(abundanceMaps(:,:,1)); colorbar; axis image;
![Figure 5: Abundance map from first N-FINDR-recovered endmember](https://raw.githubusercontent.com/davidkun/HyperSpectralToolbox/master/figures/abund-nfindr-1.png)
_Figure 5: Abundance map from first N-FINDR-recovered endmember_
These are just a few features of the Hyperspectral Toolbox.
***
[Back to top](https://github.com/davidkun/HyperSpectralToolbox#welcome-to-my-project)
***
### Algorithms to be added (requested by Dr. Gerg): ###
(Joint) Affine Matched filter
Generalization of matched filter which includes signature statistics
RAF-SAM, an improvement to SAM from: Improving the Classification Precision of Spectral Angle Mapper
ELM for radiance to reflectance conversion - http://www.cis.rit.edu/files/197_SPIE_2005_Grimm.pdf
Covariance matrix inversion methods (e.g. Dominant Mode Rejection)
Quadratic Detector
SMACC - http://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=844250
~~AMEE - http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=1046852~~
~~N-FINDR - http://proceedings.spiedigitallibrary.org/proceeding.aspx?articleid=994814~~
Fast PPI - http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=1576691
~~Joshua Broaderwater's hybrid detectors (HUD, etc)~~
Variations on ACE - e.g. adaptive covariance estimated ACE, etc
***
[Back to top](https://github.com/davidkun/HyperSpectralToolbox#welcome-to-my-project)
***
没有合适的资源?快使用搜索试试~ 我知道了~
毕业设计&课设-Sourceforge派生的HyperSpectral Matlab工具箱.zip
共71个文件
m:62个
png:5个
md:3个
0 下载量 42 浏览量
2024-01-08
22:46:01
上传
评论
收藏 1.01MB ZIP 举报
温馨提示
matlab算法,工具源码,适合毕业设计、课程设计作业,所有源码均经过严格测试,可以直接运行,可以放心下载使用。有任何使用问题欢迎随时与博主沟通,第一时间进行解答! matlab算法,工具源码,适合毕业设计、课程设计作业,所有源码均经过严格测试,可以直接运行,可以放心下载使用。有任何使用问题欢迎随时与博主沟通,第一时间进行解答! matlab算法,工具源码,适合毕业设计、课程设计作业,所有源码均经过严格测试,可以直接运行,可以放心下载使用。有任何使用问题欢迎随时与博主沟通,第一时间进行解答! matlab算法,工具源码,适合毕业设计、课程设计作业,所有源码均经过严格测试,可以直接运行,可以放心下载使用。有任何使用问题欢迎随时与博主沟通,第一时间进行解答! matlab算法,工具源码,适合毕业设计、课程设计作业,所有源码均经过严格测试,可以直接运行,可以放心下载使用。有任何使用问题欢迎随时与博主沟通,第一时间进行解答! matlab算法,工具源码,适合毕业设计、课程设计作业,所有源码均经过严格测试,可以直接运行,可以放心下载使用。有任何使用问题欢迎随
资源推荐
资源详情
资源评论
收起资源包目录
毕业设计&课设-Sourceforge派生的HyperSpectral Matlab工具箱.zip (71个子文件)
matlab_code
newFunctions
hyperDemo2.m 10KB
hyperRnfindr.m 785B
hyperTruecolor.m 2KB
hyperNfindr.m 3KB
README.md 534B
hyperAmee.m 4KB
hyperAvmax.m 3KB
figures
sampleSlice.png 167KB
abund-nfindr-1.png 218KB
endmmbrs-nfindr.png 72KB
targets_spectra.png 166KB
truecolor.png 358KB
.gitignore 51B
functions
hyperDemo_RIT_data.m 7KB
hyperMatchedFilter.m 933B
hyperFileFind.m 1KB
hyperResample.m 1KB
hyperUcls.m 649B
hyperNapc.m 2KB
hyperReadAsd.m 4KB
hyperFcls.m 2KB
hyperCorr.m 316B
hyperCem.m 1KB
hyperGlrt.m 802B
hyperDestreak.m 964B
hyperIcaEea.m 2KB
hyperOsp.m 916B
hyperNnls.m 1KB
hyperSignedAce.m 944B
hyperPct.m 791B
hyperAmsd.m 887B
hyperHud.m 2KB
hyperAtgp.m 1KB
fnnls.m 4KB
hyperDemo_ASD_reader.m 1KB
hyperHfcVd.m 2KB
hyperAce.m 1KB
hyperDemo_mams_RIT_data.m 3KB
hyperConvert3d.m 510B
hyperMax2d.m 371B
hyperDemo.m 4KB
hyperConvexHullRemoval.m 1KB
hyperSam.m 2KB
hyperVca.m 3KB
hyperNormalize.m 446B
hyperReadAvirisSpc.m 615B
hyperIcaComponentScores.m 923B
hyperReadSpecpr.m 5KB
hyperConvert2d.m 521B
hyperRxDetector.m 633B
hyperImshow.m 613B
hyperSaveFigure.m 796B
hyperRoc.m 177B
hyperFclsMatlab.m 978B
hyperOrthorectify.m 966B
hyperPlmf.m 2KB
hyperReadAvirisRfl.m 3KB
hyperPpi.m 838B
README.md 4KB
hyperImagesc.m 635B
hyperMnf.m 1KB
hyperCov.m 379B
hyperWhiten.m 971B
hyperGetEnviSignature.m 718B
hyperSid.m 641B
hyperRmf.m 2KB
hyperGetHymapWavelengthsNm.m 2KB
hyperConvert2Colormap.m 856B
hyperNormXCorr.m 776B
hyperDemo_detectors.m 4KB
README.md 7KB
共 71 条
- 1
资源评论
白话机器学习
- 粉丝: 9291
- 资源: 7681
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功