<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Remote Sensing on 热爱可抵万难</title><link>https://www.jingmengzhiyue.tech/en/tags/remote-sensing/</link><description>Recent content from 热爱可抵万难</description><generator>Hugo</generator><language>en</language><copyright>All articles on this blog are licensed under the BY license agreement unless otherwise stated. Please indicate the source when reprinting!</copyright><lastBuildDate>Mon, 25 Mar 2024 22:35:53 +0800</lastBuildDate><atom:link href="https://www.jingmengzhiyue.tech/en/tags/remote-sensing/index.xml" rel="self" type="application/rss+xml"/><item><title>Hyperspectral Image Processing Based on Prototype Analysis</title><link>https://www.jingmengzhiyue.tech/en/prototype-analysis/</link><pubDate>Mon, 25 Mar 2024 21:35:21 +0800</pubDate><guid>https://www.jingmengzhiyue.tech/en/prototype-analysis/</guid><description>
<![CDATA[<h1>Hyperspectral Image Processing Based on Prototype Analysis</h1><p>Author: Jingmengzhiyue()</p>
        
          <h2 id="introduction-to-prototype-analysis">
<a class="header-anchor" href="#introduction-to-prototype-analysis"></a>
Introduction to Prototype Analysis
</h2><p>In hyperspectral image processing, prototype analysis (Archetypal Analysis, AA) is an unsupervised method for endmember extraction and abundance estimation. Unlike conventional endmember extraction methods, prototype analysis does not assume that endmembers lie at the vertices of a simplex. Instead, it seeks the minimal convex hull of the original data points.</p>
<p>Given a hyperspectral data matrix \(\mathbf{X}=\left\{\mathbf{x}_{1}, \mathbf{x}_{2}, \ldots, \mathbf{x}_{n}, \ldots \mathbf{x}_{N}\right\} \in \mathbf{R}^{M \times N}\), where \(M\) is the number of spectral bands and \(N\) is the number of pixels, the goal of prototype analysis is to estimate \(D(D\ll N)\) prototypes \(\mathbf{Z}=\left\{\mathbf{z}_{1}, \mathbf{z}_{2}, \ldots, \mathbf{z}_{d}, \ldots \mathbf{z}_{D}\right\} \in \mathbf{R}^{M \times D}\) from \(\mathbf{X}\), so that each original observation \(\mathbf{x}_{n}\) can be written as a linear combination of the prototypes:</p>
        
        <hr><p>Published on 2024-03-25 at <a href='https://www.jingmengzhiyue.tech/'>热爱可抵万难</a>, last modified on 2024-03-25</p>]]></description><category>Remote Sensing</category></item><item><title>A Hyperspectral Image Super-Resolution Algorithm Based on Coupled Nonnegative Matrix Factorization</title><link>https://www.jingmengzhiyue.tech/en/nssr/</link><pubDate>Sun, 24 Mar 2024 16:30:33 +0800</pubDate><guid>https://www.jingmengzhiyue.tech/en/nssr/</guid><description>
<![CDATA[<h1>A Hyperspectral Image Super-Resolution Algorithm Based on Coupled Nonnegative Matrix Factorization</h1><p>Author: Jingmengzhiyue()</p>
        
          <h2 id="problem-definition">
<a class="header-anchor" href="#problem-definition"></a>
Problem Definition
</h2><p>Hyperspectral imaging acquires images of the same scene at many different wavelengths at once, and it is widely used in agriculture, astronomy, surveillance, and mineralogy. Because of various hardware limitations, however, obtaining a high-resolution (HR) hyperspectral image remains a challenge. What follows shows how to reconstruct an HR hyperspectral image from a low-resolution (LR) hyperspectral image and an HR multispectral or RGB image of the same scene.</p>
        
        <hr><p>Published on 2024-03-24 at <a href='https://www.jingmengzhiyue.tech/'>热爱可抵万难</a>, last modified on 2024-03-25</p>]]></description><category>Remote Sensing</category></item><item><title>Principles and Solution of the Vertex Component Analysis (VCA) Algorithm</title><link>https://www.jingmengzhiyue.tech/en/vca/</link><pubDate>Mon, 25 Mar 2024 21:35:21 +0800</pubDate><guid>https://www.jingmengzhiyue.tech/en/vca/</guid><description>
<![CDATA[<h1>Principles and Solution of the Vertex Component Analysis (VCA) Algorithm</h1><p>Author: Jingmengzhiyue()</p>
        
          <p>Vertex Component Analysis (VCA) is a fast unsupervised algorithm for extracting endmembers from hyperspectral images. Its main idea is to use the property that endmembers form the vertices of the data cloud in a high-dimensional feature space. By iteratively searching for spectra that lie on the boundary of the data cloud, the algorithm extracts the endmembers. The mathematical principle and the solution procedure are presented in detail below.</p>
<h2 id="problem-statement">
<a class="header-anchor" href="#problem-statement"></a>
Problem Statement
</h2><p>Consider a hyperspectral image with \(L\) bands. The spectral vector of each pixel is \(\mathbf{r} \in \mathbb{R}^{L}\), and the whole image can be written as \(\mathbf{R} = [\mathbf{r}_1, \mathbf{r}_2, \dots, \mathbf{r}_N] \in \mathbb{R}^{L \times N}\), where \(N\) is the number of pixels. Under the linear mixing model, each spectral vector \(\mathbf{r}_i\) is a linear combination of \(p\) endmembers \(\mathbf{m}_1, \mathbf{m}_2, \dots, \mathbf{m}_p \in \mathbb{R}^{L}\):</p>
        
        <hr><p>Published on 2024-03-25 at <a href='https://www.jingmengzhiyue.tech/'>热爱可抵万难</a>, last modified on 2024-03-25</p>]]></description><category>Remote Sensing</category></item><item><title>Maximum Distance (MaxD) Endmember Extraction for Hyperspectral Images</title><link>https://www.jingmengzhiyue.tech/en/maxd/</link><pubDate>Mon, 25 Mar 2024 21:35:21 +0800</pubDate><guid>https://www.jingmengzhiyue.tech/en/maxd/</guid><description>
<![CDATA[<h1>Maximum Distance (MaxD) Endmember Extraction for Hyperspectral Images</h1><p>Author: Jingmengzhiyue()</p>
        
          <p>The maximum distance (MaxD) algorithm is an unsupervised endmember extraction method used to identify pure endmember spectra in a hyperspectral image. This post presents the mathematical principle of the MaxD algorithm and a MATLAB implementation.</p>
<h2 id="algorithm-principle">
<a class="header-anchor" href="#algorithm-principle"></a>
Algorithm Principle
</h2><p>The MaxD algorithm is based on the following key ideas:</p>
<ol>
<li>
<p><strong>Maximum-distance criterion</strong>: Endmember spectra correspond to the sample points that lie farthest apart in feature space. This is because an endmember spectrum represents the spectrum of a pure surface material, whereas a mixed spectrum lies between endmember spectra.</p>
</li>
<li>
<p><strong>Sequential extraction</strong>: The algorithm extracts all endmember spectra one by one, in descending order of the distance between each endmember spectrum and the other spectra.</p>
</li>
<li>
<p><strong>Vector projection</strong>: When a new endmember spectrum is extracted, linear combinations of the known endmember spectra are used to construct a vector space. The new endmember spectrum is the one with the largest projection length in the complementary subspace of that space.</p>
</li>
</ol>
        
        <hr><p>Published on 2024-03-25 at <a href='https://www.jingmengzhiyue.tech/'>热爱可抵万难</a>, last modified on 2024-03-25</p>]]></description><category>Remote Sensing</category></item><item><title>Endmember Extraction Based on the N-FINDR Algorithm</title><link>https://www.jingmengzhiyue.tech/en/n-finder/</link><pubDate>Mon, 25 Mar 2024 21:35:21 +0800</pubDate><guid>https://www.jingmengzhiyue.tech/en/n-finder/</guid><description>
<![CDATA[<h1>Endmember Extraction Based on the N-FINDR Algorithm</h1><p>Author: Jingmengzhiyue()</p>
        
          <p>The N-FINDR algorithm extracts endmember signatures from a hyperspectral data cube. Its core idea is to search the data set for the simplex of maximum volume, whose vertices are the desired endmembers. The mathematical principle and the concrete steps are as follows.</p>
<h2 id="mathematical-principle">
<a class="header-anchor" href="#mathematical-principle"></a>
Mathematical Principle
</h2><p>Consider an \(M \times N \times C\)-dimensional hyperspectral data cube \(\mathbf{X}\). From it we need to find \(p\) endmember signatures \(\mathbf{e}_1, \mathbf{e}_2, \ldots, \mathbf{e}_p\). The matrix formed by these signature vectors is written \(\mathbf{E} = [\mathbf{e}_1, \mathbf{e}_2, \ldots, \mathbf{e}_p]^T \in \mathbb{R}^{C \times p}\).</p>
        
        <hr><p>Published on 2024-03-25 at <a href='https://www.jingmengzhiyue.tech/'>热爱可抵万难</a>, last modified on 2024-03-25</p>]]></description><category>Remote Sensing</category></item><item><title>Extracting Endmember Features with the Fast Iterative Pixel Purity Index</title><link>https://www.jingmengzhiyue.tech/en/fippi/</link><pubDate>Mon, 25 Mar 2024 21:26:16 +0800</pubDate><guid>https://www.jingmengzhiyue.tech/en/fippi/</guid><description>
<![CDATA[<h1>Extracting Endmember Features with the Fast Iterative Pixel Purity Index</h1><p>Author: Jingmengzhiyue()</p>
        
          <h1 id="introduction-to-the-fippi-algorithm">
<a class="header-anchor" href="#introduction-to-the-fippi-algorithm"></a>
Introduction to the FIPPI Algorithm
</h1><p>The Fast Iterative Pixel Purity Index (FIPPI) is an effective method for extracting endmember signatures from hyperspectral remote-sensing data. It is an iterative improvement of the pixel purity index (PPI), designed to improve computational efficiency and obtain better endmember estimates through a sound stopping condition and carefully designed iteration rules.</p>
<h1 id="mathematical-principles-of-the-algorithm">
<a class="header-anchor" href="#mathematical-principles-of-the-algorithm"></a>
Mathematical Principles of the Algorithm
</h1><h2 id="dimensionality-reduction">
<a class="header-anchor" href="#dimensionality-reduction"></a>
Dimensionality Reduction
</h2><p>In the FIPPI algorithm, the original high-dimensional data cube is first reduced in dimension in order to lower computational complexity. Common dimensionality-reduction methods include:</p>
        
        <hr><p>Published on 2024-03-25 at <a href='https://www.jingmengzhiyue.tech/'>热爱可抵万难</a>, last modified on 2024-03-25</p>]]></description><category>Remote Sensing</category></item><item><title>A Subspace-Based Change Detection Method for Hyperspectral Images</title><link>https://www.jingmengzhiyue.tech/en/change-detection/</link><pubDate>Sun, 24 Mar 2024 17:41:32 +0800</pubDate><guid>https://www.jingmengzhiyue.tech/en/change-detection/</guid><description>
<![CDATA[<h1>A Subspace-Based Change Detection Method for Hyperspectral Images</h1><p>Author: Jingmengzhiyue()</p>
        
          <p>Accurately and promptly detecting changes in land-surface features is essential for environmental monitoring and resource management. Compared with conventional single-band or multispectral remote sensing imagery, hyperspectral imagery contains a large amount of spectral information and therefore performs better at change detection. This article introduces a subspace-based change detection (SCD) method for hyperspectral images.</p>
<h2 id="subspace-analysis-for-hyperspectral-imagery">
<a class="header-anchor" href="#subspace-analysis-for-hyperspectral-imagery"></a>
Subspace Analysis for Hyperspectral Imagery
</h2><p>Subspace learning is an effective tool for analyzing hyperspectral imagery. Its strengths are that it can accurately describe linearly mixed signals and that it offers better signal separability. Orthogonal subspace projection (OSP) is a successful method for hyperspectral image analysis and is widely used across a range of applications.</p>
        
        <hr><p>Published on 2024-03-24 at <a href='https://www.jingmengzhiyue.tech/'>热爱可抵万难</a>, last modified on 2024-03-24</p>]]></description><category>Remote Sensing</category></item><item><title>A Fast Hyperspectral Image Unmixing Algorithm Based on the Minimum-Volume Simplex</title><link>https://www.jingmengzhiyue.tech/en/hypercsi/</link><pubDate>Sun, 24 Mar 2024 11:37:14 +0800</pubDate><guid>https://www.jingmengzhiyue.tech/en/hypercsi/</guid><description>
<![CDATA[<h1>A Fast Hyperspectral Image Unmixing Algorithm Based on the Minimum-Volume Simplex</h1><p>Author: Jingmengzhiyue()</p>
        
          <p>Hyperspectral remote sensing is a key technique for identifying the endmember constituents within a target region and their proportions. It is widely used in land-cover classification, environmental monitoring, and related fields. Hyperspectral unmixing (HU) is the signal-processing task of analyzing hyperspectral data to identify the endmembers in an image and their corresponding abundances. Craig&rsquo;s minimum-volume simplex criterion is a classical and effective method for blind unmixing: the vertices of the minimum-volume simplex that encloses the data point cloud are taken as the endmember estimates. Related algorithms are computationally inefficient, however, because they involve complicated volume calculations. HyperCSI (Hyperplane-based Craig-Simplex Identification) does not evaluate a complicated volume function. It uses a geometric theorem instead: an \(N\)-dimensional simplex can be determined by \(N\) hyperplanes (of dimension \(N-2\)). HyperCSI reconstructs Craig&rsquo;s minimum-volume simplex by constructing \(N\) hyperplanes associated with it. This avoids the volume calculation and is far less expensive than algorithms based on Craig&rsquo;s criterion.</p>
        
        <hr><p>Published on 2024-03-24 at <a href='https://www.jingmengzhiyue.tech/'>热爱可抵万难</a>, last modified on 2024-03-24</p>]]></description><category>Remote Sensing</category><category>Algorithm</category></item><item><title>Endmember Extraction Using the Pixel Purity Index (PPI) Algorithm</title><link>https://www.jingmengzhiyue.tech/en/ppi-en/</link><pubDate>Tue, 19 Mar 2024 22:12:45 +0800</pubDate><guid>https://www.jingmengzhiyue.tech/en/ppi-en/</guid><description>
<![CDATA[<h1>Endmember Extraction Using the Pixel Purity Index (PPI) Algorithm</h1><p>Author: Jingmengzhiyue()</p>
        
          <p>The Pixel Purity Index (PPI) algorithm extracts endmember signatures from hyperspectral data cubes. It identifies endmember pixels by projecting dimensionality-reduced data onto randomly generated unit vectors and finding the pixels with the largest projection values.</p>
<h2 id="algorithm-principle">
<a class="header-anchor" href="#algorithm-principle"></a>
Algorithm Principle
</h2><ol>
<li>Perform dimensionality reduction on the hyperspectral data cube, choosing from PCA (Principal Component Analysis), MNF (Maximum Noise Fraction), or no reduction at all.</li>
<li>Generate a set number of random unit vectors (10,000 by default).</li>
<li>Project the reduced data onto each random unit vector and record the maximum projection value obtained by each pixel.</li>
<li>Rank the pixels by their maximum projection values and select the N pixels with the largest values as the endmember signatures.</li>
</ol>
<h2 id="algorithm-implementation">
<a class="header-anchor" href="#algorithm-implementation"></a>
Algorithm Implementation
</h2><p>Here is how to implement the PPI algorithm in MATLAB.</p>
        
        <hr><p>Published on 2024-03-19 at <a href='https://www.jingmengzhiyue.tech/'>热爱可抵万难</a>, last modified on 2024-03-24</p>]]></description><category>Remote Sensing</category></item><item><title>Endmember Initialization</title><link>https://www.jingmengzhiyue.tech/en/endmember-initialization/</link><pubDate>Tue, 19 Mar 2024 21:08:00 +0800</pubDate><guid>https://www.jingmengzhiyue.tech/en/endmember-initialization/</guid><description>
<![CDATA[<h1>Endmember Initialization</h1><p>Author: Jingmengzhiyue()</p>
        
          <h1 id="the-importance-of-endmember-estimation-in-hyperspectral-image-processing">
<a class="header-anchor" href="#the-importance-of-endmember-estimation-in-hyperspectral-image-processing"></a>
The Importance of Endmember Estimation in Hyperspectral Image Processing
</h1><p>In hyperspectral image processing, one often has to initialize the endmember matrix, and an accurate estimate of that matrix has a major effect on the performance of the final result. When the problem is convex, the choice of initial value has a relatively small effect on the final result. For a nonconvex problem, however, an initial value close to the global optimum can substantially improve the algorithm&rsquo;s performance.</p>
        
        <hr><p>Published on 2024-03-19 at <a href='https://www.jingmengzhiyue.tech/'>热爱可抵万难</a>, last modified on 2024-03-24</p>]]></description><category>Remote Sensing</category></item><item><title>Endmember Number Estimation Using Eigenvalue Thresholding</title><link>https://www.jingmengzhiyue.tech/en/est-endnum-en/</link><pubDate>Tue, 19 Mar 2024 22:01:49 +0800</pubDate><guid>https://www.jingmengzhiyue.tech/en/est-endnum-en/</guid><description>
<![CDATA[<h1>Endmember Number Estimation Using Eigenvalue Thresholding</h1><p>Author: Jingmengzhiyue()</p>
        
          <p>In hyperspectral image analysis, estimating the number of endmembers</p>
<p>is a crucial step that affects the accuracy and efficiency of subsequent</p>
<p>processing. This post introduces an eigenvalue thresholding approach for</p>
<p>estimating the number of endmembers in hyperspectral data through noise</p>
<p>whitening and eigenvalue analysis.</p>
<h2 id="noise-whitening-of-hyperspectral-images">
<a class="header-anchor" href="#noise-whitening-of-hyperspectral-images"></a>
Noise Whitening of Hyperspectral Images
</h2><p>Let&rsquo;s consider a hyperspectral image \(\mathbf{X} \in \mathbb{R}^{M \times B}\),</p>
<p>where \(M\) is the number of spatial</p>
<p>pixels and \(B\) is the number of</p>
        
        <hr><p>Published on 2024-03-19 at <a href='https://www.jingmengzhiyue.tech/'>热爱可抵万难</a>, last modified on 2024-03-24</p>]]></description><category>Remote Sensing</category></item></channel></rss>