IGARSS 2011, Vancouver, Canada HYPERSPECTRAL UNMIXING USING A NOVEL CONVERSION MODEL Fereidoun A. Mianji, Member, IEEE, Shuang Zhou, Member, IEEE, Ye Zhang,

Slides:



Advertisements
Similar presentations
Random Forest Predrag Radenković 3237/10
Advertisements

Object Specific Compressed Sensing by minimizing a weighted L2-norm A. Mahalanobis.
ICONIP 2005 Improve Naïve Bayesian Classifier by Discriminative Training Kaizhu Huang, Zhangbing Zhou, Irwin King, Michael R. Lyu Oct
Olivier Duchenne , Armand Joulin , Jean Ponce Willow Lab , ICCV2011.
A ROBUST SPECTRAL TARGET RECOGNITION METHOD FOR HYPERSPECTRAL DATA BASED ON COMBINED SPECTRAL SIGNATURES IGARSS 2011 Vancouver, July Xiao Fan, Ye.
« هو اللطیف » By : Atefe Malek. khatabi Spring 90.
Learning on Probabilistic Labels Peng Peng, Raymond Chi-wing Wong, Philip S. Yu CSE, HKUST 1.
São Paulo Advanced School of Computing (SP-ASC’10). São Paulo, Brazil, July 12-17, 2010 Looking at People Using Partial Least Squares William Robson Schwartz.
Nov 4, Detection, Classification and Tracking of Targets in Distributed Sensor Networks Presented by: Prabal Dutta Dan Li, Kerry Wong,
Multiple Criteria for Evaluating Land Cover Classification Algorithms Summary of a paper by R.S. DeFries and Jonathan Cheung-Wai Chan April, 2000 Remote.
1 Learning to Detect Objects in Images via a Sparse, Part-Based Representation S. Agarwal, A. Awan and D. Roth IEEE Transactions on Pattern Analysis and.
Texture Classification Using QMF Bank-Based Sub-band Decomposition A. Kundu J.L. Chen Carole BakhosEvan Kastner Dave AbramsTommy Keane Rochester Institute.
Object Detection and Tracking Mike Knowles 11 th January 2005
Jacinto C. Nascimento, Member, IEEE, and Jorge S. Marques
Digital Imaging and Remote Sensing Laboratory Real-World Stepwise Spectral Unmixing Daniel Newland Dr. John Schott Digital Imaging and Remote Sensing Laboratory.
Information Fusion Yu Cai. Research Article “Comparative Analysis of Some Neural Network Architectures for Data Fusion”, Authors: Juan Cires, PA Romo,
Machine Learning in Simulation-Based Analysis 1 Li-C. Wang, Malgorzata Marek-Sadowska University of California, Santa Barbara.
Chapter 12 Spatial Sharpening of Spectral Image Data.
Graph-based consensus clustering for class discovery from gene expression data Zhiwen Yum, Hau-San Wong and Hongqiang Wang Bioinformatics, 2007.
Transfer Learning From Multiple Source Domains via Consensus Regularization Ping Luo, Fuzhen Zhuang, Hui Xiong, Yuhong Xiong, Qing He.
TARGETED LAND-COVER CLASSIFICATION by: Shraddha R. Asati Guided by: Prof. P R.Pardhi.
1 Template-Based Classification Method for Chinese Character Recognition Presenter: Tienwei Tsai Department of Informaiton Management, Chihlee Institute.
Active Learning for Class Imbalance Problem
Noise-Robust Spatial Preprocessing Prior to Endmember Extraction from Hyperspectral Data Gabriel Martín, Maciel Zortea and Antonio Plaza Hyperspectral.
Chenghai Yang 1 John Goolsby 1 James Everitt 1 Qian Du 2 1 USDA-ARS, Weslaco, Texas 2 Mississippi State University Applying Spectral Unmixing and Support.
CS654: Digital Image Analysis Lecture 3: Data Structure for Image Analysis.
A New Subspace Approach for Supervised Hyperspectral Image Classification Jun Li 1,2, José M. Bioucas-Dias 2 and Antonio Plaza 1 1 Hyperspectral Computing.
Tracking with Unreliable Node Sequences Ziguo Zhong, Ting Zhu, Dan Wang and Tian He Computer Science and Engineering, University of Minnesota Infocom 2009.
Jonatan Gefen 28/11/2012. Outline Introduction to classification Whole Pixel Subpixel Classification Linear Unmixing Matched Filtering (partial unmixing)
Urban Building Damage Detection From Very High Resolution Imagery By One-Class SVM and Shadow Information Peijun Li, Benqin Song and Haiqing Xu Peking.
CS 782 – Machine Learning Lecture 4 Linear Models for Classification  Probabilistic generative models  Probabilistic discriminative models.
Exploiting Context Analysis for Combining Multiple Entity Resolution Systems -Ramu Bandaru Zhaoqi Chen Dmitri V.kalashnikov Sharad Mehrotra.
Xu Huaping, Wang Wei, Liu Xianghua Beihang University, China.
A new Ad Hoc Positioning System 컴퓨터 공학과 오영준.
Scene Completion Using Millions of Photographs James Hays, Alexei A. Efros Carnegie Mellon University ACM SIGGRAPH 2007.
Interval Type-2 Fuzzy T-S Modeling For A Heat Exchange Process On CE117 Process Trainer Proceedings of 2011 International Conference on Modelling, Identification.
Plane-based external camera calibration with accuracy measured by relative deflection angle Chunhui Cui , KingNgiNgan Journal Image Communication Volume.
Endmember Extraction from Highly Mixed Data Using MVC-NMF Lidan Miao AICIP Group Meeting Apr. 6, 2006 Lidan Miao AICIP Group Meeting Apr. 6, 2006.
SAR-ATR-MSTAR TARGET RECOGNITION FOR MULTI-ASPECT SAR IMAGES WITH FUSION STRATEGIES ASWIN KUMAR GUTTA.
earthobs.nr.no Retraining maximum likelihood classifiers using a low-rank model Arnt-Børre Salberg Norwegian Computing Center Oslo, Norway IGARSS.
GENDER AND AGE RECOGNITION FOR VIDEO ANALYTICS SOLUTION PRESENTED BY: SUBHASH REDDY JOLAPURAM.
Chapter 13 (Prototype Methods and Nearest-Neighbors )
Remote Sensing Unsupervised Image Classification.
Hand Gesture Recognition Using Haar-Like Features and a Stochastic Context-Free Grammar IEEE 高裕凱 陳思安.
Review of Spectral Unmixing for Hyperspectral Imagery Lidan Miao Sept. 29, 2005.
Jeong Kanghun CRV (Computer & Robot Vision) Lab..
Mathematical Analysis of MaxEnt for Mixed Pixel Decomposition
Journal of Visual Communication and Image Representation
Comparative Analysis of Spectral Unmixing Algorithms Lidan Miao Nov. 10, 2005.
Hierarchical Mixture of Experts Presented by Qi An Machine learning reading group Duke University 07/15/2005.
Computer Vision Lecture 7 Classifiers. Computer Vision, Lecture 6 Oleh Tretiak © 2005Slide 1 This Lecture Bayesian decision theory (22.1, 22.2) –General.
Edge Preserving Spatially Varying Mixtures for Image Segmentation Giorgos Sfikas, Christophoros Nikou, Nikolaos Galatsanos (CVPR 2008) Presented by Lihan.
Zhaoxia Fu, Yan Han Measurement Volume 45, Issue 4, May 2012, Pages 650–655 Reporter: Jing-Siang, Chen.
An Automatic Method for Selecting the Parameter of the RBF Kernel Function to Support Vector Machines Cheng-Hsuan Li 1,2 Chin-Teng.
Automatic Classification for Pathological Prostate Images Based on Fractal Analysis Source: IEEE TRANSACTIONS ON MEDICAL IMAGING, VOL. 28, NO. 7, JULY.
Selected Hyperspectral Mapping Method
Pathik Thakkar University of Texas at Dallas
Guillaume-Alexandre Bilodeau
Bag-of-Visual-Words Based Feature Extraction
Amin Zehtabian, Hassan Ghassemian CONCLUSION & FUTURE WORKS
Machine Learning Basics
R-CNN region By Ilia Iofedov 11/11/2018 BGU, DNN course 2016.
Students: Meiling He Advisor: Prof. Brain Armstrong
In summary C1={skin} C2={~skin} Given x=[R,G,B], is it skin or ~skin?
Radio Propagation Simulation Based on Automatic 3D Environment Reconstruction D. He A novel method to simulate radio propagation is presented. The method.
Supervised Classification
Sparse Regression-based Hyperspectral Unmixing
Multi-Information Based GCPs Selection Method
A Novel Smoke Detection Method Using Support Vector Machine
Random Neural Network Texture Model
Presentation transcript:

IGARSS 2011, Vancouver, Canada HYPERSPECTRAL UNMIXING USING A NOVEL CONVERSION MODEL Fereidoun A. Mianji, Member, IEEE, Shuang Zhou, Member, IEEE, Ye Zhang, Member, IEEE Presentation by: Shuang Zhou School of Electronics and Information Technology Harbin Institute of Technology, Harbin, China

Hyperspectral Unmixing Using a Novel Conversion Model 1.Introduction 2. Linear Mixture Model (LMM) and Mean-Based Algorithms for Unmixing 3. Structure of the Proposed Approach 4. Data and Experimental Design 5. Experiments with an Simulated Image made of Real Hyperspectral Data 6. Experiments with Real Hyperspectral Image with Implanted Mixed Pixels 7. Experiments with Real Hyperspectral image with Many Endmembers 8. Discussion and Conclusion on the Proposed Approach

1. Introduction The precise information obtained by unmixing is very helpful in the other applications such as edge sharpening, area estimation, subpixel level target detection, etc. Fig Spectral variability (a) and mixed-pixel interference (b).

1. Introduction Continue Unmixing generally refers to a process which includes two steps: 1) an endmember extraction algorithm (EEA) to define the pure (or purest) pixels in the scene as the endmembers; 2) an endmember abundance quantification algorithm (inversion step) to define the contribution of different endmembers in each pixel.

2. Linear Mixture Model (LMM) and Mean-Based Algorithms for Unmixing There are two main models to deal with an unmixing problem: 1) the macroscopic mixture model which models a mixed pixel as a linear combination of materials with relative concentrations (LMM), 2) the intimate spectral mixture model which proposes a nonlinear mixing model of materials. Two main constraints which are often imposed in constrained Linear Spectral Mixture Analysis(LSMA) are a) abundance sum-to-one constraint (ASC) b) abundance nonnegativity constraint (ANC).

2. Linear Mixture Model (LMM) and Mean-Based Algorithms for Unmixing Mean-Based Algorithms for Unmixing Fully constrained least squares (FCLS) algorithm can efficiently meet both abundance constraints and is optimal in terms of least squares error (LSE). FCLS assumes that the spectra for all endmembers in the scene are available and unique. Therefore, it is optimized for ideal cases where the information of the pure pixels are a priori known, and where there is no spectral variability in the scene.

2. Linear Mixture Model (LMM) and Mean-Based Algorithms for Unmixing Limitations of Mean-Based Algorithms Two main obstacles for a successful LSMA-based method include: 1) the necessary a priori knowledge of the pure signatures of materials present in the image scene is usually not available. 2) these signatures rarely represent pure spectral signatures of all materials in the image scene in practical situations due to the endmember spectral variability. The main resulting drawback with the mean-based unmixing algorithms is being limited to the mean spectra of each endmemeber. They ignore the statistics associated with the training samples.

3. Structure of the Proposed Approach The rational behind the proposed method is exploiting the statistical variations of the training samples extracted from the scene rather than being only limited to their mean vectors. The central idea of the proposed method is converting the abundance quantification problem (inversion step) into a classification problem through assigning a set of artificial class labels to different possible abundances (0%-100%) of an endmember in a pixel. To realize the proposed method, an SVM approach is applied. Thus, the proposed method is referred to as unmixing-to-classification conversion model with SVM and is noted as uccm-SVM. The well known one-against-all (OAA) strategy with parallel architecture is adopted in uccm-SVM for the multiendmember unmixing tasks.

3. Structure of the Proposed Approach uccm-SVM

Fig The layout and process flow of uccm-SVM designed for unmixing with a resolution of 1%. Yes No SVM training SVM1: trained for 0% of“one” SVM101: trained for 100% of“one” SVM2: trained for 1% of“one” 1- Designating sampleset i as“one” (initialize with i=1) 2- Makingsynthetic classes using remaining p-1 training sample sets (“rest”) Quantification result for “one” in all pixels (fractional image) … p endmembers: p training sample sets including extracted pure pixel vectors Image Pixels i>p ? Endmember fractions rescaling tounity Next i (endmember)

4. Data and Experimental Design The hyperspectral data used to validate the proposed unmixing method include: a)San Diego data for making a simulated image b)Indian Pines data as real image with implanted compositions (mixed pixels) c)The University of Pavia data set as a multiendmemebr real image.

5. Experiments with a Simulated Image Made of Real Hyperspectral Data Using 30 randomly selected pixels from three classes of San Diego image, sand, concrete, and roof, we simulated a hyperspectral image. We construct the background with sand and concrete and we use roof to implant the targets. A: sand B: concrete C: roof The simulated image is a 1530 pixel vectors (a 51 by 30 hyperspectral image) Fig The simulated image constructed using samples from 3 classes of San Diego.

5. Experiments with a Simulated Image Made of Real Hyperspectral Data Results Fig. 5.2 Unmixing result for roof by FCLS. Fig Unmixing result for roof by uccm-SVM. We expect to see only 3 peaks, with the amplitue&width of 0.1&15, 0.2&10, and 0.3&5 starting in pixel locations in 300, 750, and 1200, respectively. As can be seen, our method presents a much better result.

5. Experiments with a Simulated Image Made of Real Hyperspectral Data Results From the table below it is clear that the proposed method is superior to the FCLS algorithm. Table 5.1 Comparison of uccm_SVM with FCLS in terms of average square error and computational time. TechniqueAverage square error (%)Proccessing time (s) SandConcreteRoofTrainingTest FCLS N/A7.5 Uccm-SVM

6. Experiments with Real Hyperspectral Image with Implanted Mixed Pixels Regions of Interest covering corn-notill & soybean-notill (ROI1) and wheat & corn-min (ROI2) were selected. Fig. 6.1 Two chosen ROIs on Indian Pine hyperspectral image for experiments.

6. Experiments with Real Hyperspectral Image with Implanted Mixed Pixels Implanting the Border with Mixture of Neighboring Pixels ROI1 is a 21 by 8 pixel area covering two landcovers, corn-notill and soybean-notill. We implanted the pixels of the border line with a varying mixture of the two endmembers using the neighboring pixels. To unmix the implanted pixel vectors, 21 different pixel vectors from the other areas of each landcover class were randomly chosen as the training samples. Fig Implanting a varying mixture of neighboring pixel vectors in the border line between two adjacent landcover classes

6. Experiments with Real Hyperspectral Image with Implanted Mixed Pixels Unmixing the Mixed Pixels Implanted in ROI1 Fig (from left to right) True abundance of corn-notill in the border line of ROI1, unmixing results by FCLS, and unmixing result by uccm- SVM. Table 6.1. Comparison of uccm-SVM with FCLS in terms of average square error and computational time for Indian Pine ROI1. TechniqueAverage square error (%) Training time (s) Test time (s) FCLS 17.01N/A0.44 Uccm-SVM

7. Experiments with Real Hyperspectral image with Many Endmembers 1- Using all the labeled training and test samples, and a maximum likelihood classifier, we classified the image. 2- With a visual and spectral comparison to the result of some other techniques (e.g. SVM), we insured of the high accuracy of our classification result. 3- Then, we downsampled the image by a 3x3 mean filter to obtain an image (203x113) with a priori known compositions of endmembers. 4- We unmixed the downsampled image by FCLS and uccm-SVM with very low, low, and medium number of training samples (from pure samples of the downsampled image). Fig The University of Pavia data set.

7. Experiments with Real Hyperspectral Image with Many Endmembers Results # Training samples Method Average square errors (ASE) for classes and also average ASE (%) Time (s) Ave. TrainTest 2 FCLS uccm-SVM FCLS uccm-SVM FCLS uccm-SVM ASE: uccm-SVM performs better than FCLS on majority of single classes and in ASE average over all sizes of training sets. Computationally: uccm-SVM is faster than FCLS for low number of training samples and slower for higher numbers. Table Average square error (ASE) for the obtained fractional images using FCLS and uccm-SVM for unmixing of The University of Pavia data set (downsampled).

8. Discussion and Conclusion on the Proposed Approach A new learning-based unmixing technique through SVM is proposed in this paper. An unmixing-to-classification conversion model which transfers the unmixing task into a classification problem is the main premise of the method. Abundance quantification for each endmember is performed in a one- against-rest framework using a number of artificially made classes. Uccm-SVM considerably outperformed the mean-based technique, FCLS algorithm. The next step would be improving the performance of the proposed method while reducing its computational burden for big sizes of training sets.

Thanks For Your Attention Shuang Zhou School of Electronics and Information Technology Harbin Institute of Technology, Harbin, China