Ground truth data collection for unmixing algorithm evaluation

Rivera-Borrero, Carlos; Rosario, Samuel; Hunt, Shawn; Zayas, Carmen Nieto; Mundorf, Adrienne; Cardona, Suhaily

doi:10.1117/12.779209

Cited by 3 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using the Scree test, a rank of 6 is estimated for this group of images. This result agrees with field work performed in 2007 [10], where nine main spatial classes were identified. The class map is presented in Figure 12.…”

Section: Scree Testsupporting

confidence: 95%

See 1 more Smart Citation

Understanding the impact of spatial resolution in unmixing of hyperspectral images

Santos-Garcia

Vélez-Reyes

2010

SPIE Proceedings

View full text Add to dashboard Cite

Unmixing is an ill-posed inverse problem and as such the solution computed with different unmixing algorithms depends on the underlying assumptions for the inverse problem. Ideally one would expect similar solutions for unmixing a hyperspectral image of different spatial resolutions of the same scene. In this paper, we study the results of unmixing different images of the same area at different spatial resolution using different unmixing algorithms. We also compare the estimation of the number of endmembers using the rank of a scaled correlation matrix against the positive rank estimated with the fitting error of a positive matrix factorization. The results show that algorithms that assume the pure pixels in the image given consistent results in the same scale and are limited to the number of endmembers determined from the rank of the scaled correlation matrix while algorithms that do not assume pure pixels are consistent across spatial scales and the number of endmembers is better estimated by the positive rank. One and four meter data collected with the AISA sensor over southwestern Puerto Rico is used for the study.

show abstract

Section: Scree Testsupporting

confidence: 95%

“…Figure 12. Classification map at 1 meter for Enrique reef image [10]. Figure 10 shows the normalized correlation matrices for the Enrique reef images and Figure 11 presents the corresponding Scree tests.…”

Section: Scree Testmentioning

confidence: 99%

Understanding the impact of spatial resolution in unmixing of hyperspectral images

Santos-Garcia

Vélez-Reyes

2010

SPIE Proceedings

View full text Add to dashboard Cite

show abstract

“…In order to adequately and objectively evaluate the resolution performance of the reconstructed data sets in comparison to the original when processed using the block-CS algorithm, we chose to perform testing on band 35 of a copy of the Indian Pines Hyperspectral Image data set [11], commonly used in hyperspectral image processing quality assessment, and band 140 of the Enrique Reef Hyperspectral Image data set [12], shown in Fig. 2.…”

Section: Experiments Design a Data And Platformmentioning

confidence: 99%

A Comparative Study on the Parametrization of a Block-based Compressive Sensing Algorithm for Hyperspectral Imaging Applications

Arias¹,

Maytee²,

Sierra³

et al. 2016

Proceedings of the 14th LACCEI International Multi-Conference for Engineering, Education, and Technology: “Engineering Innovati

View full text Add to dashboard Cite

Abstract-Compressive Sensing-based technologies have shown a great potential to improve the efficiency of acquisition, manipu-lation, analysis and storage processes on signals and imagery with little discernible loss in data performance. The CS framework is based on the assumption that signals are sparse in some domain and can be reconstructed from a significantly reduced amount of samples. As a result, a solution to the underdetermined linear system resulting from this paradigm makes it possible to estimate the original signal with high accuracy using linear programming techniques. This paper presents a study on the use of compressive sensing on satellite Hyperspectral Images, which provide a variety of fields and applications with data with a high information density for analysis. Hyperspectral imaging of large areas at high resolutions required for some applications can turn the image capturing, processing and storage processes into a time consuming procedure, presenting a limitation for use in resource-limited or time-sensitive settings. We present an analysis on the algorithm parametrization that may allow for a simpler capturing approach tailored specifically for a given application's needs using the well-studied l1-magic algorithm. We provide a comparative study in compressive sensing and estimate its effectiveness in terms of compression ratio vs. image reconstruction accuracy. Preliminary results show that by using as little as 25% of the original number of samples, large structures may be reconstructed with high accuracy. Abstract-Compressive Sensing-based technologies have shown a great potential to improve the efficiency of acquisition, manipulation, analysis and storage processes on signals and imagery with little discernible loss in data performance. The CS framework is based on the assumption that signals are sparse in some domain and can be reconstructed from a significantly reduced amount of samples. As a result, a solution to the underdetermined linear system resulting from this paradigm makes it possible to estimate the original signal with high accuracy using linear programming techniques. This paper presents a study on the use of compressive sensing on satellite Hyperspectral Images, which provide a variety of fields and applications with data with a high information density for analysis. Hyperspectral imaging of large areas at high resolutions required for some applications can turn the image capturing, processing and storage processes into a time consuming procedure, presenting a limitation for use in resourcelimited or time-sensitive settings. We present an analysis on the algorithm parametrization that may allow for a simpler capturing approach tailored specifically for a given application's needs using the well-studied l1-magic algorithm. We provide a comparative study in compressive sensing and estimate its effectiveness in terms of compression ratio vs. image reconstruction accuracy. Preliminary results show that by using as little as 25% of the original number of samples, large struc...

show abstract

Improving execution time for supervised sparse representation classification of hyperspectral images using the Moore–Penrose pseudoinverse

Arias

Sierra

Arzuaga

2019

J. Appl. Rem. Sens.

View full text Add to dashboard Cite

Ground truth data collection for unmixing algorithm evaluation

Cited by 3 publications

References 0 publications

Understanding the impact of spatial resolution in unmixing of hyperspectral images

Understanding the impact of spatial resolution in unmixing of hyperspectral images

A Comparative Study on the Parametrization of a Block-based Compressive Sensing Algorithm for Hyperspectral Imaging Applications

Improving execution time for supervised sparse representation classification of hyperspectral images using the Moore–Penrose pseudoinverse

Contact Info

Product

Resources

About