New Hyperspectral Endmember Extraction Algorithms using Convex Geometry (Record no. 145461)
[ view plain ]
| 000 -LEADER | |
|---|---|
| fixed length control field | 04575nam a22001577a 4500 |
| 008 - FIXED-LENGTH DATA ELEMENTS--GENERAL INFORMATION | |
| fixed length control field | 230307b |||||||| |||| 00| 0 eng d |
| 082 ## - DEWEY DECIMAL CLASSIFICATION NUMBER | |
| Classification number | TT000115 |
| Item number | SHA |
| 100 ## - MAIN ENTRY--PERSONAL NAME | |
| Personal name | Shah, Dharambhai Jayeshkumar |
| 245 ## - TITLE STATEMENT | |
| Title | New Hyperspectral Endmember Extraction Algorithms using Convex Geometry |
| Statement of responsibility, etc | by Dharambhai Jayeshkumar Shah |
| 260 ## - PUBLICATION, DISTRIBUTION, ETC. (IMPRINT) | |
| Place of publication, distribution, etc | Ahmedabad |
| Name of publisher, distributor, etc | Nirma Institute of Technology |
| Date of publication, distribution, etc | July 2021 |
| 300 ## - PHYSICAL DESCRIPTION | |
| Extent | 135p Thesis with Synopsis |
| 500 ## - GENERAL NOTE | |
| General note | Guided by: Dr. Y. N. Trivedi<br/>16FTVPHDE18<br/><br/>ABSTRACT:<br/>The decomposition of the mixed pixels into individual pure material (endmember) along with its<br/>proposition is called spectral unfixing for hyperspectral images. Spectral unfixing is considered<br/>a three-stage problem for the hyperspectral image. The first is the subspace dimension which finds<br/>the number of pure materials in the image. The second one is endmember extraction which extracts<br/>the pure material spectra from the image and the third one is abundance estimation which estimates<br/>the proportions of each material in mixing. The endmember extraction is a very challenging stage<br/>in spectral immixing as abundance mapping greatly depends on extracted endmembers. In the<br/>literature, endmember extraction is addressed using a geometrical, statistical, sparse regression,<br/>and deep learning approach. Due to simplicity and easy understanding, many researchers use the<br/>geometrical approach. In our research work, we focus on the geometrical endmember extraction<br/>approach which majorly uses the concept of convex geometry. In our work, we have developed<br/>new eight algorithms that improve the endmember extraction accuracy and abundance estimation<br/>accuracy. The first new algorithm explores entropy-based spatial information with convex set<br/>optimization-based spectral information. The second algorithm uses K-medoids clustering with<br/>convex geometry. The K-medoids clustering is used for removing redundant points that make us<br/>second algorithm as noise-robust The third algorithm uses the area maximization approach instead<br/>of conventional volume maximization. Surveyor’s formula is used for finding the area of a convex<br/>polygon in this third algorithm. The fourth algorithm uses the Rank correlation coefficient to find<br/>only effective bands for applying convex geometry. This fourth algorithm used Pearson’s correlation<br/>coefficient. The fifth algorithm combines the geometrical features with statistical features. The<br/>convex geometry is used as a geometrical feature and covariance of the band is used as a statistical<br/>feature. The sixth algorithm uses the quality bands only for applying convex geometry. The high-quality<br/>bands are selected before applying the convex geometry. The seventh proposed algorithm<br/>uses an ensemble of all Winter’s belief-based algorithms for improving accuracy. The majority<br/>voting-based ensemble is used to combine the performance of each Winter’s belief-based algorithm.<br/>The eighth algorithm uses an integration framework of maximum simplex volume and extreme<br/>projection on a subspace which are two major criteria of geometrical types of approaches. This<br/>algorithm uses the newly defined score that is based on the convexity-based purity concept. All<br/>the extracted endmembers of the proposed algorithms have been compared with the extracted<br/>i endmembers of the prevailing algorithms on benchmark real and synthetic datasets using standard<br/>evaluation parameters such as Spectral Angle Mapper (SAM), Spectral Information Divergence<br/>(SID), and Normalized Cross-Correlation (NXC). The Root Mean Square Error (RMSE) is used<br/>to test the efficacy of the extracted endmember for abundance mapping. The RMSE error is<br/>calculated between FCLS-based abundance maps by the endmember of the proposed algorithm<br/>and FCLS-based abundance maps by the Ground Truth. We have used Cuprite, Urban, Jasper,<br/>Samson, Mangalore, and Ahmedabad as a real dataset. We have used the Hyperspectral Imagery<br/>Synthesis Toolbox (HIST) for generating five types of synthetic images. The synthetic images are<br/>added with Gaussian noise to test the noise robustness. The proposed algorithms in this thesis<br/>can be used for a variety of hyperspectral applications, including classification, target detection,<br/>and many others. We have also compared our eight proposed algorithms with the benchmark<br/>geometrical algorithms. |
| 856 ## - ELECTRONIC LOCATION AND ACCESS | |
| Uniform Resource Identifier | https://repository.nirmauni.ac.in/jspui/handle/123456789/11434 |
| Public note | Institute Repository (Campus Access) |
| 856 ## - ELECTRONIC LOCATION AND ACCESS | |
| Uniform Resource Identifier | http://shodhganga.inflibnet.ac.in/handle/10603/384323 |
| Public note | Shodhganga |
| 942 ## - ADDED ENTRY ELEMENTS (KOHA) | |
| Source of classification or shelving scheme | Dewey Decimal Classification |
| Koha item type | Thesis |
No items available.