-
Tytuł:
-
Multimedia Image Data Analysis Based on KNN Algorithm.
-
Autorzy:
-
Li R; Research Institute of Finance, Hebei Finance University, Baoding, Hebei 071051, China.
Li S; School of Management, Hebei Finance University, Baoding, Hebei Province 071051, China.
-
Źródło:
-
Computational intelligence and neuroscience [Comput Intell Neurosci] 2022 Apr 12; Vol. 2022, pp. 7963603. Date of Electronic Publication: 2022 Apr 12 (Print Publication: 2022).
-
Typ publikacji:
-
Journal Article
-
Język:
-
English
-
Imprint Name(s):
-
Original Publication: New York, NY : Hindawi Pub. Corp.
-
MeSH Terms:
-
Data Analysis*
Multimedia*
Algorithms ; Image Processing, Computer-Assisted ; Remote Sensing Technology
-
References:
-
IEEE/ACM Trans Comput Biol Bioinform. 2018 Jan-Feb;15(1):38-45. (PMID: 27740494)
Guang Pu Xue Yu Guang Pu Fen Xi. 2016 Dec;36(12):4067-71. (PMID: 30256561)
Springerplus. 2016 Aug 22;5(1):1389. (PMID: 27610308)
Math Biosci Eng. 2020 Jun 23;17(5):4443-4456. (PMID: 33120512)
Guang Pu Xue Yu Guang Pu Fen Xi. 2016 Jul;36(7):2234-7. (PMID: 30035996)
-
Entry Date(s):
-
Date Created: 20220422 Date Completed: 20220425 Latest Revision: 20220716
-
Update Code:
-
20240104
-
PubMed Central ID:
-
PMC9018202
-
DOI:
-
10.1155/2022/7963603
-
PMID:
-
35449749
-
In order to improve the authenticity of multispectral remote sensing image data analysis, the KNN algorithm and hyperspectral remote sensing technology are used to organically combine advanced multimedia technology with spectral technology to subdivide the spectrum. Different classification methods are used to classify CHRIS 0°, and the results are analyzed and compared: SVM classification accuracy is the highest 72 8448%, Kappa coefficient is 0.6770, and SVM is used to classify CHRIS images from five angles, and the results are compared and analyzed: the classification accuracy is from high to low, and the order is FZA = 0 > FZA = -36 > FZA = -55 > FZA = 36 > FZA = 55; SVM is used to classify the multiangle combined image, and the result is compared with the CHRIS 0° result: the overall classification accuracy of angle-combined image types is lower than that of single-angle images; the SVM is used to classify the band-combined image, and the result is compared with CHRIS 0°: the overall classification accuracy of the band combination image forest type is very low, and the effect is not as good as the combining multiangle image classification results. It is verified that if CHRIS multiangle hyper-spectral data are used for classification, the SVM method should be used to classify spectral remote sensing image data with the best effect.
Competing Interests: The authors declare that they have no conflicts of interest.
(Copyright © 2022 Runya Li and Shenglian Li.)
Zaloguj się, aby uzyskać dostęp do pełnego tekstu.