Hyperspectral image semi-supervised classification method based on space-spectral information
A hyperspectral image and classification method technology, applied in the field of pattern recognition, image processing, and semi-supervised classification using hyperspectral image spectral information and spatial information, which can solve the problem of ignoring spatial dimension information and spectral information to improve classification accuracy. And other issues
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
specific Embodiment
[0104] The experimental data was taken in 1992 by AVIRIS at the Indian Pines experimental base in northwest Indiana, USA. The data contains images of 145×145 size in 224 bands with a spatial resolution of 20 m and a wavelength range of 0.4-2.5 μm (water absorption bands 104-108, 150-163, and 220 were removed before the experiment, and 200 bands were actually used). It contains farmland, forest, and a variety of other native vegetation.
[0105] In this embodiment, the specific parameters set by the method are:
[0106] Randomly select 25% of the total samples as the test sample set T, 75% as the training sample set, select N=9 categories for classification, use numbers 1 to 9 to represent different categories, and select s= for each category from the training sample set 15 samples, a total of n=135 labeled samples constitute the initial training sample set D l ={(x l1 ,y l1 ),…,(x l135 ,y l135 )}, the remaining samples are the unlabeled sample set U; set the maximum freq...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com