Enhancing land cover object classification in hyperspectral imagery through an efficient spectral-spatial feature learning approach
Masud Ibn Afjal, Md. Nazrul Islam Mondal, Md. Al Mamun · 2024
Read Enhancing land cover object classification in hyperspectral imagery through an efficient spectral-spatial feature learning approach by Masud Ibn Afjal, Md. Nazrul Islam Mondal, Md. Al Mamun (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11620357.txt.
Description
Masud Afjal and co-authors improve land cover sorting in hyperspectral images, in a 2024 study. Such images carry hundreds of light bands per pixel, which makes them rich but hard to classify. Training data are scarce, and the sheer number of bands strains standard networks. Flat networks miss links between bands, while full three-dimensional ones grow deep and heavy.
Their method first compresses the bands with a segmented form of principal component analysis. A relevance rule then picks the most useful features. A hybrid network mixing three-dimensional and flat layers reads the joint pattern of space and spectrum. Tests on three standard data sets beat leading methods. Land mappers gain accuracy without a bloated model.
Catalog record
| CREATOR | Masud Ibn Afjal, Md. Nazrul Islam Mondal, Md. Al Mamun |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 62 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |