The effectiveness and limitations of geometric and statistical spectral unmixing techniques for subpixel target detection in hyperspectral remote sensing

2005 ◽  
Author(s):  
Peter WT Yuen ◽  
A. Blagg ◽  
G. Bishop
Keyword(s):  
Remote Sensing ◽  
Target Detection ◽  
Hyperspectral Remote Sensing ◽  
Spectral Unmixing

scholarly journals Spectral unmixing approach in hyperspectral remote sensing: a tool for oil palm mapping

TecnoLógicas ◽  
2019 ◽  
Vol 22 (45) ◽  
pp. 129-143
Author(s):  
Hector Vargas ◽  
Ariolfo Camacho Velasco ◽  
Henry Arguello
Keyword(s):  
Remote Sensing ◽  
Oil Palm ◽  
Hyperspectral Image ◽  
Remote Sensing Data ◽  
Hyperspectral Remote Sensing ◽  
Spectral Unmixing ◽  
Spectral Variability ◽  
Computational Framework ◽  
Sensing Data ◽  
Spectral Bands

Oil palm plantations typically span large areas; therefore, remote sensing has become a useful tool for advanced oil palm monitoring. This work reviews and evaluates two approaches to analyze oil palm plantations based on hyperspectral remote sensing data: linear spectral unmixing and spectral variability. Moreover, a computational framework based on spectral unmixing for the estimation of fractional abundances of oil palm plantations is proposed in this study. Such approach also considers the spectral variability of hyperspectral image signatures. More specifically, the proposed computational framework modifies the linear mixing model by introducing a weighting vector, so that the spectral bands that contribute the least to the estimation of erroneous fractional abundances can be identified. This approach improves palm detection as it allows to differentiate them from other materials in terms of fractional abundances. Experimental results obtained from hyperspectral remote sensing data in the range 410-990 nm show improvements of 8.18 % in User Accuracy (Uacc) in the identification of oil palms by the proposed framework with respect to traditional unmixing methods. Thus, the proposed method achieved a 95% Uacc. This confirms the capabilities of the proposed computational framework and facilitates the management and monitoring of large areas of oil palm plantations.

scholarly journals Joint Local Block Grouping with Noise-Adjusted Principal Component Analysis for Hyperspectral Remote-Sensing Imagery Sparse Unmixing

2019 ◽  
Vol 11 (10) ◽  
pp. 1223 ◽  
Author(s):  
Ruyi Feng ◽  
Lizhe Wang ◽  
Yanfei Zhong
Keyword(s):  
Remote Sensing ◽  
Principal Component Analysis ◽  
Spatial Information ◽  
Principal Component ◽  
Component Analysis ◽  
Hyperspectral Remote Sensing ◽  
Spectral Unmixing ◽  
Spatial Correlations ◽  
Local Block ◽  
Unmixing Model

Spatial regularized sparse unmixing has been proved as an effective spectral unmixing technique, combining spatial information and standard spectral signatures known in advance into the traditional spectral unmixing model in the form of sparse regression. In a spatial regularized sparse unmixing model, spatial consideration acts as an important role and develops from local neighborhood pixels to global structures. However, incorporating spatial relationships will increase the computational complexity, and it is inevitable that some negative influences obtained by inaccurate estimated abundances’ spatial correlations will reduce the accuracy of the algorithms. To obtain a more reliable and efficient spatial regularized sparse unmixing results, a joint local block grouping with noise-adjusted principal component analysis for hyperspectral remote-sensing imagery sparse unmixing is proposed in this paper. In this work, local block grouping is first utilized to gather and classify abundant spatial information in local blocks, and noise-adjusted principal component analysis is used to compress these series of classified local blocks and select the most significant ones. Then the representative spatial correlations are drawn and replace the traditional spatial regularization in the spatial regularized sparse unmixing method. Compared with total variation-based and non-local means-based sparse unmixing algorithms, the proposed approach can yield comparable experimental results with three simulated hyperspectral data cubes and two real hyperspectral remote-sensing images.

Sign in / Sign up

Export Citation Format

Share Document