An automatic target detection algorithm for hyperspectral imagery based on feature-level fusion

2005 ◽  
Author(s):  
Lin He ◽  
Quan Pan ◽  
Yongqiang Zhao ◽  
Wei Di
Keyword(s):  
Target Detection ◽  
Hyperspectral Imagery ◽  
Detection Algorithm ◽  
Automatic Target Detection ◽  
Feature Level Fusion ◽  
Level Fusion

Decision-level fusion of SAR and IR sensor information for automatic target detection

2017 ◽  
Author(s):  
Young-Rae Cho ◽  
Sung-Hyuk Yim ◽  
Hyun-Woong Cho ◽  
Jin-Ju Won ◽  
Woo-Jin Song ◽  
...  
Keyword(s):  
Target Detection ◽  
Automatic Target Detection ◽  
Decision Level ◽  
Ir Sensor ◽  
Sensor Information ◽  
Decision Level Fusion ◽  
Level Fusion

scholarly journals Anomaly targets detection of hyperspectral imagery based on wavelet transform and sparse representation

2018 ◽  
Vol 232 ◽  
pp. 02054
Author(s):  
Cheng Baozhi
Keyword(s):  
Wavelet Transform ◽  
Sparse Representation ◽  
Target Detection ◽  
Hyperspectral Image ◽  
Hyperspectral Imagery ◽  
Image Data ◽  
Detection Algorithm ◽  
Discrete Wavelet ◽  
Detection Accuracy ◽  
Low Efficiency

The research of anomaly target detection algorithm in hyperspectral imagery is a hot issue, which has important research value. In order to overcome low efficiency of current anomaly target detection in hyperspectral image, an anomaly detection algorithm for hyperspectral images based on wavelet transform and sparse representation was proposed. Firstly, two-dimensional discrete wavelet transform is used to denoise the hyperspectral image, and the new hyperspectral image data are obtained. Then, the results of anomaly target detection are obtained by using sparse representation theory. The real AVIRIS hyperspectral imagery data sets are used in the experiments. The results show that the detection accuracy and false alarm rate of the propoesd algorithm are better than RX and KRX algorithm.

scholarly journals Mapping the Extent of Invasive Phragmites australis subsp. australis From Airborne Hyperspectral Imagery

2021 ◽  
Vol 9 ◽  
Author(s):  
Kathryn Elmer ◽  
Margaret Kalacska ◽  
J. Pablo Arroyo-Mora
Keyword(s):  
Remote Sensing ◽  
Invasive Species ◽  
Early Detection ◽  
Target Detection ◽  
Phragmites Australis ◽  
Hyperspectral Imagery ◽  
Ground Truth ◽  
Detection Algorithm ◽  
Visual Interpretation ◽  
Ground Truth Data

Invasive species pose one of the greatest threats to global biodiversity. Early detection of invasive species is critical in order to prevent or manage their spread before they exceed the ability of land management groups to control them. Optical remote sensing has been established as a useful technology for the early detection and mapping of invasive vegetation populations. Through the use of airborne hyperspectral imagery (HSI), this study establishes a target detection methodology used to identify and map the invasive reed Phragmites australis subsp. australis within the entire extent of Îles-de-Boucherville National Park (Quebec, ON, Canada). We applied the Spectral Angle Mapper (SAM) target detection algorithm trained with a high accuracy GNSS ground truth data set to produce a park-wide map illustrating the extent of detected Phragmites. The total coverage of detected Phragmites was 26.74 ha (0.267 km2), which represents 3.28% of the total park area of 814 ha (8.14 km2). The inherent spatial uncertainty of the airborne HSI (∼2.25 m) was accounted for with uncertainty buffers, which, when included in the measurement of detected Phragmites, lead to a total area of 59.17 ha (0.591 km2), or 7.26% of the park. The overall accuracy of the Phragmites map was 84.28%, with a sensitivity of 76.32% and a specificity of 91.57%. Additionally, visual interpretation of the validation ground truth dataset was performed by 10 individuals, in order to compare their performance to that of the target detection algorithm. The overall accuracy of the visual interpretation was lower than the target detection (i.e., 69.18%, with a sensitivity of 59.21% and a specificity of 78.31%). Overall, this study is one of the first to utilize airborne HSI and target detection to map the extent of Phragmites over a moderately large extent. The uses and limitations of such an approach are established, and the methodology described here in detail could be adapted for future remote sensing studies of Phragmites or other vegetation species, native or invasive, at study sites around the world.

Sign in / Sign up

Export Citation Format

Share Document