Abstract
Effective feature extraction (FE) has always been the focus of hyperspectral images (HSIs). For aerial remote-sensing HSIs processing and its land cover classification, in this article, an efficient two-staged hyperspectral FE method based on total variation (TV) is proposed. In the first stage, the average fusion method was used to reduce the spectral dimension. Then, the anisotropic TV model with different regularization parameters was utilized to obtain featured blocks of different smoothness, each containing multi-scale structure information, and we stacked them as the next stage’s input. In the second stage, equipped with singular value transformation to reduce the dimension again, we followed an isotropic TV model based on split Bregman algorithm for further detail smoothing. Finally, the feature-extracted block was fed to the support vector machine for classification experiments. The results, with three hyperspectral datasets, demonstrate that our proposed method can competitively outperform state-of-the-art methods in terms of its classification accuracy and computing time. Also, our proposed method delivers robustness and stability by comprehensive parameter analysis.
Highlights
Hyperspectral imaging technology is based on multi-spectral imaging, in the spectral range from ultraviolet to near-infrared, using an imaging spectrometer to continuously scan within tens or hundreds of spectral bands of the scenes
Random patches network (RPNet) directly takes the random patch obtained from the image as the convolution kernel without any training, but when the samples are few in number, the random patch pattern seems to introduce more redundant information, and the results show obvious mixed noise in the classification map
Based different poses an efficient two-staged feature extraction (FE) method based on total variation (TV) for hyperspectral images (HSIs)
Summary
Hyperspectral imaging technology is based on multi-spectral imaging, in the spectral range from ultraviolet to near-infrared, using an imaging spectrometer to continuously scan within tens or hundreds of spectral bands of the scenes. Hyperspectral images (HSIs) capture spatial features and obtains rich spectral information from each pixel, which can achieve the classification and recognition of the target objects more efficiently than traditional images. HSIs can be obtained by active remote sensing technology [5], which usually utilizes wide spectral light sources [6] to replace the sun to illuminate the scenes and which play a significant role in object detection [7] and recognition [8]. HSI classification has always been a hot topic of application among these fields. It can provide high-level intuitive judgment and interpretation, especially for land use and analysis
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
