Testing different classification methods in airborne hyperspectral imagery processing

Kozoderov, V. V.; Дмитриев, Е. В.

doi:10.1364/oe.24.00a956

Cited by 11 publications

(4 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Studies in the literature (e.g., [68][69][70]) indicate that the texture information of VHR airborne optical images can represent a proxy to describe forest structure variables (e.g., crown diameter, tree height, tree density or spacing). For these reasons we applied a Support Vector Machine (SVM) classification method [71] to our hyperspectral images, using as input the BRDF cover index [7], the three most prominent components of the hyperspectral reflectance images obtained with the Principal Components Analysis method [72], and the Gray Level Co-occurrence Matrix (GLCM) texture features [73]. The GLCM texture features (i.e., mean, variance, homogeneity, contrast, dissimilarity, entropy, second moment, and correlation; [74]) are extracted based on the first principal component data with a window size of 9 × 9 pixels.…”

Section: Land-cover Stratification (Pre-classification)mentioning

confidence: 99%

“…The GLCM texture features (i.e., mean, variance, homogeneity, contrast, dissimilarity, entropy, second moment, and correlation; [74]) are extracted based on the first principal component data with a window size of 9 × 9 pixels. In this study, coniferous forest, broadleaved forest, bare soil, urban, and unclassified were classified, with most unclassified pixels in our dataset represented by the cast-shaded areas in the deep gullies and canopy gaps [71].…”

Section: Land-cover Stratification (Pre-classification)mentioning

confidence: 99%

See 1 more Smart Citation

A Kernel-Driven BRDF Approach to Correct Airborne Hyperspectral Imagery over Forested Areas with Rugged Topography

et al. 2020

View full text Add to dashboard Cite

Airborne hyper-spectral imaging has been proven to be an efficient means to provide new insights for the retrieval of biophysical variables. However, quantitative estimates of unbiased information derived from airborne hyperspectral measurements primarily require a correction of the anisotropic scattering properties of the land surface depicted by the bidirectional reflectance distribution function (BRDF). Hitherto, angular BRDF correction methods rarely combined viewing-illumination geometry and topographic information to achieve a comprehensive understanding and quantification of the BRDF effects. This is in particular the case for forested areas, frequently underlaid by rugged topography. This paper describes a method to correct the BRDF effects of airborne hyperspectral imagery over forested areas overlying rugged topography, referred in the reminder of the paper as rugged topography-BRDF (RT-BRDF) correction. The local viewing and illumination geometry are calculated for each pixel based on the characteristics of the airborne scanner and the local topography, and these two variables are used to adapt the Ross-Thick-Maignan and Li-Transit-Reciprocal kernels in the case of rugged topography. The new BRDF model is fitted to the anisotropy of multi-line airborne hyperspectral data. The number of pixels is set at 35,000 in this study, based on a stratified random sampling method to ensure a comprehensive coverage of the viewing and illumination angles and to minimize the fitting error of the BRDF model for all bands. Based on multi-line airborne hyperspectral data acquired with the Chinese Academy of Forestry’s LiDAR, CCD, and Hyperspectral system (CAF-LiCHy) in the Pu’er region (China), the results applying the RT-BRDF correction are compared with results from current empirical (C, and sun-canopy-sensor (SCS) adds C (SCS+C)) and semi-physical (SCS) topographic correction methods. Both quantitative assessment and visual inspection indicate that RT-BRDF, C, and SCS+C correction methods all reduce the topographic effects. However, the RT-BRDF method appears more efficient in reducing the variability in reflectance of overlapping areas in multiple flight-lines, with the advantage of reducing the BRDF effects caused by the combination of wide field of view (FOV) airborne scanner, rugged topography, and varying solar illumination angle over long flight time. Specifically, the average decrease in coefficient of variation (CV) is 3% and 3.5% for coniferous forest and broadleaved forest, respectively. This improvement is particularly marked in the near infrared (NIR) region (i.e., >750 nm). This finding opens new possible applications of airborne hyperspectral surveys over large areas.

show abstract

Section: Land-cover Stratification (Pre-classification)mentioning

confidence: 99%

Section: Land-cover Stratification (Pre-classification)mentioning

confidence: 99%

A Kernel-Driven BRDF Approach to Correct Airborne Hyperspectral Imagery over Forested Areas with Rugged Topography

et al. 2020

View full text Add to dashboard Cite

show abstract

“…The maximum likelihood (MLC) algorithm is a classical classification algorithm for remotely sensed images, and the support vector machine (SVM) method has been a popular and effective classification algorithm in recent years [ 22 ]. We compare the classifier based on MLC and SVM algorithm with the classifier based on ISBDD algorithm.…”

Section: Methodsmentioning

confidence: 99%

ISBDD Model for Classification of Hyperspectral Remote Sensing Imagery

Zhao

et al. 2018

Sensors

View full text Add to dashboard Cite

The diverse density (DD) algorithm was proposed to handle the problem of low classification accuracy when training samples contain interference such as mixed pixels. The DD algorithm can learn a feature vector from training bags, which comprise instances (pixels). However, the feature vector learned by the DD algorithm cannot always effectively represent one type of ground cover. To handle this problem, an instance space-based diverse density (ISBDD) model that employs a novel training strategy is proposed in this paper. In the ISBDD model, DD values of each pixel are computed instead of learning a feature vector, and as a result, the pixel can be classified according to its DD values. Airborne hyperspectral data collected by the Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) sensor and the Push-broom Hyperspectral Imager (PHI) are applied to evaluate the performance of the proposed model. Results show that the overall classification accuracy of ISBDD model on the AVIRIS and PHI images is up to 97.65% and 89.02%, respectively, while the kappa coefficient is up to 0.97 and 0.88, respectively.

show abstract

“…Dalponte et al used hyperspectral data to classify northern forest tree species and used three classifiers (e.g., support vector machine, random forest, and maximum likelihood method) to classify tree species, proving that hyperspectral data are very effective in classifying the forest types of northern forest tree species [53]. Kozoderov and Dmitriev used different classifiers to identify tree species in hyperspectral images and found nonlinear classifiers were more suitable for hyperspectral data classification [54]. Schull et al used AVIRIS (Airborne Visible InfraRed Imaging Spectrometer) RS images to identify tree species using characteristic factors related to spectral information and crown structure and found these factors can improve the accuracy of tree species classification [55].…”

Section: Research Advances Of Plant Hyperspectral Rs Technologymentioning

confidence: 99%

Design and Application of Bionic Camouflage Materials Simulating Spectral Reflection Characteristics of Plants: A Review

Lin,

Ren,

Yang

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

Hyperspectral remote sensing (RS) has rapidly developed in recent years and has been widely used in the military field. This technology not only brings huge opportunities for military reconnaissance but also poses unprecedented challenges to military camouflage, severely complicating the development of plant hyperspectral camouflage materials and technology. In this review, the spectral reflectance characteristics of plants and the application of hyperspectral RS in plant RS and military operations are reviewed. The development status of bionic camouflage materials that simulate the spectral reflection characteristics of plants is analyzed. With the existing hyperspectral camouflage materials and technology, bionic camouflage technology is limited by the inability of bionic materials to accurately imitate the characteristic absorption peaks of green vegetation, low stability and durability, and the large overall material thickness, which complicate actual large-scale application. On this basis, a future development direction and a trend of plant hyperspectral bionic camouflage materials and technology are proposed.

show abstract

Testing different classification methods in airborne hyperspectral imagery processing

Cited by 11 publications

References 15 publications

A Kernel-Driven BRDF Approach to Correct Airborne Hyperspectral Imagery over Forested Areas with Rugged Topography

A Kernel-Driven BRDF Approach to Correct Airborne Hyperspectral Imagery over Forested Areas with Rugged Topography

ISBDD Model for Classification of Hyperspectral Remote Sensing Imagery

Design and Application of Bionic Camouflage Materials Simulating Spectral Reflection Characteristics of Plants: A Review

Contact Info

Product

Resources

About