• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.7
• /
• pp.3543-3557
• /
• 2017

Compression is a very important technique for remotely sensed hyperspectral images. The lossless compression based on the recursive least square (RLS), which eliminates hyperspectral images' redundancy using both spatial and spectral correlations, is an extremely powerful tool for this purpose, but the relatively high computational complexity limits its application to time-critical scenarios. In order to improve the computational efficiency of the algorithm, we optimize its serial version and develop a new parallel implementation on graphics processing units (GPUs). Namely, an optimized recursive least square based on optimal number of prediction bands is introduced firstly. Then we use this approach as a case study to illustrate the advantages and potential challenges of applying GPU parallel optimization principles to the considered problem. The proposed parallel method properly exploits the low-level architecture of GPUs and has been carried out using the compute unified device architecture (CUDA). The GPU parallel implementation is compared with the serial implementation on CPU. Experimental results indicate remarkable acceleration factors and real-time performance, while retaining exactly the same bit rate with regard to the serial version of the compressor.

• Korean Journal of Remote Sensing
• /
• v.25 no.3
• /
• pp.263-270
• /
• 2009

Noise band removal is a crucial step before spectral matching since the noise bands can distort the typical shape of spectral reflectance, leading to degradation on the matching results. This paper proposes a statistical noise band removal method for hyperspectral data using the correlation coefficient between two bands. The correlation coefficient measures the strength and direction of a linear relationship between two random variables. Considering each band of the hyperspectral data as a random variable, the correlation between two signal bands is high; existence of a noisy band will produce a low correlation due to ill-correlativeness and undirected ness. The unsupervised k-nearest neighbor clustering method is implemented in accordance with three well-accepted spectral matching measures, namely ED, SAM and SID in order to evaluate the validation of the proposed method. This paper also proposes a hierarchical scheme of combining those measures. Finally, a separability assessment based on the between-class and the within-class scatter matrices is followed to evaluate the applicability of the proposed noise band removal method. Also, the paper brings out a comparison for spectral matching measures. The experimental results conducted on a 228-band hyperspectral data show that while the SAM measure is rather resistant, the performance of SID measure is more sensitive to noise.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.18 no.1
• /
• pp.48-63
• /
• 2015

In this study, we introduce various hyperspectral data processing techniques for the monitoring of shallow and coastal waters to enlarge the application range and to improve the accuracy of the end results in Korea. Unlike land, more accurate atmospheric correction is needed in coastal region showing relatively low reflectance in visible wavelengths. Sun-glint which occurs due to a geometry of sun-sea surface-sensor is another issue for the data processing in the ocean application of hyperspectal imagery. After the preprocessing of the hyperspectral data, a semi-analytical algorithm based on a radiative transfer model and a spectral library can be used for bathymetry mapping in coastal area, type classification and status monitoring of benthos or substrate classification. In general, semi-analytical algorithms using spectral information obtained from hyperspectral imagey shows higher accuracy than an empirical method using multispectral data. The water depth and quality are constraint factors in the ocean application of optical data. Although a radiative transfer model suggests the theoretical limit of about 25m in depth for bathymetry and bottom classification, hyperspectral data have been used practically at depths of up to 10 m in shallow and coastal waters. It means we have to focus on the maximum depth of water and water quality conditions that affect the coastal applicability of hyperspectral data, and to define the spectral library of coastal waters to classify the types of benthos and substrates.

• Korean Journal of Remote Sensing
• /
• v.22 no.4
• /
• pp.295-304
• /
• 2006

Image fusion is defined as making new image by merging two or more images using special algorithms. In case of remote sensing, it means fusing multispectral low-resolution remotely sensed image with panchromatic high-resolution image. Generally, hyperspectral image fusion is accomplished by utilizing fusion technique of multispectral imagery or spectral unmixing model. But, the former may distort spectral information and the latter needs endmember data or additional data, and has a problem with not preserving spatial information well. This study proposes a new algorithm based on two stage spectral unmixing model for preserving hyperspectral image's spectral information. The proposed fusion technique is implemented and tested using Hyperion and ALI images. it is shown to work well on maintaining more spatial/spectral information than the PCA/GS fusion algorithms.

• International Journal of Computer Science & Network Security
• /
• v.22 no.4
• /
• pp.193-202
• /
• 2022

The classification of hyperspectral imagery (HSI) is essential in the surface of earth observation. Due to the continuous large number of bands, HSI data provide rich information about the object of study; however, it suffers from the curse of dimensionality. Dimensionality reduction is an essential aspect of Machine learning classification. The algorithms based on feature extraction can overcome the data dimensionality issue, thereby allowing the classifiers to utilize comprehensive models to reduce computational costs. This paper assesses and compares two HSI classification techniques. The first is based on the Joint Spatial-Spectral Stacked Autoencoder (JSSSA) method, the second is based on a shallow Artificial Neural Network (SNN), and the third is used the SVM model. The performance of the JSSSA technique is better than the SNN classification technique based on the overall accuracy and Kappa coefficient values. We observed that the JSSSA based method surpasses the SNN technique with an overall accuracy of 96.13% and Kappa coefficient value of 0.95. SNN also achieved a good accuracy of 92.40% and a Kappa coefficient value of 0.90, and SVM achieved an accuracy of 82.87%. The current study suggests that both JSSSA and SNN based techniques prove to be efficient methods for hyperspectral classification of snow features. This work classified the labeled/ground-truth datasets of snow in multiple classes. The labeled/ground-truth data can be valuable for applying deep neural networks such as CNN, hybrid CNN, RNN for glaciology, and snow-related hazard applications.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.8
• /
• pp.3295-3311
• /
• 2020

With the wide application of hyperspectral images, it becomes more and more important to compress hyperspectral images. Conventional recursive least squares (CRLS) algorithm has great potentiality in lossless compression for hyperspectral images. The prediction accuracy of CRLS is closely related to the correlations between the reference bands and the current band, and the similarity between pixels in prediction context. According to this characteristic, we present an improved CRLS with adaptive band selection and adaptive predictor selection (CRLS-ABS-APS). Firstly, a spectral vector correlation coefficient-based k-means clustering algorithm is employed to generate clustering map. Afterwards, an adaptive band selection strategy based on inter-spectral correlation coefficient is adopted to select the reference bands for each band. Then, an adaptive predictor selection strategy based on clustering map is adopted to select the optimal CRLS predictor for each pixel. In addition, a double snake scan mode is used to further improve the similarity of prediction context, and a recursive average estimation method is used to accelerate the local average calculation. Finally, the prediction residuals are entropy encoded by arithmetic encoder. Experiments on the Airborne Visible Infrared Imaging Spectrometer (AVIRIS) 2006 data set show that the CRLS-ABS-APS achieves average bit rates of 3.28 bpp, 5.55 bpp and 2.39 bpp on the three subsets, respectively. The results indicate that the CRLS-ABS-APS effectively improves the compression effect with lower computation complexity, and outperforms to the current state-of-the-art methods.

• Korean Journal of Remote Sensing
• /
• v.35 no.6_1
• /
• pp.959-971
• /
• 2019

CNN (Convolutional Neural Network) is one representative deep learning algorithm, which can extract high-level spatial and spectral features, and has been applied for hyperspectral image classification. However, one significant drawback behind the application of CNNs in hyperspectral images is the high dimensionality of the data, which increases the training time and processing complexity. To address this problem, several CNN based hyperspectral image classification studies have exploited PCA (Principal Component Analysis) for dimensionality reduction. One limitation to this is that the spectral information of the original image can be lost through PCA. Although it is clear that the use of PCA affects the accuracy and the CNN training time, the impact of PCA for CNN based hyperspectral image classification has been understudied. The purpose of this study is to analyze the quantitative effect of PCA in CNN for hyperspectral image classification. The hyperspectral images were first transformed through PCA and applied into the CNN model by varying the size of the reduced dimensionality. In addition, 2D-CNN and 3D-CNN frameworks were applied to analyze the sensitivity of the PCA with respect to the convolution kernel in the model. Experimental results were evaluated based on classification accuracy, learning time, variance ratio, and training process. The size of the reduced dimensionality was the most efficient when the explained variance ratio recorded 99.7%~99.8%. Since the 3D kernel had higher classification accuracy in the original-CNN than the PCA-CNN in comparison to the 2D-CNN, the results revealed that the dimensionality reduction was relatively less effective in 3D kernel.

• Current Optics and Photonics
• /
• v.5 no.3
• /
• pp.270-277
• /
• 2021

The adaptive sparse representation (ASR) can effectively combine the structure information of a sample dictionary and the sparsity of coding coefficients. This algorithm can effectively consider the correlation between training samples and convert between sparse representation-based classifier (SRC) and collaborative representation classification (CRC) under different training samples. Unlike SRC and CRC which use fixed norm constraints, ASR can adaptively adjust the constraints based on the correlation between different training samples, seeking a balance between l₁ and l₂ norm, greatly strengthening the robustness and adaptability of the classification algorithm. The correlation coefficients (CC) can better identify the pixels with strong correlation. Therefore, this article proposes a hyperspectral image classification method called correlation coefficients and adaptive sparse representation (CCASR), based on ASR and CC. This method is divided into three steps. In the first step, we determine the pixel to be measured and calculate the CC value between the pixel to be tested and various training samples. Then we represent the pixel using ASR and calculate the reconstruction error corresponding to each category. Finally, the target pixels are classified according to the reconstruction error and the CC value. In this article, a new hyperspectral image classification method is proposed by fusing CC and ASR. The method in this paper is verified through two sets of experimental data. In the hyperspectral image (Indian Pines), the overall accuracy of CCASR has reached 0.9596. In the hyperspectral images taken by HIS-300, the classification results show that the classification accuracy of the proposed method achieves 0.9354, which is better than other commonly used methods.

• Journal of Software Assessment and Valuation
• /
• v.17 no.2
• /
• pp.75-81
• /
• 2021

Since plant growth is greatly influenced by moisture, it is important to control the soil to have optimal moisture for the plant being grown. Recently, researches on automatically analyzing plant growth information including soil moisture using spectral images are being conducted. However, hyperspectral images are difficult to use due to huge amount of data appearing in spectral bands. In this paper, we propose a method to solve the complexity of hyperspectral images using a CNN. Since the proposed method automatically analyzes the entire band of the target hyperspectral using deep learning, there is no need to make an effort to find a specific band for analysis of each image. In order to show the effectiveness of the proposed system, we conduct an experiment to analyze moistures using hyperspectral images obtained from soil.

• Journal of Korean Society for Geospatial Information Science
• /
• v.25 no.1
• /
• pp.9-17
• /
• 2017

Hyperspectral imagery is used in the land cover classification with the principle component analysis and minimum noise fraction to reduce the data dimensionality and noise. Recently, studies on the supervised classification using various features having spectral information and spatial characteristic have been carried out. In this study, principle component bands and normalized difference vegetation index(NDVI) was utilized in the supervised classification for the land cover classification. To utilize additional information not included in the principle component bands by the hyperspectral imagery, we tried to increase the classification accuracy by using the NDVI. In addition, the extended attribute profiles(EAP) generated using the morphological filter was used as the input data. The random forest algorithm, which is one of the representative supervised classification, was used. The classification accuracy according to the application of various features based on EAP was compared. Two areas was selected in the experiments, and the quantitative evaluation was performed by using reference data. The classification accuracy of the proposed algorithm showed the highest classification accuracy of 85.72% and 91.14% compared with existing algorithms. Further research will need to develop a supervised classification algorithm and additional input datasets to improve the accuracy of land cover classification using hyperspectral imagery.