• Journal of information and communication convergence engineering
• /
• v.21 no.3
• /
• pp.208-215
• /
• 2023

Deep learning techniques provide powerful solutions to several pattern-recognition problems, including Raman spectral classification. However, these networks require large amounts of labeled data to perform well. Labeled data, which are typically obtained in a laboratory, can potentially be alleviated by data augmentation. This study investigated various data augmentation techniques and applied multiple deep learning methods to Raman spectral classification. Raman spectra yield fingerprint-like information about chemical compositions, but are prone to noise when the particles of the material are small. Five augmentation models were investigated to build robust deep learning classifiers: weighted sums of spectral signals, imitated chemical backgrounds, extended multiplicative signal augmentation, and generated Gaussian and Poisson-distributed noise. We compared the performance of nine state-of-the-art convolutional neural networks with all the augmentation techniques. The LeNet5 models with background noise augmentation yielded the highest accuracy when tested on real-world Raman spectral classification at 88.33% accuracy. A class activation map of the model was generated to provide a qualitative observation of the results.

• Journal of KIISE
• /
• v.43 no.6
• /
• pp.686-691
• /
• 2016

In this paper, we introduced the adaptive spectral co-clustering, a spectral clustering for multiview data, especially data with more than three views. In the adaptive spectral co-clustering, the performance is improved by sharing information from diverse views. For the efficiency in information sharing, a co-training approach is adopted. In the co-training step, a set of parameters are estimated to make all views in data maximally independent, and then, information is shared with respect to estimated parameters. This co-training step increases the efficiency of information sharing comparing with ordinary feature concatenation and co-training methods that assume the independence among views. The adaptive spectral co-clustering was evaluated with synthetic dataset and multi lingual document dataset. The experimental results indicated the efficiency of the adaptive spectral co-clustering with the performances in every iterations and similarity matrix generated with information sharing.

• Journal of Forest and Environmental Science
• /
• v.24 no.3
• /
• pp.165-169
• /
• 2008

This study attempts to analyze the relationship between forest volume and age based on 5th NFI data and spectral characteristics of satellite imagery using ASTER sensor in Korea. Forest stand volume and age had the negative correlation with the spectral reflectance in all of the band (Blue, Green, Red, SWIR). With increasing of stand volume and age, spectral reflectance decrease. The spectral reflectance of band1 showed the highest correlation between stand volume and spectral reflectance among the VNIR wavelength. The spectral reflectance band 1, 2 (visible wavelength) and stand age have high correlation compared to other bands. The correlation coefficients between forest volume and vegetation indices have low relationship. This result indicates that the reflectance of blue band may be important factor to improve the potential of optical remote sensing data to estimate forest volume and age.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.7
• /
• pp.2568-2584
• /
• 2015

Spectral clustering has become one of the most popular clustering approaches in recent years. Similarity graph constructed on the data is one of the key factors that influence the performance of spectral clustering. However, the similarity graphs constructed by existing methods usually contain some unreliable edges. To construct reliable similarity graph for spectral clustering, an efficient method based on Markov random walk (MRW) is proposed in this paper. In the proposed method, theMRW model is defined on the raw k-NN graph and the neighbors of each sample are determined by the probability of the MRW. Since the high order transition probabilities carry complex relationships among data, the neighbors in the graph determined by our proposed method are more reliable than those of the existing methods. Experiments are performed on the synthetic and real-world datasets for performance evaluation and comparison. The results show that the graph obtained by our proposed method reflects the structure of the data better than those of the state-of-the-art methods and can effectively improve the performance of spectral clustering.

• Structural Monitoring and Maintenance
• /
• v.9 no.2
• /
• pp.179-200
• /
• 2022

This paper proposes a data-driven methodology for online early damage identification under changing environmental conditions. The proposed method relies on two data analysis methods: feature-based method and hybrid principal component analysis (PCA) and kernel PCA to separate damage from environmental influences. First, spectral sub-band features, namely, spectral sub-band centroids (SSCs) and log spectral sub-band energies (LSSEs), are proposed as damage-sensitive features to extract damage information from measured structural responses. Second, hybrid modeling by integrating PCA and kernel PCA is performed on the spectral sub-band feature matrix for data normalization to extract both linear and nonlinear features for nonlinear procedure monitoring. After feature normalization, suppressing environmental effects, the control charts (Hotelling T² and SPE statistics) is implemented to novelty detection and distinguish damage in structures. The hybrid PCA-KPCA technique is compared to KPCA by applying support vector machine (SVM) to evaluate the effectiveness of its performance in detecting damage. The proposed method is verified through numerical and full-scale studies (a Bridge Health Monitoring (BHM) Benchmark Problem and a cable-stayed bridge in China). The results demonstrate that the proposed method can detect the structural damage accurately and reduce false alarms by suppressing the effects and interference of environmental variations.

• Proceedings of the KSRS Conference
• /
• v.1
• /
• pp.184-187
• /
• 2006

The radiometric correction is prerequisite to derive both land and ocean surface properties from optical remote sensing data. Radiometric calibration of remotely sensed data has traditionally been accomplished by means of vicarious ground calibration techniques. The purpose of this study is to calibrate the radiometric characteristic of Airborne Multispectral Scanner (AMS) by field campaign. In order to calibrate the AMS data, four different spectral tarps which are 3.5%, 23%, 35%, and 53% were validated by GER-3700 that is the surface reflectance measurement equipment and were utilized. After validation of the spectral tarps, each reflectance from the spectral tarps was compared with Digital Number (DN) value of AMS. There was very high correlation between tarp reflectance and DN value of AMS so that radiometric calibration of AMS data has been accomplished by those results. The calibrated AMS data were validated with in-situ measured reflectance data from artificial and natural target. Also QuickBird image data were used for verifying the results of AMS radiometric calibration. This presentation discusses the results of the above tests.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.8 no.2
• /
• pp.33-40
• /
• 2005

The objective of this research was to find an indirect method to estimate spectral reflectance and NDVI(Normalized Difference Vegetation Index) efficiently, using the spectroradiometer and NOAA AVHRR satellite data. For collecting RS base data, used spectro-radiometer that measures reflection characteristics between 300~1,100nm was used and measured the reflection of vegetation from paddy rice during the growing season at Chungbuk national university's farm in 2002. The feasibility of detecting the temporal variation in the spectral reflectance and NDVI in paddy rice were conducted on eight growth stages. AVHRR data were collected in eight different months over a one year period in 2002. The results were compared with those obtained by analyzing NDVI characteristics. The spectral reflectance and NDVI of paddy rice have a great effect on the growth condition. Considerably, NDVI was increased by developing muscle fiber tissue at the near infrared wavelength until the Booting stage. Then the NDVI increased until the Maturity stage and then decreased until harvest. The highest month was at July and the lower month was at March. The difference NDVI analysis using March and another months data was conducted, the results were provided information on the growth condition of crops.

• Korean Journal of Remote Sensing
• /
• v.22 no.5
• /
• pp.305-311
• /
• 2006

Although there have been several attempts to estimate forest LAI using optical remote sensor data, there are still not enough evidences whether the NDVI is effective to estimate forest LAI, particularly in fully closed canopy situation. In this study, we have conducted a simple correlation analysis between LAI and spectral reflectance at two different settings: 1) laboratory spectral measurements on the multiple-layers of leaf samples and 2) Landsat ETM+ reflectance in the close canopy forest stands with fieldmeasured LAI. In both cases, the correlation coefficients between LAI and spectral reflectance were higher in short-wave infrared (SWIR) and visible wavelength regions. Although the near-IR reflectance showed positive correlations with LAI, the correlations strength is weaker than in SWIR and visible region. The higher correlations were found with the spectral reflectance data measured on the simulated vegetation samples than with the ETM+ reflectance on the actual forests. In addition, there was no significant correlation between the forest.LAI and NDVI, in particular when the LAI values were larger than three. The SWIR reflectance may be important factor to improve the potential of optical remote sensor data to estimate forest LAI in close canopy situation.

• Journal of Environmental Science International
• /
• v.30 no.2
• /
• pp.173-184
• /
• 2021

Ecological disturbance plants distributed throughout the country are causing a lot of damage to us directly or indirectly in terms of ecology, economy and health. These plants are not easy to manage and remove because they have a strong fertility, and it is very difficult to express them quantitatively. In this study, drone hyperspectral sensor data and Field spectroradiometer were acquired around the experimental area. In order to secure the quality accuracy of the drone hyperspectral image, GPS survey was performed, and a location accuracy of about 17cm was secured. Spectroscopic libraries were constructed for 7 kinds of plants in the experimental area using a Field spectroradiometer, and drone hyperspectral sensors were acquired in August and October, respectively. Spectral data for each plant were calculated from the acquired hyperspectral data, and spectral angles of 0.08 to 0.36 were derived. In most cases, good values of less than 0.5 were obtained, and Ambrosia trifida and Lactuca scariola, which are common in the experimental area, were extracted. As a result, it was found that about 29.6% of Ambrosia trifida and 31.5% of Lactuca scariola spread in October than in August. In the future, it is expected that better results can be obtained for the detection of ecosystem distribution plants if standardized indicators are calculated by constructing a precise spectral angle standard library based on more data.

• Journal of Communications and Networks
• /
• v.1 no.2
• /
• pp.81-85
• /
• 1999

The regrowth of OQPSK power spectral sidelobes from AM/AM and AM/PM amplifier nonlinearity is analyzed. The time-domain expression for amplifier output shows how spectral re-growth will depend on the cubic coefficient of the Taylor's series of the amplifier nonlinearity as well as input amplitude ripple. Closed form spectrum calculations show that the spectral sidelobes produced by AM/PM take the same form as those produced by AM/AM. The rate of growth of AM/PM sidelobes is, however, not as great as for AM/AM.