Sung-June Baek

The Fast Search Algorithm for Raman Spectrum

Abstract The problem of fast search for raman spectrum has attracted much attention recently. By far the most simpleand widely used method is to calculate and compare the Euclidean distance between the given spectrum and the spectrain a database. But it is non-trivial problem because of the inherent high dimensionality of the data. One of the most seriousproblems is the high computational complexity of searching for the closet codeword. To overcome this problem, The fastcodeword search algorithm based on the mean pyramids of codewords is currently used in image coding applications. Inthis paper, we present three new methods for the fast algorithm to search for the closet codeword. the proposed algorithmuses two significant features of a vector, mean values and variance, to reject many unlikely codewords and save a great deal of computation time. The Experiment results show about 42.8-55.2% performance improvement for the 1DMPS+PDS.The results obtained confirm the effectiveness of the proposed algorithm.

A screening of Alzheimer's disease using basis synthesis by singular value decomposition from Raman spectra of platelet

Abstract In this paper, we proposed a method to screening of Alzheimers disease (AD) from Raman spectra of platelet with synthesis of basis spectra using singular value decomposition (SVD). Raman spectra of platelet from AD transgenic mice are preprocessed with denoising, removal background and normalization method. The column vectors of each data matrix consist of Raman spectrum of AD and normal (NR). The matrix is factorized using SVD algorithm and then the basis spectra of AD and NR are determined by 12 column vectors of each matrix. The classification process is completed by select the class that minimized the root-mean-square error between the validation spectrum and the linear synthesized spectrum of the basis spectra. According to the experiments involving 278 Raman spectra, the proposed method gave about 97.6% classification rate, which is better performance about 6.1% than multi-layer perceptron (MLP) with extracted features using principle components analysis (PCA). The results show that the basis spectra using SVD is well suited for the diagnosis of AD by Raman spectra from platelet.

Noisy OTDR Data Event Detection Analysis for the Real Time Optical Fiber Link Monitoring

Abstract This paper, proposes a new analysis method for the event detection of an OTDR signal. An OTDR signalwas passed through the Hamming filter to remove the high frequency noise included in the signal. The signal wasthen passed consecutively through a differential filter to detect the events of interest. The terminal position was determined using the fact that there is a large gap between the signal and the trailing noise power beyond the terminal.This study examined the local maxima of the signal up to the terminal position and determined the peak regions.The real events were determined from the peak regions using noise information and peak threshold. Finally, the userevents were found by inspecting the user peaks beyond the terminal position. The events of the OTDR signal withoutusers are located at less than 17m compared to the optical fiber link setup. The events of the JDSU device are locatedless than 25m and their users are less than 5m. For the RadianTech device, the events are detected at less than 19mand the users are found in 5m. The results suggest that the proposed method is sufficiently applicable to an opticalfiber link monitoring system.

A Fast Search Algorithm for Raman Spectrum using Singular Value Decomposition

In this paper, we propose new search algorithms using SVD(Singular Value Decomposition) for fast search of Raman spectrum. In the proposed algorithms, small number of the eigen vectors obtained by SVD are chosen in accordance with their respective significance to achieve computation reduction. By introducing pilot test, we exclude large number of data from search and then, we apply partial distance search(PDS) for further computation reduction. We prepared 14,032 kinds of chemical Raman spectrum as the library for comparisons. Experiments were carried out with 7 methods, that is Full Search, PDS, 1DMPS modified MPS for applying to 1-dimensional space data with PDS(1DMPS+PDS), 1DMPS with PDS by using descending sorted variance of data(1DMPS Sort with Variance+PDS), 250-dimensional components of the SVD with PDS(250SVD+PDS) and proposed algorithms, PSP and PSSP. For exact comparison of computations, we compared the number of multiplications and additions required for each method. According to the experiments, PSSP algorithm shows 64.8% computation reduction when compared with 250SVD+PDS while PSP shows 157% computation reduction.

Improvement in the classification performance of Raman spectra using a hierarchical tree structure

This paper proposes a method in which classes are grouped as a hierarchical tree structure for the effective classification of the Raman spectra. As experimental data, the Raman spectra of 28 chemical compounds were obtained, and pre-treated with noise removal and normalization. The spectra that induced a classification error were grouped into the same class and the hierarchical structure class was composed. Each high and low class was classified using a PCA-MAP method. According to the experimental results, the classification of 100% was achieved with 2.7 features on average when the proposed method was applied. Considering that the same classification rates were achieved with 6 features using the conventional method, the proposed method was found to be much better than the conventional one in terms of the total computational complexity and practical application.

Estimation of Optimal Mixture Number of GMM for Environmental Sounds Recognition

In this paper we applied the optimal mixture number estimation technique in GMM(Gaussian mixture model) using BIC(Bayesian information criterion) and MDL(minimum description length) as a model selection criterion for environmental sounds recognition. In the experiment, we extracted 12 MFCC(mel-frequency cepstral coefficients) features from 9 kinds of environmental sounds which amounts to 27747 data and classified them with GMM. As mentioned above, BIC and MDL is applied to estimate the optimal number of mixtures in each environmental sounds class. According to the experimental results, while the recognition performances are maintained, the computational complexity decreases by 17.8% with BIC and 31.7% with MDL. It shows that the computational complexity reduction by BIC and MDL is effective for environmental sounds recognition using GMM.

Gaussian Mixture Model using Minimum Classification Error for Environmental Sounds Recognition Performance Improvement

In this paper, we proposed the MCE as a GMM training method to improve the performance of environmental sounds recognition. We model the environmental sounds data with newly defined misclassification function using the log likelihood of the corresponding class and the log likelihood of the rest classes for discriminative training. The model parameters are estimated with the loss function using GPD(generalized probabilistic descent). For recognition performance comparison, we extracted the 12 degrees features using preprocessing and MFCC(mel-frequency cepstral coefficients) of the 9 kinds of environmental sounds and carry out GMM classification experiments. According to the experimental results, MCE training method showed the best performance by an average of 87.06% with 19 mixtures. This result confirmed us that MCE training method could be effectively used as a GMM training method in environmental sounds recognition.