• Journal of Korea Multimedia Society
• /
• v.11 no.11
• /
• pp.1471-1479
• /
• 2008

In this paper, we propose a technique to improve Out-Of-Vocabulary(OOV) rejection algorithms in variable vocabulary recognition system which is much used in ASR(Automatic Speech Recognition). The rejection system can be classified into two categories by their implementation method, keyword spotting method and utterance verification method. The utterance verification method uses the likelihood ratio of each phoneme Viterbi score relative to anti-phoneme score for deciding OOV. In this paper, we add speaker verification system before utterance verification and calculate an speaker verification probability. The obtained speaker verification probability is applied for determining the proposed variable-confidence threshold. Using the proposed method, we achieve the significant performance improvement; CA(Correctly Accepted for keyword) 94.23%, CR(Correctly Rejected for out-of-vocabulary) 95.11% in office environment, and CA 91.14%, CR 92.74% in noisy environment.

• Journal of Computational Design and Engineering
• /
• v.1 no.4
• /
• pp.272-288
• /
• 2014

This paper presents a two-step, semi-automated method for reconstructing a three-dimensional (3D) shape of the prostate from a 3D transrectal ultrasound (TRUS) image. While the method has been developed for prostate ultrasound imaging, it can potentially be applicable to any other organ of the body and other imaging modalities. The proposed method takes as input a 3D TRUS image and generates a watertight 3D surface model of the prostate. In the first step, the system lets the user visualize and navigate through the input volumetric image by displaying cross sectional views oriented in arbitrary directions. The user then draws partial/full contours on selected cross sectional views. In the second step, the method automatically generates a watertight 3D surface of the prostate by fitting a deformable spherical template to the set of user-specified contours. Since the method allows the user to select the best cross-sectional directions and draw only clearly recognizable partial or full contours, the user can avoid time-consuming and inaccurate guesswork on where prostate contours are located. By avoiding the usage of noisy, incomprehensible portions of the TRUS image, the proposed method yields more accurate prostate shapes than conventional methods that demand complete cross-sectional contours selected manually, or automatically using an image processing tool. Our experiments confirmed that a 3D watertight surface of the prostate can be generated within five minutes even from a volumetric image with a high level of speckles and shadow noises.

• Journal of Preventive Medicine and Public Health
• /
• v.7 no.2
• /
• pp.377-381
• /
• 1974

Determination of lead in urine is important in industrial hygiene and toxicology. Dithizone method has been principally used for the determination of lead in urine, which gives accurate results in skilful hands but is usually complex and time-consuming. Atomic absorption spectrophotometry is a new simple method and several procedures have been described. However, the influences of pH and the presence of chelating agents during treatment of lead poisoning are not clear. The purpose of this study was to find out the effect of pH and chelating agents on the determination of lead using Shimadzu atomic absorption/flame spectrophotometer, model AA-610. The results obtained were as follows: 1. The atomic absorption spectrophotometry(AAS) could be applied without prior acid digestion to specimens in the absence of chelating agents. The absorbance at $2,170\;{\AA}$, though more sensitive, was more noisy electronically. Therefore, we selected the wavelength of $2,833\;{\AA}$ plus scale expansion. 2. The optimal pH was in the range from 2 to 3. 3. The sensitivity was $0.075{\mu}g/ml/%$ and detection limit was about $0.2{\mu}g/ml$. 4. In the presence of EDTA, lead could not be completely determined without prior acid digestion. 5. On specimens from patients receiving penicillamine therapy, a comparison was made between the values obtained with dithizone method and AAS method with prior acid digestion. The results of comparison showed a very good agreement.

• Smart Structures and Systems
• /
• v.18 no.4
• /
• pp.643-662
• /
• 2016

The monitoring of structural integrity of pipeline tapered thread connections is of great significance in terms of safe operation in the industry. In order to detect effectively the loosening degree of tapered thread connection, an active sensing method using piezoceramic transducers was developed based on time reversal technique in this paper. As the piezoeramic transducers can be either as actuators or sensors to generate or detect stress waves, the energy transmission for tapered thread connection was analyzed. Subsequently, the detection principle for tapered thread connection based on time reversal was introduced. Finally, the inherent relationship between the contact area and tightness degree of tapered thread connection for the pipe structural model was investigated. Seven different contact area scenarios were tested. Each scenario was created by loosening connectors ranging from 3 turns to 4.5 turns in the right tapered threads when the contact area in the left tapered threads were 4.5 turns. The experiments were separately conducted with a highly noisy environment and various excitation signal amplitudes. The results show the focused peaks based on time reversal have the monotonously rising trend with the increase of the contact areas of tapered threads within an acceptable monitoring resolution for metal pipes. Compared with the energy method, the proposed time reversal based method to monitor tapered threads loosening demonstrates to be more robust in rejecting noise in Structural Health Monitoring (SHM) applications.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.12
• /
• pp.261-269
• /
• 2013

In this paper, the evoked potential(EP) was represented by additive delay model to comply with the variational noisy response of stimulus-event synchronization. The hybrid method of delay compensated-Wiener filtered-ensemble averaging(DWEA) was proposed to enhance the EP signal distortion occurred during averaging procedure due to synchronization timing mismatch. The performance of DWEA has been tested by surrogated simulation, which is composed of synthesized arbitrary delay and arbitrary level of added noise. The performance of DWEA is better than those of Wiener filtered-ensemble averaging and of conventional ensemble averaging. DWEA is endurable up to added noise gain of 7 for 10 % mean square error limit. Throughout the experimentation observation, it has been demonstrated that DWEA can be applied to enhance the evoked potential having the synchronization mismatch with added noise.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.4
• /
• pp.129-135
• /
• 2016

In this paper, we propose a novel tracking algorithm called slide window tracker (SWT) suitable for maneuvering target. To efficiently estimate trajectory of moving target, we adopt a sliding piecewise linear window which includes past trace information. By adjusting the window parameters, the proposed algorithm is to reduce measurement noise and to track fast maneuvering target with little computational increment as compared to ${\alpha}-{\beta}$ tracker. Throughout the computer simulations, we verify outstanding tracking performance of the SWT algorithm in noisy linear and nonlinear trajectories. Also, we show that the SWT algorithm is not sensitive to initial model parameter selection, which gives large degree of freedom in applying the SWT algorithm to unknown time-varying measurement environments.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.1
• /
• pp.149-156
• /
• 2014

NGE(Nested Generalized Exemplar Method) can increase the performance of the noisy data at the same time, can reduce the size of the model. It is the optimal distance-based classification method using a matching rule. NGE cross or overlap hyperrectangles generated in the learning has been noted to inhibit the factors. In this paper, We propose the DHGen(Dominant Hyperrectangle Generation) algorithm which avoids the overlapping and the crossing between hyperrectangles, uses interval weights for mixed hyperrectangles to be splited based on the mutual information. The DHGen improves the classification performance and reduces the number of hyperrectangles by processing the training set in an incremental manner. The proposed DHGen has been successfully shown to exhibit comparable classification performance to k-NN and better result than EACH system which implements the NGE theory using benchmark data sets from UCI Machine Learning Repository.

• Journal of Preventive Medicine and Public Health
• /
• v.20 no.2 s.22
• /
• pp.205-214
• /
• 1987

For the purpose of evaluating the effect of noise induced hearing loss on the blood pressure, a cross-sectional study was conducted in 481 manufacturing industrial workers in Pusan area consisted of 275 workers from noisier plant (over 90 dB(A), high exposed group) and 206 workers from less noisy plant (below 85 dB(A), low exposed group) from April to Decepmber in 1985. The summarized results were as follows; 1) The degree of hearing loss according to the audible frequency was most notable in 4,000 Hz. 2) The prevalence of hypertension was 14.8% in total examined subjects. And also in 40 dB(A) hearing lost workers, there was no significant difference between high exposed group as 15.5% and low exposed group as 15.8%. 3) In 3 models analyzed by multiple regression technique to obtain the complexed extents of risk factors related to the diastolic blood pressure, especially model III which contain age, body mass index, smoking, alcohol and family history of hypertension, duration of work, noise exposure level and degree of hearing loss in high exposed group was most remarkable compared to the others. 4) The most potential predictor related to the diastolic pressure in high exposed group was the degree of hearing loss. And the next were body mass index, familial history of hypertension and age in order. But in the case of low exposed group, the potential predictors were body mass index, age and familial history of hypertension.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.8
• /
• pp.97-105
• /
• 2015

Content-adaptive steganography methods embed secret messages in hard-to-model regions of covers such as complicated texture or noisy area. Content-adaptive steganalysis methods often need high dimensional features to capture more subtle relationships of local dependencies among adjacent pixels. However, these methods require many computational complexity and depend on the location of hidden message and the exploited distortion metrics. In this paper, we propose an improved steganalysis method for content-adaptive steganography to enhance detection rate with small number features. We first show that the features form the difference between DCT coefficients are useful for analyzing the content-adaptive steganography methods, and present feature extraction mehtod using first-order Markov probability for the the difference between DCT coefficients. The extracted features are used as input of ensemble classifier. Experimental results show that the proposed method outperforms previous schemes in terms of detection rates and accuracy in spite of a small number features in various content-adaptive stego images.

• Journal of Korea Multimedia Society
• /
• v.9 no.9
• /
• pp.1231-1246
• /
• 2006

HTML is a format for rendering Web documents visually and uses tables to present a relational information. Since HTML has limits in terms of information processing and management by a computer, it is important to transform HTML tables into XML documents, which is able to represent logical structure information. As a prerequisite for extracting information from the Web, this paper presents an efficient method for extracting logical structures from HTML tables and transforming them into XML documents. The proposed method consists of two phases: Area segmentation and structure analysis. The area segmentation step removes noisy areas and extracts attribute and value areas through visual and semantic coherency checkup. The hierarchical structure between attribute and value areas are analyzed and transformed into XML representations using a proposed table model. Experimental results with 1,180 HTML tables show that the proposed method performs better than the conventional method, resulting in an average precision of 86.7%.