• Journal of Broadcast Engineering
• /
• v.23 no.3
• /
• pp.421-430
• /
• 2018

Many voice recognition systems use methods such as MFCC, HMM to acknowledge human voice. This recognition method is designed to analyze only a targeted sound which normally appears between a human and a device one. However, the recognition capability is limited when there is a group sound formed with diversity in wider frequency range such as dog barking and indoor sounds. The frequency of overlapped sound resides in a wide range, up to 20KHz, which is higher than a voice. This paper proposes the new recognition method which provides wider frequency range by conjugating the Wideband Sound Spectrogram and the Keras Sequential Model based on DNN. The wideband sound spectrogram is adopted to analyze and verify diverse sounds from wide frequency range as it is designed to extract features and also classify as explained. The KSM is employed for the pattern recognition using extracted features from the WSS to improve sound recognition quality. The experiment verified that the proposed WSS and KSM excellently classified the targeted sound among noisy environment; overlapped sounds such as dog barking and indoor sounds. Furthermore, the paper shows a stage by stage analyzation and comparison of the factors' influences on the recognition and its characteristics according to various levels of noise.

• The Korean Journal of Petroleum Geology
• /
• v.11 no.1 s.12
• /
• pp.9-17
• /
• 2005

In the main target area of the block II, Targe-scale faults occur below the unconformity developed around 1 km in depth. The contrast of seismic velocity around the unconformity is generally so large that the strong multiples and the radical velocity variation would deteriorate the quality of migrated section due to serious distortion. More than 15 kinds of data processing techniques have been applied to improve the image resolution for the structures farmed from this active crustal activity. The bad and noisy traces were edited on the common shot gathers in the first step to get rid of acquisition problems which could take place from unfavorable conditions such as climatic change during data acquisition. Correction of amplitude attenuation caused from spherical divergence and inelastic attenuation has been also applied. Mild F/K filter was used to attenuate coherent noise such as guided waves and side scatters. Predictive deconvolution has been applied before stacking to remove peg-leg multiples and water reverberations. The velocity analysis process was conducted at every 2 km interval to analyze migration velocity, and it was iterated to get the high fidelity image. The strum noise caused from streamer was completely removed by applying predictive deconvolution in time space and ${\tau}-P$ domain. Residual multiples caused from thin layer or water bottom were eliminated through parabolic radon transform demultiple process. The migration using curved ray Kirchhoff-style algorithm has been applied to stack data. The velocity obtained after several iteration approach for MVA (migration velocity analysis) was used instead or DMO for the migration velocity. Using various testing methods, optimum seismic processing parameter can be obtained for structural and stratigraphic interpretation in the Block II, Yellow Sea Basin.

• Korean Institute of Interior Design Journal
• /
• v.19 no.5
• /
• pp.217-225
• /
• 2010

Today's public libraries in communities are on the processes of changes to integrate information and communication technology into traditional library system in order to support current users' demands for the new digital era. The purpose of this study is to examine the changing characters on space design and space uses of community based public libraries by conducting case studies of three branch libraries which were built after 2004 in Ann Arbor, Michigan in the United States. As the conclusion of this research, the findings of the case studies are utilized as basic data for planning and design guidelines for public libraries as community resources. The study summarizes the characteristics of space design and space uses in public libraries as follow; first, the floor plans of small-scale public libraries are open visually as well as spatially. The space organization of the libraries is arranged by potential noise levels, as placing noisy spaces near the entrance halls and quiet spaces at the back. Main book shelves are located in the middle of the library buildings, while seats are arranged along the window sides. By placing various kinds of furniture in open reading areas, library users can select different types of seats and tables for their comforts. Second. the survey of observation also finds that a large number of users often use library computers and personal computers to connect the internet at the libraries. These personal computer users who are new user group in community based libraries preferred to sit in casual study areas and individual tables with one or two seats only. Third, the libraries, in addition, develop and provide various programs and events for people in communities. Especially, the programs for children, the elderly and new comers from the abroad are well prepared, thus provide opportunities for them to visit the libraries in regular bases. The survey finds that family entertainment and leisure activities are the important parts of the program as well as renting music CD and movie DVD are also important reasons for people to come. Thus, the libraries prepare high quality children's space and CD shelves near the entrance hall.

• Journal of KIISE:Software and Applications
• /
• v.35 no.9
• /
• pp.562-571
• /
• 2008

For the purpose of building domain ontology, this paper proposes a methodology for building core ontology first, and then enriching the core ontology with the concepts and relations in the domain thesaurus. First, the top-level concept taxonomy of the core ontology is built using domain dictionary and general domain thesaurus. Then, the concepts of the domain thesaurus are classified into top-level concepts in the core ontology, and relations between broader terms (BT) - narrower terms (NT) and related terms (RT) are classified into semantic relations defined for the core ontology. To classify concepts, a two-step approach is adopted, in which a frequency-based approach is complemented with a similarity-based approach. To classify relations, two techniques are applied: (i) for the case of insufficient training data, a rule-based module is for identifying isa relation out of non-isa ones; a pattern-based approach is for classifying non-taxonomic semantic relations from non-isa. (ii) For the case of sufficient training data, a maximum-entropy model is adopted in the feature-based classification, where k-NN approach is for noisy filtering of training data. A series of experiments show that performances of the proposed systems are quite promising and comparable to judgments by human experts.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.2
• /
• pp.180-186
• /
• 2004

The presence of noise in speech signals degrades the performance of recognition systems in which there are mismatches between the training and test environments. To make a speech recognizer robust, it is necessary to compensate these mismatches. In this paper, we studied about an improvement of stochastic feature extraction based on band-SNR for robust speech recognition. At first, we proposed a modified version of the multi-band spectral subtraction (MSS) method which adjusts the subtraction level of noise spectrum according to band-SNR. In the proposed method referred as M-MSS, a noise normalization factor was newly introduced to finely control the over-estimation factor depending on the band-SNR. Also, we modified the architecture of the stochastic feature extraction (SFE) method. We could get a better performance when the spectral subtraction was applied in the power spectrum domain than in the mel-scale domain. This method is denoted as M-SFE. Last, we applied the M-MSS method to the modified stochastic feature extraction structure, which is denoted as the MMSS-MSFE method. The proposed methods were evaluated on isolated word recognition under various noise environments. The average error rates of the M-MSS, M-SFE, and MMSS-MSFE methods over the ordinary spectral subtraction (SS) method were reduced by 18.6%, 15.1%, and 33.9%, respectively. From these results, we can conclude that the proposed methods provide good candidates for robust feature extraction in the noisy speech recognition.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.8
• /
• pp.3748-3759
• /
• 2013

The purpose of this study was to examine the classification accuracies of ML, MAP, and EAP methods under DINA model. For this purpose, this study examined the classification accuracies of the classification methods under the various conditions: the number of attributes, the ability distribution of examinees, and test length. To accomplish this purpose, this study used a simulation method. For the simulation study, data was simulated under the various simulation conditions including the number of attributes (K= 5, 7), the ability distribution of examinees (high, middle, low), and test length (J= 15, 30, 45). Additionally, the percent of agreements between true skill patterns(true ${\alpha}$) and skill patterns estimated by the ML, MAP, and EAP methods were calculated. The summary of the main results of this study is as follows: First, When the number of attributes was 5 and 7, the EAP method showed relatively higher average in the percent of exact agreement than the ML and MAP methods. Second, under the same conditions, as the number of attributes increased, the average percent of exact agreement decreased in ML, MAP, and EAP methods. Third, when the prior distribution of examinees ability was different from low to high under the conditions of the same test length, the EAP method showed relatively higher average in the percent of exact agreement than those of the ML and MAP methods. Fourth, the average percent of exact agreement increased in all methods, ML, MAP, and EAP when the test length increased from 15 to 30 and 45 under the conditions of the same the ability distribution of examinees.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.1
• /
• pp.21-31
• /
• 2017

In this paper, the Zynq system-on-chip (SoC) platform is used to design an adaptive noise reduction and edge-detection system using partial reconfiguration. Filters are implemented in a partially reconfigurable (PR) region to provide high computational complexity in real-time, 1080p video processing. In addition, partial reconfiguration enables better utilization of hardware resources in the embedded system from autonomous replacement of filters in the same PR region. The proposed edge-detection system performs adaptive noise reduction if the noise density level in the incoming video sequences exceeds a given threshold value. Results of implementation show that the proposed system improves the accuracy of edge-detection results (14~20 times in Pratt's Figure of Merit) through self-reconfiguration of filter bitstreams triggered by noise density level in the video sequences. In addition, the ZyCAP controller implemented in this paper enables about 2.1 times faster reconfiguration when compared to a PCAP controller.

• Journal of Intelligence and Information Systems
• /
• v.10 no.1
• /
• pp.109-123
• /
• 2004

Corporate financial distress and bankruptcy prediction is one of the major application areas of artificial neural networks (ANNs) in finance and management. ANNs have showed high prediction performance in this area, but sometimes are confronted with inconsistent and unpredictable performance for noisy data. In addition, it may not be possible to train ANN or the training task cannot be effectively carried out without data reduction when the amount of data is so large because training the large data set needs much processing time and additional costs of collecting data. Instance selection is one of popular methods for dimensionality reduction and is directly related to data reduction. Although some researchers have addressed the need for instance selection in instance-based learning algorithms, there is little research on instance selection for ANN. This study proposes a genetic algorithm (GA) approach to instance selection in ANN for bankruptcy prediction. In this study, we use ANN supported by the GA to optimize the connection weights between layers and select relevant instances. It is expected that the globally evolved weights mitigate the well-known limitations of gradient descent algorithm of backpropagation algorithm. In addition, genetically selected instances will shorten the learning time and enhance prediction performance. This study will compare the proposed model with other major data mining techniques. Experimental results show that the GA approach is a promising method for instance selection in ANN.

• Journal of Korea Multimedia Society
• /
• v.8 no.11
• /
• pp.1496-1509
• /
• 2005

In this paper, an adaptive license plate recognition system which detects and recognizes license plate at real-time by using projected plane convolution and Decision Tree Classifier is proposed. And it was tested in circumstances which presence of complex background. Generally, in expressway tollgate or gateway of parking lots, it is very difficult to detect and segment license plate because of size, entry angle and noisy problem of vehicles due to CCD camera and road environment. In the proposed algorithm, we suggested to extract license plate candidate region after going through image acquisition process with inputted real-time image, and then to compensate license size as well as gradient of vehicle with change of vehicle entry position. The proposed algorithm can exactly detect license plate using accumulated edge, projected convolution and chain code labeling method. And it also segments letter of license plate using adaptive binary method. And then, it recognizes license plate letter by applying hybrid pattern vector method. Experimental results show that the proposed algorithm can recognize the front and rear direction license plate at real-time in the presence of complex background environments. Accordingly license plate detection rate displayed $98.8\%$ and $96.5\%$ successive rate respectively. And also, from the segmented letters, it shows $97.3\%$ and $96\%$ successive recognition rate respectively.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.5
• /
• pp.518-523
• /
• 2017

This paper proposes a novel denoising method based on non-local(NL) means. The NL-means algorithm is effective for removing an additive Gaussian noise, but the denoising parameter should be controlled depending on the noise level for proper noise elimination. Therefore, the proposed method optimizes the denoising parameter according to the noise levels. The proposed method consists of two processes: off-line and on-line. In the off-line process, the relations between the noise level and the denoising parameter of the NL-means filter are analyzed. For a given noise level, the various denoising parameters are applied to the NL-means algorithm, and then the qualities of resulting images are quantified using a structural similarity index(SSIM). The parameter with the highest SSIM is chosen as the optimal denoising parameter for the given noise level. In the on-line process, we estimate the noise level for a given noisy image and select the optimal denoising parameter according to the estimated noise level. Finally, NL-means filtering is performed using the selected denoising parameter. As shown in the experimental results, the proposed method accurately estimated the noise level and effectively eliminated noise for various noise levels. The accuracy of noise estimation is 90.0% and the highest Peak Signal-to-noise ratio(PSNR), SSIM value.