• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.3
• /
• pp.464-470
• /
• 2016

This paper presents a speech recognition database verification system using speech measures, and describes a speech measure extraction algorithm which is applied to this system. In our previous study, to produce an effective speech quality measure for the system, we propose a combination of various speech measures which are highly correlated with WER (Word Error Rate). The new combination of various types of speech quality measures in this study is more effective to predict the speech recognition performance compared to each speech measure alone. In this paper, we increase the system independency by employing GMM acoustic score instead of HMM score which is obtained by a secondary speech recognition system. The combination with GMM score shows a slightly lower correlation with WER compared to the combination with HMM score, however it presents a higher relative improvement in correlation with WER, which is calculated compared to the correlation of each speech measure alone.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.2
• /
• pp.157-168
• /
• 1993

In this paper, it is shown that neural networks trained with back propagation algorithm and dictionary can be applied to recognize handwritten English characters. To eliminate the useless data part and to minimize the variety of characters from the scanned image file, various preprocessings : that is, segmentation, centering, noise filtering, sealing and thinning are performed. After these, characteristic features are derived from thinned character pattern. The neural network is trained by using the extracted features for sample data, and all test data are classified into English alphabets according to their features through the neural network. Finally, the ways of reducing learning time and improving recognition rate, and the relationship between learning time and hidden layer nodes are considered. As a result of this study, after successful training, a high recognition rate has been obtained with this system for the trained patterns and about 93% for test patterns. Using dictionary, the recognition rate was about 97% for test pattern.

• Journal of the Korean Society of Radiology
• /
• v.10 no.8
• /
• pp.645-652
• /
• 2016

In this study, based on the analysis of texture feature values of statistical properties. And we examined the normal and the applicability of the computer-aided diagnosis of cerebral infarction in the brain computed tomography images. The experiment was analyzed to evaluate the ROC curve recognition rate of disease using six parameters representing the feature values of the texture. As a result, it showed average mean 88%, variance 92%, relative smoothness 94%, uniformity of 88%, a high disease recognition rate of entropy 84%. However, it showed a slightly lower disease recognition rate and 58% for skewness. In the analysis using ROC curve, the area under the curve for each parameter indicates 0.886 (p = 0.0001) or more, resulted in a meaningful recognition of the disease. Further, to determine the cut-off values for each parameter are determined to be the prediction of disease through the computer-aided diagnosis.

• Journal of Korean society of Dental Hygiene
• /
• v.12 no.4
• /
• pp.647-654
• /
• 2012

Objectives : To propose the data for prevention of oral health problems through assessment on the knowledge, cognition and practice about oral health care of the clients who visit dental clinics. Methods : The subjects were total 400 clients from the four dental clinics in Busan city. Their knowledge, cognition and performances about oral healthcare collected with questionnaire survey from February to March, 2011. The rate and mean compared by chi-square test, t-test, and ANOVA. Results : Mean level of their knowledge, recognition and practice were 72.5, 80.2, and 65.6, respectively. Theses levels were higher in women (p<0.01), in the group of higher interested (p<0.001), immediately treated (p<0.001), correct tooth brushing (p<0.05), toothbrushing educated (p<0.001), take regular scaling (p<0.001), use assistant hygiene items (p<0.001), instead no differences by age and education level. And, the rate of correct toothbrushing was higher as 1.24 (1.03~1.50) times in the group with higher knowledge level, the rate of immediate treatment and regular scaling were higher as 1.35 (1.07~1.70) times, and 2.26 (1.41~3.64) times in the group with higher recognition level, respectively. Conclusions : These results demonstrate that insufficient knowledge and recognition of oral health care are related to a lack of its practice. Though their primary goal of the visits was treatment, the clients' attention needed to raise the comprehensive knowledge and recognition levels for their oral health promotion, especially reach to the performance level rather than simple knowledge acquisition. And more, to achieve the efficient oral health promotion, the importance of early treatment and strengthened efforts for prevention should be emphasized.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.5
• /
• pp.13-20
• /
• 2015

In this paper, we propose an algorithm that can recognize not only the number of stretched fingers but also determination of attached fingers for extracting features required for hand gesture recognition. The proposed algorithm detects the hand area in the input image by the skin color range filter based on a color model and labeling, and then recognizes various hand gestures by extracting the number of stretched fingers and determination of attached fingers using curvature information extracted from outlines and feature points. Experiment results show that the recognition rate and the frame rate are similar to those of the conventional algorithm, but the number of gesture cases that can be defined by the extracted characteristics is about four times higher than the conventional algorithm, so that the proposed algorithm can recognize more various gestures.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.10 no.6
• /
• pp.32-39
• /
• 2011

Even if the traffic light is green, if vehicles enter a jammed intersection, they are violation of the law. The police is enforcing law as a part of a nation wide campaign. Because, using the camcorder, the police can not do enforcement the offending vehicle, there are other techniques. Our research group proposed automated photographic equipment enable to enforce moving violation in intersection. Using new license plate recognition technology and backtracking technology to trace the offending vehicle, once the system detects a violator, it records 8 wide pictures and 1picture from the front vehicle, showing it enter and proceed through the intersection. Field experimental results obtained in the system, the following conclusions. The three measures of effectiveness investigated were recognition rate 83.5, mis-match recognition rate 1.5%.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.12
• /
• pp.333-338
• /
• 2020

This paper proposes an augmented reality-based electronic circuit education system as a way for electronic circuit education, which is the basis of ICT convergence technology field. It consists of a hardware module that can identify the actual circuit and a mobile educational content that can check the current flow, input, output, and measured value by applying augmented reality technology. An experiment was conducted on image recognition, which is the main performance, for the purpose of stable operation of the system, and as the experimental method the recognition rate was measured by changing the distance between the hardware module and the mobile device to a certain interval. As a result of the experiment, the recognition rate was 100 percent at a distance of 25[Cm] or higher, and it was confirmed that the recognition rate decreased by 12% at a distance below 25[Cm], which can be said to be the effect of an error that results in image loss taken due to close distance. In the future, we plan to apply the education system presented in this paper to classes, which increases the efficiency of classes and improve students' interest and understanding of the subject.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.3
• /
• pp.13-22
• /
• 2024

Currently, with the proliferation of digital devices, the significance of handwritten texts in daily lives is gradually diminishing. As the use of keyboards and touch screens increase, a decline in Korean handwriting quality is being observed across a broad spectrum of Korean documents, from young students to adults. However, Korean handwriting still remains necessary for many documentations, as it retains individual unique features while ensuring readability. To this end, this paper aims to implement an application designed to improve and correct the quality of handwritten Korean script The implemented application utilizes the CRAFT (Character-Region Awareness For Text Detection) model for handwriting area detection and employs the VGG-Feature-Extraction as a deep learning model for learning features of the handwritten script. Simultaneously, the application presents the user's handwritten Korean script's reliability on a syllable-by-syllable basis as a recognition rate and also suggests the most similar fonts among candidate fonts. Furthermore, through various experiments, it can be confirmed that the proposed application provides an excellent recognition rate comparable to conventional commercial character recognition OCR systems.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.30 no.5
• /
• pp.380-388
• /
• 2019

Jet engine modulation(JEM) signals are widely used for target recognition. These signals coming from a potentially hostile aircraft provide specific information about the jet engine. In order to obtain the number of blades, which is uniquely provided by the JEM signal, one must extract the spool rate, which is the rotation speed of the blades. In this paper, we propose an algorithm to extract the spool rate from a weak JEM signal. A criterion is developed to extract the spool rate from the JEM signal by analyzing the intensity of the JEM signal component. The weak signal is first subjected to a high-pass filtering-based process, which modifies it to facilitate spool rate extraction. We then apply a peak detection process and extract the spool rate. The technique is simpler than the existing CEMD or WD method, is accurate, and greatly reduces the time required.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.8 no.4
• /
• pp.185-194
• /
• 2013

Abstract- In distant-talking environments, speech recognition performance degrades significantly due to noise and reverberation. Recent work of Michael L. Selzer shows that in microphone array speech recognition, the word error rate can be significantly reduced by adapting the beamformer weights to generate a sequence of features which maximizes the likelihood of the correct hypothesis. In this approach, called Likelihood Maximizing Beamforming algorithm (Limabeam), one of the method to implement this Limabeam is an UnSupervised Limabeam(USL) that can improve recognition performance in any situation of environment. From our investigation for this USL, we could see that because the performance of optimization depends strongly on the transcription output of the first recognition step, the output become unstable and this may lead lower performance. In order to improve recognition performance of USL, some post-filter techniques can be employed to obtain more correct transcription output of the first step. In this work, as a post-filtering technique for first recognition step of USL, we propose to add a Wiener-Filter combined with Feature Weighted Malahanobis Distance to improve recognition performance. We also suggest an alternative way to implement Limabeam algorithm for Hidden Markov Network (HM-Net) speech recognizer for efficient implementation. Speech recognition experiments performed in real distant-talking environment confirm the efficacy of Limabeam algorithm in HM-Net speech recognition system and also confirm the improved performance by the proposed method.