• Journal of Korea Multimedia Society
• /
• v.13 no.10
• /
• pp.1494-1505
• /
• 2010

A variety of sounds take place in medium and small-sized manufactories producing many kinds of parts in a small quantity with one press. We developed the identification system for the quality of parts using HTM(Hierarchical Temporal Memory)-based sound recognition. HTM is the theory that the operation principle of human brain's neocortex is applied to computer, suggested by Jeff Hopkins. This theory memorizes temporal and spatial patterns hierarchically about the real world, which is known for its cognitive power superior to the previous recognition technologies in many cases. By applying the HTM model to the sound recognition, we developed the identification system for the quality of molding parts. In order to verify its performance we recorded the various sounds at the moment of producing parts in the real factory, constructed the HTM network of sound, and then identified the quality of parts by repeating learning and training. It reveals that this system gets an excellent and accurate results at the noisy factory.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.4
• /
• pp.478-484
• /
• 2009

Automatic speech recognition(ASR) is attractive technique in trend these day that seek convenient life. Although many approaches have been proposed for ASR but the performance is still not good in noisy environment. Now-a-days in the state of art in speech recognition, ASR uses not only the audio information but also the visual information. In this paper, We present a novel lip detection method for visual speech recognition in mobile environment. In order to apply visual information to speech recognition, we need to extract exact lip regions. Because eye-detection is more easy than lip-detection, we firstly detect positions of left and right eyes, then locate lip region roughly. After that we apply K-means clustering technique to devide that region into groups, than two lip corners and lip center are detected by choosing biggest one among clustered groups. Finally, we have shown the effectiveness of the proposed method through the experiments based on samsung AVSR database.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2006.05a
• /
• pp.1334-1337
• /
• 2006

System identification is the field of modeling dynamic systems from experimental data. As a modeling technique, we can mention finite element method (FEM). In addition, we are able to measure modal data as the experimental data. The system can be generally categorized into a gray box and black box. In the gray box, we know mathematical model of a system, but we don't know structural parameters exactly, so we need to estimate structural parameters. In the black box, we don't know a system completely, so we need to identify system from nothing. To date, various system identification methods have been developed. Among them, we introduce system realization theory which uses Hankel matrix and Eigensystem Realization Algorithm (ERA) that enable us to identify modal parameters from noisy measurement data. Although we obtain noise-free data, however, we are likely to face difficulties in identifying a structure with hidden modes. Hidden modes can be occurred when the input or output position comes to a nodal point. If we change a system using a mode decoupling controller, the hidden modes can be revealed. Because we know the perturbation quantities in a closed loop system with the controller, we can realize an original system by subtracting perturbation quantities from the closed loop system. In this paper, we propose a novel method to identify a structure with hidden modes using the mode decoupling controller and the associated example is given for illustration.

• Journal of the Institute of Convergence Signal Processing
• /
• v.1 no.1
• /
• pp.23-31
• /
• 2000

Iterative curve evolution techniques are powerful methods for image segmentation. Classical methods proposed curve evolutions which guarantee close contours at convergence and, combined with the level set method, they easily handled curve topology changes. In this paper, we present a new geometric active contour model based on level set methods introduced by Osher & Sethian for detection of object boundaries or shape and we adopt anisotropic diffusion filtering method for removing noise from original image. Classical methods allow only one-way curve evolutions : shrinking or expanding of the curve. Thus, the initial curve must encircle all the objects to be segmented or several curves must be used, each one totally inside one object. But our method allows a two-way curve evolution : parts of the curve evolve in the outward direction while others evolve in the inward direction. It offers much more freedom in the initial curve position than with a classical geodesic search method. Our algorithm performs accurate and precise segmentations from noisy images with complex objects(jncluding sharp angles, deep concavities or holes), Besides it easily handled curve topology changes. In order to minimize the processing time, we use the narrow band method which allows us to perform calculations in the neighborhood of the contour and not in the whole image.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.19 no.2
• /
• pp.152-159
• /
• 2005

In this paper, we propose a diagnosis algorithm using fusion wかd based on PCA and LDA to detect fault states of the induction motor that is applied to various industrial fields. After yielding a feature vector from the current value measured by an experiment using PCA and LDA, training data is made to produce each matching value. In a diagnostic step, two matching values yielded by PCA and LDA are fused by probability model and finally verified. Since the proposed diagnosis algorithm takes only merits of PCA and LDA it shows excellent results under noisy environments. The simulation results to verify the usability of the proposed algorithm showed better performance than the case just using conventional PCA or LDA.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.10
• /
• pp.1867-1872
• /
• 2016

In the area of digital signal processing, it is necessary to improve the quality of the speech signal after removing the background noise which exists in a various real environments. The important thing to consider when removing the background noise acoustically is that to solve the problem, depending on the information of the human auditory mechanism is mainly the amplitude spectrum of the speech signal. This paper introduces the characteristics of a frequency-weighted spectrum shaping filter for the extraction of the amplitude spectrum of the speech signal with the primary purpose. Therefore, this paper proposes an algorithm using the methods of a Wiener filter and the frequency-weighted spectrum shaping filter according to the acoustic model, after extracted the amplitude spectral information in the noisy speech signal. The spectral distortion (SD) output of the proposed algorithm is experimentally improved more than 5.28 dB compared to a conventional method.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.24 no.1
• /
• pp.112-121
• /
• 2016

AFLS (Adaptive front lighting system) is being applied to improve safety in driving automotive at night. Safe embedded system design for controlling head-lamps is required to improve noise robust ECU hardware and software simultaneously by considering safety requirement of hardware-dependent software under severe environmental noise. In this paper, we propose an adaptive headlight controller with a newly-designed symmetric angle sensor compensator, especially based on the proposed steering-swivel angle lookup table to determine whether the current controlling target is safe. The proposed system includes an additional backup hardware to compare the system status and provides safe swivel-angle management using a controlling algorithm based on the pre-defined lookup table (LUT), which is a symmetric mapping relationship between the requested steering angle and expected swivel angle target. The implemented system model shows that the proposed architecture effectively detects abnormal situations and restores safe status of controlling the light-angle in AFLS operations under severe noisy environment.

• Journal of Biomedical Engineering Research
• /
• v.29 no.3
• /
• pp.198-204
• /
• 2008

Uroflowmetry is a non-invasive clinical test useful for screening benign prostatic hyperplasia(BPH) common in the aged men. The current standard way to obtain the urinary flow rate is to continuously acquire the urine weight signal proportional to volume over time. The present study proposed an alternative technique measuring pressure to overcome noise problems present in the standard weight measuring technique. Experiments were performed to simultaneously acquire both weight and pressure changes during urination of 9 normal men. Noise components were separated from volume signals converted from both weight and pressure signals based on the polynomial signal model. Signal-to-noise ratio was defined as the ratio of the energies between signal and noise components of the measured volume changes, which was 8.5 times larger in the pressure measuring technique, implying that cleaner signal could be obtained, more immune to noisy environments. When four important diagnostic parameters were estimated, excellent correlation coefficients higher than 0.99 were resulted with mean relative errors less than 5%. Therefore, the present pressure measurement seemed valid as an alternative technique for uroflowmetry.

• Smart Media Journal
• /
• v.6 no.3
• /
• pp.41-48
• /
• 2017

Event detection using social media has been widespread since social network services have been an active communication channel for connecting with others, diffusing news message. Especially, the real-time characteristic of social media has created the opportunity for supporting for real-time applications/systems. Social network such as Twitter is the potential data source to explore useful information by mining messages posted by the user community. This paper proposed a novel system for temporal event detection by analyzing social data. As a result, this information can be used by first responders, decision makers, or news agents to gain insight of the situation. The proposed approach takes advantages of deep learning methods that play core techniques on the main tasks including informative data identifying from a noisy environment and temporal event detection. The former is the responsibility of Convolutional Neural Network model trained from labeled Twitter data. The latter is for event detection supported by Recurrent Neural Network module. We demonstrated our approach and experimental results on the case study of earthquake situations. Our system is more adaptive than other systems used traditional methods since deep learning enables to extract the features of data without spending lots of time constructing feature by hand. This benefit makes our approach adaptive to extend to a new context of practice. Moreover, the proposed system promised to respond to acceptable delay within several minutes that will helpful mean for supporting news channel agents or belief plan in case of disaster events.

• The Journal of Korean Institute of Information Technology
• /
• v.16 no.11
• /
• pp.97-103
• /
• 2018

In this paper, the estimation method for the power signal harmonics is proposed by using the time-varying optimal finite impulse response (FIR) filter. To estimate the magnitude and phase-angle of the harmonic components, the time-varying optimal FIR filter is designed for the state space representation of the noisy power signal which the magnitude and phase is considered as a stochastic process. Since the time-varying optimal FIR filter used in the proposed method does not use any priori information of the initial condition and has FIR structure, the proposed method could overcome the demerits of Kalman filter based method such as poor estimation and divergence problem. Due to the FIR structure, the proposed method is more robust against to the model uncertainty than the Kalman filter. Moreover, the proposed method gives more general solution than the time-invariant optimal FIR filter based harmonic estimation method. To verify the performance and robustness of the proposed method, the proposed method is compared with time-varying Kalman filter based method through simulation.