• The Journal of the Korea institute of electronic communication sciences
• /
• v.9 no.5
• /
• pp.543-548
• /
• 2014

Multiple-model based speech recognition framework (MMSR) has been known to be very successful in speech recognition. Since it uses multiple hidden Markov modes (HMMs) that corresponds to various noise types and signal-to-noise ratio (SNR) values, the selected acoustic model can have a close match with the test noisy speech. However, since the number of HMM sets is limited in practical use, the acoustic mismatch still remains as a problem. In this study, we experimentally determined the optimal SNR mapping between the test noisy speech and the HMM set to mitigate the mismatch between them. Improved performance was obtained by employing the SNR mapping instead of using the estimated SNR from the test noisy speech. When we applied the proposed method to the MMSR, the experimental results on the Aurora 2 database show that the relative word error rate reduction of 6.3% and 9.4% was achieved compared to a conventional MMSR and multi-condition training (MTR), respectively.

• Journal of Animal Science and Technology
• /
• v.65 no.2
• /
• pp.365-376
• /
• 2023

Pig breeding management directly contributes to the profitability of pig farms, and pregnancy diagnosis is an important factor in breeding management. Therefore, the need to diagnose pregnancy in sows is emphasized, and various studies have been conducted in this area. We propose a computer-aided diagnosis system to assist livestock farmers to diagnose sow pregnancy through ultrasound. Methods for diagnosing pregnancy in sows through ultrasound include the Doppler method, which measures the heart rate and pulse status, and the echo method, which diagnoses by amplitude depth technique. We propose a method that uses deep learning algorithms on ultrasonography, which is part of the echo method. As deep learning-based classification algorithms, Inception-v4, Xception, and EfficientNetV2 were used and compared to find the optimal algorithm for pregnancy diagnosis in sows. Gaussian and speckle noises were added to the ultrasound images according to the characteristics of the ultrasonography, which is easily affected by noise from the surrounding environments. Both the original and noise added ultrasound images of sows were tested together to determine the suitability of the proposed method on farms. The pregnancy diagnosis performance on the original ultrasound images achieved 0.99 in accuracy in the highest case and on the ultrasound images with noises, the performance achieved 0.98 in accuracy. The diagnosis performance achieved 0.96 in accuracy even when the intensity of noise was strong, proving its robustness against noise.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.37 no.2
• /
• pp.32-45
• /
• 2000

This paper presents the analytic derivation of the SINR, when a linear array antenna is accommodated into the cellular CDMA basestation receiver, in relation to the two major performance effecting factors in beamforming(BF) applications, i. e., the direction selectivity, which refers to the narrowness of the mainbeam width, and the direction-of-arrival(DOA) estimation accuracy. The analytically derived results are compared with the operation simulation of the receiver realized with the several BF algorithms and their agreements are confirmed, consequently verifying the correctness of the analysis and the operation simulation. In order to investigate separately the effects of the errors occurring in the direction estimation and in the interference suppression, which are the two major functional components of general BF algorithms, both the algorithms of steering BF and the minimum- variance- distortionless-response(MVDR) BF are applied to the analysis. A signal model to reflect the spatially scattering phenomenon of the RF waves entering into the .:nay antenna, which directly affects on the accuracy of the BF algorithm's direction estimation, is also suggested in this paper and applied to the analysis and the operation simulation. It is confirmed from the results that the enhancement of the direction selectivity of the away antenna is not desirable in view of both the implementation economy and the BF algorithm's robustness to the erroneous factors. Such a trade-off characteristics is significant in the sense that it can be capitalized to obtain an economic means of BF implementation that does not severely deteriorate its performance while ensuring the robustness to the erroneous effects, consequently manifesting the significance of the analysis results of this paper that can be used as a design reference in developing BF algorithms to the cellular CDMA system.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2009.10a
• /
• pp.258-263
• /
• 2009

This paper describes an analysis of the stability and performance of a large-capacity flywheel energy storage system (FESS) supported by active magnetic bearings. We designed and manufactured the system that can store up to 5kWh of usable energy at the maximum speed of 18,000 rpm. In order to analyze the stability of the systems accurately, we derived a rigid body rotor model, flexible rotor model using finite-element method, and a reduced-order model using modal truncation. The rotor model is combined with those of active magnetic bearings, amplifiers, and position sensors, resulting in a system simulation model. This simulation model is validated against experimental measurements. The stability of the system is checked from the pole locations of the closed-loop transfer functions. We also investigated the sensitivity function to quantify the robustness of the systems to the disturbances such as mass imbalance and sensor noises.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.7
• /
• pp.487-494
• /
• 2013

Medical image watermarking has received extensive attention as wide security services in the healthcare information system. This paper proposes a blind medical image watermarking approach on the segmented gray-matter (GM) images by utilizing discrete wavelet transform (DWT) and discrete cosine transform (DCT) along with enhanced suppressed fuzzy C-means (EnSFCM) for the optimal selection of sub-blocks position to insert a watermark. Experimental results show that the proposed approach outperforms other methods in terms of peak signal to noise ratio (PSNR) and M-SVD. In addition, the proposed approach shows better robustness than other methods in normalized correlation (NC) values against several attacks, such as noise addition, filtering, JPEG compression, blurring, histogram equalization, and cropping.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.14 no.12
• /
• pp.1347-1353
• /
• 2004

With respect to the researches of the optical flying head(OFH) , the head-gimbal assembly should be analyzed to guarantee the stable fabrication and the characteristics of shock resistance. The suitable design is proved through the Probabilistic analysis of the design parameters and material properties of the model. Probabilistic analysis is a technique that be used to assess the effect of uncertain input parameters and assumptions on your analysis model. Using a probabilistic analysis you can find out how much the results of a finite elements analysis are affected by uncertainties in the model. Another factor is analysis of the dynamic shock analysis. For the mobile application, one of the important requirements is durability under severe environmental condition, especially, resistance to mechanical shock. An important challenge in the disk recording is to improve disk drive robustness in shock environments. If the system comes in contact with outer shock disturbance. the system gets critical damage in head-gimbal assembly or disk. This paper describes probabilistic and dynamic shock analysis of head-gimbal assembly in micro MO drives using OFH slider.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.2087-2092
• /
• 2005

Using disturbance observer (DOB) is effective in enhancing the performance of dynamic system in the presence of disturbances. Recently the definition of disturbance has been expanded to modeling uncertainty including parameter variation, internal disturbance. Various structures of DOB have been proposed to improve sensitivity of system for better disturbance rejection performance. However in the case of improvement of sensitivity function, it tends to bring poor transient response due to cross-coupling and phase lag. Furthermore it could be very sensitive to measurement noise due to increased peak of complementary sensitivity function. In this paper, a dual disturbance observer (Dual-DOB) is proposed to reduce the effect of such cross-coupling. It is possible for us to improve the sensitivity function with additional external DOB with hardly affecting complementary sensitivity function. Thus it is able to have robustness against measurement noise. Since we are able to design DOBs of internal and external loop independently, we could prevent transient response quality from degrading while improving the sensitivity function. The proposed Dual-DOB is applied to a commercial optical disk drive tracking servo system. The experimental result shows that the Dual-DOB is an effective method in rejecting the disturbance as well as improving the tracking performance.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.37 no.2
• /
• pp.85-95
• /
• 2000

This paper describes a sensorless control scheme for brushless dc motors(BLDCMs) using a phase shifter(FIPS) which can shift the zero-crossing point of the input signal with a specified amount of phase. The detection performance of the proposed FIPS is independent of the frequency of the input signal and quite robust with respect to the measurement noise. It is shown that the proposed sensorless control scheme using the FIPS is more effective in the respects of noise-robustness and cost than the previously known schemes. The generality and practicality of the proposed sensorless control scheme is demonstrated through performance analysis and experiments under various operating conditions.

• The KIPS Transactions:PartB
• /
• v.8B no.2
• /
• pp.155-163
• /
• 2001

In this paper, we propose an image registration method by optimization of mutual information to provide a significant infonnation from multimodality images. The method applies mutual infonnation to measure the statistical dependency'r information redundancy between the image intensities of corresponding pixels in both images, which is assumed to be maximal if the images are geometrically aligned. We show the registration results optimizing mutual information between brain MR image and brain CT image and the comparison results with additive gaussian noise. Since our method uses the native image rather than prior segmentation or feature extraction, no user interaction is required and the accuracy of registration is improved. In addition, it shows the robustness against the noise.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.40 no.9
• /
• pp.658-665
• /
• 2003

In this paper, we propose an improved image encryption and the shift-tolerance method in the Fourier space using a virtual phase image. The encrypted image is obtained by the Fourier transform of the product of a phase-encoded virtual image, not an original image, and a random phase image. Therefore, even if unauthorized users analyze the encrypted image, we can prevent the possibility of counterfeiting from unauthorized people using virtual image which dose not contain any information from the original image. The decryption technique is simply performed by inverse Fourier transform of the interference pattern between the encrypted image and the Fourier decrypting key, made of proposed phase assignment rule, in frequency domain. We demonstrate the robustness to noise, to data loss and shift of the encrypted image or the Fourier decryption key in the proposed technique.