• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.10a
• /
• pp.824-826
• /
• 2015

최근 음성인식 시스템의 성능 향상은 많이 개선되었지만 아직도 잡음과 같은 문제로 인하여 문제점이 나타나고 있다. 음성인식 시스템에 있어서의 잡음 문제를 해결함으로써 인식 성능을 향상할 목적으로 본 논문에서는 단시간 스펙트럼에 기초한 주파수특성을 고려한 위너필터를 사용한 잡음 차감 알고리즘을 제안한다. 제안한 알고리즘은 먼저 각 프레임에서 문턱값을 검출한 후에 비묵음 구간과 묵음 구간을 식별한다. 각 프레임에 대해서 비묵음 구간에서는 위너필터법에 의한 잡음 차감법을 실시하며, 묵음 구간에 대해서는 일반적인 잡음 차감법을 적용한다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.4
• /
• pp.945-952
• /
• 2013

The R wave of QRS complex is the most prominent feature in ECG because of its specific shape; therefore it is taken as a reference in ECG feature extraction. But R wave detection suffers from the fact that frequency bands of the noise/other components such as P/T waves overlap with that of QRS complex. ECG signal processing must consider efficiency for hardware and software resources available in processing for miniaturization and low power. In other words, the design of algorithm that exactly detects QRS region using minimal computation by analyzing the person's physical condition and/or environment is needed. Therefore, efficient QRS detection based on SOM(Subtractive Operation Method) is presented in this paper. For this purpose, we detected R wave through the preprocessing method using morphological filter, empirical threshold, and subtractive signal. Also, we applied dynamic backward searching method for efficient detection. The performance of R wave detection is evaluated by using MIT-BIH arrhythmia database. The achieved scores indicate the average of 99.41% in R wave detection.

• Journal of the Korean Society of Radiology
• /
• v.79 no.6
• /
• pp.340-347
• /
• 2018

Purpose: To evaluate value of image subtraction for identifying hepatocellular carcinoma (HCC) capsule on gadoxetic acid-enhanced MR images. Materials and Methods: This study involved 108 patients at risk of HCC preoperatively examined using gadoxetic acid-enhanced MRI with hepatic resection between May 2015 and February 2017. We evaluated qualities of subtraction images and presence of capsular appearance on portal venous or transitional phases conventional and subtraction images. We assessed effect of capsular appearance on subtraction images on HCC. Results: After excluding 1 patient who had treated by transarterial chemoembolization prior to surgery and 33 patients with unsatisfactory subtraction image qualities, 82 focal hepatic lesions (73 HCC, 5 non-HCC malignancies, and 4 benign) from 74 patients were analyzed. Regarding detection of capsules, sensitivity, accuracy, and area under the receiver operating characteristic curve (AUC) on subtraction images were significantly higher than those on conventional images (95.4%, 89.0%, and 0.80, respectively; p < 0.001), though specificities were same (64.7%). For diagnosis of HCC, sensitivity, accuracy, and AUC on subtraction images were significantly higher than on conventional images (82.2%, 79.3%, and 0.69, respectively; p = 0.011), though specificities were identical (55.6%). Conclusion: Portal venous or transitional phase gadoxetic acid-enhanced MRI subtraction images could improve detection of HCC capsule.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.2
• /
• pp.447-454
• /
• 2011

Power-line interference(PLI) can distort certain regions in analysing the ECG signal. In particular, the regions such as P and R wave that are important element in diagnosing with arrhythmia is expressed as different type of noise according to the case whether power-line frequency is multiples of sampling frequency and or not. Noise characteristics is also divided into linearity and non-linearity. In this paper, adaptive subtraction method for removing variable PLI of ECG signal is proposed. We classify the multiple relationship between power line and sampling frequency as Multiple and Non-multiple. PLI of Linear segment is extracted through moving average filter, PLI of non-linear segment is extracted through the interference component that is extracted in the linear segment and stored in the temporary buffer. The performance of P wave and R wave detection is evaluated by using 119 data record of MIT-BIH arrhythmia database. The achieved scores indicate P wave detection rate of 97.91%, R wave detection rate of 96.66% and P wave detection rate of 99.01%, R wave detection rate of 97.93% accuracy respectively for Notch filter and proposed subtraction method.

• The Journal of the Korea Contents Association
• /
• v.9 no.1
• /
• pp.139-146
• /
• 2009

This paper addresses a modified spectral subtraction schemes which is suitable to speech recognition under low signal-to-noise ratio (SNR) noisy environment such as the automatic speech recognition (ASR) system in car. The conventional spectral subtraction schemes rely on the SNR such that attenuation is imposed on that part of the spectrum that appears to have low SNR, and accentuation is made on that part of high SNR. However, such postulation is adequate for high SNR environment, it is grossly inadequate for low SNR scenarios such as that of car environment. Proposed methods focused specifically to low SNR noisy environment by using weighting function for enhancing speech dominant region in speech spectrum. Experimental results by using voice commands for car show the superior performance of the proposed method over conventional methods.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.572-573
• /
• 2021

By optimizing the development of IoT-type thermal image-based photovoltaic fault detection equipment and interworking with drones using a drone with an intelligent path movement function, real-time analysis of the acquired image data facilitates fault reading of solar power plants. , design a system that can read out the failure of a solar panel using the image subtraction analysis technique and the presentation of the basic technology that can improve the power generation rate of the solar power plant and make an efficient maintenance model.

• Journal of KIISE:Software and Applications
• /
• v.36 no.12
• /
• pp.967-976
• /
• 2009

In this paper, we propose an automatic lower extremity vessel extraction based on rigid registration and bone elimination techniques in CT and CT angiography images. First, automatic partitioning of the lower extremity based on the anatomy is proposed to consider the local movement of the bone. Second, rigid registration based on distance map is performed to estimate the movement of the bone between CT and CT angiography images. Third, bone elimination and vessel masking techniques are proposed to remove bones in CT angiography image and to prevent the vessel near to bone from eroding. Fourth, post-processing based on vessel tracking is proposed to reduce the effect of misalignment and noises like a cartilage. For the evaluation of our method, we performed the visual inspection, accuracy measures and processing time. For visual inspection, the results of applying general subtraction, registered subtraction and proposed method are compared using volume rendering and maximum intensity projection. For accuracy evaluation, intensity distributions of CT angiography image, subtraction based method and proposed method are analyzed. Experimental result shows that bones are accurately eliminated and vessels are robustly extracted without the loss of other structure. The total processing time of thirteen patient datasets was 40 seconds on average.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.2
• /
• pp.11-17
• /
• 1999

This paper proposes an observation probability smoothing technique for the robustness of a discrete hidden Markov(DHMM) model based speech recognizer. Also, an appropriate noise robust processing in car environment is suggested from experimental results. The noisy speech is often mislabeled during the vector quantization process. To reduce the effects of such mislabelings, the proposed technique increases the observation probability of similar codewords. For the noise robust processing in car environment, the liftering on the distance measure of feature vectors, the high pass filtering, and the spectral subtraction methods are examined. Recognition experiments on the 14-isolated words consists of the Korean digits and command words were performed. The database was recorded in a stopping car and a running car environments. The recognition rates of the baseline recognizer were 97.4% in a stopping situation and 59.1% in a running situation. Using the proposed observation probability smoothing technique, the liftering, the high pass filtering, and the spectral subtraction the recognition rates were enhanced to 98.3% in a stopping situation and to 88.6% in a running situation.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.6
• /
• pp.445-449
• /
• 2009

In this paper, we propose a computationally efficient multi GPU accelerated image registration technique to correct the motion difference between the pre-contrast CT image and post-contrast CTA image. Our method consists of two steps: multi GPU based image registration and a cerebrovascular visualization. At first, it computes a similarity measure considering the parallelism between both GPUs as well as the parallelism inside GPU for performing the voxel-based registration. Then, it subtracts a CT image transformed by optimal transformation matrix from CTA image, and visualizes the subtracted volume using GPU based volume rendering technique. In this paper, we compare our proposed method with existing methods using 5 pairs of pre-contrast brain CT image and post-contrast brain CTA image in order to prove the superiority of our method in regard to visual quality and computational time. Experimental results show that our method well visualizes a brain vessel, so it well diagnose a vessel disease. Our multi GPU based approach is 11.6 times faster than CPU based approach and 1.4 times faster than single GPU based approach for total processing.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.3
• /
• pp.342-350
• /
• 2022

In this paper, actual drone noise samples are collected for speech processing in disaster environments to build noise-corrupted speech database, and speech enhancement performance is evaluated by applying spectrum subtraction and mask-based speech enhancement techniques. To improve the performance of VoiceFilter (VF), an existing deep neural network-based speech enhancement model, we apply the Self-Attention operation and use the estimated noise information as input to the Attention model. Compared to existing VF model techniques, the experimental results show 3.77%, 1.66% and 0.32% improvements for Source to Distortion Ratio (SDR), Perceptual Evaluation of Speech Quality (PESQ), and Short-Time Objective Intelligence (STOI), respectively. When trained with a 75% mix of speech data with drone sounds collected from the Internet, the relative performance drop rates for SDR, PESQ, and STOI are 3.18%, 2.79% and 0.96%, respectively, compared to using only actual drone noise. This confirms that data similar to real data can be collected and effectively used for model training for speech enhancement in environments where real data is difficult to obtain.