• 전자공학회논문지SC
• /
• 제44권1호
• /
• pp.100-107
• /
• 2007

자동화 청성뇌간반응검사(automated auditory brainstem response; AABR)는 ABR 파형을 자동으로 검출하여 신생아의 청각선별검사에 사용되고 있다. 본 논문은 앙상블 평균된 ABR 파형에 대해서 롤의 정리를 이용한 새로운 자동화 ABR 파형 검출 알고리즘을 제안하였다. 정상 청력을 가진 신생아의 55개의 귀를 대상으로 30, 40, 50, 60 dBnHL의 다양한 강도를 가진 클릭 자극음에 대한 청성뇌간반응 파형을 측정하였다. 수동 검출법(manual detection method)과 제안된 자동 검출법을 이용하여 파형 III 과 V의 평균 잠복기(average latency time) 차를 분석하였는데, 동일한 파형(잠복기 차 < 0.2 ms)으로 관측되어 두 방법 간에는 유의한 차이가 없었다. 또한 미분 자동 검출법(automated detection method using derivative estimation)과 제안된 자동 검출법을 파형 III과 V로 판명될 후보 파형의 개수에 대해 비교하였다. 미분 자동 검출법에 비해 제안한 자동 검출법에서 후보 파형의 개수가 47 % 감소되어 검출되었다. 또한 수동 검출법에 대한 제안된 자동 검출법의 잠복기 오차율은 미분 자동 검출법에 비해 60 dBnHL의 자극강도에서 낮은 잠복기 오차율(<0.01 %)을 보였다. 따라서 제안된 알고리즘으로 청각전문가가 기존의 수동 검출 방법보다 객관적이고 정량적으로 파형 III과 V를 검출하고 표시할 수 있게 된 데에 의의가 있다.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1992년도 춘계학술대회
• /
• pp.175-178
• /
• 1992

This paper proposes a method called linear prediction (a high performant technique in digital speech processing) for analyzing digital ECG signals. There are several significant properties indicating that ECG signals have an important feature in the residual error signal obtained after processing by Durbin's linear prediction algorithm. The ECG signal classification puts an emphasis on the residual error signal. For each ECG's QRS complex. the feature for recognition is obtained from a nonlinear transformation which transforms every residual error signal to set of three states pulse-cord train relative to the original ECG signal. The pulse-cord train has the advantage of easy implementation in digital hardware circuits to achive automated ECG diagnosis. The algorithm performs very well feature extraction in arrythmia detection. Using this method, our studies indicate that the PVC (premature ventricular contration) detection has a at least 90 percent sensityvity for arrythmia data.

• Structural Engineering and Mechanics
• /
• 제87권3호
• /
• pp.211-219
• /
• 2023

The structural health of a pipeline is usually assessed by visual inspection. In addition to the fact that this method is expensive and time consuming, inspection of the whole structure is not possible due to limited access to some points. Therefore, adopting a damage detection method without the mentioned limitations is important in order to increase the safety of the structure. In recent years, vibration-based methods have been used to detect damage. These methods detect structural defects based on the fact that the dynamic responses of the structure will change due to damage existence. Therefore, the location and extent of damage, before and after the damage, are determined. In this study, fuzzy genetic algorithm has been used to monitor the structural health of the pipeline to create a fuzzy automated system and all kinds of possible failure scenarios that can occur for the structure. For this purpose, the results of an experimental model have been used. Its numerical model is generated in ABAQUS software and the results of the analysis are used in the fuzzy genetic algorithm. Results show that the system is more accurate in detecting high-intensity damages, and the use of higher frequency modes helps to increase accuracy. Moreover, the system considers the damage in symmetric regions with the same degree of membership. To deal with the uncertainties, some error values are added, which are observed to be negligible up to 10% of the error.

• 대한안전경영과학회지
• /
• 제1권1호
• /
• pp.69-77
• /
• 1999

One of the objectives of any task design is to provide a safe and helpful workplace for the employees. The safety and health module may include means for confronting the design with safety and health regulations and standards as well as tools for obstacles and collisions detection (such as error models and simulators), Virtual Reality is a leading edge technology which has only very recently become available on platforms and at prices accessible to the majority of simulation engineers. The design of an automated manufacturing system is a complicated, multidisciplinary task that requires involvement of several specialists. In this paper, a design procedure that facilitates the safety and ergonomic considerations of an automated manufacturing system are described. The procedure consists of the following major steps. Data collection and analysis of the data, creation of a three-dimensional simulation model of the work environment, simulation for safety analysis and risk assessment, development of safety solutions, selection of the preferred solutions, implementation of the selected solutions, reporting, and training. When improving the safety of an existing system the three-dimensional simulation model helps the designer to perceive the work from operators point of view objectively and safely without the exposure to hazards of the actual system.

• 대한의용생체공학회:의공학회지
• /
• 제36권5호
• /
• pp.169-176
• /
• 2015

Implantable closed-loop epilepsy controllers require ideally both accurate epileptic seizure detection and low power consumption. On-chip oscillators can be used in implantable devices because they consume less power than other oscillators such as crystal oscillators. In this study, we investigated the tolerable error range of a lower power on-chip oscillator without losing the accuracy of seizure detection. We used 24 ictal and 14 interictal intracranial electroencephalographic segments recorded from epilepsy surgery patients. The performance variations with respect to oscillator frequency errors were estimated in terms of specificity, modified sensitivity, and detection timing difference of seizure onset using Generic Osorio Frei Algorithm. The frequency errors of on-chip oscillators were set at ${\pm}10%$ as the worst case. Our results showed that an oscillator error of ${\pm}10%$ affected both specificity and modified sensitivity by less than 3%. In addition, seizure onsets were detected with errors earlier or later than without errors and the average detection timing difference varied within less than 0.5 s range. The results suggest that on-chip oscillators could be useful for low-power implantable devices without error compensation circuitry requiring significant additional power. These findings could help the design of closed-loop systems with a seizure detector and automated stimulators for intractable epilepsy patients.

• ETRI Journal
• /
• 제37권6호
• /
• pp.1199-1210
• /
• 2015

Automatic emotion recognition based on facial cues, such as facial action units (AUs), has received huge attention in the last decade due to its wide variety of applications. Current computer-based automated two-phase facial emotion recognition procedures first detect AUs from input images and then infer target emotions from the detected AUs. However, more robust AU detection and AU-to-emotion mapping methods are required to deal with the error accumulation problem inherent in the multiphase scheme. Motivated by our key observation that a single AU detector does not perform equally well for all AUs, we propose a novel two-phase facial emotion recognition framework, where the presence of AUs is detected by group decisions of multiple AU detectors and a target emotion is inferred from the combined AU detection decisions. Our emotion recognition framework consists of three major components - multiple AU detection, AU detection fusion, and AU-to-emotion mapping. The experimental results on two real-world face databases demonstrate an improved performance over the previous two-phase method using a single AU detector in terms of both AU detection accuracy and correct emotion recognition rate.

• 대한치과교정학회지
• /
• 제51권2호
• /
• pp.77-85
• /
• 2021

Objective: To evaluate the accuracy of a multi-stage convolutional neural network (CNN) model-based automated identification system for posteroanterior (PA) cephalometric landmarks. Methods: The multi-stage CNN model was implemented with a personal computer. A total of 430 PA-cephalograms synthesized from cone-beam computed tomography scans (CBCT-PA) were selected as samples. Twenty-three landmarks used for Tweemac analysis were manually identified on all CBCT-PA images by a single examiner. Intra-examiner reproducibility was confirmed by repeating the identification on 85 randomly selected images, which were subsequently set as test data, with a two-week interval before training. For initial learning stage of the multi-stage CNN model, the data from 345 of 430 CBCT-PA images were used, after which the multi-stage CNN model was tested with previous 85 images. The first manual identification on these 85 images was set as a truth ground. The mean radial error (MRE) and successful detection rate (SDR) were calculated to evaluate the errors in manual identification and artificial intelligence (AI) prediction. Results: The AI showed an average MRE of 2.23 ± 2.02 mm with an SDR of 60.88% for errors of 2 mm or lower. However, in a comparison of the repetitive task, the AI predicted landmarks at the same position, while the MRE for the repeated manual identification was 1.31 ± 0.94 mm. Conclusions: Automated identification for CBCT-synthesized PA cephalometric landmarks did not sufficiently achieve the clinically favorable error range of less than 2 mm. However, AI landmark identification on PA cephalograms showed better consistency than manual identification.

• Smart Structures and Systems
• /
• 제32권2호
• /
• pp.111-121
• /
• 2023

This study proposes an automated assembly performance evaluation method for prefabricated steel structures (PSSs) using machine learning methods. Assembly component images were segmented using a modified version of the receptive field pyramid. By factorizing channel modulation and the receptive field exploration layers of the convolution pyramid, highly accurate segmentation results were obtained. After completing segmentation, the positions of the bolt holes were calculated using various image processing techniques, such as fuzzy-based edge detection, Hough's line detection, and image perspective transformation. By calculating the distance ratio between bolt holes, the assembly performance of the PSS was estimated using the k-nearest neighbors (kNN) algorithm. The effectiveness of the proposed framework was validated using a 3D PSS printing model and a field test. The results indicated that this approach could recognize assembly components with an intersection over union (IoU) of 95% and evaluate assembly performance with an error of less than 5%.

• 한국ITS학회 논문지
• /
• 제21권1호
• /
• pp.273-286
• /
• 2022

자율주행 4단계 이상에서는 차량 스스로가 목적지까지 주행하기 위해 차선 단위의 전역경로를 생성하는 것이 필수적이며, 이를 위해 정밀도로지도 활용에 관한 연구가 활발히 진행되고 있다. 정밀도로지도 기반 전역경로 생성을 위해서는 정확한 링크 정보를 통해 도로 네트워크를 구축하는 것이 필수적인데, 현재 공개된 정밀도로지도는 이 부분의 구현을 어렵게 하는 다양한 오류를 포함하는 것을 볼 수 있다. 이러한 배경을 바탕으로 본 연구에서는 정밀도로지도 기반 전역경로 생성을 위한 링크 오류 개선 및 도로 네트워크 구축에 관한 연구를 수행하였다. 전역경로 생성에 치명적일 수 있는 오류를 검출하고 링크를 포함한 정밀도로지도의 정보들을 활용하여 도로 네트워크를 구축하는 전처리 알고리즘을 개발하였다. 제안하는 방법을 통하여 실제 정밀도로지도로부터 정확한 전역경로를 생성할 수 있는 것을 확인함으로써 본 연구의 유효성을 검증하였다.

• 대한원격탐사학회지
• /
• 제39권4호
• /
• pp.371-393
• /
• 2023

Crack detection in structures plays a vital role in ensuring their safety, durability, and reliability. Traditional crack detection methods sometimes need significant manual inspections, which are laborious, expensive, and prone to error by humans. Deep learning algorithms, which can learn intricate features from large-scale datasets, have emerged as a viable option for automated crack detection recently. This study presents an in-depth review of crack detection methods used till now, like image processing, traditional machine learning, and deep learning methods. Specifically, it will provide a comparative analysis of crack detection methods using deep learning, aiming to provide insights into the advancements, challenges, and future directions in this field. To facilitate comparative analysis, this study surveys publicly available crack detection datasets and benchmarks commonly used in deep learning research. Evaluation metrics employed to check the performance of different models are discussed, with emphasis on accuracy, precision, recall, and F1-score. Moreover, this study provides an in-depth analysis of recent studies and highlights key findings, including state-of-the-art techniques, novel architectures, and innovative approaches to address the shortcomings of the existing methods. Finally, this study provides a summary of the key insights gained from the comparative analysis, highlighting the potential of deep learning in revolutionizing methodologies for crack detection. The findings of this research will serve as a valuable resource for researchers in the field, aiding them in selecting appropriate methods for crack detection and inspiring further advancements in this domain.