• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.6
• /
• pp.157-165
• /
• 2020

Aiming at the development of neural network and self-driving data set, it is also an idea to improve the performance of network model to detect moving objects by dividing the data set. In Darknet network framework, the YOLOv4 (You Only Look Once v4) network model was used to train and test Udacity data set. According to 7 proportions of the Udacity data set, it was divided into three subsets including training set, validation set and test set. K-means++ algorithm was used to conduct dimensional clustering of object boxes in 7 groups. By adjusting the super parameters of YOLOv4 network for training, Optimal model parameters for 7 groups were obtained respectively. These model parameters were used to detect and compare 7 test sets respectively. The experimental results showed that YOLOv4 can effectively detect the large, medium and small moving objects represented by Truck, Car and Pedestrian in the Udacity data set. When the ratio of training set, validation set and test set is 7:1.5:1.5, the optimal model parameters of the YOLOv4 have highest detection performance. The values show mAP50 reaching 80.89%, mAP75 reaching 47.08%, and the detection speed reaching 10.56 FPS.

• Journal of the Korean Society of Radiology
• /
• v.18 no.5
• /
• pp.557-566
• /
• 2024

Ultrasound-guided regional anesthesia is one of the most common techniques used in peripheral nerve blockade by enhancing pain control and recovery time. However, accurate Brachial Plexus (BP) nerve detection and identification remains a challenging task due to the difficulty in data acquisition such as speckle and Doppler artifacts even for experienced anesthesiologists. To mitigate the issue, we introduce a BP nerve small target segmentation network by incorporating BP object detection and U-Net based semantic segmentation into a single deep learning framework based on the multi-scale approach. To this end, the current BP detection and identification was estimated: 1) A RetinaNet model was used to roughly locate the BP nerve region using multi-scale based feature representations, and 2) U-Net was then used by feeding plural BP nerve features for each scale. The experimental results demonstrate that our proposed model produces high quality BP segmentation by increasing the accuracies of the BP nerve identification with the assistance of roughly locating the BP nerve area compared to competing methods such as segmentation-only models.

• Korean Journal of Food Science and Technology
• /
• v.39 no.1
• /
• pp.71-76
• /
• 2007

Foodborne illness caused by Noroviruses (NVs) is increasing rapidly in Korea. This study developed an effective detection protocol for NVs found in contaminated oysters and lettuce through an investigation using the major steps of virus particle separation, concentration and RT-PCR. As a surrogate model for NVs, the cultivable feline calicivirus (FCV) that belongs to the same Caliciviridae family was used. Instead of using a time-consuming ultracentrifugation method, efficient methods based on solvent extraction and PEG precipitation procedure were applied. Direct homogenization of a 25g sample of whole oyster and lettuce in 175mL PBS provided the simplicity that would be needed in the actual field of food product examination. The overnight PEG precipitation step at $4^{\circ}C$ was reduced to 3 h by placing the reaction tube in ice and by adjusting the PEG concentrations. The application of the use of chloroform and 0.2 ${\mu}m$ syringe filtration together showed a better detection efficiency than the use of chloroform alone in removing PCR inhibitors for both oyster and lettuce samples. Also, dilution of the extracted RNA solution before PCR provided increased sensitivity. The improved detection protocol developed in this study could be efficiently applied to detect FCV and most likely NVs from oysters and lettuce.

• The KIPS Transactions:PartB
• /
• v.10B no.4
• /
• pp.381-388
• /
• 2003

This paper presents a method of giving weights to garbage class clustering and Filler model to improve performance of keyword spotting system and a time-saving method of dialogue speech processing system for keyword spotting by calculating keyword transition probability through speech analysis of task domain users. The point of the method is grouping phonemes with phonetic similarities, which is effective in sensing similar phoneme groups rather than individual phonemes, and the paper aims to suggest five groups of phonemes obtained from the analysis of speech sentences in use in Korean morphology and in stock-trading speech processing system. Besides, task-subject Filler model weights are added to the phoneme groups, and keyword transition probability included in consecutive speech sentences is calculated and applied to the system in order to save time for system processing. To evaluate performance of the suggested system, corpus of 4,970 sentences was built to be used in task domains and a test was conducted with subjects of five people in their twenties and thirties. As a result, FOM with the weights on proposed five phoneme groups accounts for 85%, which has better performance than seven phoneme groups of Yapanel [1] with 88.5% and a little bit poorer performance than LVCSR with 89.8%. Even in calculation time, FOM reaches 0.70 seconds than 0.72 of seven phoneme groups. Lastly, it is also confirmed in a time-saving test that time is saved by 0.04 to 0.07 seconds when keyword transition probability is applied.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10d
• /
• pp.502-504
• /
• 2002

군중잡음(crowd noise)이 발생하는 환경에서 음성 통화 및 화자 인식을 할 때에는 음성에 파열음이나 마찰음과 같은 유색잡음(colored noise)이 부가되어 원래 음성이 왜곡된다. 이와 같이 왜곡된 음성 신호를 처리할 때에는 군중잡음을 제거하는 과정이 반드시 필요하다. 본 논문에서는 군중잡음의 특성을 분석하고, 그 결과를 이용하여 음성 신호처리 시에 효과적으로 군중잡음만을 제거할 수 있는 모델을 제안한다. 제안된 모델은 시간 영역에서는 침묵 구간을 검출하여 마찰음과 파열음을 제거하는 과정과 주파수 영역에서는 잡음 평균을 생성하고 이를 이용한 스펙트럼 차감법(spectral subtraction)으로 군중 잡음을 제거하는 과정으로 이루어진다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10d
• /
• pp.472-474
• /
• 2002

본 논문에서는 복잡한 영상에서 얼굴 영역의 윤곽선을 검출하는 방법을 제안하였다. 이를 위하여 얼굴의 칼라 정보에 기반한 액티브 컨투어 모델을 이용하였다. 얼굴의 칼라 정보는 색채칼라 공간(chromatic color space)에서 2D-Gaussian모델로 나타내어지는 스킨 칼로 모델로 표현 되었다. 실험결과 제안된 방법은 복잡한 영상뿐 아니라 잡음이 많은 영상에서 하나 또는 여러 개의 얼굴 영역을 추출할 수 있었다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04a
• /
• pp.859-861
• /
• 2001

본 논문은 오디오의 masking특성을 적용한 심리음향 모델을 이용하여 오디오의 고음질을 보장하면서 잡음과 압축 등의 공격에 강한 오디오 워터마킹 방법을 제안한다. 제안하는 워터마킹 방법은 심리음향 모델에 의해 생산되는 masking thresholds와 원신호의 power spectral density의 각 주파수별 차이 에너지를 이용하여 시간도메인에서 워터마크를 삽입하는 방법으로 오디오의 품질을 유지할 수 있다. 워터마크로는 자기상관성이 강한 PN-시퀀스를 이용하여 강인한 워터마킹을 구현한다. 그리고 PN-시퀀스와 같은 이진 시퀀스 워터마크의 검출을 위한 유사도 측정식을 제안한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10a
• /
• pp.409-411
• /
• 2001

본 논문에서는 소프트웨어 테스트 단계중에 발생되는 테스트노력 소요량을 고려한 소프트웨어 신뢰도 성장 모델을 제시하여 시간종속적인 테스트 노력소요량 동태를 일정 테스트 노력일 때와 웨이블 테스트 노력일 때를 비교하여 연구한다. 테스트 단계중에 소요되는 테스트노력의 양에 대한 결함 검출비를 현재의 절함 내용에 비례하는 것으로 가정하여 모델을 NHPP로 공식화하되, 이 모델을 이용하여 소프트웨어 신뢰도 척도에 대한 데이터 분식기법을 개발한다. 테스트 시간의 경과와 신뢰도와의 관계를 연구한다. 목표신뢰도를 만족시키는 최적발행시각을 정한다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.4 no.5
• /
• pp.1017-1025
• /
• 2000

The human auditory system has been shown to posses remarkable abilities in the localization and tracking of sound sources. The localization is the result of processing two primary acoustics cues. These are the interaural time difference(ITD) cues and interaural intensity difference(IID) cues at the two ears. In this paper, we propose TEPILD(Time Energy Previous Integration Location Detector) model. TEPILD model is constructed with time function generator, energy function generator, previous location generator and azimuth detector. Time function generator is to process ITD and energy function generator is to process IID. Total average accuracy rate is 99.2%. These result are encouraging and show that proposed model can be applied to the sound source location detector.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.12C
• /
• pp.1157-1164
• /
• 2006

In this paper, we proposed a method for measuring accurate positions and sizes of package and balls in a micro BGA. To find defects of BGA accurately, we focused on finding positions of package and balls. After labeling, we detected connected components of package and balls using feature parameters. After the detection of package component, we measured position and size of package by employing rectangular model which was constructed by the package information. After the detection of the ball components, we measured positions and diameters of balls by employing circular models which were constructed by the ball informations. We did calibration based on landmarks to measure real length, and we compared the measured results with the SEM data. Finally, we found that the accuracy of the proposed method is 94% in terms of ball's radius.