• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.5-6
• /
• 2021

본 논문에서는 Semi-supervised Learning 방식의 이상감지 방법을 제안한다. 취득한 소음 데이터를 이미지화 시킨 후 Convolution AutoEncoder 학습 방법을 이용하여 모델을 학습한다. 고장 데이터와 정상 데이터 간의 데이터 불균형 문제가 대두되기 때문에 정상 데이터만을 활용한 이상감지는 실제 산업현장의 상황에 알맞게 사용할 수 있을 것이라 기대한다.

• Journal of Korea Multimedia Society
• /
• v.8 no.11
• /
• pp.1496-1509
• /
• 2005

In this paper, an adaptive license plate recognition system which detects and recognizes license plate at real-time by using projected plane convolution and Decision Tree Classifier is proposed. And it was tested in circumstances which presence of complex background. Generally, in expressway tollgate or gateway of parking lots, it is very difficult to detect and segment license plate because of size, entry angle and noisy problem of vehicles due to CCD camera and road environment. In the proposed algorithm, we suggested to extract license plate candidate region after going through image acquisition process with inputted real-time image, and then to compensate license size as well as gradient of vehicle with change of vehicle entry position. The proposed algorithm can exactly detect license plate using accumulated edge, projected convolution and chain code labeling method. And it also segments letter of license plate using adaptive binary method. And then, it recognizes license plate letter by applying hybrid pattern vector method. Experimental results show that the proposed algorithm can recognize the front and rear direction license plate at real-time in the presence of complex background environments. Accordingly license plate detection rate displayed $98.8\%$ and $96.5\%$ successive rate respectively. And also, from the segmented letters, it shows $97.3\%$ and $96\%$ successive recognition rate respectively.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.8C
• /
• pp.1184-1192
• /
• 2004

A real-time pupil detection system that detects the pupil movement from the real-time video data achieved by the visual light camera for general purpose personal computer is proposed. It is implemented with three steps; at first, face region is detected using the Haar-like feature detection scheme, and then eye region is detected within the face region using the template-based scheme. Finally, pupil movement is detected within the eye region by convolution of the horizontal and vertical histogram profiling and Gaussian filter. As results, we obtained more than 90% of the detection rate from 2375 simulation images and the data processing time is about 160㎳, that detects 7 times per second.

• International journal of advanced smart convergence
• /
• v.9 no.2
• /
• pp.20-27
• /
• 2020

Sound event detection in real-world environments suffers from the interference of non-stationary and time-varying noise. This paper presents an adaptive noise reduction method for sound event detection based on non-negative matrix factorization (NMF). In this paper, we proposed a deep learning model that integrates Convolution Neural Network (CNN) with Non-Negative Matrix Factorization (NMF). To improve the separation quality of the NMF, it includes noise update technique that learns and adapts the characteristics of the current noise in real time. The noise update technique analyzes the sparsity and activity of the noise bias at the present time and decides the update training based on the noise candidate group obtained every frame in the previous noise reduction stage. Noise bias ranks selected as candidates for update training are updated in real time with discrimination NMF training. This NMF was applied to CNN and Hidden Markov Model(HMM) to achieve improvement for performance of sound event detection. Since CNN has a more obvious performance improvement effect, it can be widely used in sound source based CNN algorithm.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.1
• /
• pp.115-126
• /
• 2023

Due to the fluctuating random and periodical nature of renewable energy generation power quality disturbances occurred more frequently in power generation transformation transmission and distribution. Various power quality disturbances may lead to equipment damage or even power outages. Therefore it is essential to detect and classify different power quality disturbances in real time automatically. The traditional PQD identification method consists of three steps: feature extraction feature selection and classification. However, the handcrafted features are imprecise in the feature selection stage, resulting in low classification accuracy. This paper proposes a deep neural architecture based on Convolution Neural Network and Long Short Term Memory combining the time and frequency domain features to recognize 16 types of Power Quality signals. The frequency-domain data were obtained from the Fast Fourier Transform which could efficiently extract the frequency-domain features. The performance in synthetic data and real 6kV power system data indicate that our proposed method generalizes well compared with other deep learning methods.

• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.149-152
• /
• 1999

This paper addresses major issues in localising sound sources in space using the experimental data set of head-related responses in the time or frequency domain. They come from the technical realisation steps for implementing the convolution of HRIR's with sound sources, the cross-talk cancellation for transaural filtering, the matched time delay compensation, etc. in real, those technical matters seem to be minor because they can be realised in off-line signal processing schemes. This paper puts much emphasis on what we misunderstood about the sets of HRTF's or HRIR's, More specifcaily, the sets of HRTF's or HRIR's of course supply relevant information to sound localisation but include much useless 'rubbish' that have made for us to fail to put spatial image into real souno signals such as voices and music's. This paper proposes possible reasons for such failure and, furthermore, introduces detained subjects that should be challenged so as to resolve them.

• Journal of Biomedical Engineering Research
• /
• v.8 no.2
• /
• pp.117-117
• /
• 1987

The medical image processing system is intended for a diverse set of users in the medical Imaging Parts. This system consists of a 640 Kbyte IBM-PC/AT with 30 Mbyte hard disk, special purpose image processor with video input devices and display monitor. Image may be recorded and processed in real time at sampling rate up to 10 MHz. This system provides a wide range of image enhancement processing facilities via a menu-driven software packages. These facilities include point by point processing, image averaging, convolution filter and subtraction.

• Journal of the Korean Institute of Navigation
• /
• v.16 no.4
• /
• pp.55-63
• /
• 1992

This computer simulation is the basic research for realize a real-time hardware of the reproduction system in original sound field with two loudspeakers based on the OSS(Ortho Stereophonic System) method which was proposed by Hamada of Japan in 1983. Through the computer simulation, presumed the system function of OSS equalizer using HRTF(Head Related Transfer Function), constructed the model of OSS equalizer and , evaluated the modelling OSS equalizer by evaluation formula. The obtained results are summarized as follows : 1) By the modelling OSS equalize operate as inverse filter of HRTF, an input signal reproduced effectively. 2) Known that the real-time hardware of OSS equalizer can be made by the fast convolution between the impulse response of OSS equalizer and input speech signal. 3) Since the system function of OSS equalizer presumed from HRTF, the study on the measuring of HRTF have to proceed.

• Journal of the Semiconductor & Display Technology
• /
• v.19 no.2
• /
• pp.88-91
• /
• 2020

Image segmentation is the task of partitioning an image into multiple sets of pixels based on some characteristics. The objective is to simplify the image into a representation that is more meaningful and easier to analyze. In this paper, we apply deep-learning to pre-train the learning model, and implement an algorithm that performs image segmentation in real time by extracting frames for the stream input from the Android device. Based on the open source of DeepLab-v3+ implemented in Tensorflow, some convolution filters are modified to improve real-time operation on the Android platform.

• Journal of Korea Multimedia Society
• /
• v.25 no.11
• /
• pp.1582-1590
• /
• 2022

Due to the situation of the widespread of the coronavirus, which causes the problem of lack of face image data occluded by masks at recent time, in order to solve the related problems, this paper proposes a method to generate face images with masks using a combination of generative adversarial networks and spatial transformation networks based on CNN model. The system we proposed in this paper is based on the GAN, combined with multi-scale convolution kernels to extract features at different details of the human face images, and used Wasserstein divergence as the measure of the distance between real samples and synthetic samples in order to optimize Generator performance. Experiments show that the proposed method can effectively put masks on face images with high efficiency and fast reaction time and the synthesized human face images are pretty natural and real.