• The Journal of Korea Robotics Society
• /
• v.15 no.3
• /
• pp.269-276
• /
• 2020

As we become an aging society, the number of elderly patients continues to increase. Pressure sores that can easily occur in patients with trauma cause serious socio-economic problems. In general, prevention of bedsores through predicting the patient's posture is being developed. Developed method usually use artificial intelligence techniques to estimate the patient's posture by measured pressure images in the mattress. In this method, it has a problem the reduction of estimation accuracy when posture of patient is changed. Therefore, it is necessary to use the filter of pressure images in the position transition of patient. In this paper, we propose an algorithm to predict the patient's posture, and an algorithm to reduce the ambiguity that can occur in the patient's posture transition section. By obtaining stable data through this algorithm, learning/prediction stability of the neural network can be expected, and prediction performance is improved accordingly. Through experiments, the effectiveness of the algorithm was verified.

• Journal of the Institute of Convergence Signal Processing
• /
• v.5 no.1
• /
• pp.32-39
• /
• 2004

Biological neurons generally have chaotic characteristics for permanent or transient period. The effects of chaotic response of biological neuron have not yet been verified by using analytical methods. Even though the transient chaos of neuron could be beneficial to overcoming the local minimum problem, the permanent chaotic response gives adverse effect on optimization problems in general. To solve optimization problems, which are needed in almost all neural network applications such as pattern recognition, identification or prediction, and control, the neuron should have one stable fixed point. In this paper, the dynamic characteristics of the chaotic dynamic neuron and the condition that produces the chaotic response are analyzed, and the convergence conditions are presented.

• Annual Conference of KIPS
• /
• 2017.11a
• /
• pp.781-784
• /
• 2017

Face recognition systems are characterized by low invasiveness of acquisition, and increasingly better reliability. Such systems may not be applied effectively, when the images are in low resolution (LR) as in the case that photos are taken from long distances, typically public surveillance. In theory, the high resolution (HR) image reconstructed from an LR face image, applying a super resolution (SR) method, can be used for face recognition. However, existing face SR algorithms may not give satisfactory results in face recognition. This article investigates the very low resolution face recognition problem and introduces a partial differential equation (PDE)-based SR method for a face recognition system of convolutional neural network (CNN).

• Proceedings of the Korean Association of Geographic Inforamtion Studies Conference
• /
• 2004.03a
• /
• pp.487-492
• /
• 2004

본 연구에서는 인공신경망 기법을 이용하여 산사태 발생원인에 대한 가중치를 구하였다. 여름철 집중호우시 산사태가 많이 발생하는 강원도 강릉시 사천면 사기막리 와 주문진읍 삼교리에 해당한다. 산사태가 발생할 수 있는 요인으로 지형도로부터 경사, 경사방향, 곡률, 수계추출을, 정밀토양도로부터 토질, 모재, 배수, 유효토심, 지형을, 임상도로부터 임상, 경급, 영급, 밀도를, 지질도로부터 암상을, Landsat TM 영상으로부터 토지이용도와 추출하여 격자화 하였으며, 아리랑1호 영상으로부터 선구조를 추출하여 l00m 간격으로 버퍼링 한 후 격자화 하였다. 이렇게 구축된 산사태 발생 위치 및 발생요인 데이터 베이스를 이용하여 인공신경망 기법을 적용하여 산사태 발생 원인에 대한 상대적인 가중치를 구하였다. 인공신경망의 역전파 알고리즘을 이용한 사기막리 지역과 삼교리 지역의 산사태 가중치를 보면 GPS를 이용한 현장조사와 위성영상을 이용한 변화탐지 기법모두의 경우모두와 훈련지역을 실제 산사태 발생 지역과 경사도가 0°인 지역, 실제 산사태 발생 지역과 Frequence ratio를 이용하여 작성한 취약성도에서 산사태 발생이 낮을 것으로 예상되는 지역, Frequence ratio를 이용한 취약성도에서 산사태 발생이 높을 것으로 예상되는 지역 과 낮을 것으로 예상되는 지역의 경우에서도 경사도는 1.5～2.5배정도 높은 상대적 가중치를 나타냈다. 이러한 가중치는 산사태 취약성도를 작성하는데 활용될 수 있다.

• International Journal of Control, Automation, and Systems
• /
• v.4 no.3
• /
• pp.372-381
• /
• 2006

The purpose of this study was to develop a process management system to manage ingot fabrication and improve ingot quality. The ingot is the first manufactured material of wafers. Trace parameters were collected on-line but measurement parameters were measured by sampling inspection. The quality parameters were applied to evaluate the quality. Therefore, preprocessing was necessary to extract useful information from the quality data. First, statistical methods were used for data generation. Then, modeling was performed, using the generated data, to improve the performance of the models. The function of the models is to predict the quality corresponding to control parameters. Secondly, rule extraction was performed to find the relation between the production quality and control conditions. The extracted rules can give important information concerning how to handle the process correctly. The dynamic polynomial neural network (DPNN) and decision tree were applied for data modeling and rule extraction, respectively, from the ingot fabrication data.

• Journal of Institute of Control, Robotics and Systems
• /
• v.14 no.2
• /
• pp.168-177
• /
• 2008

Path finding is a key element in the navigation of a mobile robot. To find a path, robot should know their position exactly, since the position error exposes a robot to many dangerous conditions. It could make a robot move to a wrong direction so that it may have damage by collision by the surrounding obstacles. We propose a method obtaining an accurate robot position. The localization of a mobile robot in its working environment performs by using a vision system and Virtual Reality Modeling Language(VRML). The robot identifies landmarks located in the environment. An image processing and neural network pattern matching techniques have been applied to find location of the robot. After the self-positioning procedure, the 2-D scene of the vision is overlaid onto a VRML scene. This paper describes how to realize the self-positioning, and shows the overlay between the 2-D and VRML scenes. The suggested method defines a robot's path successfully. An experiment using the suggested algorithm apply to a mobile robot has been performed and the result shows a good path tracking.

• Proceedings of the IEEK Conference
• /
• 2002.07c
• /
• pp.1424-1427
• /
• 2002

This paper proposes a novel postprocessing algorithm for reducing the blocking artifacts in low bit rate block-based transform coded images, that use adaptive neural network filter (NNF) in wavelet transform domain. n this algorithm, after performing a 2-level wavelet transform of the decompressed image, the existence of locking artifacts is determined using statistical characteristic of neighborhood blocks. And then a different one-dimensional (1-D) or 2-D NNF is used to reduce the locking artifacts according to the classified regions. That is, for HL and LH subbands regions with the blocking artifacts, a different 1-D NNF is used. And 2-D NNF is used in HH subband. Experimental results show that the proposed algorithm produced better results than those of conventional algorithms both subjectively and objectively.

• International Journal of Advanced Culture Technology
• /
• v.6 no.1
• /
• pp.60-67
• /
• 2018

In this paper, the pattern of extracting the same expression is proposed by using the Android intelligent device to identify the facial expression. The understanding and expression of expression are very important to human computer interaction, and the technology to identify human expressions is very popular. Instead of searching for the symbols that users often use, you can identify facial expressions with a camera, which is a useful technique that can be used now. This thesis puts forward the technology of the third data is available on the website of the set, use the content to improve the infrastructure of the facial expression recognition accuracy, to improve the synthesis of neural network algorithm, making the facial expression recognition model, the user's facial expressions and similar expressions, reached 66%. It doesn't need to search for symbols. If you use the camera to recognize the expression, it will appear symbols immediately. So, this service is the symbols used when people send messages to others, and it can feel a lot of convenience. In countless symbols, there is no need to find symbols, which is an increasing trend in deep learning. So, we need to use more suitable algorithm for expression recognition, and then improve accuracy.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• v.31 no.3
• /
• pp.105-120
• /
• 2020

Although autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by social impairments, patients with ASD frequently manifest atypical sensory behaviors. Recently, atypical sensory perception in ASD has received much attention, yet little is known about its cause or neurobiology. Herein, we review the findings from neuroimaging studies related to visual perception in ASD. Specifically, we examined the neural underpinnings of visual detection, motion perception, and face processing in ASD. Results from neuroimaging studies indicate that atypical visual perception in ASD may be influenced by attention or higher order cognitive mechanisms, and atypical face perception may be affected by disrupted social brain network. However, there is considerable evidence for atypical early visual processing in ASD. It is likely that visual perceptual abnormalities are independent of deficits of social functions or cognition. Importantly, atypical visual perception in ASD may enhance difficulties in dealing with complex and subtle social stimuli, or improve outstanding abilities in certain fields in individuals with Savant syndrome. Thus, future research is required to elucidate the characteristics and neurobiology of autistic visual perception to effectively apply these findings in the interventions of ASD.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.8
• /
• pp.105-113
• /
• 1996

In this paper, a new korean digita speech recognition technique was proposed using muktolayer perceptron (MLP). In spite of its weakness in dynamic signal recognition, MLP was adapted for this model, cecause korean syllable could give static features. It is so simle in its structure and fast in its computing that MLP was used to the suggested system. MLP's input vectors was transformed using karhunen-loeve transformation (KLT), which compress signal successfully without losin gits separateness, but its physical properties is changed. Because the suggested technique could extract static features while it is not affected from the changes of syllable lengths, it is effectively useful for korean numeric recognition system. Without decreasing classification rates, we can save the time and memory size for computation using KLT. The proposed feature extraction technique extracts same size of features form the tow same parts, front and end of a syllable. This technique makes frames, where features are extracted, using unique size of windows. It could be applied for continuous speech recognition that was not easy for the normal neural network recognition system.