• Journal of IKEEE
• /
• v.22 no.3
• /
• pp.805-809
• /
• 2018

In this paper, we analyze the change of accuracy when fixed point arithmetic is used instead of floating point arithmetic in binary weight network(BWN). We observed the change of accuracy by varying total bit size and fraction bit size. If the integer part is not changed after fixed point approximation, there is no significant decrease in accuracy compared to the floating-point operation. When overflow occurs in the integer part, the approximation to the maximum or minimum of the fixed point representation minimizes the decrease in accuracy. The results of this paper can be applied to the minimization of memory and hardware resource requirement in the implementation of FPGA-based BWN accelerator.

• Journal of Practical Engineering Education
• /
• v.9 no.1
• /
• pp.7-13
• /
• 2017

This study will provide a practical plan of engineering education through the study of major activities connected with the production of works to accomplish the graduation conditions by completing the comprehensive design subject and the result of the performance. The designed subject is to measure the minimum safety distance during driving using the obstacle detection function of the ultrasonic sensor and to perform the avoidance algorithm based on the measurement value of the acceleration gyro sensor. It is proposed an access surveillance system that minimizes the damage of drones, surrounding objects, and people, and improves air mobility. Experimental results show that the obstacles around the drone are detected by five ultrasonic sensors and the difference of output value is applied to each motor of the drone and obstacle avoidance is confirmed. In addition, the content and level of the data for measuring the achievement of learning achievement in the engineering education certification program were used and the results were confirmed to be consistent with the description of the engineering problem level required for the graduates of 4-year engineering college.

• KIPS Transactions on Computer and Communication Systems
• /
• v.8 no.4
• /
• pp.93-102
• /
• 2019

As interest in wellness grows, There is a lot of research about monitoring individual health using wearable devices. Accordingly, a variety of methods have been studied to distinguish exercise from daily activities using wearable devices. Most of these existing studies are machine learning methods. However, there are problems with over-fitting on individual person's learning, data discontinuously recognition by independent segmenting and fake activity. This paper suggests a detection method for exercise activity based on the physiological response principle of heart rate up and down during exercise. This proposed method calculates activity intensity and heart rate from triaxial and photoplethysmography sensor to determine a heart rate recovery, then detects exercise by estimating activity intensity or detecting a heart rate rising state. Experimental results show that our proposed algorithm has 98.64% of averaged accuracy, 98.05% of averaged precision and 98.62% of averaged recall.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.6
• /
• pp.137-141
• /
• 2022

The method of estimating the azimuth by measuring the geomagnetism has been used for a very long time. However, there are many cases where an error occurs in the estimated azimuth due to disturbances in the earth's magnetic field due to metal structures inside and outside the room. Although many studies have been conducted to correct this, there is a limit to reducing the error. In this paper, we propose a method of estimating the azimuth by applying the measured geomagnetic sensor data to the neural network of the LSTM structure. Data preprocessing is very important for learning a neural network. In this paper, data is collected using the built-in acceleration sensor, gyro sensor, and geomagnetic sensor in the smartphone, and the geomagnetic sensor data is uniformly sampled using EKF. As a result, an average azimuth estimation error of 0.9 degrees was obtained using four hidden layers.

• Science of Emotion and Sensibility
• /
• v.13 no.1
• /
• pp.103-110
• /
• 2010

TPR(Total Physical Response) is a new representative learning method for children's education. Today's approach to TPR has focused on signals from a user which becomes input data in a human-computer interaction, but the accuracy of sensing from body signals(e. g. motion and voice) isn't so perfect that it seems difficult to apply on an education system. To overcome these limits, we suggest a mediating interface device which can detect the user's motion using correct numerical values such as acceleration and angular speed. In addition, we suggest new design requirements for the mediating device through analyzing children's behavior as human factors by ethnography research and protocol analysis. As a result, we found that; children are unskilled in physical control when they use objects; tend to lean on an object unconsciously with touch. Also their behaviors are restricted, when they use objects. Therefore a mediating device should satisfy new design requirements which are make up for unskilled handling, support familiar and natural physical activity.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.16 no.4
• /
• pp.472-478
• /
• 2006

This paper presents a new fuzzy-neural-network based interacting multiple model (FNNBIMM) algorithm for tracking a maneuvering target. To effectively handle the unknown target acceleration, this paper regards it as additional noise, time-varying variance to target model. Each sub model characterized by the variance of the overall process noise, which is obtained on the basis of each acceleration interval. Since it is hard to approximate this time-varying variance adaptively owing to the unknown acceleration, the FNN is utilized to precisely approximate this time-varying variance. The error back-propagation method is utilized to optimize each FNN. To show the feasibility of the proposed algorithm, a numerical example is provided.

• Journal of Biomedical Engineering Research
• /
• v.21 no.5
• /
• pp.527-535
• /
• 2000

For a reliable patient set-up verification, better portal films are needed to track relevant features. Simulator films are compared with portal films as a reference image in radiotherapy planning. This shows some possibilities of the use of image information of simulator images for enhancement and restorations of portal images which are very poor in quality compared with the simulator images. This paper present an approach that combine an associative memory, a kind of artificial neural networks with fuzzy image enhancement technique using genetic algorithm which determines the fuzzy region of membership function by the use of maximum entropy principles. A higher portal image quality than conventional technique is achieved.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.1153-1156
• /
• 2019

본 논문에서 제안하는 음주로 인한 사회 연결망 와해 방지를 위한 스마트 애플리케이션 '나 취했니?'는 과도한 음주 후 스마트폰 사용시 발생할 수 있는 사용자의 의도치 않은 실수를 예방하는 것을 목표로 한다. 가속도 변화, 터치 속도 및 오타 수 확인, 심박수 변화 등을 통하여 사용자의 데이터를 얻고 기계학습을 통해 사용자의 음주 여부를 판단한다. 이때, 음주를 한 것으로 판단되면 사용자가 설정해 두었던 애플리케이션들의 사용을 제한하게 된다. 또한 애플리케이션에 사용자가 음주를 한 날짜를 체크하는 자가진단 서비스 및 음주를 한 것으로 확인된 직후 GPS 를 이용해 사용자의 위치 및 이동 경로를 기록하여 스마트폰 내의 사고 뿐만 아니라 외부적인 사고에도 도움을 줄 수 있도록 구현하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.1384-1387
• /
• 2015

고령화 사회로 진입하면서 노인들은 노화과정에 의한 보행능력의 감소 및 근력 약화와 같은 신체적 변화로 인해 잦은 낙상을 경험한다. 이에 따라 낙상 사고를 감지하는 연구가 활발히 진행되고 있다. 낙상은 사전 예방도 중요하지만 사고 발생 후의 신속한 대처도 중요하다. 낙상을 감지하고 의료진에게 즉시 낙상정보를 제공하여 후속적 조치를 취하는 것은 사고 후 대처의 핵심이다. 본 논문에서는 스마트폰 환경에서 사용자의 낙상 후 방향을 판별하기 위해 두 가지 센서 데이터의 특정 값들을 추출하였으며, 이에 5 가지 기계학습 알고리즘을 적용하였다. 사용자는 스마트폰을 착용한 상태로 전후좌우 4 방향 낙상 실험을 진행하며 스마트폰 내에 내장된 3 축 가속도 센서와 3 축 자이로 센서값을 측정한다. 피험자 11 명을 대상으로 낙상 실험 결과, 5 가지의 분류기 중 k-NN에서 98.6%의 인식률을 나타내었다. 뽑아낸 특징 값과 분류 알고리즘은 낙상의 방향 검출에 유용한 것으로 판단된다.

• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.11
• /
• pp.537-542
• /
• 2017

Aggressive driving is a major cause of car accidents. Previous studies have mainly analyzed young driver's aggressive driving tendency, yet they were only done through pure clustering or classification technique of machine learning. However, since elderly people have different driving habits due to their fragile physical conditions, it is necessary to develop a new method such as enhancing the characteristics of driving data to properly analyze aggressive driving of elderly drivers. In this study, acceleration data collected from a smartphone of a driving vehicle is analyzed by a newly proposed ECA(Enhanced Clustering method for Acceleration data) technique, coupled with a conventional clustering technique (K-means Clustering, Expectation-maximization algorithm). ECA selects high-intensity data among the data of the cluster group detected through K-means and EM in all of the subjects' data and models the characteristic data through the scaled value. Using this method, the aggressive driving data of all youth and elderly experiment participants were collected, unlike the pure clustering method. We further found that the K-means clustering has higher detection efficiency than EM method. Also, the results of K-means clustering demonstrate that a young driver has a driving strength 1.29 times higher than that of an elderly driver. In conclusion, the proposed method of our research is able to detect aggressive driving maneuvers from data of the elderly having low operating intensity. The proposed method is able to construct a customized safe driving system for the elderly driver. In the future, it will be possible to detect abnormal driving conditions and to use the collected data for early warning to drivers.