• Journal of the Korea Society of Computer and Information
• /
• v.11 no.1 s.39
• /
• pp.19-25
• /
• 2006

Iris recognition is a bio metric personal identification which uses iris pattern of the pupil, and it is recognized as one of the best technology in personal identification and information security field. Before iris recognition, it is very important to search center of pupil. In recent years, there was developed many searching algorithms of center of pupil, but all most method are too many processing time. In this paper, we proposed a new method for searching center of pupil. This method is greatly reduced processing time about 30% compared with other algorithm using Hough transformation.

• Journal of Sensor Science and Technology
• /
• v.31 no.1
• /
• pp.16-23
• /
• 2022

Cardiac disease is the most common cause of death worldwide. Therefore, detection and classification of electrocardiogram (ECG) signals are crucial to extend life expectancy. In this study, we aimed to implement an artificial intelligence signal recognition system in field programmable gate array (FPGA), which can recognize patterns of bio-signals such as ECG in edge devices that require batteries. Despite the increment in classification accuracy, deep learning models require exorbitant computational resources and power, which makes the mapping of deep neural networks slow and implementation on wearable devices challenging. To overcome these limitations, spiking neural networks (SNNs) have been applied. SNNs are biologically inspired, event-driven neural networks that compute and transfer information using discrete spikes, which require fewer operations and less complex hardware resources. Thus, they are more energy-efficient compared to other artificial neural networks algorithms.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.1001-1004
• /
• 2021

Pain is an unpleasant experience for the patient. The recognition and assessment of pain help tailor the treatment to the patient, and they are also challenging in the medical. In this paper, we propose an approach for pain recognition through a deep neural network applied to pre-processed physiological. The proposed approach applies the idea of shortcut connections to concatenate the spatial information of a convolutional neural network and the temporal information of a recurrent neural network. In addition, our proposed approach applies the attention mechanism and achieves competitive performance on the BioVid Heat Pain dataset.

• Journal of Sensor Science and Technology
• /
• v.31 no.3
• /
• pp.139-144
• /
• 2022

Advanced tactile sensors are receiving significant attention in various industries such as extended reality, electronic skin, organic user interfaces, and robotics. The capabilities of advanced tactile sensors require a variety of functions, including position sensing, pressure sensing, and material recognition. Moreover, they should comsume less power and be bio-friendly with human contact. Recently, a tactile sensor based on the triboelectrification effect was developed. Triboelectric tactile sensors have the advantages of wide material availability, simple structure, and low manufacturing cost. Because they generate electricity by contact, they have low power consumption compared to conventional tactile sensors such as capacitive and piezoresistive. Furthermore, they have the ability to recognize the contact material as well as execute position and pressure sensing functions using the triboelectrification effect. The aim of this study is to introduce the progress of research on triboelectrification-based tactile sensors with various functions such as position sensing, pressure sensing and contact material recognition.

• Journal of Internet of Things and Convergence
• /
• v.1 no.1
• /
• pp.39-44
• /
• 2015

Organization is increasing the authorizing process to use public certificate and bio information. Certificate, has evolved to be able to parallel distributes the bio authentication and portable bio-authentication device. Authentication using an individual's PC and smart devices continue to generalize, while convenience for authentication is increased by comparison Study on cooperation with the security at the network level's a weak situation. If ask authentication method through the cooperation of the public certificate and bio information work with current network access control, there is a possibility to develop a more powerful security policy. by cooperation weaknesses against vulnerable personal authentication techniques on security token in a reliable and secure personal authentication techniques, such as bio-recognition, Bio Information for identification and to prevent exposing a methodology suggest to validate whether or not to carry out in this paper. In addition, organize the scenario that can work with the 802.1x network authentication method, and presented a proposal aimed at realization.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2005.06a
• /
• pp.888-891
• /
• 2005

In this paper, the prototype of a walking guide robot with tactile display is introduced, and the psychophysical experiment of the tactile recognition for a tactile display is carried out and analyzed. The objective of this research is the development of a walking guide robot for the blind to walk safely. A walking guide robot consists of a guide vehicle and a tactile display device. A guide vehicle, located in the front of the walking blind, detects the obstacle using ultrasonic sensors and offers the information of position and walking direction acquired from GPS module to the walking blind by voice. The tactile display device, located in the handle which is connected with the guide vehicle by cane, offers the processed obstacle information such as position, size, moving, shape of obstacle and safe path, etc. The psychophysical experiments for the threshold of perception and recognition ability of tactile stimulation are carried out by the estimation of the subject group. As a result the appropriate tactile stimulus intensity and frequency to recognize tactile stimulation effectively are discussed and derived.

• Bulletin of the Korean Chemical Society
• /
• v.33 no.11
• /
• pp.3625-3628
• /
• 2012

Fluorescent organic molecules that respond to changes in the $Fe^{2+}$ concentration with selectivity to other abundant di-valent metal ions will offer the ability to understand the dynamic fluctuations of the $Fe^{2+}$ ion in interesting media. The use of 6-Br-ppmbi, derived from 2-pyridin-2-yl-benzimidazole, for metal ion-selective fluorescence recognition was investigated. Screening of the main group and transition metal ions showed exclusive selectivity for $Fe^{2+}$ ions even in the presence of competing metal ions. In addition, the requirement for low concentrations of probe molecules to detect certain amounts of $Fe^{2+}$ ions make this sensor unique compared to previously reported $Fe^{2+}$ ion sensors.

• Food Science and Preservation
• /
• v.24 no.7
• /
• pp.885-890
• /
• 2017

For the construction of the microbial monitoring method, anti-Salmonella polyclonal antibodies (pAbs) were produced from a rabbit and purified by saturated ammonium sulfate precipitation and protein A affinity column. The reactivity of anti-Salmonella pAbs was compared to that of commercial ones by using an indirect ELISA. The specificity of anti-Salmonella pAbs was investigated using 20 Salmonella serotypes and 20 non-Salmonella strains. A capturing ability of anti-Salmonella pAbs was investigated by exposing antibody-immobilized gold biosensor to different concentration of Salmonella mixture. Anti-Salmonella pAbs were successfully produced and purified with an antibody concentration of 2.0 mg/mL The reactivity of purified anti-Salmonella pAbs was greater than that of commercial one at all tested concentrations. All Salmonella serotypes, except S. Diarizonae, showed excellent binding efficiency with purified anti-Salmonella pAbs. Moreover, the purified anti-Salmonella pAbs showed excellent specificity against all non-Salmonella strains. The anti-Salmonella pAbs immobilized on the gold biosensor demonstrated the successful capturing capability against Salmonella with a dose-response manner. Therefore, the anti-Salmonella pAbs exhibited sufficient reactivity, specificity, as well as capturing capability against Salmonella to be considered as a bio-recognition element.

• Proceedings of the KIEE Conference
• /
• 1999.11c
• /
• pp.757-759
• /
• 1999

여러 사람의 얼굴들 중에 특정 개인 얼굴을 찾는 문제나 인식하는 문제는 최근들어 법 집행이나 상업적 목적 등 여러 응용분야에서 요구되고 있어 학문적으로 활발히 연구되어지고 있다. 얼굴 인식 기술은 여러 가지 방법으로 연구되어 왔는데 그 중 'PCA를 이용한 얼굴 인식 방법'이 가장 효율적인 방법으로 알려졌다. 그러나 이 방법은 조명 변화에 따라 정확성이 떨어지는 단점이 있다. 그래서 Histogram equalization을 이용해 조명 변화에 영향을 줄였다. 그리고 인식의 정확성을 유지하면서 eigenface를 추출하는데 시간을 줄이기 위해 웨이블렛 변환을 이용해 저주파 성분이 포함된 영역만을 추출, 그 부분을 입력영상으로 사용해 입력 영상에서 처리해야하는 차원을 줄여 특징 추출하는데 시간을 감소시켰다. 그 결과 특징 추출하는데 시간을 크게 줄어든 반면, 심한 조명 변화에서도 90%이상의 높은 인식률을 유지할 수 있었다.

• Transactions of the KSME C: Technology and Education
• /
• v.5 no.1
• /
• pp.23-52
• /
• 2017

Bio-fusion and medical device industry is the multidisciplinary engineering application technology industries, which are fused, such as electricity, electronics, machinery, materials. It aims to improve the quality of human life by using bio-fusion and medical devices, and is an important industry recognition and brand power for the product. However, there is a period that is required from the development of products with technology-dependent industries. Therefore, it is necessary to have continuous effort for industrial investment in research and development at the national level including technical support, institutional support, and human resources.