• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.4
• /
• pp.41-46
• /
• 2023

Human urine is a process of excreting waste products in the blood, and it is easy to collect and contains various substances. Urinalysis is used to check for diseases, health conditions, and urinary tract infections. There are three methods of urinalysis: physical property test, chemical test, and microscopic test, and chemical test results can be easily confirmed using urine test strips. A variety of items can be tested on the urine test strip, through which various diseases can be identified. Recently, with the spread of smart phones, research on reading urine test strips using smart phones is being conducted. There is a method of detecting and reading the color change of a urine test strip using a smartphone. This method uses the RGB values and the color difference formula to discriminate. However, there is a problem in that accuracy is lowered due to various environmental factors. This paper applies a deep learning model to solve this problem. In particular, color discrimination of a urine test strip is improved in a smartphone using a lightweight CNN (Convolutional Neural Networks) model. CNN is a useful model for image recognition and pattern finding, and a lightweight version is also available. Through this, it is possible to operate a deep learning model on a smartphone and extract accurate urine test results. Urine test strips were taken in various environments to prepare deep learning model training images, and a urine test service application was designed using MobileNet V3.

• Journal of Internet of Things and Convergence
• /
• v.10 no.2
• /
• pp.1-15
• /
• 2024

The construction of smart communities is a new method and important measure to ensure the security of residential areas. In order to solve the problem of low accuracy in face recognition caused by distorting facial features due to monitoring camera angles and other external factors, this paper proposes the following optimization strategies in designing a face recognition network: firstly, a global graph convolution module is designed to encode facial features as graph nodes, and a multi-scale feature enhancement residual module is designed to extract facial keypoint features in conjunction with the global graph convolution module. Secondly, after obtaining facial keypoints, they are constructed as a directed graph structure, and graph attention mechanisms are used to enhance the representation power of graph features. Finally, tensor computations are performed on the graph features of two faces, and the aggregated features are extracted and discriminated by a fully connected layer to determine whether the individuals' identities are the same. Through various experimental tests, the network designed in this paper achieves an AUC index of 85.65% for facial keypoint localization on the 300W public dataset and 88.92% on a self-built dataset. In terms of face recognition accuracy, the proposed network achieves an accuracy of 83.41% on the IBUG public dataset and 96.74% on a self-built dataset. Experimental results demonstrate that the network designed in this paper exhibits high detection and recognition accuracy for faces in surveillance videos.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.22 no.3
• /
• pp.1-7
• /
• 2018

As part of a research dedicated to the field evaluation of the durability of concrete subjected to freezing-thawing, this study analyzes the relationship between the surface roughness and the relative dynamic elastic modulus through image analysis. Four mix compositions with water-to-binder ratios (W/B) of 40%, 50%, 60% and 70% and without AE agent were considered to provoke early freezing. The basic physical properties of the mixes including the relative dynamic elastic modulus and the compressive strength were first evaluated experimentally according to W/B. Then, tests were performed to measure the surface roughness followed by photographs and SEM image analysis. The measured surface roughness tended to increase with larger number of freezing-thawing cycles regardless of W/B. The relative dynamic elastic modulus appeared to increase gradually with the number of cycles for the relatively denser mixes with W/B of 40% and 50%. Besides, the surface roughness increased only at rupture for the mixes with W/B of 60% and 70%. Moreover, the analysis of the photographs of the surface of the mixes with W/B of 40% and 50% revealed that the degradation progressed gradually from the surface with the freezing-thawing cycles. However, for the mixes with W/B of 60% and 70%, apparent change of the surface remained very insignificant until rupture at which damage like cracking could be observed. Consequently, the analysis of surface photograph or the measurement of the surface roughness presented some limitation in assessing the degree of freezing-thawing-induced degradation in case of relatively porous specimens. On the other hand, the photograph and surface roughness appeared to be sufficient for assessing such degradation for the mixes with W/B of 40% and 50%. Accordingly, the image of the surface and the surface roughness are potentially applicable on site for the assessment of freezing-thawing damages in relatively dense mixes.

• Informatization Policy
• /
• v.19 no.4
• /
• pp.46-62
• /
• 2012

This research intends to find factors to analyze users, non-users, and the relation between how many sites people use and the users' tendencies, in regards of social network services such as Facebook, Twitter, and Cyworld. The major results are as follows: firstly, the user group seems to be composed of people who are younger, have higher self-efficacy, and use smartphones, compared to the non-user group. Secondly, a single-site user group has lower motivation of 'expert search and expression' and a lower degree of personal tendency of innovation (novelty seeking and self-efficacy), compared to multiple-site user group. Thirdly, foreign SNS user groups have a higher degree of personal tendency of innovation and motivation of 'experts search and expression' than the Cyworld user group. On the other hand, the Cyworld user group seems to be affected by reference group in their use, while they have high motivation of 'connection'. In conclusion, there were significant relationships between personality and multiple SNS uses. Also, Cyworld users tended to use it for maintaining established relationships.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2005.05a
• /
• pp.190-197
• /
• 2005

본 논문에서는 출입국자 관리의 효율성과 체계적인 출입국 관리를 위하여 여권 코드를 자동으로 인식하고 위조 여권을 판별할 수 있는 여권 인식 및 얼굴 인증 방법을 제안한다. 여권 이미지는 기울어진 상태로 스캔되어 획득되어질 수도 있으므로 기울기 보정은 문자 분할 및 인식, 얼굴 인증에 있어 매우 중요하다. 따라서 본 논문에서는 여권 영상을 스미어링한 후, 추출된 문자열 중에서 가장 긴 문자열을 선택하고 이 문자열의 좌측과 우측 부분의 두께 중심을 연결하는 직선과 수평선과의 기울기를 이용하여 여권 여상에 대한 각도 보정을 수행한다. 여권 코드 추출은 소벨 연산자와 수평 스미어링, 8방향 윤곽선 추적 알고리즘을 적용하여 여권 코드의 문자열 영역을 추출하고, 추출된 여권 코드 문자열 영역에 대해 반복 이지화 방법을 적용하여 코드의 문자열 영역을 이진화한다. 이진화된 문자열 영역에 대해 CDM 마스크를 적용하여 문자열의 코드들을 복원하고 8방향 윤곽선 추적 알고리즘을 적용하여 개별 코드를 추출한다. 추출된 개별 코드는 ART2 알고리즘을 적용하여 인식한다. 얼굴 인증을 위해 템플릿 매칭 알고리즘을 이용하여 얼굴 템플릿 데이터베이스를 구축하고 여권에서 추출된 얼굴 영역과의 유사도 측정을 통하여 여권 얼굴 영역의 위조 여부를 판별한다. 얼굴 인증을 위해서 Hue, YIQ-I, YCbCr-Cb 특징들의 유사도를 종합적으로 분석하여 얼굴 인증에 적용한다. 제안된 여권 인식 및 얼굴 인증 방법의 성능을 평가를 위하여 원본 여권에 얼굴 부분을 위조한 여권과 노이즈, 대비 증가 및 감소, 밝기 증가 및 감소 및 여권 영상을 흐리게 하여 실험한 결과, 제안된 방법이 여권 코드 인식 및 얼굴 인증에 있어서 우수한 성능이 있음을 확인하였다.권 영상에서 획득되어진 얼굴 영상의 특징벡터와 데이터베이스에 있는 얼굴 영상의 특징벡터와의 거리 값을 계산하여 사진 위조 여부를 판별한다. 제안된 여권 인식 및 얼굴 인증 방법의 성능을 평가를 위하여 원본 여권에서 얼굴 부분을 위조한 여권과 기울어진 여권 영상을 대상으로 실험한 결과, 제안된 방법이 여권의 코드 인식 및 얼굴 인증에 있어서 우수한 성능이 있음을 확인하였다.진행하고 있다.태도와 유아의 창의성간에는 상관이 없는 것으로 나타났고, 일반 유아의 아버지 양육태도와 유아의 창의성간의 상관에서는 아버지 양육태도의 성취-비성취 요인에서와 창의성제목의 추상성요인에서 상관이 있는 것으로 나타났다. 따라서 창의성이 높은 아동의 아버지의 양육태도는 일반 유아의 아버지와 보다 더 애정적이며 자율성이 높지만 창의성이 높은 아동의 집단내에서 창의성에 특별한 영향을 더 미치는 아버지의 양육방식은 발견되지 않았다. 반면 일반 유아의 경우 아버지의 성취지향성이 낮을 때 자녀의 창의성을 향상시킬 수 있는 것으로 나타났다. 이상에서 자녀의 창의성을 향상시키는 중요한 양육차원은 애정성이나 비성취지향성으로 나타나고 있어 정서적인 측면의 지원인 것으로 밝혀졌다.징에서 나타나는 AD-SR맥락의 반성적 탐구가 자주 나타났다. 반성적 탐구 척도 두 그룹을 비교 했을 때 CON 상호작용의 특징이 낮게 나타나는 N그룹이 양적으로 그리고 내용적으로 더 의미 있는 반성적 탐구를 했다용을 지원하는 홈페이지를 만들어 자료 제공 사이트에 대한 메타 자료를 데이터베이스화했으며 이를 통해 학생들이 원하는 실시간 자료를 검색하여 찾을 수 있고 홈페이지를 방분했을 때 이해하기 어려운 그래프나 각 홈페이지가 제공하는 자료들에 대한 처리 방법을 도움말로 제공받을 수 있게 했다. 실

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.3 no.1
• /
• pp.17-21
• /
• 2005

Shielded laser ablation system composed of laser system, image analyser, XYZ translator with motion controller, ablation chamber, manipulator and various optics was designed. Nd:YAG laser which can be tunable from 1064 nm to 266 m was selected as light source. CCD camera(< $\pm$200) was chosen to analyze a crater less than 50 un in diameter. XYZ translator was composed of three linear stage which can travel 50 w with a minimum movement of 1 um and motion controller. Before the performance test, each part of system was optically aligned. To perform the ablation test, the specimen was ablated by 50 um interval and observed by image analyser The shape of crater was almost round, indicating laser beam has homogeneous energy distribution. The resolution and magnification of image system were compatible with the design.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.29 no.5
• /
• pp.493-500
• /
• 2009

An ultrasonic signal processing algorithm was developed for extracting the information of cracks generated around the keyway of a turbine rotor disk. B-scan images were obtained by using keyway specimens and an ultrasonic scan system with x-y position controller. The B-scan images were used as input images for 2-Dimensional signal processing, and the algorithm was constructed with four processing stages of pre-processing, crack candidate region detection, crack region classification and crack information extraction. It is confirmed by experiments that the developed algorithm is effective for the quantitative evaluation of cracks generated around the keyway of turbine rotor disk.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.4
• /
• pp.43-53
• /
• 2014

In this paper, we propose an ontology-based approach to measuring degree-of-interest in country by analyzing wikipedia usage data. First, we developed the degree-of-interest ontology called DOI ontology by extracting concept hierarchies from wikipedia categories. Second, we map the title of frequently edited articles into DOI ontology, and we measure degree-of-interest based on DOI ontology by analyzing wikipedia page views. Finally, we perform chi-square test of independence to figure out if interesting fields are independent or not by country. This approach shows interesting fields are closely related to each country, and provides degree of interests by country timely and flexibly as compared with conventional questionnaire survey analysis.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.7
• /
• pp.149-156
• /
• 2013

This paper presents a hardware edge detector of image signal at pixel level of CMOS image sensor (CIS). The circuit detects edges of an image based on a bump circuit combining with the pixels. The APS converts light into electrical signals and the bump circuit compares the brightness between the target pixel and its neighbor pixels. Each column on CIS 64 by 64 pixels array shares a comparator. The comparator decides a peak level of the target pixel comparing with a reference voltage. The proposed edge detector is implemented using 0.18um CMOS technology. The circuit shows higher fill factor 34% and power dissipation by 0.9uW per pixel at 1.8V supply.

• Journal of IKEEE
• /
• v.14 no.3
• /
• pp.224-231
• /
• 2010

Automatic detection of objects in images has been one of core challenges in the areas such as computer vision and pattern analysis. Especially, with the recent deployment of personal mobile devices such as smart phone, such technology is required to be transported to them. Usually, these smart phone users are equipped with devices such as camera, GPS, and gyroscope and provide various services through user-friendly interface. However, the smart phones fail to give excellent performance due to limited system resources. In this paper, we propose a new scheme to improve object recognition performance based on pre-computation and simple local features. In the pre-processing, we first find several representative parts from similar type objects and classify them. In addition, we extract features from each classified part and train them using regression functions. For a given query image, we first find candidate representative parts and compare them with trained information to recognize objects. Through experiments, we have shown that our proposed scheme can achieve resonable performance.