• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2023년도 제67차 동계학술대회논문집 31권1호
• /
• pp.403-404
• /
• 2023

패턴의 복잡도와 정보량을 계산하는 것은 음성 및 영상 인식을 위하여 향후 더 중요한 정보를 제공하는 단계로 발전할 것으로 기대된다. 패턴의 복잡도를 표현하는 정보 엔트로피의 개념은 정보량 측정외에 데이터의 압축 복원 과정, 데이터의 복잡도 등 다양한 목적으로 활용되고 있다. 본 논문에서는 영상 패턴의 복잡도를 영상 화질의 차이를 분석함으로써 영상 인식 시 지표 가능성을 파악하고자 한다.

• 한국멀티미디어학회논문지
• /
• 제23권6호
• /
• pp.756-764
• /
• 2020

A typical algorithm used for image analysis is the Convolutional Neural Network(CNN). R-CNN, Fast R-CNN, Faster R-CNN, etc. have been studied to improve the performance of the CNN, but they essentially require large amounts of data and high algorithmic complexity., making them inappropriate for small and medium-sized services. Therefore, in this paper, the image label prediction algorithm based on CNN with collaborative layer with low complexity, high accuracy, and small amount of data was proposed. The proposed algorithm was designed to replace the part of the neural network that is performed to predict the final label in the existing deep learning algorithm by implementing collaborative filtering as a layer. It is expected that the proposed algorithm can contribute greatly to small and medium-sized content services that is unsuitable to apply the existing deep learning algorithm with high complexity and high server cost.

• 한국국방경영분석학회지
• /
• 제27권1호
• /
• pp.10-27
• /
• 2001

This paper deals with the application of PRICE model in estimating the proper acquisition cost for weapon budgeting phase. The PRICE(Parametric Review of Information for Costing and Evaluation) Hardware model is a computerized method for deriving cost estimates of electronic and mechanical hardware assemblies and systems. The model can be used in obtaining not only initial cost estimates in conceptual phase, but also detailed cost estimates in budgeting phase depending on available historical and empirical data. We analyzed first step cost estimate parameters and derived cost equations using PRICe output dta. Using weight and complexity, We can find cost variation. Sensitivity analysis shows that cost increases exponentially as complexity increases exponentially as complexity increases. We estimated KAAV\`s (Korea Amphibious Assault Vehicle) production cost using the PRICE model and compare with engineering cost estimates which is based on actual production data submitted by the production company. The result shows that tow estimates are close within $\pm2%$ differences.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1990년도 하계학술대회 논문집
• /
• pp.532-534
• /
• 1990

홉필드 신경회로망을 VLSI로 구현하는데 문제가 되는 area complexity를 개선한다. 적용한 응용은 layer assignment 문제이며, 기존의 layer assignment 신경회로망의 N ${\times}$ L개의 노드를 N ${\times}$ log L개로 감소시킴으로써 O( $(log\;L/L)^2$ )로 area complexity를 개선한다 (N : net의 수, L : layer의 수). 이를 위한 새로운 에너지 함수를 제안한다. 기존의 layer assignment 신경회로망과 동일하게 본래의 홉필드 신경회로망의 진동을 방지하기 위하여 self-feedback이 있는 수정된 홉필드 모델을 사용한다.

• 한국초등과학교육학회지:초등과학교육
• /
• 제30권3호
• /
• pp.379-393
• /
• 2011

The purpose of this study was to analyze the complexity of scientific reasoning during open inquiry activities of pre-service elementary school teachers. In this study, 6 pre-service elementary teachers who participated in open-inquiry activities were selected. The data of scientific reasoning during their inquiry process was collected from the video recording of reporting about inquiry process and results, their reports and researcher's notetaking. CSRI Matrix (Dolan & Grady, 2010) was used to analyze the complexity of participants' scientific reasoning. The result showed that the degree of the complexity of their scientific reasoning varied in participants. Particularly the low degree of the complexity of scientific reasoning presented in posing preliminary hypotheses, providing suggestions for future research, communicating and defending finding. Also, The more pre-service teachers' epistemology of inquiry are similar to that of scientists, the more complex scientific reasoning represents. This results suggest that teachers should impress on students the importance of doing the precedent study and providing suggestions for future research, and provide a place for communicating and defending findings.

• 유통과학연구
• /
• 제17권12호
• /
• pp.13-21
• /
• 2019

Purpose - It is very important to examine customer's behavior about AR shopping either practically or academically. Thus, it will be worthwhile to discuss more in details about AR utility which is even in early stage of distribution industry now. Research design, data, and methodology - This study has designed in consideration of control effects of perceived complexity based on customer's flow as dependent variable, and on AR characteristics and technology readiness as independent variables. Study data has been collected from questionnaires after using AR shopping service directly by those who are 20-30 years old of male and female respondents, which has been analyzed with 167 questionnaires. Hypothesis is verified using by hierarchical regression analysis. Results - After results of hypothesis verified, positive influence has been shown in terms of sensory immersion, manipulation, and optimism, however, it is rejected in relation to navigation and innovativeness. Control effect of perceived complexity has not been appeared. Conclusions - Implications of this study are as follows. First, AR shopping service has to provide an informational value. Second, by providing AR service to customer group, marketing activities will be in effects. Third, recognized complexity is not connected with significant control effect in terms of customer's devotion of service.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권7호
• /
• pp.3433-3446
• /
• 2019

Aiming at the high complexity of traditional single-channel demodulation algorithm for PCMA signals, a new demodulation algorithm based on neural network is proposed to reduce the complexity of demodulation in the system of non-cooperative PCMA communication. The demodulation network is trained in this paper, which combines the preprocessing module and decision module. Firstly, the preprocessing module is used to estimate the initial parameters, and the auxiliary signals are obtained by using the information of frequency offset estimation. Then, the time-frequency characteristic data of auxiliary signals are obtained, which is taken as the input data of the neural network to be trained. Finally, the decision module is used to output the demodulated bit sequence. Compared with traditional single-channel demodulation algorithms, the proposed algorithm does not need to go through all the possible values of transmit symbol pairs, which greatly reduces the complexity of demodulation. The simulation results show that the trained neural network can greatly extract the time-frequency characteristics of PCMA signals. The performance of the proposed algorithm is similar to that of PSP algorithm, but the complexity of demodulation can be greatly reduced through the proposed algorithm.

• 정보보호학회논문지
• /
• 제19권2호
• /
• pp.39-48
• /
• 2009

본 논문에서는 미 연방 표준 블록 암호 AES에 대한 기존의 9 라운드 연관키 렉탱글 공격을 10 라운드로 향상시킨다. 256개의 연관키를 사용하는 12 라운드 AES-192의 첫 10 라운드는 $2^{124}$의 데이터 복잡도와 $2^{183}$의 시간 복잡도로 공격되며, 64개의 연관키를 사용하는 AES-192의 첫 10 라운드는 $2^{122}$의 데이터 복잡도와 $2^{183.6}$의 시간 복잡도로 공격된다. 본 논문의 공격은 AES-192에 대한 기존 공격 중 최상의 공격이다.

• 한국통신학회논문지
• /
• 제27권10C호
• /
• pp.887-901
• /
• 2002

소프트웨어 관리 모델은 크게 소프트웨어 프로젝트 견적 모델과 소프트웨어 설계평가 모델, 소프트웨어 복잡성 모델, 소프트웨어 신뢰도 성장 모델, 소프트웨어 프로세스 개선 모델 등으로 나누어진다. 그 중에서도 개발된 소프트웨어를 정량적으로 분석하여 평가하는 모델이 소프트웨어 복잡도 모델이다. 즉, 소프트웨어를 객관적으로 관리하는데 있어서 정량적인 판단 기준이 필요하다. 그렇기 때문에 소프트웨어 복잡성 모델(software complexity model)은 소프트웨어의 복잡성에 대한 정량적인 평가에 응용할 목적으로 도입되었으며, 소프트웨어 복잡성 측정시 이 결과를 소프트웨어 관리(software management)의 기준에 이용하는데 있어서 품질이나 생산성에 대한 영향을 제공할 때 정량적이고 명확히 할 필요가 있기 때문이다. 다시 말해서 복잡한 프로그램에는 에러가 많이 포함될 가능성이 높고 또 개발비용(source cost)도 높다는 것을 경험적으로 알 수 있다. 이러한 관계가 성립되지 않는 경우에는 소프트웨어의 효과적인 관리가 곤란하다. 본 논문은 이런 관점에서 대표적인 소프트웨어 복잡성 모델에 대한 적용법에 대해 기술하고 개발중인 교환시스템의 소프트웨어에 대해 volume metrics와 process complexity metrics 방법, 제어구조 및 인터페이스 구조 방법에 대한 분석 결과와 시스템 개발을 수행하는 과정에서 발생되고 있는 문제점들에 대해 다각도로 분석을 하여 이를 유사한 시스템 연구개발 및 프로젝트 관리에 활용하고자 한다.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 2007년도 추계학술대회 및 정기총회
• /
• pp.388-394
• /
• 2007

In wireless network, we propose a predictive location update scheme which considers mobile user's(MU's) mobility patterns. MU's mobility patterns can be found from a movement history data. The prediction accuracy and model complexity depend on the degree of application of history data. The more data we use, the more accurate the prediction is. As a result, the location management cost is reduced, but complexity of the model increases. In this paper, we classify MU's mobility patterns into four types. For each type, we find the respective optimal number of application of history data, and predictive location area by using the simulation. The optimal numbers of four types are shown to be different. When we use more than three application of history data, the simulation time and data storage are shown to increase very steeply.