• Journal of Korea Multimedia Society
• /
• 제24권7호
• /
• pp.868-879
• /
• 2021

PCB (Printed Circuit Board) inspection using a deep learning model requires a large amount of data and storage. When the amount of stored data increases, problems such as learning time and insufficient storage space occur. In this study, the existing object detection model is changed to a continual learning model to enable the recognition and classification of PCB components that are constantly increasing. By changing the structure of the object detection model to a knowledge distillation model, we propose a method that allows knowledge distillation of information on existing classified parts while simultaneously learning information on new components. In classification scenario, the transfer learning model result is 75.9%, and the continual learning model proposed in this study shows 90.7%.

• Journal of Korea Multimedia Society
• /
• 제25권2호
• /
• pp.287-296
• /
• 2022

Today, air pollution is becoming a severe issue worldwide and various policies are being implemented to solve environmental pollution. In major cities, public bicycles are installed and operated to reduce pollution and solve transportation problems, and operational information is collected in real time. However, research using public bicycle operation information data has not been processed. This study uses the daily weather data of Korea Meteorological Agency and real-time air pollution data of Korea Environment Corporation to predict the amount of daily rental bicycles. Cross- validation, principal component analysis and multiple regression analysis were used to determine the independent variables of the predictive model. Then, the study selected the elements that satisfy the significance level, constructed a model, predicted the amount of daily rental bicycles, and measured the accuracy.

• Archives of Craniofacial Surgery
• /
• 제25권1호
• /
• pp.17-21
• /
• 2024

Background: The pain caused by local anesthetic injection can lead to patient anxiety prior to surgery, potentially necessitating sedation or general anesthesia during the excision procedure. In this study, we aim to compare the pain relief efficacy and safety of using a digital automatic anesthetic injector for local anesthesia. Methods: Thirty-three patients undergoing excision of a benign soft tissue tumor under local anesthesia were prospectively enrolled from September 2021 to February 2022. A single-blind, randomized controlled study was conducted. Patients were divided into two groups by randomization: the experimental group with digital automatic anesthetic injector method (I-JECT group) and the control group with conventional injection method. Before surgery, the Amsterdam preoperative anxiety information scale was used to measure the patients' anxiety. After local anesthetic was administered, the Numeric Pain Rating Scale was used to measure the pain. The amount of anesthetic used was divided by the surface area of the lesion was recorded. Results: Seventeen were assigned to the conventional group and 16 to the I-JECT group. The mean Numeric Pain Rating Scale was 1.75 in the I-JECT group and 3.82 in conventional group. The injection pain was lower in the I-JECT group (p< 0.01). The mean Amsterdam preoperative anxiety information scale was 11.00 in the I-JECT group and 9.65 in conventional group. Patient's anxiety did not correlate to injection pain regardless of the method of injection (p= 0.47). The amount of local anesthetic used per 1 cm² of tumor surface area was 0.74 mL/cm² in the I-JECT group and 2.31 mL/cm² in the conventional group. The normalization amount of local anesthetic was less in the I-JECT group (p< 0.01). There was no difference in the incidence of complications. Conclusion: The use of a digital automatic anesthetic injector has shown to reduce pain and the amount of local anesthetics without complication.

• Journal of the Korean Data and Information Science Society
• /
• 제25권1호
• /
• pp.245-254
• /
• 2014

The traditional methods of simulating daily precipitation have paid little attention to the inherent dependence structure between the total precipitation amount and the precipitation frequency for a fixed period of time. To address this issue, we propose a new simulation algorithm using copula in order to incorporate the dependence into the traditional methods. The algorithm consists of two parts: First, while reflecting the observed dependence, we generate the total precipitation amount (S) and the frequency (N) during the period of interest; then we simulate the daily precipitation whose aggregation matches the pair of (N; S) generated in the first part. Our result shows that the proposed method substantially improves the traditional methods.

• Journal of Korea Multimedia Society
• /
• 제18권7호
• /
• pp.834-842
• /
• 2015

We propose a method to construct composite feature vector based on discriminant analysis for face recognition. For this, we first extract the holistic- and local-features from whole face images and local images, which consist of the discriminant pixels, by using a discriminant feature extraction method. In order to utilize both advantages of holistic- and local-features, we evaluate the amount of the discriminative information in each feature and then construct a composite feature vector with only the features that contain a large amount of discriminative information. The experimental results for the FERET, CMU-PIE and Yale B databases show that the proposed composite feature vector has improvement of face recognition performance.

• Journal of Korea Society of Digital Industry and Information Management
• /
• 제8권4호
• /
• pp.13-18
• /
• 2012

In this paper, we propose a noise estimation method for noise reduction. It is based on block and pixel-based noise estimation. We assume that an input image is contaminated by the additive white Gaussian noise. Thus, we use an adaptive Gaussian filter and estimate the amount of noise. It computes the standard deviation of each block and estimation is performed on pixel-based operation. The proposed algorithm divides an input image into blocks. This method calculates the standard deviation of each block and finds the minimum standard deviation block. The block in flat region shows well noise and filtering effects. Blocks which have similar standard deviation are selected as test blocks. These pixels are filtered by adaptive Gaussian filtering. Then, the amount of noise is calculated by the standard deviation of the differences between noisy and filtered blocks. Experimental results show that our proposed estimation method has better results than those by existing estimation methods.

• Journal of the Korea Society of Computer and Information
• /
• 제16권8호
• /
• pp.189-196
• /
• 2011

To develop a creative design concept, amount of idea thinking between team members is important. In the process of planning to proposal brainstorming helps clarification of problems and allows a creative solution. Here, idea generation is to express thinking into words, and greater amount of idea is more valued than to choose one fine idea at first. This paper studies actual processes of brainstorming, and we utilize network services upgraded in Web and multimedia techniques to develop idea with arrangement. We compare the amount of idea generation by groups and topics with Brainwriting and Braindrawing of Webstorming, without limitation of space and time. Then, we propose a creative idea thinking by analyzing the characteristics of design team members in various aspects. This result will contribute in improving national potential value and design competitiveness by the help of design and IT, on product research and marketing.

• Journal of information and communication convergence engineering
• /
• 제7권4호
• /
• pp.496-500
• /
• 2009

In EBCOT, the context modeling process takes excessive calculation time and this paper proposed a method to reduce this calculation time. That is, if the finest resolution coefficient is less than a pre-defined transfer factor the coefficient and its descendents skip the context modeling process. There is a trade-off relationship between the calculation time and the image quality or the amount of output data such that as this threshold value increases, the calculation time and the amount of output data decreases, but the image degradation increases. The experimental results showed that in this range the resulting reduction rate in calculation time was from 3% to 64% in average, the reduction rate in output data was from 32% to 73% in average.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 한국컴퓨터정보학회 2022년도 제66차 하계학술대회논문집 30권2호
• /
• pp.203-204
• /
• 2022

본 연구에서는 게임관련 학업을 전공으로 하는 남학생 27명, 여학생 18명, 총 45명을 대상으로 게임이용 한달 지출 금액을 조사하였다. 그 결과 남학생의 경우는 응답자의 약 74%가 게임이용에 금액을 지출한다고 응답하고 있으며, 여학생의 경우는 응답자의 50%가 게임이용에 금액을 지출한다고 응답하고 있다. T-검정 결과, 양측검정 유의수준 P값이 0.13(P>0.05)로 남학생과 여학생간의 게임이용에 대한 지출금액의 평균에 대한 차이가 없는 것으로 나타났다. 결론적으로 게임관련 학업을 전공으로 하는 학생들은 남학생과 여학생 구분없이 게임이용에 지출하는 금액의 평균의 차이가 없다고 할 수 있다.

• Asia Marketing Journal
• /
• 제26권1호
• /
• pp.1-10
• /
• 2024

This study explores incomplete decision-making dynamics on reward-based crowdfunding platforms, focusing on temporal and social factors influencing backers' decisions. Utilizing the temporal aspect (i.e., pledging campaign phase) and social aspect (i.e., current pledged amount ratio) as stimuli within the stimulus-organism-response framework, our findings reveal that nearly 50.9% of respondents change their initial decisions, highlighting widespread incomplete information processing. Backers are more prone to altering decisions under heightened time pressure and display herding behaviors. Furthermore, backers exhibit an increased likelihood of changing decisions under heightened time pressure, coupled with a greater chance that the pledged goal amount will not be achieved. The study discusses theoretical and practical implications.