• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.11
• /
• pp.403-409
• /
• 2016

Recently, harmful content (such as images and photos of nudes) has been widely distributed. Therefore, there have been various studies to detect and filter out such harmful image content. In this paper, we propose a new method using Haar-like features and an AdaBoost algorithm for robustly extracting navel areas in a color image. The suggested algorithm first detects the human nipples through color information, and obtains candidate navel areas with positional information from the extracted nipple areas. The method then selects real navel regions based on filtering using Haar-like features and an AdaBoost algorithm. Experimental results show that the suggested algorithm detects navel areas in color images 1.6 percent more robustly than an existing method. We expect that the suggested navel detection algorithm will be usefully utilized in many application areas related to 2D or 3D harmful content detection and filtering.

• Journal of Broadcast Engineering
• /
• v.26 no.6
• /
• pp.778-789
• /
• 2021

Recently, various lightweight methods for using Convolutional Neural Network(CNN) models in mobile devices have emerged. Weight quantization, which lowers bit precision of weights, is a lightweight method that enables a model to be used through integer calculation in a mobile environment where GPU acceleration is unable. Weight quantization has already been used in various models as a lightweight method to reduce computational complexity and model size with a small loss of accuracy. Considering the size of memory and computing speed as well as the storage size of the device and the limited network environment, this paper proposes a method of compressing integer weights after quantization using a video codec as a method. To verify the performance of the proposed method, experiments were conducted on VGG16, Resnet50, and Resnet18 models trained with ImageNet and Places365 datasets. As a result, loss of accuracy less than 2% and high compression efficiency were achieved in various models. In addition, as a result of comparison with similar compression methods, it was verified that the compression efficiency was more than doubled.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.05a
• /
• pp.482-484
• /
• 2017

Over the past 70 years, movies and television have revolutionized the way people communicate. However, even with this development, TV has been used only as a means of communication targeting an unspecified number of people due to the restriction of media such as radio waves and movies. However, the development of the Internet and online video has come to a time when 100 million people watch YouTube videos uploaded from the other side of the world by eliminating these restrictions. The message that you want to deliver now can be delivered to anyone, but making the image with these messages remains the last obstacle to communication. To solve these problems, we implemented a web application and a video production program through AWS. This system basically provides the administrator with the video production through the easy interface, the information management and the program on the server on the internet through AWS, the assigned lecture including the computer and the smart phone, the learning materials, And implemented to increase the efficiency of educational video production service.

• Journal of Broadcast Engineering
• /
• v.27 no.3
• /
• pp.378-390
• /
• 2022

With the recent development of deep composite multiplication neural network learning, deep learning techniques applied to single-image super-resolution have shown good results, and the strong expression ability of deep networks has enabled complex nonlinear mapping between low-resolution and high-resolution images. However, there are limitations in applying it to real-time or low-power devices with increasing parameters and computational amounts due to excessive use of composite multiplication neural networks. This paper uses blocks that extract hierarchical characteristics little by little using information distillation and suggests the Recursive Distillation Super Resolution Network (RDSRN), a lightweight network that improves performance by making more accurate high frequency components through high frequency residual purification blocks. It was confirmed that the proposed network restores images of similar quality compared to RDN, restores images 3.5 times faster with about 32 times fewer parameters and about 10 times less computation, and produces 0.16 dB better performance with about 2.2 times less parameters and 1.8 times faster processing time than the existing lightweight network CARN.

• Journal of Korea Multimedia Society
• /
• v.24 no.11
• /
• pp.1552-1559
• /
• 2021

Convolution neural networks (CNNs) show notable performance in image processing and are used as representative core models. CNNs extract and learn features from large amounts of train dataset. In general, it has a structure in which a convolution layer and a fully connected layer are stacked. The core of CNN is the convolution layer. The size of the kernel used for feature extraction and the number that affect the depth of the feature map determine the amount of weight parameters of the CNN that can be learned. These parameters are the main causes of increasing the computational complexity and memory usage of the entire neural network. The most computationally expensive components in CNNs are fully connected and spatial convolution computations. In this paper, we propose a Fourier Convolution Neural Network that performs the operation of the convolution layer in the Fourier domain. We work on modifying and improving the amount of computation by applying the fast fourier transform method. Using the MNIST dataset, the performance was similar to that of the general CNN in terms of accuracy. In terms of operation speed, 7.2% faster operation speed was achieved. An average of 19% faster speed was achieved in experiments using 1024x1024 images and various sizes of kernels.

• Journal of KIBIM
• /
• v.13 no.4
• /
• pp.45-53
• /
• 2023

The educational benefits and potential of XR as a new medium are well recognized. However, there are still limitations in understanding the specific effects of XR compared to the more widely utilized representation of images on computer monitors. This study therefore aims to demonstrate the differences in effectiveness between the two technologies and to draw implications from a cognitive comparison of three-dimensional objects represented on a flat surface and virtually. The study was conducted a quantitative research method with an experiment involving two independent groups, and the results were tested using regression analysis. The results showed that for low-level, two-dimensional objects, the computer monitor method may be more effective, but above a certain level of complexity, the effectiveness of learning through the monitor tends to decrease rapidly. On the other hand, the group that used extended reality technology showed relatively high comprehension compared to the monitor group even as the complexity increased, and in particular, unlike the monitor group's rapidly decreasing comprehension level, the extended reality technology group showed a trend of decreasing comprehension with the level of complexity, suggesting the potential for compatibility and predictability in the use of technology.

• Industrial Engineering and Management Systems
• /
• v.7 no.3
• /
• pp.214-227
• /
• 2008

Today's dynamic competitiveness requires an organization to improve its performance measurement and management. Quality Management Systems (QMS) abound, the main ones being: ISO series, Malcolm Baldridge National Quality Award (MBNQA), European Forum for Quality Management (EFQM), Six Sigma Business Scorecard and the Balanced Scorecard. Based on the literature, the IPMMM (Integrated Performance Measurement and Management Model) identified 7 key synthesized factors: leadership, strategy management and policy, customer and market, learning and growth, partnership and resources, internal processes and business results that are employed to investigate the key performance indicators of a car assembler using the Delphi methodology. In the 2 rounds of Delphi panels consisting of 20 senior management personnel, the $1^{st}$ round of 198 indicators in the IPMMM yielded 90 indicators. The $2^{nd}$ round yielded 43 performance indicators with 18 rated as critical based on the % assigned in the $1^{st}$ and $2^{nd}$ priority rating of "very important factor" and "key performance indicator" that must be ranked high on both of the priorities. The very critical indicators appeared to be: defect percentage and first time capability (tie in $1^{st}$ place) and revenue, goal setting, customer satisfaction index, on-time delivery, brand image, return on investment, Claim Occurrence Ratio, and debt being ranked from $3^{rd}$ to $10^{th}$. It can be surmised that an organization can identify and develop an appropriate set of performance indicators through the Delphi methodology and implement and manage them based on the Balanced Scorecard.

• Journal of Korea Multimedia Society
• /
• v.24 no.8
• /
• pp.1114-1121
• /
• 2021

In this paper, we implement a recipe recommendation system using ingredient harmonization analysis based on recipe data. The proposed system receives an image of a food ingredient purchase receipt to recommend ingredients and recipes to the user. Moreover, it performs preprocessing of the receipt images and text extraction using the OCR algorithm. The proposed system can recommend recipes based on the combined data of ingredients. It collects recipe data to calculate the combination for each food ingredient and extracts the food ingredients of the collected recipe as training data. And then, it acquires vector data by learning with a natural language processing algorithm. Moreover, it can recommend recipes based on ingredients with high similarity. Also, the proposed system can recommend recipes using replaceable ingredients to improve the accuracy of the result through preprocessing and postprocessing. For our evaluation, we created a random input dataset to evaluate the proposed recipe recommendation system's performance and calculated the accuracy for each algorithm. As a result of performance evaluation, the accuracy of the Word2Vec algorithm was the highest.

• Journal of Korea Multimedia Society
• /
• v.25 no.10
• /
• pp.1499-1511
• /
• 2022

In the context of globalization, circulation between cities has become more frequent. The airport is no longer just a place for boarding, disembarking, and transportation, but a public place that serves as the communication function of the "aviation city". The intervention of public art in the airport space not only gives users a sense of space experience, but also becomes a unique carrier for city and country image shaping. The purpose of this paper is to study the emotional value brought by airport public art to users, and to investigate the correlation analysis of public art design elements and user preferences based on this premise. The research methods are machine learning method and SPSS 21.0. The user's emotional value is introduced in the big data evaluation, and the preference and inclination of airport users to various elements of public art are analyzed by questionnaire. Through the research conclusion, the preference and main contradiction of users in the airport for the four dimensions of public art design elements are obtained. Opinions and optimization methods to provide reference data and theoretical support for public art design.

• Journal of Korea Multimedia Society
• /
• v.24 no.2
• /
• pp.285-294
• /
• 2021

This study developed a responsive healthcare system that users can easily use in real life to prevent turtle neck syndrome by posture correction. We propose a system that naturally induces direct posture improvement by adjusting the height with a responsive cradle through a turtle neck discrimination algorithm detecting the turtle neck posture in real time using a webcam. The turtle neck algorithm was developed based on machine learning, using the points that the distance relationship between the jaw line and the shoulder varies depending on the posture. For the younger age group, which is particularly problematic due to the increase in the use of IT devices, image data in different situations according to the height and posture of the cradle was collected and learned as a support vector machine classifier. In addition, a height-adjustable cradle that can support a laptop has been created and expanded into a responsive cradle that can be controlled with software by interlocking with the Arduino. Therefore, this service enables posture correction of many modern people suffering from turtle neck syndrome and will become an essential platform in the increasing online environment in the non-contact era.