• Journal of Korea Multimedia Society
• /
• v.22 no.5
• /
• pp.535-546
• /
• 2019

Recently, there have been a lot of researches on the whole face replacement system, but it is not easy to obtain stable results due to various attitudes, angles and facial diversity. To produce a natural synthesis result when replacing the face shown in the video image, technologies such as face area detection, feature extraction, face alignment, face area segmentation, 3D attitude adjustment and facial transposition should all operate at a precise level. And each technology must be able to be interdependently combined. The results of our analysis show that the difficulty of implementing the technology and contribution to the system in facial replacement technology has increased in facial feature point extraction and facial alignment technology. On the other hand, the difficulty of the facial transposition technique and the three-dimensional posture adjustment technique were low, but showed the need for development. In this paper, we propose four facial replacement models such as 2-D Faceswap, OpenPose, Deekfake, and Cycle GAN, which are suitable for the Cox platform. These models have the following features; i.e. these models include a suitable model for front face pose image conversion, face pose image with active body movement, and face movement with right and left side by 15 degrees, Generative Adversarial Network.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.9 no.1
• /
• pp.179-187
• /
• 2019

With the recent development of small computing devices, IoT sensor network can be widely deployed and is now readily available with sensing, calculation and communi-cation functions at low cost. Sensor data management is a major component of the Internet of Things environment. The huge volume of data produced and transmitted from sensing devices can provide a lot of useful information but is often considered the next big data for businesses. New column-wise compression technology is mounted to the large data server because of its superior space efficiency. Since sensor nodes have narrow bandwidth and fault-prone wireless channels, sensor-based storage systems are subject to incomplete data services. In this study, we will bring forth a short overview through providing an analysis on IoT sensor networks, and will propose a new storage management scheme for IoT data. Our management scheme is based on RAID storage model using column-wise segmentation and compression to improve space efficiency without sacrificing I/O performance. We conclude that proposed storage control scheme outperforms the previous RAID control by computer performance simulation.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.3
• /
• pp.389-395
• /
• 2022

Because of improvement of deep learning techniques, deep learning using computer vision such as classification, detection and segmentation has also been used widely at many fields. Expecially, automatic driving is one of the major fields that applies computer vision systems. Also there are a lot of works and researches to combine multiple tasks in a single network. In this study, we propose the network that predicts the individual depth of pedestrians and vehicles. Proposed model is constructed based on YOLOv3 for object detection and Monodepth for depth estimation, and it process object detection and depth estimation consequently using encoder and decoder based on hard parameter sharing. We also used attention module to improve the accuracy of both object detection and depth estimation. Depth is predicted with monocular image, and is trained using self-supervised training method.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.35 no.5
• /
• pp.259-266
• /
• 2022

In this study, we developed a deep learning and vision sensor-based assembly performance evaluation method isfor prefabricated steel structures. The assembly parts were segmented using a modified version of the receptive field block convolution module inspired by the eccentric function of the human visual system. The quality of the assembly was evaluated by detecting the bolt holes in the segmented assembly part and calculating the bolt hole positions. To validate the performance of the evaluation, models of standard and defective assembly parts were produced using a 3D printer. The assembly part segmentation network was trained based on the 3D model images captured from a vision sensor. The sbolt hole positions in the segmented assembly image were calculated using image processing techniques, and the assembly performance evaluation using the k-nearest neighbor algorithm was verified. The experimental results show that the assembly parts were segmented with high precision, and the assembly performance based on the positions of the bolt holes in the detected assembly part was evaluated with a classification error of less than 5%.

• Journal of Korea Entertainment Industry Association
• /
• v.14 no.3
• /
• pp.83-91
• /
• 2020

The purpose of this study is to develop creativity-based creative convergence subjects for nursing students. For the purpose of this study, the procedures are conducted that the needs analysis, setting educational goals, segmentation of educational goals, selection of educational contents and organization by F. Bobbitt's curriculum development model and the creative convergence subject was developed through the verification process of the validity of experts. Through a theoretical review, the contents of education in creative convergence subjects consisted of converging with other areas, focusing on creativity. It was presented as a liberal arts subject with two credits, and as an educational method, an online class utilizing blended learning and offline classes centered on activities by teams were presented. In addition, the curriculum was divided into understanding, application, synthesis, and deepening so that students could understand the concept of creative convergence thinking and apply it through thinking techniques and strategies, and finally improve their creative convergence thinking abilities through team projects.

• Journal of the Korea Fashion and Costume Design Association
• /
• v.24 no.2
• /
• pp.59-72
• /
• 2022

The online fashion market in the 21st century has shown rapid growth. Against this backdrop, using consumer activity data to provide customized customer services has emerged as a viable business model that draws attention. Algorithm-based personalized recommendation services are a good example. But their application in fashion products has clear limitations. It is not easy to identify consumers' perceptions of the attributes of fashion, which are various, hard to define, and very sensitive to trends. So there is a need to compile data on consumers' underlying awareness and to carry out defined research to increase the utilization of such services in the fashion industry and further engage consumers. This research aims to classify the attributes and types of fashion products and to identify consumers' perceptions of a given situation where a recommendation service is offered. To find out consumers' perceptions of and satisfaction with recommendation services, an online and mobile survey was conducted on women in their 20s and 30s, a group that uses recommendation services frequently. A total of 455 responses were used for analysis. SPSS 28.0 was used, combined with Conjoint Analysis and multiple regression, to analyze data. The study results could provide insights into a better understanding of recommendation services and be used as basic data for companies to identify consumers' preferences and draw up a detailed strategy for market segmentation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.12
• /
• pp.4292-4307
• /
• 2021

For the research field of night infrared surveillance video, the target imaging in the video is easily affected by the light due to the characteristics of the active infrared camera and the classical ViBe algorithm has some problems for moving target extraction because of background misjudgment, noise interference, ghost shadow and so on. Therefore, an improved ViBe algorithm (I-ViBe) for moving target extraction in night infrared surveillance video is proposed in this paper. Firstly, the video frames are sampled and judged by the degree of light influence, and the video frame is divided into three situations: no light change, small light change, and severe light change. Secondly, the ViBe algorithm is extracted the moving target when there is no light change. The segmentation factor of the ViBe algorithm is adaptively changed to reduce the impact of the light on the ViBe algorithm when the light change is small. The moving target is extracted using the region growing algorithm improved by the image entropy in the differential image of the current frame and the background model when the illumination changes drastically. Based on the results of the simulation, the I-ViBe algorithm proposed has better robustness to the influence of illumination. When extracting moving targets at night the I-ViBe algorithm can make target extraction more accurate and provide more effective data for further night behavior recognition and target tracking.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.374-376
• /
• 2021

For the diagnosis of cancer-related diseases in clinical practice, pathological examination using biopsy is essential after basic diagnosis using imaging equipment. In order to proceed with such a biopsy, the assistance of an oncologist, clinical pathologist, etc. with specialized knowledge and the minimum required time are essential for confirmation. In recent years, research related to the establishment of a system capable of automatic classification of cancer cells using artificial intelligence is being actively conducted. However, previous studies show limitations in the type and accuracy of cells based on a limited algorithm. In this study, we propose a method to identify a total of 4 cancer cells through a convolutional neural network, a kind of deep learning. The optical images obtained through cell culture were learned through EfficientNet after performing pre-processing such as identification of the location of cells and image segmentation using OpenCV. The model used various hyper parameters based on EfficientNet, and trained InceptionV3 to compare and analyze the performance. As a result, cells were classified with a high accuracy of 96.8%, and this analysis method is expected to be helpful in confirming cancer.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.43 no.4
• /
• pp.511-523
• /
• 2023

Research on the integration of unmanned aerial vehicles and deep learning for reinforced concrete damage detection is actively underway. Convolutional neural networks have a high impact on the performance of image classification, detection, and segmentation as backbones. The MobileNet, a pre-trained convolutional neural network, is efficient as a backbone for an unmanned aerial vehicle-based damage detection model because it can achieve sufficient accuracy with low computational complexity. Analyzing vanilla convolutional neural networks and MobileNet under various conditions, MobileNet was evaluated to have a verification accuracy 6.0~9.0% higher than vanilla convolutional neural networks with 15.9~22.9% lower computational complexity. MobileNetV2, MobileNetV3Large and MobileNetV3Small showed almost identical maximum verification accuracy, and the optimal conditions for MobileNet's reinforced concrete damage image feature extraction were analyzed to be the optimizer RMSprop, no dropout, and average pooling. The maximum validation accuracy of 75.49% for 7 types of damage detection based on MobilenetV2 derived in this study can be improved by image accumulation and continuous learning.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.3D
• /
• pp.535-540
• /
• 2006

This paper presented an effective strategy to extract the buildings and to reconstruct 3-D buildings using high-resolution multispectral stereo satellite images. Proposed scheme contained three major steps: building enhancement and segmentation using both BDT (Background Discriminant Transformation) and ISODATA algorithm, conjugate building identification using the object matching with Hausdorff distance and color indexing, and 3-D building reconstruction using photogrammetric techniques. IKONOS multispectral stereo images were used to evaluate the scheme. As a result, the BDT technique was verified as an effective tool for enhancing building areas since BDT suppressed the dominance of background to enhance the building as a non-background. In building recognition, color information itself was not enough to identify the conjugate building pairs since most buildings are composed of similar materials such as concrete. When both Hausdorff distance for edge information and color indexing for color information were combined, most segmented buildings in the stereo images were correctly identified. Finally, 3-D building models were successfully generated using the space intersection by the forward RFM (Rational Function Model).