• Convergence Security Journal
• /
• v.23 no.5
• /
• pp.57-64
• /
• 2023

In this study, we analyzed how to steal the target model without training data. Input data is generated using the generative model, and a similar model is created by defining a loss function so that the predicted values of the target model and the similar model are close to each other. At this time, the target model has a process of learning so that the similar model is similar to it by gradient descent using the logit (logic) value of each class for the input data. The tensorflow machine learning library was used as an experimental environment, and CIFAR10 and SVHN were used as datasets. A similar model was created using the ResNet model as a target model. As a result of the experiment, it was found that the model stealing method generated a similar model with an accuracy of 86.18% for CIFAR10 and 96.02% for SVHN, producing similar predicted values to the target model. In addition, considerations on the model stealing method, military use, and limitations were also analyzed.

• Convergence Security Journal
• /
• v.19 no.1
• /
• pp.145-152
• /
• 2019

In the era of the 3rd Offset Strategy led by the 4th Industrial Revolution, the use of robots with AI and autonomous abilities is becoming more active in military field. The 3rd Offset Strategy is based on the technology of the 4th Industrial Revolution, and S. Korea is heavily dependent on US military technology and is directly and indirectly influenced by the military revolutionary strategy and the alliance relationship. There are many issues that need to be addressed beyond technical maturity for both strategies to be successfully applied in the military. However, there are few discussions about these limitations in many studies and media reports in comparison with the advantages of the techniques. This research describes robot ethics & technology unbalance, problems of autonomous functions, display fatigue induced by VR/AR/MR, cyber/network security to be solved for successful strategies, also the solutions are addressed.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.4
• /
• pp.165-172
• /
• 2020

In this paper we propose an uses on-device-based edge computing technology and big data analysis methods through the use of on-device-based edge computing technology and analysis of big data, which are distributed computing paradigms that introduce computations and storage devices where necessary to solve problems such as transmission delays that occur when data is transmitted to central centers and processed in current general smart factories. However, even if edge computing-based technology is applied in practice, the increase in devices on the network edge will result in large amounts of data being transferred to the data center, resulting in the network band reaching its limits, which, despite the improvement of network technology, does not guarantee acceptable transfer speeds and response times, which are critical requirements for many applications. It provides the basis for developing into an AI-based facility prediction conservation analysis tool that can apply deep learning suitable for big data in the future by supporting intelligent facility management that can support productivity growth through research that can be applied to the field of facility preservation and smart factory industry with integrated hardware technology that can accommodate these requirements and factory management and control technology.

• Journal of Broadcast Engineering
• /
• v.28 no.3
• /
• pp.314-328
• /
• 2023

In this paper, we propose a texture map compression method based on the hierarchical coding method of SHVC to support the scalability function of dynamic mesh compression. The proposed method effectively eliminates the redundancy of multiple-resolution texture maps by downsampling a high-resolution texture map to generate multiple-resolution texture maps and encoding them with SHVC. The dynamic mesh decoder supports the scalability of mesh data by decoding a texture map having an appropriate resolution according to receiver performance and network environment. To evaluate the performance of the proposed method, the proposed method is applied to V-DMC (Video-based Dynamic Mesh Coding) reference software, TMMv1.0, and the performance of the scalable encoder/decoder proposed in this paper and TMMv1.0-based simulcast method is compared. As a result of experiments, the proposed method effectively improves in performance the average of -7.7% and -5.7% in terms of point cloud-based BD-rate (Luma PSNR) in AI and LD conditions compared to the simulcast method, confirming that it is possible to effectively support the texture map scalability of dynamic mesh data through the proposed method.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.1
• /
• pp.45-53
• /
• 2020

Conventional Convolutional Neural Network(CNN) algorithms have limitations in detecting small objects in large image. In this paper, we propose an improved model which is based on Region Of Interest(ROI) selection and image dividing technique. We prepared YOLOv3 / Faster R-CNN algorithms which are transfer-learned by airfield and aircraft datasets. Also we prepared large images for testing. In order to verify our model, we selected airfield area from large image as ROI first and divided it in two power n orders. Then we compared the aircraft detection rates by number of divisions. We could get the best size of divided image pieces for efficient small object detection derived from the comparison of aircraft detection rates. As a result, we could verify that the improved CNN algorithm can detect small object in large images.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.1
• /
• pp.163-169
• /
• 2020

Recently, research on automation and unmanned operation of machines in the industrial field has been conducted with the advent of AI, Big data, and the IoT, which are the core technologies of the Fourth Industrial Revolution. The machines for these automation processes are controlled based on the data collected from the sensors attached to them, and further, the processes are managed. Conventionally, the abnormalities of sensors are periodically checked and managed. However, due to various environmental factors and situations in the industrial field, there are cases where the inspection due to the failure is not missed or failures are not detected to prevent damage due to sensor failure. In addition, even if a failure occurs, it is not immediately detected, which worsens the process loss. Therefore, in order to prevent damage caused by such a sudden sensor failure, it is necessary to identify the failure of the sensor in an embedded system in real-time and to diagnose the failure and determine the type for a quick response. In this paper, a deep neural network-based fault diagnosis system is designed and implemented using Raspberry Pi to classify typical sensor fault types such as erratic fault, hard-over fault, spike fault, and stuck fault. In order to diagnose sensor failure, the network is constructed using Google's proposed Inverted residual block structure of MobilieNetV2. The proposed scheme reduces memory usage and improves the performance of the conventional CNN technique to classify sensor faults.

• Journal of the Korea Convergence Society
• /
• v.10 no.7
• /
• pp.15-21
• /
• 2019

Research in detecting fake information gained a lot of interest after the US presidential election in 2016. Information from unknown sources are produced in the shape of news, and its rapid spread is fueled by the interest of public drawn to stimulating and interesting issues. In addition, the wide use of mass communication platforms such as social network services makes this phenomenon worse. Poynter Institute created the International Fact Checking Network (IFCN) to provide guidelines for judging the facts of skilled professionals and releasing "Code of Ethics" for fact check agencies. However, this type of approach is costly because of the large number of experts required to test authenticity of each article. Therefore, research in automated fake news detection technology that can efficiently identify it is gaining more attention. In this paper, we investigate fake news detection systems and researches that are rapidly developing, mainly thanks to recent advances in deep learning technology. In addition, we also organize shared tasks and training corpus that are released in various forms, so that researchers can easily participate in this field, which deserves a lot of research effort.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.5
• /
• pp.23-28
• /
• 2024

So far, most of the military trainings were carried out in the field, and were influenced by the various parameters of the weather, the climate and the civil complaints regarding the noise. Also, it's the reality that the considerable time and resources are required to maneuver the weapon system used for the military training. Furthermore, the serious damage and casualties during tha military training are important parameters that can't be ignored. Recently, with the development of 5G communication networks and XR technologies, XR technologies are used in various fields that participate with multilateral parts, i.e. in military technology and training. In this paper, to implement the military education, 5G communication network and military education training system were established. The military education training system were composed that over 10 persons were possible to train in the various circumstances such as counter combat, mountains combat, urban combat and beaches combat. Also it is possible to fight with AI combatants, and train the gun disassembly and assembly, and train the various firing exercise. The military training system of using XR technologies were applied to the multilateral military training, and we analyzed the satisfaction results for the experienced persons of this XR system.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.6
• /
• pp.850-858
• /
• 2022

To deploy Gate Recurrent Units (GRU) on resource-constrained embedded devices, this paper presents a reconfigurable FPGA-based GRU accelerator that enables structured compression. Firstly, a dense GRU model is significantly reduced in size by hybrid quantization and structured top-k pruning. Secondly, the energy consumption on external memory access is greatly reduced by the proposed reuse computing pattern. Finally, the accelerator can handle a structured sparse model that benefits from the algorithm-hardware co-design workflows. Moreover, inference tasks can be flexibly performed using all functional dimensions, sequence length, and number of layers. Implemented on the Intel DE1-SoC FPGA, the proposed accelerator achieves 45.01 GOPs in a structured sparse GRU network without batching. Compared to the implementation of CPU and GPU, low-cost FPGA accelerator achieves 57 and 30x improvements in latency, 300 and 23.44x improvements in energy efficiency, respectively. Thus, the proposed accelerator is utilized as an early study of real-time embedded applications, demonstrating the potential for further development in the future.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2022.06a
• /
• pp.304-305
• /
• 2022

Recently, as digital communication technology is widely applied to the maritime field, large amounts of maritime data are being accumulated. Accordingly, attempts to create new value by applying data science and Artificial Intelligence(AI) technologies are emerging. Typically, Ministry of Oceans and Fisheries has been providing korean e-Navigation service since 2021 based on LTE-Maritime communication network, as well as R&D for creating value-added service through analyzing huge-sized maritime traffic data is underway. By the way, to do any data-based research, legal system, as a research infra, that researchers can get the data whenever they need is essential. This paper looked at types of data in maritime fields, checked related legal system about scientific analysis and utilization. It is confirmed that there are some legal factors which restrict its scientific analysis and utilization, and suggested ways of improvement to boost R&D using maritime data as a conclusion.