• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.89-106
• /
• 2022

Recently, the amount of video data collected from smartphones, CCTVs, black boxes, and high-definition cameras has increased rapidly. According to the increasing video data, the requirements for analysis and utilization are increasing. Due to the lack of skilled manpower to analyze videos in many industries, machine learning and artificial intelligence are actively used to assist manpower. In this situation, the demand for various computer vision technologies such as object detection and tracking, action detection, emotion detection, and Re-ID also increased rapidly. However, the object detection and tracking technology has many difficulties that degrade performance, such as re-appearance after the object's departure from the video recording location, and occlusion. Accordingly, action and emotion detection models based on object detection and tracking models also have difficulties in extracting data for each object. In addition, deep learning architectures consist of various models suffer from performance degradation due to bottlenects and lack of optimization. In this study, we propose an video analysis system consists of YOLOv5 based DeepSORT object tracking model, SlowFast based action recognition model, Torchreid based Re-ID model, and AWS Rekognition which is emotion recognition service. Proposed model uses single-linkage hierarchical clustering based Re-ID and some processing method which maximize hardware throughput. It has higher accuracy than the performance of the re-identification model using simple metrics, near real-time processing performance, and prevents tracking failure due to object departure and re-emergence, occlusion, etc. By continuously linking the action and facial emotion detection results of each object to the same object, it is possible to efficiently analyze videos. The re-identification model extracts a feature vector from the bounding box of object image detected by the object tracking model for each frame, and applies the single-linkage hierarchical clustering from the past frame using the extracted feature vectors to identify the same object that failed to track. Through the above process, it is possible to re-track the same object that has failed to tracking in the case of re-appearance or occlusion after leaving the video location. As a result, action and facial emotion detection results of the newly recognized object due to the tracking fails can be linked to those of the object that appeared in the past. On the other hand, as a way to improve processing performance, we introduce Bounding Box Queue by Object and Feature Queue method that can reduce RAM memory requirements while maximizing GPU memory throughput. Also we introduce the IoF(Intersection over Face) algorithm that allows facial emotion recognized through AWS Rekognition to be linked with object tracking information. The academic significance of this study is that the two-stage re-identification model can have real-time performance even in a high-cost environment that performs action and facial emotion detection according to processing techniques without reducing the accuracy by using simple metrics to achieve real-time performance. The practical implication of this study is that in various industrial fields that require action and facial emotion detection but have many difficulties due to the fails in object tracking can analyze videos effectively through proposed model. Proposed model which has high accuracy of retrace and processing performance can be used in various fields such as intelligent monitoring, observation services and behavioral or psychological analysis services where the integration of tracking information and extracted metadata creates greate industrial and business value. In the future, in order to measure the object tracking performance more precisely, there is a need to conduct an experiment using the MOT Challenge dataset, which is data used by many international conferences. We will investigate the problem that the IoF algorithm cannot solve to develop an additional complementary algorithm. In addition, we plan to conduct additional research to apply this model to various fields' dataset related to intelligent video analysis.

• The Korean Society of Law and Medicine
• /
• v.24 no.4
• /
• pp.67-101
• /
• 2023

The utilization of generative AI in the medical field is also being rapidly researched. Access to vast data sets reduces the time and energy spent in selecting information. However, as the effort put into content creation decreases, there is a greater likelihood of associated issues arising. For example, with generative AI, users must discern the accuracy of results themselves, as these AIs learn from data within a set period and generate outcomes. While the answers may appear plausible, their sources are often unclear, making it challenging to determine their veracity. Additionally, the possibility of presenting results from a biased or distorted perspective cannot be discounted at present on ethical grounds. Despite these concerns, the field of generative AI is continually advancing, with an increasing number of users leveraging it in various sectors, including biomedical and life sciences. This raises important legal considerations regarding who bears responsibility and to what extent for any damages caused by these high-performance AI algorithms. A general overview of issues with generative AI includes those discussed above, but another perspective arises from its fundamental nature as a large-scale language model ('LLM') AI. There is a civil law concern regarding "the memorization of training data within artificial neural networks and its subsequent reproduction". Medical data, by nature, often reflects personal characteristics of patients, potentially leading to issues such as the regeneration of personal information. The extensive application of generative AI in scenarios beyond traditional AI brings forth the possibility of legal challenges that cannot be ignored. Upon examining the technical characteristics of generative AI and focusing on legal issues, especially concerning the protection of personal information, it's evident that current laws regarding personal information protection, particularly in the context of health and medical data utilization, are inadequate. These laws provide processes for anonymizing and de-identification, specific personal information but fall short when generative AI is applied as software in medical devices. To address the functionalities of generative AI in clinical software, a reevaluation and adjustment of existing laws for the protection of personal information are imperative.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.11
• /
• pp.94-103
• /
• 2017

As automatic hacking tools and techniques have been improved, the number of new vulnerabilities has increased. The CVE registered from 2010 to 2015 numbered about 80,000, and it is expected that more vulnerabilities will be reported. In most cases, patching a vulnerability depends on the developers' capability, and most patching techniques are based on manual analysis, which requires nine months, on average. The techniques are composed of finding the vulnerability, conducting the analysis based on the source code, and writing new code for the patch. Zero-day is critical because the time gap between the first discovery and taking action is too long, as mentioned. To solve the problem, techniques for automatically detecting and analyzing software (SW) vulnerabilities have been proposed recently. Cyber Grand Challenge (CGC) held in 2016 was the first competition to create automatic defensive systems capable of reasoning over flaws in binary and formulating patches without experts' direct analysis. Darktrace and Cylance are similar projects for managing SW automatically with artificial intelligence and machine learning. Though many foreign commercial institutions and academies run their projects for automatic binary analysis, the domestic level of technology is much lower. This paper is to study developing automatic detection of SW vulnerabilities and defenses against them. We analyzed and compared relative works and tools as additional elements, and optimal techniques for automatic analysis are suggested.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.12
• /
• pp.491-498
• /
• 2019

Digital twin is a technology that virtualizes physical objects of the real world on a computer. It is used by collecting sensor data through IoT, and using the collected data to connect physical objects and virtual objects in both directions. It has an advantage of minimizing risk by tuning an operation of virtual model through simulation and responding to varying environment by exploiting experiments in advance. Recently, artificial intelligence and machine learning technologies have been attracting attention, so that tendency to virtualize a behavior of physical objects, observe virtual models, and apply various scenarios is increasing. In particular, recognition of each robot's motion is needed to build digital twin for co-robot which is a heart of industry 4.0 factory automation. Compared with modeling based research for recognizing motion of co-robot, there are few attempts to predict motion based on sensor data. Therefore, in this paper, an experimental environment for collecting current and inertia data in co-robot to detect the motion of the robot is built, and a motion prediction model based on the collected sensor data is proposed. The proposed method classifies the co-robot's motion commands into 9 types based on joint position and uses current and inertial sensor values to predict them by accumulated learning. The data used for accumulating learning is the sensor values that are collected when the co-robot operates with margin in input parameters of the motion commands. Through this, the model is constructed to predict not only the nine movements along the same path but also the movements along the similar path. As a result of learning using SVM, the accuracy, precision, and recall factors of the model were evaluated as 97% on average.

• Journal of Convergence for Information Technology
• /
• v.8 no.5
• /
• pp.77-83
• /
• 2018

A modular robot that memorizes motion can be easily created and operated because it expresses by hand. However, since there is not enough storage space in the module to store the user-created operation, it is impossible to reuse the created operation, and when the modular robot again memorizes the operation, it changes to another operation. There is no main controller capable of operating a plurality of modular robots at the same time, and thus there is a disadvantage that the user must input directly to the modular robot. To overcome these disadvantages, a remote controller has been proposed that can be operated in the surrounding smart devices by designing web server and component based software using wired and wireless network. In the proposed method, various types of structures are created by connecting to a modular robot, and the reconstructed operation is performed again after storing, and the usefulness is confirmed by regenerating the stored operation effectively. In addition, the reliability of the downloaded trajectory data is verified by analyzing the difference between the trajectory data and the actual trajectory. In the future, the trajectory stored in the remote controller will be standardized using the artificial intelligence technique, so that the operation of the modular robot will be easily implemented.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.17 no.4
• /
• pp.49-59
• /
• 2017

Smart sensing plays a key role in a variety of IoT applications, and its importance is growing more and more together with the development of artificial intelligence. Therefore the importance of smart sensors cannot be overemphasized. However, most studies related to smart sensors have been focusing on specific application purposes, for example, security, energy saving, monitoring, and there are not much effort on researches on how to efficiently configure various types of smart sensors to be needed in the future. In this paper, a component-based framework with hierarchical structure for efficient construction of smart sensor is proposed and its application to smart home is designed and implemented. The proposed method shows that various types of smart sensors to be appeared in the near future can be configured through the design and development of necessary components within the proposed software framework. In addition, since it has a layered architecture, the configuration of the smart sensor can be expanded by inserting the internal or external layers. In particular, it is possible to independently design the internal and external modules when designing an IoT application service through connection with the external device layer. A small-scale smart home system is designed and implemented using the proposed method, and a home cloud operating as an external layer, is further designed to accommodate and manage multiple smart homes. By developing and thus adding the components of each layer, it will be possible to efficiently extend the range of applications such as smart cars, smart buildings, smart factories an so on.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.6
• /
• pp.672-680
• /
• 2019

This paper proposes "A Smart Farm Environment Optimization and Yield Prediction Platform based on IoT and Deep Learning" which gathers bio-sensor data from farms, diagnoses the diseases of growing crops, and predicts the year's harvest. The platform collects all the information currently available such as weather and soil microbes, optimizes the farm environment so that the crops can grow well, diagnoses the crop's diseases by using the leaves of the crops being grown on the farm, and predicts this year's harvest by using all the information on the farm. The result shows that the average accuracy of the AEOM is about 15% higher than that of the RF and about 8% higher than the GBD. Although data increases, the accuracy is reduced less than that of the RF or GBD. The linear regression shows that the slope of accuracy is -3.641E-4 for the ReLU, -4.0710E-4 for the Sigmoid, and -7.4534E-4 for the step function. Therefore, as the amount of test data increases, the ReLU is more accurate than the other two activation functions. This paper is a platform for managing the entire farm and, if introduced to actual farms, will greatly contribute to the development of smart farms in Korea.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.5
• /
• pp.189-194
• /
• 2018

Unmanned technology is a representative technology that integrates various technologies like electric, electronic, mechanical, artificial intelligence, ICT technology, ect. In special emphasize, ground technology has been developing exponentially in the military field and expanding its utilization area. The patent information analysis method presented in this study, proposes a new patent analysis methodology for patent information analysis and patent information on unmanned ground technology. The patent information analysis processor has 6 levels to extract core patents and technologies. The process consists of: selection of technology to be analyzed, classification of detailed technology / key keyword selection, patent information collection / noise reduction, selection of patent information analysis method, patent information analysis, finally, core patents and key technologies that are extracted. Patent information on unmanned ground technology is also analyzed in this study. First, the technical classification of ground unmanned technology is carried out in detail. The core technology and core patents of ground unmanned technology were extracted through CPP and IPC code connectivity analysis. The results of patent information analysis using proposed patent information analysis method that can be applied to various fields of technology and analysis. These can be used as a material to forecast the direction of future research and development on the technology to be analyzed.

• The Journal of the Convergence on Culture Technology
• /
• v.6 no.4
• /
• pp.789-795
• /
• 2020

Many deaths have been reported due to the worldwide pandemic of COVID-19. In order to prevent the further spread of COVID-19, it is necessary to quickly and accurately read images of suspected patients and take appropriate measures. To this end, this paper introduces a deep learning-based COVID-19 chest X-ray reading technique that can assist in image reading by providing medical staff whether a patient is infected. First of all, in order to learn the reading model, a sufficient dataset must be secured, but the currently provided COVID-19 open dataset does not have enough image data to ensure the accuracy of learning. Therefore, we solved the image data number imbalance problem that degrades AI learning performance by using a Stacked Generative Adversarial Network(StackGAN++). Next, the DenseNet-based classification model was trained using the augmented data set to develop the reading model. This classification model is a model for binary classification of normal chest X-ray and COVID-19 chest X-ray, and the performance of the model was evaluated using part of the actual image data as test data. Finally, the reliability of the model was secured by presenting the basis for judging the presence or absence of disease in the input image using Grad-CAM, one of the explainable artificial intelligence called XAI.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.1
• /
• pp.674-681
• /
• 2021

Cyber attacks targeting critical infrastructure are on the rise. Critical infrastructure is defined as core infrastructures within a country with a high degree of interdependence between the different structures; therefore, it is difficult to sufficiently protect it using outdated cybersecurity techniques. In particular, the distinction between the physical and logical risks of critical infrastructure is becoming ambiguous; therefore, risk management from a comprehensive perspective must be implemented. Accordingly, as a means of further actively protecting critical infrastructure, major countries have begun to apply their security and cybersecurity systems by design, as a more expanded concept is now being considered. This proactive security approach (CSbD, Cybersecurity by Design) includes not only securing the stability of software (SW) safety design and management, but also physical politics and device (HW) safety, precautionary and blocking measures, and overall resilience. It involves a comprehensive security system. Therefore, this study compares and analyzes security by design measures towards critical infrastructure that are leading the way in the US, Europe, and Singapore. It reflects the results of an analysis of optimal cybersecurity solutions for critical infrastructure. I would like to present a plan for applying by Design.