KJC Korea
Journal Central

Recently, illegal copyright infringement websites have been utilizing cloud services such as Cloudflare, which makes investigation and enforcement challenging. They evade domain blocking by regularly changing domains. Existing domain blocking technologies follow the same process even when a modified illegal site is discovered and these illegal sites bypass blocking by reducing the domain change interval. In this paper, we propose an approach to define and collect feature information about target sites, such as site logos, menus, categories, content, and advertisement banners, based on publicly available information within the sites that can be utilized to identify target sites engaged in illegal activities. Through experiments, it was found that the feature information values of sites before and after domain change were the same. Accordingly, it is expected to contribute to establishing a system that can take immediate action by reducing the unnecessary verification and screening period in blocking illegal copyright infringement sites.

As the creation of various types of copyrighted works increases, the license for the use and utilization of original works is increasing accordingly. Since the license contract is not a general contract used by the general public, but applies to specific technologies, it is limited to only those who need a license, so contracts are not generally concluded. Currently, there are several license models, but in order to use licenses owned by individual platforms, contract provisions are unified, and there are limitations in contract technology for managing and restricting users. In the case of existing copyright contracts, the terms of the contents of the contract are extensive, so it is necessary to develop a license contract template that enables the use of copyrighted works under specific conditions to be reviewed by interested parties. Therefore, this paper suggests a license contract model for each condition of use that can reduce the burden of reviewing the license contract and easily obtain license under specific conditions.

In quantitative similarity process, the role of the engineer and the role of the judge overlap partly. In the past, there was a problem that this overlapping area remained blank in which no one was involved. However, many practical and theoretical studies on copyright infringement litigation are being conducted. Therefore, in recent years, the gap as in the past has been decreasing. In litigation practice, the use of quantitative similarity to determine practical similarity, recognition of its limitations, and the way the filtration process works to specify expressions protected by copyright law are treated very importantly. The Supreme Court recently responded to these matters through the Supreme Court Judgment 2021Da246804 Sentenced December 1, 2022. The judgment approached the overall process of judging copyright infringement faithfully, but there was some carelessness found. So, this point was examined in this paper.

Although the proportion of using other people's works in the creative activities of individual creators is increasing, users are feeling the burden of going through legal procedures such as licensing agreement procedures suitable for the purpose of using the works. Therefore, it is necessary to develop a platform that can reduce the burden of contracts by limiting the conditions and rights of creators and users, and approve and automate requests under conditions desired by creators and users during contracts. In this study, we propose a platform model that can automatically approve a license agreement request for a copyrighted work for a limited purpose defined in advance by the copyright owner. Through this, it is expected to contribute to vitalizing the content market by supporting direct contracts between creators and users and by increasing re-usability of copyrighted works through simple licensing agreements.

In recent times, with the rapid proliferation of the metaverse environment, there has been a notable upswing in instances of performance within virtual realms, capitalizing on the inherent characteristics of this environment. These performances, conducted within virtual settings, are commonly referred to as virtual performances. The distinctive advantage of these virtual performances lies in their unrestricted accessibility, transcending the spatial confines of the physical world. As such, they are emerging as a viable substitution for overseas concerts that might be impeded by spatial constraints brought about by infectious disease pandemics or geographical remoteness. These virtual performances integrate a myriad of digital components, termed digital assets, to coalesce into a singular content. Consequently, the inherent trait of digital data, characterized by its ease of duplication and reuse, can be directly leveraged. Nevertheless, there is a concurrent rise in instances of copyright infringement, stemming from the unauthorized replication, reuse, and manipulation of these features. In this study, we propose technologies centered around redundancy and similarity determination for digital data, aiming to address these issues. This technology is employed to ascertain whether a given dataset corresponds to a previously registered digital asset or if it has undergone partial modifications when an author endeavors to register a new digital asset within the digital asset market. Through this, this study blocks digital asset data, divides it into several pieces, and determines the redundancy and similarity of digital assets by analyzing whether the blocks match existing registered digital assets. Through this, it is possible to prevent unauthorized registration of digital assets that have been copied or modified by rights infringers in advance.

In order to prevent safety accidents in urban underground spaces, an integrated underground space map is being generated. To shorten the update time of the map, a changed object detection and extraction technology is applied, which selects only updated objects from newly input maps. Updated objects are classified into a new object, a deleted object, and an attribute changed object. However, when a partial map including only buried facilities that are targets of excavation work is input, a problem of not finding a deleted object occurs. In this paper, we discussed three methods to find deleted objects in the partial map. In order to detect the deleted object, it must be determined whether an updated object exists in which only part of the original objects has been changed. DE-9IM, Relate function, and Intersection function were used to extract partially changed objects, and the performances of these methods were compared by using linear objects. All three methods determined objects including overlapping shapes as the same object without errors, and the Relate function showed the fastest performance.

Although Deep Neural Network (DNN) shows excellent performance in pattern recognition such as image processing and voice recognition, it has the disadvantage of requiring intensive computing powers for real-time processing. This paper presents a programmable On-a-chip hardware for embedded smart applications based on Deep Neural Network (DNN) computations. To implement and test the proposed on-chip system, an FPGA-based development platform is produced and used. Since the main processor and related hardwares can be implemented in the same FPGA using the platform, all hardware blocks can be integrated and verified easily. A operation program is provided for On-a-chip system core initialization, operation, and software development environment. The effectiveness of the proposed On-a-chip system is shown through an experiment implementing an actual application system based on DNNs.

In this paper, a method for improving the resolution of digital-to-analog conversion with the minimum number of ports using the three-state characteristics of micro-controller ports is presented. To verify the validity of the proposed method, a digital-to-analog conversion circuit consisting of two-state digital that is currently being used was analyzed and compared with the proposed three-state method. In order to prove the effectiveness of the proposed method, an experimental circuit was constructed and an experiment was conducted. As a result of the experiment, it was confirmed that the resolution increased by 225% from step 4 to step 9 when using 2 bits compared to the existing two-state method, and increased by 338% from step 8 to step 27 when using 3 bits.

In this paper, we proposed a method of applying ROI to the lane prediction area in the lane recognition algorithm through the camera in autonomous driving. In the lane recognition method using a camera, lane recognition may be difficult due to environmental factors such as vehicles and obstacles on the road. In order to solve this problem, after getting an image through a camera, an ROI is allocated by estimating where lane detection is needed, and after removing the area of another vehicle adopting Deeplab-V3, lane recognition is performed through perspective, pre-processing, and Hough transformation. An algorithm that implements above transformations is proposed. A test image for the experiment used an urban driving image with many vehicle elements that can affect lane recognition. As a result of the experiment which setting the ROI for a specific area, it was possible to increase the accuracy of lane recognition by removing other vehicle areas and recognizing lanes.

The detection of buried objects using Ground Penetrating Radar utilizes hyperbolic curves visualized by the variation of reflected signals due to the relative distance difference between the antenna and the buried objects. However, these hyperbolic signals are also created by various objects buried underground, making it difficult to identify and detection target. This difficulty can be solved through a plan-view image viewed from above along with the same hyperbolic image for each channel as the existing one created based on the 3D data of multi-channel GPR. In this paper, we discussed the transmitter-receiver antenna array and signal processing procedure for constructing plan-view images using multi-channel GPR for underground object detection. And we also verified the effectiveness of the plan-view images constructed using multi-channel GPR survey data through detection tests on object with known sizes.

Ground-Penetrating Radar(GPR) system can detect buried objects such as underground cavities and mines by emitting electromagnetic waves and visualizing signals reflected from underground facilities by emitting electromagnetic waves. But in order to distinguish buried objects, it is necessary to understand signal patterns for each target and it is difficult to detect signal patterns for each target in real time. In this study, adaptive signal processing technique is applied to reduce the effect of non-uniform clutter signals on images to increase the visibility of target signals, and energy profile extraction technique is used to simplify the identification of target signals. In addition, the 2.5D template matching technique is applied to calculate matching rate of the detected object. Several objects selected in soil-filled test sites were buried in various locations to confirm their performance through real-time detection software, and it is expected to be applied to mine detection systems using real-time automatic exploration in the future.

To mitigate security risks, organizational assets must be identified and security vulnerabilities eliminated. active scanning and passive scanning techniques have been applied in existing IT environments to identify assets.. In this paper, we first examine the problems of applying the existing network scanning methods to OT systems. Then, to select network traffic features required when identifying OT devices by applying a passive scanning, we analyze network traffic features (TTL, TCP Window Size, IAT, etc.) used in prior studies on identifying IoT devices which have similar properties to OT devices. In addition, we recognize the types of network traffic features, and present research issues to be addressed for effective identification of OT devices.