• The Journal of Korea Robotics Society
• /
• v.15 no.2
• /
• pp.184-189
• /
• 2020

This paper is a study on data augmentation for small dataset by using deep learning. In case of training a deep learning model for recognition and classification of non-mainstream objects, there is a limit to obtaining a large amount of training data. Therefore, this paper proposes a data augmentation method using perspective transform and image synthesis. In addition, it is necessary to save the object area for all training data to detect the object area. Thus, we devised a way to augment the data and save object regions at the same time. To verify the performance of the augmented data using the proposed method, an experiment was conducted to compare classification accuracy with the augmented data by the traditional method, and transfer learning was used in model learning. As experimental results, the model trained using the proposed method showed higher accuracy than the model trained using the traditional method.

• Journal of the Korean Society for Railway
• /
• v.4 no.3
• /
• pp.102-109
• /
• 2001

In this paper, it has a proposal of the RFID reader antenna design that expand the dedicated short-range communication distance between a static object on the ground and a mobile object attached on the moving article. The static reader equipped with micro-processor makes it possible to have a serial communication with a main system, so that much data can be transfer to the main system. An antenna is adjusted in order to a communication, the scale is designed by results values of simulation using matlab. It is achieved to systematically manage logistics, person resource and security system by grasping the information and location of mobile object on the basis that this system receives the information between a static reader and a mobile object tag at 134.2kHz band on real time, also to make it possible the main system to process. Therefore, the reader antenna scale is controlled on the foundation of a magnetic field theory in order to expand a recognition distance of reader and tag, so that can be optimistically recognized with minimizing the direction influence of reader and tag.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.9
• /
• pp.137-145
• /
• 2015

The Criminal Law of South Korea has needed to cope with new kinds of crimes such as the fraud by use of computer efficiently in a legislative way because the society has witnessed the rapid progress of the industrialization and informatization after established in 1953. As a result, the Criminal Law revised on December 29, 1995, created the regulations of the crimes related to fraud by use of computer, work disturbance, and secret piracy by using information processing units. The regulation stipulated in Clause 347, Article 2 of Criminal Law is the most typical one against the new crimes. However, the new regulation of fraud by use of computer, established and revised to supplement the lacking parts of the current rules of the punishment of fraud, limits its object to "any benefits to property." not to "property" itself, and so cannot achieve the purpose of the revision of the law. This paper aims to suggest a new legislative measure about the object of the regulation of fraud by use of computer to solve this kind of problem efficiently.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.222-224
• /
• 2022

Deep learning-based computer vision applications in the field of object segmentation take a transfer learning method using hyperparameters and models pretrained and distributed by STOA techniques to improve performance. Custom datasets used in this process require a lot of resources, such as time and labeling, in labeling tasks to generate Ground Truth information. In this paper, we present an automatic labeling construction method using object segmentation techniques so that resources such as time and labeling can be used less to build custom datasets used in deep learning neural networks.

• Journal of the Korean Society for Heat Treatment
• /
• v.35 no.6
• /
• pp.327-336
• /
• 2022

In this study, a model for automatically recognizing several steel parts through a camera before charging materials was developed under the assumption that the temperature distribution in the pre-air atmosphere was known. For model development, datasets were collected in random environments and factories. In this study, the YOLO-v5 model, which is a YOLO model with strengths in real-time detection in the field of object detection, was used, and the disadvantages of taking a lot of time to collect images and learning models was solved through the transfer learning methods. The performance evaluation results of the derived model showed excellent performance of 0.927 based on mAP 0.5. The derived model will be applied to the model development study, which uses the model to accurately recognize the material and then match it with the temperature distribution in the atmosphere to determine whether the material layout is suitable before charging materials.

• THE INTERNATIONAL COMMERCE & LAW REVIEW
• /
• v.59
• /
• pp.3-26
• /
• 2013

The purpose of this paper is to examine the several points at issue in international technology transfer contract from licensor's and licensee's perspectives, and to refer them when the licensor and the licensee draw up the contract. This author analyzed the critical points of the related provisions of ICC Model International Technology Transfer Contract(2009) by citing the explanations of the Introduction of the Model Contract. The provisions of the Model Contract are generally divided into two categolies; specific conditions and general conditions. This author selected four topics in the specific conditions; Contents of the Contract, Royalty, Modification and/or Improvements of Products, and Territory and Competition. Likewise this author selected three topics in the general conditions; Resolutions of Disputes, Applicable Law and Taxes. Both parties need to be mindful of the following points in the above topics, when they draw up the contract. First, both parties should make the definitions of special terminologies clear, which are included in the Contract. Second, before the parties sign the Contract, they should check any approvals to be necessary by the both countries' governments. Third, for the calculation of the royalty, they should clear the criteria, the scope, and the object. Fourth, as the local laws or regional laws regarding the territory limitation and taxation are mandatory, any provisions of the Contract should not be inconsistent with them. Therefore, both parties should check before-hand the local laws or rules related with the provisions of the Contract. Fifth, when the parties draw up the Contract, they should examine the Provisions of Dispute Resolution in consideration of the Governing Law. Thus both parties decide to make the technology transfer contract, the three aspects namely profitability, legal conflict with mandatory rules, and sustainability of the business resulting from the Contract should be examined in advance, and then proceed the business using the technology transfer.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.25 no.10
• /
• pp.1346-1354
• /
• 2001

The heat transfer characteristics off swirling air jet impinging on a heated flat plate have been investigated experimentally. The main object is to enhance the heat transfer rate by increasing turbulence intensity of impinging jet with a specially designed swirl generator. The mean velocity and turbulent intensity profiles of swirling jet were measured using a hot-wire anemomety. The temperature distribution on the heated flat surface was measured with thermocouples. As a result the swirl effect on the local heat transfer rate on the impinging plate is confined mainly in the small nozzle-to-plate spacings such as L/D<3 at the stagnation region. For small nozzle-to-plate spacings, the local heat transfer in the stagnation region is enhanced from the increased turbulence intensity due to swirl motion, compared with the conventional axisymmetric impinging jet without swirl. For example, the local Nusselt number of swirling jet with swirl number Sw=0.75 and Sw=1 is about 9.7-76% higher than that of conventional impinging jet at the radial location of R/D=0.5. With the increase of the nozzle-to-plate distance, the stagnation heat transfer rate is decreased due to the diminishing axial momentum of the swirling jet. However, the swirling impinging jet for all nozzle-to-plate spacings tested in this study does not enhance the average heat transfer rate.

• Journal of computational fluids engineering
• /
• v.19 no.1
• /
• pp.1-6
• /
• 2014

Since experiments on the actual operational status are said to be very impractical because of their economic and repeatability problems, it is difficult to understand the thermal profiles of aerospace or military equipments. Thus, the CFD codes with considering the radiation heat transfer are used to compensate the defect. In case, analyzing the radiation exchanges between the object surfaces are very important. Because the temperature and the IR signal distributions of the object surface are significantly affected by the radiative heat transfer. To achieve accurate thermal radiation exchange between surfaces, it is important to calculate the radiation view factor precisely. Finer subdivision of meshes can be used to increase the accuracy of radiation view factor, but if the mesh is subdivided infinitely, the time required for calculation increases significantly and thus decreasing the efficiency. If the subdivision is not sufficient, assurance of accuracy is not guaranteed. In this paper, optimal mesh subdivision method using the solid angle has been successfully tested and found to be useful in increasing the efficiency of calculating the shape factors.

• Korean Journal of Optics and Photonics
• /
• v.31 no.3
• /
• pp.134-141
• /
• 2020

Optical three-dimensional (3D) measurement systems based on digital fringe projection are used in many contactless measurement applications. The system which can measure a dozen micrometers uses a liquid-crystal display (LCD) as the projection unit for generating a digital fringe pattern, because a flexible fringe pattern can be easily made by computer software. According to the gamma nonlinearity of the LCD projection unit, the digital fringe projection error on the object affects the accuracy of 3D object measurement. An improved method of LCD gamma-nonlinearity error reduction is proposed, by using the inverse function of the intensity transfer function to improve the accuracy. The improvement due to the proposed method is shown by measuring the difference in precision between a computer-generated sine wave and a camera-obtained sine wave for a standard semiconductor specimen.

• Journal of Platform Technology
• /
• v.11 no.4
• /
• pp.3-12
• /
• 2023

In the post-COVID era, the importance of quarantine measures is greatly emphasized, and accordingly, research related to the detection of mask wearing conditions and prevention of other infectious diseases using deep learning is being conducted. However, research on the detection and tracking of visitors to cultural facilities to prevent the spread of diseases is equally important, so research on this should be conducted. In this paper, a convolutional neural network-based object detection model is trained through transfer learning using a pre-collected dataset. The weights of the trained detection model are then applied to a multi-object tracking model to monitor visitors. The visitor detection model demonstrates results with a precision of 96.3%, recall of 85.2%, and an F1-score of 90.4%. Quantitative results of the tracking model include a MOTA (Multiple Object Tracking Accuracy) of 65.6%, IDF1 (ID F1 Score) of 68.3%, and HOTA (Higher Order Tracking Accuracy) of 57.2%. Furthermore, a qualitative comparison with other multi-object tracking models showcased superior results for the model proposed in this paper. The research of this paper can be applied to the hygiene systems within cultural facilities in the post-COVID era.