• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06a
• /
• pp.116-118
• /
• 2012

안드로이드 앱 시장이 활성화되면서, 안드로이드 앱의 불법복제나 역공학 공격으로 인한 피해가 증가하고 있다. 앱 불법복제는 앱 판매 수익의 저하뿐만 아니라 개발자의 의지를 뺏고 개발 노력에 대해 상대적 박탈감을 주게 된다. 자바 프로그램의 경우 역공학으로 인해 바이트 코드에 존재하는 핵심 알고리즘이 쉽게 노출되어 지적재산권이 유출될 수 있다는 점에서 개발자나 개발사에게 심각한 위협이 되고 있다. 본 논문에서는 안드로이드 환경에서 앱에 대한 역공학 공격의 위협을 보이고, 역공학 방지 기법인 RACC를 제안한다. RACC는 보호할 핵심 클래스를 앱으로 부터 추출하여 바이트 코드 형태로 안전한 원격 서버에 관리하여 수행하며, 스마트폰(클라이언트)에는 저장하지 않는다. 스마트폰 앱이 해당 핵심 클래스를 호출하면, 그 호출이 원격 서버로 전송되어 수행된 후 결과가 스마트폰에 반환된다. 이처럼 핵심 클래스 코드가 클라이언트에 직접 노출 없이 원격지에서 관리되고 수행됨으로써 역공학 공격을 원천적으로 방지한다.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.3
• /
• pp.107-115
• /
• 2006

This paper presents a new class of activation functions for Cascade Correlation learning algorithm, which herein will be called CosGauss function. This function is a cosine modulated gaussian function. In contrast to the sigmoidal, hyperbolic tangent and gaussian functions, more ridges can be obtained by the CosGauss function. Because of the ridges, it is quickly convergent and improves a pattern recognition speed. Consequently it will be able to improve a learning capability. This function was tested with a Cascade Correlation Network on the two spirals problem and results are compared with those obtained with other activation functions.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.1
• /
• pp.23-33
• /
• 1995

This paper describes a visual programming tool which provides novice programmer as well expert with the abilites to capture real physical world of problem domain and to manipulate it user-friendly using icons and symbols. Therefore, novice can understand object-oriented features of C＋＋ incrementally and construct classes easily. For this, we introduce some visual metaphors which are displayed as tables. The tables can not only represent objects and classes, but also be considered themselves as icons. We have named these tables as table-icons. Three levels of table-icons(i.e., Super Table-Icons, Intermediate Table-Icons and Detailed Table-Icons) are proposed to follow up appropriate evolution of object-oriented concepts. Table-icons are not simple pictographs but are activated and expanded to table forms. And then, developer can inset necessary entities into table body and delete useless entities from it. These table-icons are applied to a diagramming technique, C＋＋gram［18］, which is suggested for designing and implementing C＋＋ programs.

• Journal of Korea Multimedia Society
• /
• v.13 no.7
• /
• pp.1076-1087
• /
• 2010

Together with a new advent of embedded processor developed to support specific application area and it evolution, a new research of software development to support the embedded processor and its commercial challenge has been revitalized. Retargetability is typically achieved by providing target machine information, ADL, as input. The ADLs are used to specify processor and memory architectures and generate software toolkit including compiler, simulator, assembler, profiler, and debugger. The EXPRESSION ADL follows a mixed level approach-it can capture both the structure and behavior supporting a natural specification of the programmable architectures consisting of processor cores, coprocessors, and memories. And it was originally designed to capture processor/memory architectures and generate software toolkit to enable compiler-in-the-loop exploration of SoC architecture. In this paper, we designed the class-based ADL based on the EXPRESSION ADL to promote the write-ability, extensibility and verified the validation of grammar. For this works, we defined 6 core classes and generated the EXPRESSION's compiler and simulator through the MIPS R4000 description.

• Journal of the Korea Computer Industry Society
• /
• v.4 no.12
• /
• pp.937-942
• /
• 2003

Component development of EJB base is activated newly been using widely object oriented analysis and design method recently. Usually, case that use permanent data that use database in EJB base bean creation is much. Separated DBMS connection that various DBMS is applied been subordinate to each entity bean used in the environment in this paper applying DAO pattern, and is normalized object oriented design and analysis because using UML and standardization so that can create done class optimized modelling do.

• Journal of KIISE
• /
• v.45 no.1
• /
• pp.94-98
• /
• 2018

We propose Wide Inception ResNet (WIR Net) an optimized neural network architecture as a real-time semantic segmentation method for autonomous driving. The neural network architecture consists of an encoder that extracts features by applying a residual connection and inception module, and a decoder that increases the resolution by using transposed convolution and a low layer feature map. We also improved the performance by applying an ELU activation function and optimized the neural network by reducing the number of layers and increasing the number of filters. The performance evaluations used an NVIDIA Geforce GTX 1080 and TX1 boards to assess the class and category IoU for cityscapes data in the driving environment. The experimental results show that the accuracy of class IoU 53.4, category IoU 81.8 and the execution speed of $640{\times}360$, $720{\times}480$ resolution image processing 17.8fps and 13.0fps on TX1 board.

• Journal of Internet Computing and Services
• /
• v.4 no.1
• /
• pp.75-86
• /
• 2003

In this paper, we extended hierarchial structure of managed object class to support Role-Based Access Control, and described constraint conditions that have support dynamic temporal function as well as statical temporal function established by management process. And we defined about violation notifications should report to manager when rules violate constraint conditions. Also we presented system architecture that support RBAC with MIB(Management Information Base) of ITU-T recommendation. By access control enforcement and decision function, constraint conditions and activated translation procedure of each roles are described, our system presents dynamic temporal property systematically.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.4
• /
• pp.145-151
• /
• 2021

In this paper, we propose a novel visualization technique to explain the predictions of deep neural networks. We use knowledge distillation (KD) to identify the interior of a black-box model for which we know only inputs and outputs. The information of the black box model will be transferred to a white box model that we aim to create through the KD. The white box model will learn the representation of the black-box model. Second, the white-box model generates attention maps for each of its layers using Grad-CAM. Then we combine the attention maps of different layers using the pixel-wise summation to generate a final saliency map that contains information from all layers of the model. The experiments show that the proposed technique found important layers and explained which part of the input is important. Saliency maps generated by the proposed technique performed better than those of Grad-CAM in deletion game.

• Journal of Drive and Control
• /
• v.20 no.2
• /
• pp.15-23
• /
• 2023

In this study, two-dimensional location of crops for auto weeding was detected using deep learning. To construct a dataset for soybean detection, an image-capturing system was developed using a mono camera and single-board computer and the system was mounted on a weeding robot to collect soybean images. A dataset was constructed by extracting RoI (region of interest) from the raw image and each sample was labeled with soybean and the background for classification learning. The deep learning model consisted of four convolutional layers and was trained with a weakly supervised learning method that can provide object localization only using image-level labeling. Localization of the soybean area can be visualized via CAM and the two-dimensional position of the soybean was estimated by clustering the pixels associated with the soybean area and transforming the pixel coordinates to world coordinates. The actual position, which is determined manually as pixel coordinates in the image was evaluated and performances were 6.6(X-axis), 5.1(Y-axis) and 1.2(X-axis), 2.2(Y-axis) for MSE and RMSE about world coordinates, respectively. From the results, we confirmed that the center position of the soybean area derived through deep learning was sufficient for use in automatic weeding systems.

• The Journal of the Korea Contents Association
• /
• v.16 no.2
• /
• pp.152-162
• /
• 2016

The purpose of this study was to find the mediation effect of school adjustment between perceived parental attitude and mental health in adolescents. 544 adolescents who live in Chungbuk filled out questionnaires. Data were analyzed by regression analysis with SPSS 18.0 version. Results showed significant relationships between socio-economic variables and main variables. Female showed higher depression and suicidal ideation than male. Higher performance group perceived parental attitude more positively, showed lower level of depression and suicidal ideation, and higher level of school adjustment than lower performance group. In addition, perceived parental attitude was a predictor of adolescent' mental health. School adjustment proved to have the mediating effect on depression and suicidal ideation overall. The specific interventions are associated with the increase of professionals and prevention program with parents, and the activation of school welfare.