• 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.

• International Journal of Computer Science & Network Security
• /
• v.22 no.2
• /
• pp.81-90
• /
• 2022

There are several ways to improve software performance by using existing software. So, the developments of some programs are the most promising ways. However, traditional part programming studies usually assume that the components are recycled "as is". Existing models of component objects only provide limited support for partial adjustments, namely white box technologies ( copy-paste & inheritance) and the black-box methods (such as mixing and encapsulation). These technologies have problems related to recovery, efficiency, implementation of indirect costs, or their own problems. This paper suggests as JALTREE, The Black Box adaptation technology, which allows us for the implementation of previous components, but we need configurable the interface types, for measuring the adaptability. In this article we discussed the types of adjustments including component interfaces and component composition. An example of customizing JALTREE and component can be illustrated in several examples

• Journal of the Korean Wood Science and Technology
• /
• v.32 no.5
• /
• pp.86-92
• /
• 2004

This research was carried out to examine the ethylene gas adsorption and preservation life of white charcoal boards for packaging. Two types of white charcoal board were made of #40-60 charcoal particles and mixed charcoal particles with PVA and MDI resin by wet process. For not only beauty effect but also avoiding stain from touching, white charcoal boards were overlaid with two kinds of overlay material, thin printed paper and non woven fabric. A charcoal board adsorbed ethylene gas much more than the paper and even white charcoal itself. There was no difference between two board types, between overlay treatments, and between overlay materials. Gray mold growth of strawberry arised after 2 days in a paper box and after 4 days in a white charcoal board. Preservation life of white charcoal board box was approximately twice more than that. of a paper box. There was no difference in the ability of keeping freshness between two board types, between overlay treatments, and between overlay materials.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.3
• /
• pp.580-588
• /
• 1999

The reuse of software components from existing software system enhances productivity and reliability, decreases the cost of production in software development. The extraction of software components with high reusability from existing software system is very important in software reuse. The reuse of a class is classified into white-box reuse to reuse with modification and black-box reuse to reuse without modification. A class has the property of procedural language and object-oriented language. Therefore, it must measure reusability in consideration of two properties. In black-box reuse, independence of class is important quality. It can quantify through information hiding, coupling between objects, cohesion, etc. In white-box reuse, modification is the best important quality. It can quality through class complexity, coupling, cohesion, documentation, etc. We propose a new model for measurement of class reusability and the measure criteria in object-oriented program. A class that is measured by proposed model can judge whether the reuse with modification has the advantage or the reuse without modification has the advantage.

• Journal of Digital Convergence
• /
• v.13 no.10
• /
• pp.343-351
• /
• 2015

With the diverse services of the current M2M environment being expanded to the organizations, the corporations, and the daily lives, the possibility of the occurrence of the vulnerabilities of the security of the related technologies have become an issue. In order to solve such a problem of the vulnerability of the security, this thesis proposes the technique for applying the cryptography algorithm for the protection of the device key of the M2M environment. The proposed technique was based on the elliptic curve cryptography Through the key exchange and the signature exchange in the beginning, the security session was created. And the white box cipher was applied to the encryption that creates the white box table using the security session key. Application results cipher algorithm, Elliptic Curve Cryptography provides a lightweight mutual authentication, a session key for protecting the communication session and a conventional white-box cipher algorithm and was guaranteed the session key used to encrypt protected in different ways. The proposed protocol has secure advantages against Data modulation and exposure, MITM(Man-in-the-middle attack), Data forgery and Manipulation attack.

• Journal of the Korean School Mathematics Society
• /
• v.18 no.2
• /
• pp.223-240
• /
• 2015

This paper deals with how to design and develop white-box based e-learning contents which are equipped with conceptual understanding and step-by-step computational procedures for studying vector calculus for science-engineering majors who might need supplementary mathematics learning. Noting that rewriting rules are often used in school mathematics for students' problem solving, the theoretical aspects of rewriting rules are reviewed for developing supplementary e-learning contents for them. The software design of step-by-step problem solving requires careful arrangement of rewriting rules and pattern matching techniques for white-box procedures using a computer algebra system such as Mathematica. Several modules for step-by-step problem solving as well as producing dynamic display of e-learning contents was coded by Mathematica in order to find the length of a curve in vector calculus after implementing several rules for differentiation and integration. The developed contents are equipped with diagnostic modules and immediate feedback for supplementary learning in terms of a tutorial. At the end, this paper indicates the strengths and features of the developed contents for college students who need to increase math learning capabilities, and suggests future research directions.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.12
• /
• pp.1173-1177
• /
• 2015

This paper deals with estimating the acid concentration of pickling process using the Volterra inputs. To estimate the acid concentration, the whole pickling process is represented by the grey box model consists of the white box dealing with known system and the black box dealing with unknown system. Because there is a possibility of nonlinear term in the unknown system, the Volterra series are used to estimate the acid concentration. For the white box modeling, the acid tank solution level and concentration equations are used, and for the black box modeling, the acid concentration is estimated using the Volterra Least Mean Squares (LMS) algorithm and Least Squares (LS) algorithm. The LMS algorithm has the advantage of the simple structure and the low computation, and the LS algorithm has the advantage of lowest error. The simulation results compared to the measured data are included.

• Journal of KIISE:Software and Applications
• /
• v.31 no.6
• /
• pp.741-749
• /
• 2004

Component users must customize components they obtain from providers, in order to fit them to their own purposes. Normally, a component consists of black-box parts and white-box parts. Component users customize a component by modifying white-box parts of a component, and the customization faults appear through the interaction between black-box parts and white-box parts. Customization testing could be an integration testing of these two parts of a component. Also, customization testing in CBSD should select effective test data to reduce the testing cost, since CBSD aims to reduce the development cost. Therefore, this paper proposes a customization testing technique based on COM architecture through analyzing many COM components, and the technique selects effective test data. This paper evaluates the effectiveness of the test data selected by the proposed technique through an empirical study. It applies the techlique to a large-scale component-based system, Chamois, and it shows that the technique enables us to test customized COM components that run in a real component-based system

• Journal of KIISE:Software and Applications
• /
• v.29 no.10
• /
• pp.694-702
• /
• 2002

An application in Component-Based Software Development (CBSD) is built by 'composing'two kinds of components; One is a component that is made by current developer himself, and the other is a component that is from other developments. We define the former as a 'White-box component' and the latter as a 'Black-box component.' The error from the composition can be said to be caused by interactions of Black-box components and White-box components. This paper proposes a new testing technique for composition errors, and applies the technique to Enterprise Java Beans component architecture. Our technique selects test cases by injecting a fault only into the specific parts of a White-box component. This specific parts for injecting a fault are selected by analyzing composition patterns, and lead to make our test cases have a good effectiveness. We show the effectiveness of our test cases through an experiment. Moreover, we also mention an automation tool for our technique.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.2
• /
• pp.19-29
• /
• 2008

In case of unit testing, White-box test can be used to closely check source code and to analyze logic and statement errors. On the other hand, in case of function testing of system level, Black-box test can be mainly used to compare actual and expected results by inputting test data because the scale of function is large. This Black-Box test in system testing level has problem in tracing errors in source code when we find errors. Moreover applying White-box test is not easy for system testing level because the levels of test target are different. Therefore this paper suggests the vertical test method of a practical and integrated system level which can checks up to source code level using White-box test style although it aims to test the highly abstract level like a system function. In addition, the experiment explains how to apply the vertical test by displaying an example which traces from UML specification model to the source code and also shows efficiency of error trace.