• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.6
• /
• pp.567-572
• /
• 2019

In this paper, we propose an algorithm to perform convolution, pooling, and ReLU operations in CNN using binary image and binary kernel. It decomposes 256 gray-scale images into 8 bit planes and uses a binary kernel consisting of -1 and 1. The convolution operation of binary image and binary kernel is performed by addition and subtraction. Logically, it is a binary operation algorithm using the XNOR and comparator. ReLU and pooling operations are performed by using XNOR and OR logic operations, respectively. Through the experiments to verify the usefulness of the proposed algorithm, We confirm that the CNN operation can be performed by converting it to binary logic operation. It is an algorithm that can implement deep running even in a system with weak computing power. It can be applied to a variety of embedded systems such as smart phones, intelligent CCTV, IoT system, and autonomous car.

• Journal of The Korean Association of Information Education
• /
• v.7 no.2
• /
• pp.227-237
• /
• 2003

Conventional studies of Computer Assisted Instruction(CAI) and Intelligent Tutoring System(ITS) have been general patterns to solve problems, so to solve specialized problems, the learner which has the attitude of passiveness should to solve problems including unnecessary processing to the need of the system. Consequently, those are not support the process of creativity and individual problems for the learner to solve the fractional number operations as this study. This study is the design and implementation of ITS on the fractional number addition and subtraction for the supplementary student. Our system can diagnosis mistakes of learning and guide the student to know their errors of learning process automatically And our system assist the learners to study with self-initiative learning, replacement their lacking of learning and control the process of fractional addition and subtraction operation with creativity according to their level. We showed that this system had improved problems of lacking care to supplementary student result in are not enough teachers involved their school and that the learner had achieved the higher learning effect according to the improved self-initiative learning causing this system.

• Journal of Korean Neurosurgical Society
• /
• v.48 no.4
• /
• pp.354-356
• /
• 2010

Partial thrombosis of giant aneurysms is not uncommon however, complete angiographic occlusion occurs less frequently. In the case of non-giant aneurysms, complete thrombosis and recanalization has been rarely reported. A 31-year-old man presented to the emergency department with sudden bursting headache. Brain computed tomography (CT) revealed diffuse subarachnoid hemorrhage on the left side. Both CT angiography (CTA) and digital subtraction angiography showed suspicion of small left anterior choroidal artery aneurysm. We performed surgical exploration. In the operation field, anterior choroidal artery aneurysm of $2{\times}2\;mm$ with broad neck and friable appearance was observed. Because we could not clip without sacrificing the anterior choroidal artery, we performed wrapping only. Follow up CTA after 7 months demonstrated 4 mm right internal carotid artery bifurcation aneurysm. The patient underwent aneurismal neck clipping. During the operation, $9{\times}13\;mm$ sized thrombosed aneurysm was detected and completely clipped. We initially thought this aneurysm to be a de novo aneurysm however, it was an aneurysm that had recanalized from a completely thrombosed aneurysm. This case report provides an insight into the potential for complete thrombosis and recanalization of non-giant aneurysms.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.9 no.3
• /
• pp.13-20
• /
• 2013

In this paper, we propose a multi-scale crack detection method. This method uses decomposition, composition, and shape properties. It is based on morphology algorithm, crack features. We use a morphology operator which extracts patterns of crack. It segments cracks and background using opening and closing operations. Morphology based segmentation is better than existing integration methods using subtraction in detecting a crack it has small width. However, morphology methods using only one structure element could detect only fixed width crack. Thus, we use decomposition and composition methods. We use a decimation method for decomposition. After decomposition and morphology operation, we get edge images given by binary values. Our method calculates values of properties such as the number of pixels and the maximum length of the segmented region. We decide whether the segmented region belongs to cracks according to those data. Experimental results show that our proposed multi-scale crack detection method has better results than those of existing detection methods.

• Proceedings of the IEEK Conference
• /
• 2005.11a
• /
• pp.477-480
• /
• 2005

In this paper, we propose a preprocess method to needs for Car License Plate Recognition on poor environment. This preprocess method use multi images to get low value to compare images value. Last method was Opening operation that Using Edge pixel to add and subtraction. The Result was removed White pixel and very mini feather. But This method needs many process times and License Plate Recognition is low quality problem. Another method is median filter and conversion. This paper key idea that rain & snow is high value. So This paper propose get low value to compare image value.

• Research in Mathematical Education
• /
• v.19 no.2
• /
• pp.89-100
• /
• 2015

The study investigated pre-service teachers' knowledge and computational skills by using Triangular Arithmagon. Participants included 90 pre-service teachers from two schools in the United States and Taiwan. The Triangular Arithmagons Test (TAT) was used to measure pre-service teachers' performance in whole number, fractions, and decimals operations (i.e., addition, subtraction, multiplication, and division), each of which included level-1 (basic) and level-2 (advanced) tests. MANOVA analysis was performed to compare the performance between teachers from the United States and Taiwan. Results indicated that overall, pre-service teachers in Taiwan outperformed those in the United States, especially on the advanced-level tests. Pre-service teachers in the United States were found to have poor ability of solving complex operation problems. Different curriculum plans and teaching methods may lead to the performance gap between the two countries.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2002.05a
• /
• pp.833-835
• /
• 2002

In this paper, we propose multi-resolutional representation of B-rep solid models using the selective Boolean operations on non-manifold geometric models. Since the union and subtraction operations of the selective Boolean operations are commutative, the integrity of the model is guaranteed for reordering design features. A multi-resolution representation is established using a non-manifold merged set model and a feature modeling tree reordered according to some criterion of level of detail (LOD). Then, a solid model for a specified LOD can be extracted from this multi-resolution model using the selective Boolean operations.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2002.05a
• /
• pp.836-839
• /
• 2002

This paper describes the selective Boolean operations on non-manifold geometric models whose union and subtraction operations are communicative. These operations guarantee the same resulting shape in spite of change of the order of Boolean operations, and the integrity of the model for omission of some features. In addition, a B-rep model for a modified modeling history is obtained in a short time, as no boundary evaluation is necessary. These features enable easy implementation of multi-resolution representation of B-rep models.

• Proceedings of the KOSOMBE Conference
• /
• v.1994 no.12
• /
• pp.26-27
• /
• 1994

The first step implementation of a hospital-wide Picture Archiving Communications System (PACS) at a newly built hospital Samsung Medical Center (SMC), is described. Current clinical operation encompasses the fiber optics delivery of direct-interfaced magnetic resonance imager (LRI), X-ray computed tomography (CT). digital subtraction angiography (DSA) and computed radiography (CR) digital images via high performance file server to the departments of radiology, neurosurgery, orthopedics surgery, neurology, emergency room and the surgical intensive care unit.

• Journal of Mechanical Science and Technology
• /
• v.19 no.2
• /
• pp.558-566
• /
• 2005

We propose a feature-based multi-resolution representation of B-rep solid models using history-based Boolean operations based on the merge-and-select algorithm. Because union and subtraction are commutative in the history-based Boolean operations, the integrity of the models at various levels of detail (LOD) is guaranteed for the reordered features regardless of whether the features are subtractive or additive. The multi-resolution solid representation proposed in this paper includes a non-manifold topological merged-set model of all feature primitives as well as a feature-modeling tree reordered consistently with a given LOD criterion. As a result, a B-rep solid model for a given LOD can be provided quickly, because the boundary of the model is evaluated without any geometric calculation and extracted from the merged set by selecting the entities contributing to the LOD model shape.