• Journal of the Society of Naval Architects of Korea
• /
• v.57 no.3
• /
• pp.140-151
• /
• 2020

It is important to identify the location of ship hull blocks with exact block identification number when scheduling the shipbuilding process. The wrong information on the location and identification number of some hull block can cause low productivity by spending time to find where the exact hull block is. In order to solve this problem, it is necessary to equip the system to track the location of the blocks and to identify the identification numbers of the blocks automatically. There were a lot of researches of location tracking system for the hull blocks on the stockyard. However there has been no research to identify the hull blocks on the stockyard. This study compares the performance of 5 Convolutional Neural Network (CNN) models with multi-view image set on the classification of the hull blocks to identify the blocks on the stockyard. The CNN models are open algorithms of ImageNet Large-Scale Visual Recognition Competition (ILSVRC). Four scaled hull block models are used to acquire the images of ship hull blocks. Learning and transfer learning of the CNN models with original training data and augmented data of the original training data were done. 20 tests and predictions in consideration of five CNN models and four cases of training conditions are performed. In order to compare the classification performance of the CNN models, accuracy and average F1-Score from confusion matrix are adopted as the performance measures. As a result of the comparison, Resnet-152v2 model shows the highest accuracy and average F1-Score with full block prediction image set and with cropped block prediction image set.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.5
• /
• pp.619-624
• /
• 2023

As the importance of digital competency education is highlighted, this study is a case study on block coding classes for elementary school students during vacation for the purpose of bridging the information education gap among students. The purpose of this study is to design and operate a play-centered block coding class program and find out if it is effective in improving students' interest. As a result of completing the teaching plan through the second consultation and revision, running the class, and analyzing the change in learning interest of the students through the t-test, the play-oriented block coding class designed in this study was effective in improving students' interest. In addition, it was possible to discover interesting elements such as student-led learning process and immersion through realistic play activities, friendship, collaboration, and communication through group activities. This study is significant in suggesting a plan to increase learning interest for students who are new to coding.

• Journal of Korea Multimedia Society
• /
• v.19 no.2
• /
• pp.325-333
• /
• 2016

The development of smart technology has brought the conversion of closed traditional e-learning contents into open flexible smart learning contents consisting of learner-centered modules, without the constraints of time and space by use of smart devices from the uniformed and passive classroom between teachers and learners. It has been demanded an open, personalized and customized teaching and learning contents of smart education and training systems according to wide supply of various smart devices. In this paper, we discuss about the status of the smart teaching and learning systems and analyze the characteristics and structure of the web contents for smart education and training systems by use of smart devices. And we propose a method how to block web contents, to extract them, and adapt personalized segments of web contents by adaptive algorithm into smart learning devices. We extract blocks from the web contents based on the smart device information and the preference information of the learners from existing web contents without the hassle of learners environment. After specifying a block priority from the extracted web contents by the adaptive segment algorithm, it can be displayed directly to the screen to fit the individual learning progress of the learners.

• Education of Primary School Mathematics
• /
• v.26 no.1
• /
• pp.65-82
• /
• 2023

The purpose of this study is to examine how pre-service teachers' digital capabilities and content knowledge for teaching pi appear and are strengthened in the process of developing digital-based teaching and learning materials of pi, and to derive implications for pre-service teacher education. To this end, the researchers analyzed the process of two pre-service teachers developing exploratory activity materials for teaching pi using block coding of AlgeoMath program. Through the analysis results, it was confirmed that AlgeoMath' block coding activities provided an experience of expressing and expanding the digital capabilities of pre-service teachers, an opportunity to deepen the content knowledge of pi, and to recognize the problems and limitations of the digital learning environment. It was also suggested that the development of digital materials using block coding needs to be used to strengthen digital capabilities of pre-service teachers, and that the curriculum knowledge needs to be emphasized as knowledge necessary for the development of digital teaching and learning materials in pre-service teacher education.

• The Journal of Korean Physical Therapy
• /
• v.20 no.3
• /
• pp.1-7
• /
• 2008

Purpose: Implicit motor learning is the capacity to acquire skill through physical practice without conscious awareness of what elements of performance improved. This study investigated whether subacute stroke patients can implicitly learn a perceptual-motor task. Methods: We recruited 12 patients with subacute stroke and 12 age-matched controls. All participants performed a perceptual-motor task that involved pressing a button corresponding with colored circles (blue, green, yellow, red) on a computer screen. The task consists of 7 blocks composed of 10 repetitions for a repeating 12-element sequence (total 120 responses). Results: Both groups demonstrated significant improvement in acquisition performance. Reaction times deceased in both groups at similar rate within the sequential block trials (2-5 blocks), and reaction times increased at a similar rate when the task paradigm was transferred from the sequential block trial to the random block trial (5-6-7 blocks). Conclusion: The results of this study suggest that patients with sub-actue stroke can implicitly learn a perceptual motor skill. Although explicit instructions should be used to focus the learner's attention rather than provide information about the task, the application of implicit motor learning strategies in the rehabilitation setting may be beneficial.

• Information Systems Review
• /
• v.22 no.1
• /
• pp.1-16
• /
• 2020

Blockchain is emerging as a technology that can build trust between users participating in the system. As interest of Blockchain has increased, previous studies have mainly focused on cryptocurrency and application methods related to Blockchain technology. On the other hand, the studies on the stable implementation of Blockchain were rarely conducted. Typically, uncle block in the Blockchain plays an important role in the stable implementation of the Blockhain system, but no study was conducted on this. Drawing on this recognition, this study attempts to predict the uncle block of Blockchain using machine learning method, Blockchain information, and macro-economic factors. The results of artificial neural network and support vector machine analysis, Blockchain information and macro-economic factors contributed to the prediction of uncle block of Blockchain. In addition, artificial neural network using only Blockchain information provided the best performance for predicting the occurrence of uncle block. This study suggests ways to lead and contribute to Blockchain research in information systems filed.

• Physical Therapy Korea
• /
• v.11 no.2
• /
• pp.1-8
• /
• 2004

Motor skill learning can be acquired implicitly without consciousness of what is being learned. The purpose of this study was to examine the characteristics of implicit motor learning in young and elderly people using a perceptual-motor task. Forty normal young and elderly subjects participated. A modified version of the Serial Reaction Time Task (SRTT) using six blocks of twelve perceptual motor sequences was administered. The paradigm consisted of the first random sequence block followed by the four patterned blocks and another random block. In each block, the go signal consisted of an asterisk displayed in the one of the four parallel arrayed boxes in the middle of the screen. Subjects were instructed to push the corresponding response buttons as quickly as possible. Young subjects demonstrated shorter reaction times during the consecutive patterned blocks reflecting appropriate learning accomplished. Elderly subjects were able to learn a perceptual-motor task with implicit knowledge, but the performance was lower than that of the young persons. These results indicated that implicit sequence learning is still preserved in elderly adults, but the rate of learning is slower.

• Journal of the Semiconductor & Display Technology
• /
• v.21 no.3
• /
• pp.1-6
• /
• 2022

Modern computer systems usually have special hardware for operations used in deep learning workload even edge computing environment. Non-volatile memories (NVMs) have been considered for alternative memory storage because they consume little static energy and occupy small area. However, there is a problem for NVMs to be directly adopted. An NVM cell has limited write endurance, so that the lifetime of NVM-based memory system is much shorter than that of conventional memory system. To overcome this problem for the deep learning system, this paper proposes a novel method to extend the lifetime based on the analysis of the deep learning workloads. If an incoming block has more than a predefined number of frequently used values, the cacheline is defined as write friendly block. During the victim selection, the cacheline has lower possibility to be chosen as victim. The experimental results show that the lifetime is increased by about 50% and energy consumption is decreased by 3% with a little performance hurt.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.4
• /
• pp.1-9
• /
• 2023

In this paper, we propose a method to recognize the braille blocks for embedded devices in real time through deep learning. First, a deep learning model for braille block recognition is trained on a high-performance computer, and the learning model is applied to a lightweight tool to apply to an embedded device. To recognize the walking information of the braille block, an algorithm is used to determine the path using the distance from the braille block in the image. After detecting braille blocks, bollards, and crosswalks through the YOLOv8 model in the video captured by the embedded device, the walking information is recognized through the braille block path discrimination algorithm. We apply the model lightweight tool to YOLOv8 to detect braille blocks in real time. The precision of YOLOv8 model weights is lowered from the existing 32 bits to 8 bits, and the model is optimized by applying the TensorRT optimization engine. As the result of comparing the lightweight model through the proposed method with the existing model, the path recognition accuracy is 99.05%, which is almost the same as the existing model, but the recognition speed is reduced by 59% compared to the existing model, processing about 15 frames per second.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1091-1100
• /
• 2022

As deep learning technology advances and various high-resolution remote sensing images are available, interest in using deep learning technology and remote sensing big data to detect buildings and change in urban areas is increasing significantly. In this paper, for semantic building segmentation of high-resolution remote sensing images, we propose a new building segmentation model, Convolutional Block Attention Module (CBAM)-DRUNet that uses the DeepResUNet model, which has excellent performance in building segmentation, as the basic structure, improves the residual learning unit and combines a CBAM with the basic structure. In the performance evaluation using WHU dataset and INRIA dataset, the proposed building segmentation model showed excellent performance in terms of F1 score, accuracy and recall compared to ResUNet and DeepResUNet including UNet.