• Journal of Sensor Science and Technology
• /
• v.30 no.1
• /
• pp.15-19
• /
• 2021

Image processing is one of the major techniques that are used for computer vision. Nowadays, researchers are using machine learning and deep learning for the aforementioned task. In recent years, digit recognition tasks, i.e., automatic meter recognition approach using electric or water meters, have been studied several times. However, two major issues arise when we talk about previous studies: first, the use of the deep learning technique, which includes a large number of parameters that increase the computational cost and consume more power; and second, recent studies are limited to the detection of digits and not storing or providing detected digits to a database or mobile applications. This paper proposes a system that can detect the digital number of meter readings using a lightweight deep neural network (DNN) for low power consumption and send those digits to an Android mobile application in real-time to store them and make life easy. The proposed lightweight DNN is computationally inexpensive and exhibits accuracy similar to those of conventional DNNs.

• Journal of Convergence for Information Technology
• /
• v.10 no.8
• /
• pp.23-34
• /
• 2020

Recently, the artificial intelligence deep learning field has been hard to commercialize due to the high computing power and the price problem of computing resources. In this paper, we apply a double pruning techniques to evaluate the performance of the in-depth neural network and various datasets. Double pruning combines basic Network-slimming and Parameter-prunning. Our proposed technique has the advantage of reducing the parameters that are not important to the existing learning and improving the speed without compromising the learning accuracy. After training various datasets, the pruning ratio was increased to reduce the size of the model.We confirmed that MobileNet-V3 showed the highest performance as a result of NetScore performance analysis. We confirmed that the performance after pruning was the highest in MobileNet-V3 consisting of depthwise seperable convolution neural networks in the Cifar 10 dataset, and VGGNet and ResNet in traditional convolutional neural networks also increased significantly.

• Journal of Broadcast Engineering
• /
• v.27 no.3
• /
• pp.341-350
• /
• 2022

In this paper, we propose a deep learning based on-device augmented reality (AR) system in which multiple input images are used to implement the correct occlusion in a real environment. The proposed system is composed of three technical steps; camera pose estimation, depth estimation, and object augmentation. Each step employs various mobile frameworks to optimize the processing on the on-device environment. Firstly, in the camera pose estimation stage, the massive computation involved in feature extraction is parallelized using OpenCL which is the GPU parallelization framework. Next, in depth estimation, monocular and multiple image-based depth image inference is accelerated using the mobile deep learning framework, i.e. TensorFlow Lite. Finally, object augmentation and occlusion handling are performed on the OpenGL ES mobile graphics framework. The proposed augmented reality system is implemented as an application in the Android environment. We evaluate the performance of the proposed system in terms of augmentation accuracy and the processing time in the mobile as well as PC environments.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.9 no.3
• /
• pp.195-201
• /
• 2015

In this study, IT education contents have been developed based on the u-Learning approach for people with hearing impairment, focusing on allowing the user to learn from anywhere and anytime. Specifically, this study applies HTML5 to implementing IT education contents(JSP, Oracle) for the deaf because HTML5 enables the learner to access the contents through both web and mobile device on various platforms including android, Mac OS, and PC etc. The results of this study are as follows: First, the online computer courses are mostly supposed to be compatible with diverse types of mobile devices. However, some of the contents could not be run on applications residing in web and mobile devices because the contents tend to be developed using FLASH. HTML5 is the effective way to overcome the compatibility problem. Second, FLASH and HTML5 contents authoring tools have been compared in terms of their strong and weak points by applying the developed contents to those tools. The study also suggests that the future work would be needed in order to implement wide variety of event functions with HTML5. Lastly, design strategies enabling access through web and mobile devices have been analyzed in accordance with u-Learning design guidelines for the deaf and mobile application accessibility guidelines. However, in the latter case, the future work regarding design guidelines needs to be conducted to improve the educational accessibility depending on the level of impairment.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.4
• /
• pp.25-34
• /
• 2024

This paper presents the development of a mobile application that detects and identifies canine skin diseases by training a deep learning-based U-Net model to infer the presence and location of skin lesions from images. U-Net, primarily used in medical imaging for image segmentation, is effective in distinguishing specific regions of an image in a polygonal form, making it suitable for identifying lesion areas in dogs. In this study, six major canine skin diseases were defined as classes, and the U-Net model was trained to differentiate among them. The model was then implemented in a mobile app, allowing users to perform lesion analysis and prediction through simple camera shots, with the results provided directly to the user. This enables pet owners to monitor the health of their pets and obtain information that aids in early diagnosis. By providing a quick and accurate diagnostic tool for pet health management through deep learning, this study emphasizes the significance of developing an easily accessible service for home use.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.6 no.11
• /
• pp.2849-2865
• /
• 2012

In order to block mobile junk messages automatically, many studies on spam filters have applied machine learning algorithms. Most previous research focused only on the accuracy rate of spam filters from the view point of the algorithm used, not on individual user's preferences. In terms of individual taste, the spam filters implemented on a mobile device have the advantage over spam filters on a network node, because it deals with only incoming messages on the users' phone and generates no additional traffic during the filtering process. However, a spam filter on a mobile phone has to consider the consumption of resources, because energy, memory and computing ability are limited. Moreover, as time passes an increasing number of feature words are likely to exhaust mobile resources. In this paper we propose a spam filter model distributed between a users' computer and smart phone. We expect the model to follow personal decision boundaries and use the uniform resources of smart phones. An authorized user's computer takes on the more complex and time consuming jobs, such as feature selection and training, while the smart phone performs only the minimum amount of work for filtering and utilizes the results of the information calculated on the desktop. Our experiments show that the accuracy of our method is more than 95% with Na$\ddot{i}$ve Bayes and Support Vector Machine, and our model that uses uniform memory does not affect other applications that run on the smart phone.

• Journal of the Korea Convergence Society
• /
• v.11 no.7
• /
• pp.1-7
• /
• 2020

As the performance of mobile devices has dramatically improved, users can perform many tasks in a mobile environment. This means that the use of behavior information stored in the mobile device can tell a lot of users. For example, a user's text message and frequently used application information (behavioral information) can be utilized to create useful information, such as whether the user is interested in parenting(profile prediction). In this study, I investigate the behavior information of the user that can be collected in the mobile device and propose the item that can profile the user. And I also suggest ideas about how to utilize profiling information.

• Research in Community and Public Health Nursing
• /
• v.32 no.2
• /
• pp.150-161
• /
• 2021

Purpose: The purpose of this study was to develop a mobile health lifestyle program for university students and to verify its effectiveness. Methods: The program was developed based on Jung's teaching-learning system design model. The research used a non-equivalent control group pretest-posttest non-synchronized design. Data were collected from October 20 to December 5, 2018. To verify the effects of the program, the knowledge, self-efficacy, and intention to plan health lifestyle and health lifestyle behavior were measured. A two hour health lecture and a mobile health lifestyle program were delivered for 3 weeks to 23 students in the experimental group. 19 students in the control group received only a two hour health lecture. Results: The experimental group showed significantly higher scores on knowledge (F=4.63, p=.038), intention to plan health lifestyle (F=14.44, p<.001), and health lifestyle behavior (F=46.80, p<.001). However, the score on self-efficacy was not significantly different (F=2.65, p=.112). Conclusion: It was confirmed that the mobile health lifestyle program can be useful in increasing the level of knowledge, intention and behavior of health lifestyle among university students. Therefore, the mobile health lifestyle application can be used as a supporting resource to enhance the health promotion for university students.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.10a
• /
• pp.173-176
• /
• 2013

This paper's questions, and an automatic scoring system written in HTML, and XML-based system that is at issue, the issue questions in a convenient offline automatically how to register, Easy to manage questions of issues, questions and problems of merging the PC and the mobile device in a place that can be obtained without taking the test system study. Server systems, and real-time registration questions merging problem, such as difficulty adjusting to the test required to build the system. Clients communicate with the server using the mobile device and the PC is required to take the exam in the View application, and responses are sent for treatment research.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.6
• /
• pp.639-645
• /
• 2018

This paper is intended to develop a decision model that can be applied to autonomous vehicles and autonomous mobile vehicles. The developed module has an independent configuration for application in various driving environments and is based on a platform for organically operating them. Each module is studied for decision making on lane changes and for securing safety through reinforcement learning using a deep learning technique. The autonomous mobile moving body operating to change the driving state has a characteristic where the next operation of the mobile body can be determined only if the definition of the speed determination model (according to its functions) and the lane change decision are correctly preceded. Also, if all the moving bodies traveling on a general road are equipped with an autonomous driving function, it is difficult to consider the factors that may occur between each mobile unit from unexpected environmental changes. Considering these factors, we applied the decision model to the platform and studied the lane change decision system for implementation of the platform. We studied the decision model using a modular learning method to reduce system complexity, to reduce the learning time, and to consider model replacement.