• Smart Media Journal
• /
• v.9 no.4
• /
• pp.9-16
• /
• 2020

Recently, various studies are being conducted to integrate Image Segmentation into smart factory industries and autonomous driving fields. In particular, Image Segmentation systems using deep learning algorithms have been researched and developed enough to learn from large volumes of data with higher accuracy. In order to use image segmentation in the autonomous driving sector, sufficient amount of learning is needed with large amounts of data and the streaming environment that processes drivers' data in real time is important for the accuracy of safe operation through highways and child protection zones. Therefore, we proposed a novel DFCN algorithm that enhanced existing FCN algorithms that could be applied to various road environments, demonstrated that the performance of the DFCN algorithm improved 1.3% in terms of "loss" value compared to the previous FCN algorithms. Moreover, the proposed DFCN algorithm was applied to the existing U-Net algorithm to maintain the information of frequencies in the image to produce better results, resulting in a better performance than the classical FCN algorithm in the autonomous environment.

• Journal of the Korea Convergence Society
• /
• v.13 no.1
• /
• pp.141-147
• /
• 2022

In a broad sense, the definition of digital health care is an industrial area that manages personal health and diseases through the convergence of the health care industry and ICT. In a narrow sense, various medical technologies are used to manage medical services to improve patient health. This paper aims to provide design guidelines so that artificial intelligence technology can be applied stably and efficiently to more diverse digital health care fields in the future by introducing use cases of artificial intelligence and machine learning techniques applied in the digital health care field. For this purpose, in this thesis, the medical field and the daily life field are divided and examined. The two regions have different data characteristics. By further subdividing the two areas, we looked at the use cases of artificial intelligence algorithms according to data characteristics and problem definitions and characteristics. Through this, we will increase our understanding of artificial intelligence technologies used in the digital health care field and examine the possibility of using various artificial intelligence technologies.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.3
• /
• pp.571-580
• /
• 2022

Marine deposited waste is a major cause of problems such as a lot of damage and an increase in the estimated amount of garbage due to abandoned fishing grounds caused by ghost fishing. In this paper, we implement a real-time marine deposited waste detection artificial intelligence system to understand the actual conditions of waste fishing gear usage, distribution, loss, and recovery, and study methods for performance improvement. The system was implemented using the yolov5 model, which is an excellent performance model for real-time object detection, and the 'data screening process' and 'class segmentation' method of learning data were applied as performance improvement methods. In conclusion, the object detection results of datasets that do screen unnecessary data or do not subdivide similar items according to characteristics and uses are better than the object recognition results of unscreened datasets and datasets in which classes are subdivided.

• Membrane and Water Treatment
• /
• v.14 no.4
• /
• pp.155-162
• /
• 2023

Global climate change and urbanization have various demerits, such as water pollution, flood damage, and deterioration of water circulation. Thus, attention is drawn to Nature-based Solution (NbS) that solve environmental problems in ways that imitate nature. Among the NbS, urban wetlands are facilities that perform functions, such as removing pollutants from a city, improving water circulation, and providing ecological habitats, by strengthening original natural wetland pillars. Frequent monitoring and maintenance are essential for urban wetlands to maintain their performance; therefore, there is a need to apply the Internet of Things (IoT) technology to wetland monitoring. Therefore, in this study, we attempted to develop a real-time wetland monitoring device and interface. Temperature, water temperature, humidity, soil humidity, PM1, PM2.5, and PM10 were measured, and the measurements were taken at 10-minute intervals for three days in both indoor and wetland. Sensors suitable for conditions that needed to be measured and an Arduino MEGA 2560 were connected to enable sensing, and communication modules were connected to transmit data to real-time databases. The transmitted data were displayed on a developed web page. The data measured to verify the monitoring device were compared with data from the Korea meteorological administration and the Korea environment corporation, and the output and upward or downward trend were similar. Moreover, findings from a related patent search indicated that there are a minimal number of instances where information and communication technology (ICT) has been applied in wetland contexts. Hence, it is essential to consider further research, development, and implementation of ICT to address this gap. The results of this study could be the basis for time-series data analysis research using automation, machine learning, or deep learning in urban wetland maintenance.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.17 no.1
• /
• pp.261-268
• /
• 2017

The development of ICT technology centered on mobile smart platforms have been emerging as a shared economic platform based on collaborative consumption. In this study, we analyze what factors affect success and failure in commercialized shared economic platforms from 2008 to 2016, and present what policy factors are needed to activate shared economic platform. To do this, we analyze successful cases of shared economic platforms and failed cases, derive key variables that affect success and failure, and conduct inductive analysis based on ID 3 algorithm based on them. Through this, we present the policy factors for the commercial success of the shared economic platform by deriving the rules for the success and failure of the shared economic platform.

• Journal of Digital Contents Society
• /
• v.6 no.4
• /
• pp.217-222
• /
• 2005

The purpose of this study is to design and implement a sign language dictionary for the deaf to understand information communication terminologies. When the deafs who have difficulties in communication use the internet, they an get help from this dictionary in accessing various types of information and expressing their intension. In order for the deaf to utilize the internet as efficiently as ordinary people, they must understand information communication terminologies first In order to implement the dictionary, we defined the concepts of the deaf and examined their characteristics. In addition, we established principles in designing this dictionary and selected some terminologies. When explaining the terminologies. we tried to use expressions common to the deaf, but sometimes modified them to keep the original meanings of the terms in producing sign language videos. This studies are applied as learning aid to information education for the deaf, and the deaf's understanding of ICT was measured through two tests.

• Journal of Platform Technology
• /
• v.10 no.3
• /
• pp.69-78
• /
• 2022

Job offer and job search data related to employment are in the form of highly-unstructured texts that occur in real-time, NCS duty, learning modules, and job dictionaries. Job announcements and training information have a high data value amid changes in industrial technology, such as the Fourth Industrial Evolution. This study developed a job data dictionary by defining relevant data to intuitively understand and harness information on job offers and job searches. This study also designed, constructed, and evaluated a data map based on ontology to enable linking and inferring data about public announcement-job-training. Through this, it was found that the inference function centered on work ability enables QoS support that can satisfy users by minimizing mismatch between consumers and optimizing the data dictionary.

• Journal of Platform Technology
• /
• v.11 no.1
• /
• pp.72-84
• /
• 2023

With the rise of the Internet of Things, the security of such lightweight computing environments has become a hot topic. Lightweight block ciphers that can provide efficient performance and security by having a relatively simpler structure and smaller key and block sizes are drawing attention. Due to these characteristics, they can become a target for new attack techniques. One of the new cryptanalytic attacks that have been attracting interest is Neural cryptanalysis, which is a cryptanalytic technique based on neural networks. It showed interesting results with better results than the conventional cryptanalysis method without a great amount of time and cryptographic knowledge. The first work that showed good results was carried out by Aron Gohr in CRYPTO'19, the attack was conducted on the lightweight block cipher SPECK-/32/64 and showed better results than conventional differential cryptanalysis. In this paper, we first apply the Differential Neural Distinguisher proposed by Aron Gohr to the block ciphers HIGHT and GOST to test the applicability of the attack to ciphers with different structures. The performance of the Differential Neural Distinguisher is then analyzed by replacing the neural network attack model with five different models (Multi-Layer Perceptron, AlexNet, ResNext, SE-ResNet, SE-ResNext). We then propose a Related-key Neural Distinguisher and apply it to the SPECK-/32/64, HIGHT, and GOST block ciphers. The proposed Related-key Neural Distinguisher was constructed using the relationship between keys, and this made it possible to distinguish more rounds than the differential distinguisher.

• Journal of Broadcast Engineering
• /
• v.25 no.1
• /
• pp.1-12
• /
• 2020

In the medical field, numerical imbalance of data due to differences in disease prevalence is a common problem. It reduces the performance of a artificial intelligence network, leading to difficulties in learning a network with good performance. Recently, generative adversarial network (GAN) technology has been introduced as a way to address this problem, and its ability has been demonstrated by successful applications in various fields. However, it is still difficult to achieve good results in solving problems with performance degraded by numerical imbalances because the image resolution of the previous studies is not yet good enough and the structure in the image is modeled locally. In this paper, we propose a multi-scale conditional generative adversarial network based on attention mechanism, which can produce high resolution images to solve the numerical imbalance problem of chest X-ray image data. The network was able to produce images for various diseases by controlling condition variables with only one network. It's efficient and effective in that the network don't need to be learned independently for all disease classes and solves the problem of long distance dependency in image generation with self-attention mechanism.

• Journal of Animal Science and Technology
• /
• v.65 no.3
• /
• pp.627-637
• /
• 2023

As the population and income levels rise, meat consumption steadily increases annually. However, the number of farms and farmers producing meat decrease during the same period, reducing meat sufficiency. Information and Communications Technology (ICT) has begun to be applied to reduce labor and production costs of livestock farms and improve productivity. This technology can be used for rapid pregnancy diagnosis of sows; the location and size of the gestation sacs of sows are directly related to the productivity of the farm. In this study, a system proposes to determine the number of gestation sacs of sows from ultrasound images. The system used the YOLOv7-E6E model, changing the activation function from sigmoid-weighted linear unit (SiLU) to a multi-activation function (SiLU + Mish). Also, the upsampling method was modified from nearest to bicubic to improve performance. The model trained with the original model using the original data achieved mean average precision of 86.3%. When the proposed multi-activation function, upsampling, and AutoAugment were applied, the performance improved by 0.3%, 0.9%, and 0.9%, respectively. When all three proposed methods were simultaneously applied, a significant performance improvement of 3.5% to 89.8% was achieved.