• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.11
• /
• pp.411-418
• /
• 2018

With the proliferation of artificial intelligence technology, it is becoming important to obtain, store, and utilize scientific data in research and science sectors. A number of methods for extracting meaningful objects such as graphs and tables from research articles have been proposed to eventually obtain scientific data. Existing extraction methods using heuristic approaches are hardly applicable to electronic documents having heterogeneous manuscript formats because they are designed to work properly for some targeted manuscripts. This paper proposes a prototype of an object extraction system which exploits a recent deep-learning technology so as to overcome the inflexibility of the heuristic approaches. We implemented our trained model, based on the Faster R-CNN algorithm, using the Google TensorFlow Object Detection API and also composed an annotated data set from 100 research articles for training and evaluation. Finally, a performance evaluation shows that the proposed system outperforms a comparator adopting heuristic approaches by 5.2%.

• Journal of Information Processing Systems
• /
• v.14 no.6
• /
• pp.1445-1456
• /
• 2018

Environment perception and three-dimensional (3D) reconstruction tasks are used to provide unmanned ground vehicle (UGV) with driving awareness interfaces. The speed of obstacle segmentation and surrounding terrain reconstruction crucially influences decision making in UGVs. To increase the processing speed of environment information analysis, we develop a CPU-GPU hybrid system of automatic environment perception and 3D terrain reconstruction based on the integration of multiple sensors. The system consists of three functional modules, namely, multi-sensor data collection and pre-processing, environment perception, and 3D reconstruction. To integrate individual datasets collected from different sensors, the pre-processing function registers the sensed LiDAR (light detection and ranging) point clouds, video sequences, and motion information into a global terrain model after filtering redundant and noise data according to the redundancy removal principle. In the environment perception module, the registered discrete points are clustered into ground surface and individual objects by using a ground segmentation method and a connected component labeling algorithm. The estimated ground surface and non-ground objects indicate the terrain to be traversed and obstacles in the environment, thus creating driving awareness. The 3D reconstruction module calibrates the projection matrix between the mounted LiDAR and cameras to map the local point clouds onto the captured video images. Texture meshes and color particle models are used to reconstruct the ground surface and objects of the 3D terrain model, respectively. To accelerate the proposed system, we apply the GPU parallel computation method to implement the applied computer graphics and image processing algorithms in parallel.

• International Journal of Advanced Culture Technology
• /
• v.9 no.2
• /
• pp.100-105
• /
• 2021

Contactless service is rapidly emerging as a new growth strategy due to consumers who are reluctant to the face-to-face situation in the global pandemic of coronavirus disease 2019 (COVID-19), and various technologies are being developed to support the fast-growing contactless service market. In particular, the restaurant industry is one of the most desperate industrial fields requiring technologies for contactless service, and the representative technical case should be a kiosk, which has the advantage of reducing labor costs for the restaurant owners and provides psychological relaxation and satisfaction to the customer. In this paper, we propose a solution to the restaurant's store operation through the unmanned kiosk using a state-of-the-art artificial intelligence (AI) technology of image recognition. Especially, for the products that do not have barcodes in bakeries, fresh foods (fruits, vegetables, etc.), and autonomous restaurants on highways, which cause increased labor costs and many hassles, our proposed system should be very useful. The proposed system recognizes products without barcodes on the ground of image-based AI algorithm technology and makes automatic payments. To test the proposed system feasibility, we established an AI vision system using a commercial camera and conducted an image recognition test by training object detection AI models using donut images. The proposed system has a self-learning system with mismatched information in operation. The self-learning AI technology allows us to upgrade the recognition performance continuously. We proposed a fully automated payment system with AI vision technology and showed system feasibility by the performance test. The system realizes contactless service for self-checkout in the restaurant business area and improves the cost-saving in managing human resources.

• Journal of Korean Academy of Oral Health
• /
• v.41 no.2
• /
• pp.154-161
• /
• 2017

Objectives: The purpose of this study was to investigate changes in public recognition of parabens on Twitter and the research status of parabens related to toothpaste. Methods: Tweet information between 2010 and October 2016 was collected by an automatic web crawler and examined according to tweet frequency, key words (2012-October 2016), and issue tweet detection analyses to reveal changes in public recognition of parabens on Twitter. To investigate the research status of parabens related to toothpaste, queries such as "paraben," "paraben and toxicity," "paraben and (toothpastes or dentifrices)," and "paraben and (toothpastes or dentifrices) and toxicity" were used. Results: The number of tweets concerning parabens sharply increased when parabens in toothpaste emerged as a social issue (October 2014), and decreased from 2015 onward. However, toothpaste and its related terms were continuously included in the core key words extracted from tweets from 2015. They were not included in key words before 2014, indicating that the emergence of parabens in toothpaste as a social issue plays an important role in public recognition of parabens in toothpaste. The issue tweet analysis also confirmed the change in public recognition of parabens in toothpaste. Despite the expansion of public recognition of parabens in toothpaste, there are only seven research articles on the topic in PubMed. Conclusions: The general public clearly recognized parabens in toothpaste after emergence of parabens in toothpaste as a social issue. Nevertheless, the scientific information on parabens in toothpaste is very limited, suggesting that the efforts of dental scientists are required to expand scientific knowledge related to parabens in oral hygiene measures.

• Journal of Biomedical Engineering Research
• /
• v.42 no.3
• /
• pp.100-106
• /
• 2021

Measuring Intima-media thickness (IMT) with ultrasound images can help early detection of coronary artery disease. As a result, numerous machine learning studies have been conducted to measure IMT. However, most of these studies require several steps of pre-treatment to extract the boundary, and some require manual intervention, so they are not suitable for on-site treatment in urgent situations. in this paper, we propose to use deep learning networks U-Net, Attention U-Net, and Pretrained U-Net to automatically segment the intima-media complex. This study also applied the HE, HS, and CLAHE preprocessing technique to wireless portable ultrasound diagnostic device images. As a result, The average dice coefficient of HE applied Models is 71% and CLAHE applied Models is 70%, while the HS applied Models have improved as 72% dice coefficient. Among them, Pretrained U-Net showed the highest performance with an average of 74%. When comparing this with the mean value of IMT measured by Conventional wired ultrasound equipment, the highest correlation coefficient value was shown in the HS applied pretrained U-Net.

• Journal of Internet of Things and Convergence
• /
• v.6 no.4
• /
• pp.65-72
• /
• 2020

In this paper, we developed a web-based DApp system based on a private blockchain by applying machine learning techniques to automatically identify Android malicious apps that are continuously increasing rapidly. The optimal machine learning model that provides 96.2587% accuracy for Android malicious app identification was selected to the authorized experimental data, and automatic identification results for Android malicious apps were recorded/managed in the Hyperledger Fabric blockchain system. In addition, a web-based DApp system was developed so that users who have been granted the proper authority can use the blockchain system. Therefore, it is possible to further improve the security in the Android mobile app usage environment through the development of the machine learning-based Android malicious app identification block chain DApp system presented. In the future, it is expected to be able to develop enhanced security services that combine machine learning and blockchain for general-purpose data.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.5
• /
• pp.677-684
• /
• 2021

This paper presents a time-to-digital converter for measuring absolute time differences. The time-to-digital converter was designed and fabricated in 0.18-um CMOS technology and it can be applied to Light Detection and Ranging system which requires long time-cover range and 50ps time resolution. Since designed time-to-digital converter adopted the reference clock of 625MHz generated by phase locked loop, it could have absolute time resolution of 50ps after automatic calibration and its cover range was over than 800ns. The time-to-digital converter adopted a counter and chain delay lines for time measurement. The counter is used for coarse time measurement and chain delay lines are used for fine time measurement. From many times experiments, fabricated time-to-digital converter has 50 ps time resolution with maximum INL of 0.8 LSB and its power consumption is about 70 mW.

• Atmosphere
• /
• v.32 no.2
• /
• pp.135-148
• /
• 2022

This study investigates the influence of atmospheric river (AR) on precipitation over South Korea with a focus on regional characteristics. The 42-year-long catalog of ARs, which is obtained by applying the automatic AR detection algorithm to ERA5 reanalysis data and the insitu precipitation data recorded at 56 weather stations across the country are used to quantify their relationship. Approximately 51% of the climatological annual precipitation is associated with AR. The AR-related precipitation is most pronounced in summer by approximately 58%, while only limited fraction of precipitation (26%) is AR-related in winter. The heavy precipitation (> 30 mm day^-1) is more prone to AR activity (59%) than weak precipitation (5~30 mm day^-1; 33%) in all seasons. By grouping weather stations into the four sub-regions based on orography, it is found that the contribution of AR precipitation to the total is largest in the southern coast (57%) and smallest in the eastern coast (36%). Similar regional variations in AR precipitation fractions also occur in weak precipitation events. The regional contrast between the northern and southern stations is related to the seasonal variation of AR-frequency. In addition, the regional contrast between the western and eastern stations is partly modulated by the orographic forcing. The fractional contribution of AR to heavy precipitation exceeds 50% in all seasons, but this is true only in summer along the eastern coast. This result indicates that ARs play a critical role in heavy precipitation in South Korea, thus routine monitoring of ARs is needed for improving operational hydrometeorological forecasting.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.3
• /
• pp.422-429
• /
• 2022

In this paper, the intermediate frequency transmission signal level between the network system-based baseband and RF unit consisting of multi-channel airborne relay devices and a lot of mission devices, which are currently undergoing technology development tasks, is kept constant at the reference signal level. Considering the other party's receiving input range, despite changes in the short-range long-range wireless communication environment, it presents a multi-link protection and MLP-A algorithm that allows signals to be transmitted stably and reliably through signal detection automatic gain control, and experiments and analysis considering short-distance and long-distance wireless environments were performed by designing, manufacturing, and implementing RF units to which MLP-A algorithms were applied, and applying distance calculation equations to the configuration of multiple air movements and verification networks. Through this, it was confirmed that a stable and reliable RF communication system can be operated.

• The Journal of the Korea Contents Association
• /
• v.21 no.12
• /
• pp.322-330
• /
• 2021

In modern aquaculture, mass mortality is a very important issue that determines the success of aquaculture business. If a fish disease is not detected at an early stage in the farm, the disease spreads quickly because the farm is a closed environment. Therefore, early detection of diseases is crucial to prevent mass mortality of fish raised in farms. Recently deep learning-based automatic identification of fish diseases has been widely used, but there are many difficulties in identifying objects due to insufficient images of fish diseases. Therefore, this paper suggests a method to generate a large number of fish disease images by synthesizing normal images and disease images using SinGAN deep learning model in order to to solve the lack of fish disease images. We generate images from the three most frequently occurring Paralichthys Olivaceus diseases such as Scuticociliatida, Vibriosis, and Lymphocytosis and compare them with the original image. In this study, a total of 330 sheets of scutica disease, 110 sheets of vibrioemia, and 110 sheets of limphosis were made by synthesizing 10 disease patterns with 11 normal halibut images, and 1,320 images were produced by quadrupling the images.