• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.38 no.6
• /
• pp.635-644
• /
• 2020

Object recognition, detection and instance segmentation based on DL (Deep Learning) have being used in various practices, and mainly optical images are used as training data for DL models. The major objective of this paper is object segmentation and building detection by utilizing multimodal datasets as well as optical images for training Detectron2 model that is one of the improved R-CNN (Region-based Convolutional Neural Network). For the implementation, infrared aerial images, LiDAR data, and edges from the images, and Haralick features, that are representing statistical texture information, from LiDAR (Light Detection And Ranging) data were generated. The performance of the DL models depends on not only on the amount and characteristics of the training data, but also on the fusion method especially for the multimodal data. The results of segmenting objects and detecting buildings by applying hybrid fusion - which is a mixed method of early fusion and late fusion - results in a 32.65% improvement in building detection rate compared to training by optical image only. The experiments demonstrated complementary effect of the training multimodal data having unique characteristics and fusion strategy.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.6
• /
• pp.2086-2097
• /
• 2021

The current cultural assets are being restored depending on the opinions of experts (craftsmen). We intend to introduce digitalized artificial intelligence techniques, excluding the personal opinions of experts on reconstruction of such cultural properties. The first step toward restoring digitized cultural properties is separation. The restoration of cultural properties should be reorganized based on recorded documents, period historical backgrounds and regional characteristics. The cultural properties in the form of photographs or images should be collected by separating the background. In addition, when restoring cultural properties most of them depend a lot on the tendency of the restoring person workers. As a result, it often occurs when there is a problem in the accuracy and reliability of restoration of cultural properties. In this study, we propose a search method for learning stored digital cultural assets using AI technology. Pagoda was selected for restoration of Cultural Properties. Pagoda data collection was collected through the Internet and various historical records. The pagoda data was classified by period and region, and grouped into similar buildings. The collected data was learned by applying the well-known CNN algorithm for artificial intelligence learning. The pagoda search used Yolo Marker to mark the tower shape. The tower was used a total of about 100-10,000 pagoda data. In conclusion, it was confirmed that the probability of searching for a tower differs according to the number of pagoda pictures and the number of learning iterations. Finally, it was confirmed that the number of 500 towers and the epochs in training of 8000 times were good. If the test result exceeds 8,000 times, it becomes overfitting. All so, I found a phenomenon that the recognition rate drops when the enemy repeatedly learns more than 8,000 times. As a result of this study, it is believed that it will be helpful in data gathering to increase the accuracy of tower restoration.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.3
• /
• pp.159-168
• /
• 2021

As the information age enters, the use of biometrics using the body is gradually increasing because it is very important to accurately recognize and authenticate each individual's identity for information protection. Among them, finger vein authentication technology is receiving a lot of attention because it is difficult to forge and demodulate, so it has high security, high precision, and easy user acceptance. However, the accuracy may be degraded depending on the algorithm for identification or the surrounding light environment. In this paper, we designed and manufactured a side-type finger vein recognizer that is highly versatile among finger vein measuring devices, and authenticated using the deep learning model of DenseNet-201 for high accuracy and recognition rate. The performance of finger vein authentication technology according to the influence of the infrared light source used and the surrounding visible light was analyzed through simulation. The simulations used data from MMCBNU_6000 of Jeonbuk National University and finger vein images taken directly were used, and the performance were compared and analyzed using the EER.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.16 no.6
• /
• pp.111-121
• /
• 2006

As Internet lastly grows, network attack techniques are transformed and new attack types are appearing. The existing network-based intrusion detection systems detect well known attack, but the false-positive or false-negative against unknown attack is appearing high. In addition, The existing network-based intrusion detection systems is difficult to real time detection against a large network pack data in the network and to response and recognition against new attack type. Therefore, it requires method to heighten the detection rate about a various large dataset and to reduce the false-positive. In this paper, we propose method to reduce the false-positive using multi-level detection algorithm, that is combine the multidimensional Apriori algorithm and the modified Negative Selection algorithm. And we apply this algorithm in intrusion detection and, to be sure, it has a good performance.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.6
• /
• pp.41-50
• /
• 2020

This paper deals with the Braille Education System which complements the shortcomings of the existing Braille Learning Products. An application dedicated to the blind is configured to perform full functions through touch gestures and voice guidance for user convenience. Braille kit is produced for educational purposes through Arduino and 3D printing. The system supports the following functions. First, the learning of the most basic braille, such as initial consonants, final consonant, vowels, abbreviations, etc. Second, the ability to check learned braille by solving step quizzes. Third, translation of braille. Through the experiment, the recognition rate of touch gestures and the accuracy of braille expression were confirmed, and in case of translation, the translation was done as intended. The system allows blind people to learn braille efficiently.

• Journal of Veterinary Clinics
• /
• v.38 no.4
• /
• pp.204-209
• /
• 2021

A seven-month-old castrated male Chihuahua weighing 1.6 kg presented with generalized tonic-clonic seizure following ingestion of isoniazid. Emergency treatment with three doses of diazepam (total 1.5 mg/kg, intravenous [IV]) and phenobarbital (15 mg/kg IV) was administered. The seizure stopped after administration of propofol (constant rate infusion [CRI]; 0.2 mg/kg/min). Blood analyses showed mildly increased serum blood glucose concentration, hyperkalemia, and hyperphosphatemia. On suspicion of isoniazid toxicity, activated charcoal (1 g/kg, orally), lipid emulsion (CRI; 9 mL/hr), and pyridoxine hydrochloride (70 mg/kg IV) were added to the treatment regimen. Twelve hours after presentation, the dog showed increased serum liver enzyme activities, serum blood urea nitrogen, and creatinine concentrations indicating hepatic and renal failure. Twenty-two hours after presentation, blood analysis still revealed increased liver enzyme activities, blood urea nitrogen, and creatinine concentrations with low blood glucose concentration. Twenty-six hours after presentation, the dog's vital signs deteriorated and the owner elected for the dog to be euthanized. This is the first report of the clinical course of isoniazid toxicosis in a dog in South Korea. Furthermore, to our best knowledge, this is the first report where secondary multiple organ failure was observed due to isoniazid toxicosis. Clinicians should be aware of the possibility of isoniazid toxicosis in dogs. Rapid initiation of treatment after clinical recognition is warranted in such cases.

• Journal of Korea Water Resources Association
• /
• v.54 no.8
• /
• pp.599-606
• /
• 2021

Water Distribution Networks, one of the social infrastructures buried underground, has the function of transporting and supplying purified water to customers. In recent years, as measurement capability is improved, a number of studies related to leak recognition and detection by applying a deep learning technique based on flow rate data have been conducted. In this study, a cognitive model for leak occurrence was developed using an LSTM-based deep learning algorithm that has not been applied to the waterworks field until now. The model was verified based on the assumed data, and it was found that all cases of leaks of 2% or more can be recognized. In the future, based on the proposed model, it is believed that more precise results can be derived in the prediction of flow data.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.4
• /
• pp.89-96
• /
• 2021

The number of confirmed cases of coronavirus-19 is soaring around the world and has caused numerous deaths. Wearing a mask is very important to prevent infection. Incidents and accidents have occurred due to the recommendation to wear a mask in public places such as buses and subways, and it has emerged as a serious social problem. To solve this problem, this paper proposes an open source-based face mask wearing monitoring system. We used open source software, web-based artificial intelligence tool teachable machine and open source hardware Arduino. It judges whether the mask is worn, and performs commands such as guidance messages and alarms. The learning parameters of the teachable machine were learned with the optimal values of 50 learning times, 32 batch sizes, and 0.001 learning rate, resulting in an accuracy of 1 and a learning error of 0.003. We designed and implemented a mask wearing monitoring system that can perform commands such as guidance messages and alarms by determining whether to wear a mask using a web-based artificial intelligence tool teachable machine and Arduino to prove its validity.

• Human Ecology Research
• /
• v.59 no.4
• /
• pp.449-463
• /
• 2021

The aim of this study was to analyze the perceptions and entering status of small business online fashion retailers on portal shopping and fashion shopping malls. Case studies were conducted on a total of 10 research samples. The results were as follows: first, regarding the strategic factors of online fashion stores, 'price competitiveness' is important, especially in portal shopping and low-cost brands; 'product assortment' is important but not essential in all platforms; and 'differentiation' is important to continuously secure loyal customers in fashion shopping malls. Customer satisfaction leads to customer loyalty, and customer loyalty affects the sales conversion rate and brand growth of online sales channels. Factors that promoted sales activities in online sales channels were exposure, advertisements, SNS, events, special exhibitions, and events. Hindrance factors were low price competition, overheated competition, and the MD of sales channels. Second, the research samples used multiple online sales channels, including portal shopping malls and fashion shopping malls, in addition to their own malls. The selection factors were platform reputation and commission, branding, and customer inflow through exposure. Portal shopping malls were perceived as providing easy access, advertising/customer communication, exposure/search, price competitiveness, scalability, and intense competition, whereas fashion shopping malls were perceived as providing a brand image and concept, brand promotion, high commissions, difficult entry, and low profits. The factors for success in portal shopping malls were exposure/search, price competitiveness, and brand recognition, whereas the factors for success in fashion shopping malls were differentiation, brand, exposure/advertisement, product assortment, and MD.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.6
• /
• pp.264-279
• /
• 2021

Most automated vehicle evaluation scenarios are developed based on the typical driving situations that automated vehicles will face. However, various situations occur during actual driving, and sometimes complex judgments are required. This study is to define a situation that requires complex judgment for safer driving of an automated vehicle as a dilemma situation, and to suggest a driving strategy necessary to secure driving safety in each situation. To this end, we defined dilemma situations based on the automated vehicle ethics guidelines, the criteria for recognition of error rate in automobile accidents, and suggestions from the automated vehicle developers. In addition, in the defined dilemma situations, the factors affecting movement for establishing driving strategies were explored, and the priorities of factors affecting driving according to the Road Traffic Act and driving strategies were derived accordingly.