• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.101-125
• /
• 2022

Recently, with the development of computing technology and the improvement of the cloud environment, deep learning technology has developed, and attempts to apply deep learning to various fields are increasing. A typical example is anomaly detection, which is a technique for identifying values or patterns that deviate from normal data. Among the representative types of anomaly detection, it is very difficult to detect a contextual anomaly that requires understanding of the overall situation. In general, detection of anomalies in image data is performed using a pre-trained model trained on large data. However, since this pre-trained model was created by focusing on object classification of images, there is a limit to be applied to anomaly detection that needs to understand complex situations created by various objects. Therefore, in this study, we newly propose a two-step pre-trained model for detecting abnormal situation. Our methodology performs additional learning from image captioning to understand not only mere objects but also the complicated situation created by them. Specifically, the proposed methodology transfers knowledge of the pre-trained model that has learned object classification with ImageNet data to the image captioning model, and uses the caption that describes the situation represented by the image. Afterwards, the weight obtained by learning the situational characteristics through images and captions is extracted and fine-tuning is performed to generate an anomaly detection model. To evaluate the performance of the proposed methodology, an anomaly detection experiment was performed on 400 situational images and the experimental results showed that the proposed methodology was superior in terms of anomaly detection accuracy and F1-score compared to the existing traditional pre-trained model.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.2
• /
• pp.343-350
• /
• 2022

It is worth verifying the effectiveness of data integration between data with different features. This study investigated whether the data integration affects the accuracy of deep neural network (DNN), and which integration method shows the best improvement. This study used two different public datasets. One public dataset was taken in an actual farm in India. And another was taken in a laboratory environment in Korea. Leaf images were selected from two different public datasets to have five classes which includes normal and four different types of plant diseases. DNN used pre-trained VGG16 as a feature extractor and multi-layer perceptron as a classifier. Data were integrated into three different ways to be used for the training process. DNN was trained in a supervised manner via the integrated data. The trained DNN was evaluated by using a test dataset taken in an actual farm. DNN shows the best accuracy for the test dataset when DNN was first trained by images taken in the laboratory environment and then trained by images taken in the actual farm. The results show that data integration between plant images taken in a different environment helps improve the performance of deep neural networks. And the results also confirmed that independent use of plant images taken in different environments during the training process is more effective in improving the performance of DNN.

• Journal of the Korean Society of Food Culture
• /
• v.25 no.2
• /
• pp.179-188
• /
• 2010

In order to aide in the design of an improved menu book, which could play an important role as a marketing tool, the current version of the menu books and managers (subjects) of 295 restaurants in the Incheon area were examined. These were managers of Korean (36.3%), Western (25.8%), Japanese (14.6%), cafeteria (12.5%) and Chinese (10.8%) style restaurants. The level of service (self-evaluation, 3-point scale) was average $2.25{\pm}0.45$. The general colorings of the menu books were green (19.0%), brown (18.6%), black (17.6%), yellow (15.9%), red (13.6%) and blue (13.2%). The material of the menu book cover was mainly leather (35.9%), and the internal material was mainly coated paper (59.7%). Physically, the design was two-panel fold (38.3%), two-panel multi-page (35.6%), die style (10.2%), single panel (8.1%) and tent style (7.8%). The type sizes were unchanged in 49.9% of the menu books and in 61.7% photos were not used. 53.9% of menu books did not explain the menus, and 13.2% did not classify the items into groups. Emphasis of profit-making menus was not done in 66.8%. 51.5% of menu books were irreplaceable in parts. The emphasis of profit-making menus was less among the Korean style restaurants (p<0.001). The possibility of partial replacement of menu books was lower in both Korean and Chinese restaurants (p<0.001). The explanation of the items was lower in the Japanese restaurants (p<0.001). The classification of items into groups was lower in cafeteria (p<0.001). In cases in which there were both seasonal and event menus, the possibility of partial replacements of menu books was higher (p<0.001). Restaurants of which service level was less than ordinary were lower in the differentiation of type sizes (p<0.001), the use of photos (p<0.001), the explanation of menus (p<0.001), the classification of menus by groups (p<0.05), the emphasis of profit-making menus (p<0.001) and the possibility of partial replacement of menu books (p<0.001). If these study findings are applied to the designing of menu books, the role of the menu book as an important tool for marketing could be greatly improved.

• Journal of The Geomorphological Association of Korea
• /
• v.18 no.1
• /
• pp.41-56
• /
• 2011

This study analyzes the change in land-cover based on satellite imagery to draw up land-cover map in the future, and estimates the change in land category using statistical data of the land category. To estimate land category, this study applied the double exponentially smoothing method. The result of the land cover classification according to year using satellite imagery showed that the type with the largest increase in area of land cover change in the cities of Cheonan and Asan was artificial structure, followed by water, grass field and bare land. However forest, paddy, marsh and dry field were reduced. Further, the result of the time-series analysis of the land category was found to be similar to the result of the land cover classification using satellite imagery. Especially, the result of the estimation of the land category change using the double exponentially smoothing method showed that paddy, dry field, forest and marsh are anticipated to consistently decrease in area from 2010 to 2100, whereas artificial structure, water, bare land and grass field are anticipated to consistently increase. Such results can be utilized as basic data to estimate the change in land cover according to climate change in order to prepare climate change response strategies.

• Journal of Wetlands Research
• /
• v.26 no.3
• /
• pp.197-203
• /
• 2024

The algal blooms in rivers can negatively affect water source management and water treatment processes, necessitating continuous management. In this study, a multi-classification model was developed to predict the concentration of chlorophyll-a (chl-a), one of the key indicators of algal blooms, using Tabular Prior Fitted Networks (TabPFN), a novel deep learning algorithm known for its relatively superior performance on small tabular datasets. The model was developed using daily observation data collected at Buyeo water quality monitoring station from January 1, 2014, to December 31, 2022. The collected data were averaged to construct input data sets with measurement frequencies of 1 day, 3 days, 6 days, 12 days. The performance comparison of the four models, constructed with input data on observation frequencies of 1 day, 3 days, 6 days, and 12 days, showed that the model exhibits stable performance even when the measurement frequency is longer and the number of observations is smaller. The macro average for each model were analyzed as follows: Precision was 0.77, 0.76, 0.83, 0.84; Recall was 0.63, 0.65, 0.66, 0.74; F1-score was 0.67, 0.69, 0.71, 0.78. For the weighted average, Precision was 0.76, 0.77, 0.81, 0.84; Recall was 0.76, 0.78, 0.81, 0.85; F1-score was 0.74, 0.77, 0.80, 0.84. This study demonstrates that the chl-a prediction model constructed using TabPFN exhibits stable performance even with small-scale input data, verifying the feasibility of its application in fields where the input data required for model construction is limited.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.45 no.2
• /
• pp.1-9
• /
• 2008

As the security requirements of mobile phones have been increasing, there have been extensive researches using one biometric feature (e.g., an iris, a fingerprint, or a face image) for authentication. Due to the limitation of uni-modal biometrics, we propose a method that combines face and iris images in order to improve accuracy in mobile environments. This paper presents four advantages and contributions over previous research. First, in order to capture both face and iris image at fast speed and simultaneously, we use a built-in conventional mega pixel camera in mobile phone, which is revised to capture the NIR (Near-InfraRed) face and iris image. Second, in order to increase the authentication accuracy of face and iris, we propose a score level fusion method based on SVM (Support Vector Machine). Third, to reduce the classification complexities of SVM and intra-variation of face and iris data, we normalize the input face and iris data, respectively. For face, a NIR illuminator and NIR passing filter on camera are used to reduce the illumination variance caused by environmental visible lighting and the consequent saturated region in face by the NIR illuminator is normalized by low processing logarithmic algorithm considering mobile phone. For iris, image transform into polar coordinate and iris code shifting are used for obtaining robust identification accuracy irrespective of image capturing condition. Fourth, to increase the processing speed on mobile phone, we use integer based face and iris authentication algorithms. Experimental results were tested with face and iris images by mega-pixel camera of mobile phone. It showed that the authentication accuracy using SVM was better than those of uni-modal (face or iris), SUM, MAX, NIN and weighted SUM rules.

• Journal of KIISE:Databases
• /
• v.33 no.3
• /
• pp.310-326
• /
• 2006

In this paper, we deal with a method of efficiently processing continuous queries in a data stream environment. We classify previous query processing methods into two dual categories - data-initiative and query-initiative - depending on whether query processing is initiated by selecting a data element or a query. This classification stems from the fact that data and queries have been treated asymmetrically. For processing continuous queries, only data-initiative methods have traditionally been employed, and thus, the performance gain that could be obtained by query-initiative methods has been overlooked. To solve this problem, we focus on an observation that data and queries can be treated symmetrically. In this paper, we propose the duality model of data and queries and, based on this model, present a new viewpoint of transforming the continuous query processing problem to a multi-dimensional spatial join problem. We also present a continuous query processing algorithm based on spatial join, named Spatial Join CQ. Spatial Join CQ processes continuous queries by finding the pairs of overlapping regions from a set of data elements and a set of queries defined as regions in the multi-dimensional space. The algorithm achieves the effects of both of the two dual methods by using the spatial join, which is a symmetric operation. Experimental results show that the proposed algorithm outperforms earlier methods by up to 36 times for simple selection continuous queries and by up to 7 times for sliding window join continuous queries.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.1
• /
• pp.31-38
• /
• 2011

A RAM-based Neural Net is a weightless neural network based on binary neural network. 3-D neural network using this paper is binary neural network with multiful information bits and store counts of training. Recognition method by MRD technique is based on the supervised learning. Therefore neural network by itself can not distinguish between the categories and well-separated categories of training data can achieve only through the performance. In this paper, unsupervised learning algorithm is proposed which is trained existing 3-D neural network without distinction of data, to distinguish between categories depending on the only input training patterns. The training data for proposed unsupervised learning provided by the NIST handwritten digits of MNIST which is consist of 0 to 9 multi-pattern, a randomly materials are used as training patterns. Through experiments, neural network is to determine the number of discriminator which each have an idea of the handwritten digits that can be interpreted.

• Journal of the Institute of Convergence Signal Processing
• /
• v.19 no.1
• /
• pp.7-13
• /
• 2018

In the current study, we proposed an optimized brain-computer interface (BCI) which employed blind source separation (BBS) approach to remove noises. Thus motor imagery (MI) signal and steady state visual evoked potential (SSVEP) signal were easily to be detected due to enhancement in signal-to-noise ratio (SNR). Moreover, a combination between MI and SSVEP which is typically can increase the number of commands being generated in the current BCI. To reduce the computational time as well as to bring the BCI closer to real-world applications, the current system utilizes a single-channel EEG signal. In addition, a convolutional neural network (CNN) was used as the multi-class classification model. We evaluated the performance in term of accuracy between a non-BBS+BCI and BBS+BCI. Results show that the accuracy of the BBS+BCI is achieved $16.15{\pm}5.12%$ higher than that in the non-BBS+BCI by using BBS than non-used on. Overall, the proposed BCI system demonstrate a feasibility to be applied for multi-dimensional control applications with a comparable accuracy.

• International Commerce and Information Review
• /
• v.15 no.1
• /
• pp.227-256
• /
• 2013

Existing multinational subsidiary typologies seem to have limitations in two respects. First, the prevalence of subsidiary classification along two-dimensions fails to capture many distinct subsidiary types. Failure to reflect a sufficient richness in dimensionality can give rise to a partial picture of subsidiary typologies in the international business literature. A new typology developed from multi-dimensional approach will be required for reflecting various subsidiary roles in the multinational enterprise. Second, multinational subsidiary performing a number of activities is hard to be defined functionally across the value chain activities. In addition, multinational subsidiary roles can vary dramatically. In conclusion, despite a growing amount of work on subsidiary typologies, there seems to be limited convergence of results. the study regarding subsidiary roles still remain a challenge. In this respect, the purpose of this study is to develop a new typology based on multi-dimensional approach in order to overcome the limitations of traditional typologies. To classify subsidiary types, we propose 8 types of multinational subsidiary according to three dimensions that are adopted: (1) number of required value chain activities (2) subsidiary's sourcing capability and autonomy (3) global orientation (3) The case study analyzing Korean foreign subsidiaries appropriate for 8 types is performed to establish the validity of this study.