• Journal of the Korean earth science society
• /
• v.40 no.5
• /
• pp.447-463
• /
• 2019

Sea surface wind is a fundamental element for understanding the oceanic phenomena and for analyzing changes of the Earth environment caused by global warming. Global research institutes have developed and operated scatterometers to accurately and continuously observe the sea surface wind, with the accuracy of approximately ${\pm}20^{\circ}$ for wind direction and ${\pm}2m\;s^{-1}$ for wind speed. Given that the spatial resolution of the scatterometer is 12.5-25.0 km, the applicability of the data to the coastal area is limited due to complicated coastal lines and many islands around the Korean Peninsula. In contrast, Synthetic Aperture Radar (SAR), one of microwave sensors, is an all-weather instrument, which enables us to retrieve sea surface wind with high resolution (<1 km) and compensate the sparse resolution of the scatterometer. In this study, we investigated the Geophysical Model Functions (GMF), which are the algorithms for retrieval of sea surface wind speed from the SAR data depending on each band such as C-, L-, or X-band radar. We reviewed in the simulation of the backscattering coefficients for relative wind direction, incidence angle, and wind speed by applying LMOD, CMOD, and XMOD model functions, and analyzed the characteristics of each GMF. We investigated previous studies about the validation of wind speed from the SAR data using these GMFs. The accuracy of sea surface wind from SAR data changed with respect to observation mode, GMF type, reference data for validation, preprocessing method, and the method for calculation of relative wind direction. It is expected that this study contributes to the potential users of SAR images who retrieve wind speeds from SAR data at the coastal region around the Korean Peninsula.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1505-1514
• /
• 2022

Image matching is a crucial preprocessing step for effective utilization of multi-temporal and multi-sensor very high resolution (VHR) satellite images. Deep learning (DL) method which is attracting widespread interest has proven to be an efficient approach to measure the similarity between image pairs in quick and accurate manner by extracting complex and detailed features from satellite images. However, Image matching of VHR satellite images remains challenging due to limitations of DL models in which the results are depending on the quantity and quality of training dataset, as well as the difficulty of creating training dataset with VHR satellite images. Therefore, this study examines the feasibility of DL-based method in matching pair extraction which is the most time-consuming process during image registration. This paper also aims to analyze factors that affect the accuracy based on the configuration of training dataset, when developing training dataset from existing multi-sensor VHR image database with bias for DL-based image matching. For this purpose, the generated training dataset were composed of correct matching pairs and incorrect matching pairs by assigning true and false labels to image pairs extracted using a grid-based Scale Invariant Feature Transform (SIFT) algorithm for a total of 12 multi-temporal and multi-sensor VHR images. The Siamese convolutional neural network (SCNN), proposed for matching pair extraction on constructed training dataset, proceeds with model learning and measures similarities by passing two images in parallel to the two identical convolutional neural network structures. The results from this study confirm that data acquired from VHR satellite image database can be used as DL training dataset and indicate the potential to improve efficiency of the matching process by appropriate configuration of multi-sensor images. DL-based image matching techniques using multi-sensor VHR satellite images are expected to replace existing manual-based feature extraction methods based on its stable performance, thus further develop into an integrated DL-based image registration framework.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.25 no.4
• /
• pp.337-345
• /
• 2023

Estimating the rice heading date is one of the most crucial agricultural tasks related to productivity. However, due to abnormal climates around the world, it is becoming increasingly challenging to estimate the rice heading date. Therefore, a more objective classification method for estimating the rice heading date is needed than the existing methods. This study, we aimed to classify the rice heading stage from various images using a CNN classification model. We collected top-view images taken from a drone and a phenotyping tower, as well as slanted-view images captured with a RGB camera. The collected images underwent preprocessing to prepare them as input data for the CNN model. The CNN architectures employed were ResNet50, InceptionV3, and VGG19, which are commonly used in image classification models. The accuracy of the models all showed an accuracy of 0.98 or higher regardless of each architecture and type of image. We also used Grad-CAM to visually check which features of the image the model looked at and classified. Then verified our model accurately measure the rice heading date in paddy fields. The rice heading date was estimated to be approximately one day apart on average in the four paddy fields. This method suggests that the water head can be estimated automatically and quantitatively when estimating the rice heading date from various paddy field monitoring images.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_1
• /
• pp.1283-1297
• /
• 2023

Mid-wave infrared (MWIR) imagery, due to its ability to capture the temperature of land cover and objects, serves as a crucial data source in various fields including environmental monitoring and defense. The KOMPSAT-3A satellite acquires MWIR imagery with high spatial resolution compared to other satellites. However, the limited spatial resolution of MWIR imagery, in comparison to electro-optical (EO) imagery, constrains the optimal utilization of the KOMPSAT-3A data. This study aims to create a highly visible MWIR fusion image by leveraging the edge information from the KOMPSAT-3A panchromatic (PAN) image. Preprocessing is implemented to mitigate the relative geometric errors between the PAN and MWIR images. Subsequently, we employ a pre-trained pixel difference network (PiDiNet), a deep learning-based edge information extraction technique, to extract the boundaries of objects from the preprocessed PAN images. The MWIR fusion imagery is then generated by emphasizing the brightness value corresponding to the edge information of the PAN image. To evaluate the proposed method, the MWIR fusion images were generated in three different sites. As a result, the boundaries of terrain and objects in the MWIR fusion images were emphasized to provide detailed thermal information of the interest area. Especially, the MWIR fusion image provided the thermal information of objects such as airplanes and ships which are hard to detect in the original MWIR images. This study demonstrated that the proposed method could generate a single image that combines visible details from an EO image and thermal information from an MWIR image, which contributes to increasing the usage of MWIR imagery.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.163-177
• /
• 2019

As smartphones are getting widely used, human activity recognition (HAR) tasks for recognizing personal activities of smartphone users with multimodal data have been actively studied recently. The research area is expanding from the recognition of the simple body movement of an individual user to the recognition of low-level behavior and high-level behavior. However, HAR tasks for recognizing interaction behavior with other people, such as whether the user is accompanying or communicating with someone else, have gotten less attention so far. And previous research for recognizing interaction behavior has usually depended on audio, Bluetooth, and Wi-Fi sensors, which are vulnerable to privacy issues and require much time to collect enough data. Whereas physical sensors including accelerometer, magnetic field and gyroscope sensors are less vulnerable to privacy issues and can collect a large amount of data within a short time. In this paper, a method for detecting accompanying status based on deep learning model by only using multimodal physical sensor data, such as an accelerometer, magnetic field and gyroscope, was proposed. The accompanying status was defined as a redefinition of a part of the user interaction behavior, including whether the user is accompanying with an acquaintance at a close distance and the user is actively communicating with the acquaintance. A framework based on convolutional neural networks (CNN) and long short-term memory (LSTM) recurrent networks for classifying accompanying and conversation was proposed. First, a data preprocessing method which consists of time synchronization of multimodal data from different physical sensors, data normalization and sequence data generation was introduced. We applied the nearest interpolation to synchronize the time of collected data from different sensors. Normalization was performed for each x, y, z axis value of the sensor data, and the sequence data was generated according to the sliding window method. Then, the sequence data became the input for CNN, where feature maps representing local dependencies of the original sequence are extracted. The CNN consisted of 3 convolutional layers and did not have a pooling layer to maintain the temporal information of the sequence data. Next, LSTM recurrent networks received the feature maps, learned long-term dependencies from them and extracted features. The LSTM recurrent networks consisted of two layers, each with 128 cells. Finally, the extracted features were used for classification by softmax classifier. The loss function of the model was cross entropy function and the weights of the model were randomly initialized on a normal distribution with an average of 0 and a standard deviation of 0.1. The model was trained using adaptive moment estimation (ADAM) optimization algorithm and the mini batch size was set to 128. We applied dropout to input values of the LSTM recurrent networks to prevent overfitting. The initial learning rate was set to 0.001, and it decreased exponentially by 0.99 at the end of each epoch training. An Android smartphone application was developed and released to collect data. We collected smartphone data for a total of 18 subjects. Using the data, the model classified accompanying and conversation by 98.74% and 98.83% accuracy each. Both the F1 score and accuracy of the model were higher than the F1 score and accuracy of the majority vote classifier, support vector machine, and deep recurrent neural network. In the future research, we will focus on more rigorous multimodal sensor data synchronization methods that minimize the time stamp differences. In addition, we will further study transfer learning method that enables transfer of trained models tailored to the training data to the evaluation data that follows a different distribution. It is expected that a model capable of exhibiting robust recognition performance against changes in data that is not considered in the model learning stage will be obtained.

• Journal of Korean Society of Environmental Engineers
• /
• v.35 no.2
• /
• pp.115-123
• /
• 2013

Stoker type incinerator is one of the most popular one used as municipal solid waste (MSW) incineration because, in general, it is quite suitable for large capacity and need no preprocessing facility. Nowadays, however, since the combustible portion of incoming MSW increases together with the decrease of the moisture content due to prohibition of directly burying food waste in landfill, the heating value of waste is remarkably increasing in comparison with the early stage of incinerator installation. Consequently, the increased heating value in incinerator operation causes a number of serious problems such as reduction of waste amount to be burned due to the boiler heat capacity together with the significant NO generation in high temperature environment. Therefore, in this study, a series of numerical simulation have been made as parameters of waste amount and the fraction of moisture in air stream in order to investigate optimal operating condition for the resolution of the problems associated with the high heating value of waste mentioned above. In specific, a detailed turbulent reaction flow field calculation with NO model was made for the full scale incinerator of D city. To this end, the injection method of moisturized air as oxidizer was intensively reviewed by the addition of moisture water amount from 10% and 20%. The calculation result, in general, showed that the reduction of maximum flame temperature appears consistently due to the combined effects of the increased specific heat of combustion air and vaporization heat by the addition of water moisture. As a consequence, the generation of NOx concentration was substantially reduced. Further, for the case of 20% moisture amount stream, the afterburner region is quite appropriate in temperature range for the operation of SNCR. This suggests the SNCR facility can be considered for reoperation. which is not in service at all due to the increased heating value of MSW.

• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.179-200
• /
• 2019

Word-of-mouth (WoM) is defined by consumer activities that share information concerning consumption. WoM activities have long been recognized as important in corporate marketing processes and have received much attention, especially in the marketing field. Recently, according to the development of the Internet, the way in which people exchange information in online news and online communities has been expanded, and WoM is diversified in terms of word of mouth, score, rating, and liking. Social media makes online users easy access to information and online WoM is considered a key source of information. Although various studies on WoM have been preceded by this phenomenon, there is no meta-analysis study that comprehensively analyzes them. This study proposed a method to extract major researches by applying text mining techniques and to grasp the main issues of researches in order to find the trend of WoM research using scholarly big data. To this end, a total of 4389 documents were collected by the keyword 'Word-of-mouth' from 1941 to 2018 in Scopus (www.scopus.com), a citation database, and the data were refined through preprocessing such as English morphological analysis, stopwords removal, and noun extraction. To carry out this study, we adopted main path analysis (MPA) and word co-occurrence network analysis. MPA detects key researches and is used to track the development trajectory of academic field, and presents the research trend from a macro perspective. For this, we constructed a citation network based on the collected data. The node means a document and the link means a citation relation in citation network. We then detected the key-route main path by applying SPC (Search Path Count) weights. As a result, the main path composed of 30 documents extracted from a citation network. The main path was able to confirm the change of the academic area which was developing along with the change of the times reflecting the industrial change such as various industrial groups. The results of MPA revealed that WoM research was distinguished by five periods: (1) establishment of aspects and critical elements of WoM, (2) relationship analysis between WoM variables, (3) beginning of researches of online WoM, (4) relationship analysis between WoM and purchase, and (5) broadening of topics. It was found that changes within the industry was reflected in the results such as online development and social media. Very recent studies showed that the topics and approaches related WoM were being diversified to circumstantial changes. However, the results showed that even though WoM was used in diverse fields, the main stream of the researches of WoM from the start to the end, was related to marketing and figuring out the influential factors that proliferate WoM. By applying word co-occurrence network analysis, the research trend is presented from a microscopic point of view. Word co-occurrence network was constructed to analyze the relationship between keywords and social network analysis (SNA) was utilized. We divided the data into three periods to investigate the periodic changes and trends in discussion of WoM. SNA showed that Period 1 (1941~2008) consisted of clusters regarding relationship, source, and consumers. Period 2 (2009~2013) contained clusters of satisfaction, community, social networks, review, and internet. Clusters of period 3 (2014~2018) involved satisfaction, medium, review, and interview. The periodic changes of clusters showed transition from offline to online WoM. Media of WoM have become an important factor in spreading the words. This study conducted a quantitative meta-analysis based on scholarly big data regarding WoM. The main contribution of this study is that it provides a micro perspective on the research trend of WoM as well as the macro perspective. The limitation of this study is that the citation network constructed in this study is a network based on the direct citation relation of the collected documents for MPA.

• Journal of IKEEE
• /
• v.27 no.1
• /
• pp.116-121
• /
• 2023

In this paper, we develop a deep learning structure for a complex microbial incubator that applies deep learning prediction result information. The proposed complex microbial incubator consists of pre-processing of complex microbial data, conversion of complex microbial data structure, design of deep learning network, learning of the designed deep learning network, and GUI development applied to the prototype. In the complex microbial data preprocessing, one-hot encoding is performed on the amount of molasses, nutrients, plant extract, salt, etc. required for microbial culture, and the maximum-minimum normalization method for the pH concentration measured as a result of the culture and the number of microbial cells to preprocess the data. In the complex microbial data structure conversion, the preprocessed data is converted into a graph structure by connecting the water temperature and the number of microbial cells, and then expressed as an adjacency matrix and attribute information to be used as input data for a deep learning network. In deep learning network design, complex microbial data is learned by designing a graph convolutional network specialized for graph structures. The designed deep learning network uses a cosine loss function to proceed with learning in the direction of minimizing the error that occurs during learning. GUI development applied to the prototype shows the target pH concentration (3.8 or less) and the number of cells (10⁸ or more) of complex microorganisms in an order suitable for culturing according to the water temperature selected by the user. In order to evaluate the performance of the proposed microbial incubator, the results of experiments conducted by authorized testing institutes showed that the average pH was 3.7 and the number of cells of complex microorganisms was 1.7 × 10⁸. Therefore, the effectiveness of the deep learning structure for the complex microbial incubator applying the deep learning prediction result information proposed in this paper was proven.

• Asia pacific journal of information systems
• /
• v.18 no.3
• /
• pp.123-145
• /
• 2008

Recently, as the information communication technology develops, the discussion regarding the ubiquitous environment is occurring in diverse perspectives. Ubiquitous environment is an environment that could transfer data through networks regardless of the physical space, virtual space, time or location. In order to realize the ubiquitous environment, the Pervasive Sensing technology that enables the recognition of users' data without the border between physical and virtual space is required. In addition, the latest and diversified technologies such as Context-Awareness technology are necessary to construct the context around the user by sharing the data accessed through the Pervasive Sensing technology and linkage technology that is to prevent information loss through the wired, wireless networking and database. Especially, Pervasive Sensing technology is taken as an essential technology that enables user oriented services by recognizing the needs of the users even before the users inquire. There are lots of characteristics of ubiquitous environment through the technologies mentioned above such as ubiquity, abundance of data, mutuality, high information density, individualization and customization. Among them, information density directs the accessible amount and quality of the information and it is stored in bulk with ensured quality through Pervasive Sensing technology. Using this, in the companies, the personalized contents(or information) providing became possible for a target customer. Most of all, there are an increasing number of researches with respect to recommender systems that provide what customers need even when the customers do not explicitly ask something for their needs. Recommender systems are well renowned for its affirmative effect that enlarges the selling opportunities and reduces the searching cost of customers since it finds and provides information according to the customers' traits and preference in advance, in a commerce environment. Recommender systems have proved its usability through several methodologies and experiments conducted upon many different fields from the mid-1990s. Most of the researches related with the recommender systems until now take the products or information of internet or mobile context as its object, but there is not enough research concerned with recommending adequate store to customers in a ubiquitous environment. It is possible to track customers' behaviors in a ubiquitous environment, the same way it is implemented in an online market space even when customers are purchasing in an offline marketplace. Unlike existing internet space, in ubiquitous environment, the interest toward the stores is increasing that provides information according to the traffic line of the customers. In other words, the same product can be purchased in several different stores and the preferred store can be different from the customers by personal preference such as traffic line between stores, location, atmosphere, quality, and price. Krulwich(1997) has developed Lifestyle Finder which recommends a product and a store by using the demographical information and purchasing information generated in the internet commerce. Also, Fano(1998) has created a Shopper's Eye which is an information proving system. The information regarding the closest store from the customers' present location is shown when the customer has sent a to-buy list, Sadeh(2003) developed MyCampus that recommends appropriate information and a store in accordance with the schedule saved in a customers' mobile. Moreover, Keegan and O'Hare(2004) came up with EasiShop that provides the suitable tore information including price, after service, and accessibility after analyzing the to-buy list and the current location of customers. However, Krulwich(1997) does not indicate the characteristics of physical space based on the online commerce context and Keegan and O'Hare(2004) only provides information about store related to a product, while Fano(1998) does not fully consider the relationship between the preference toward the stores and the store itself. The most recent research by Sedah(2003), experimented on campus by suggesting recommender systems that reflect situation and preference information besides the characteristics of the physical space. Yet, there is a potential problem since the researches are based on location and preference information of customers which is connected to the invasion of privacy. The primary beginning point of controversy is an invasion of privacy and individual information in a ubiquitous environment according to researches conducted by Al-Muhtadi(2002), Beresford and Stajano(2003), and Ren(2006). Additionally, individuals want to be left anonymous to protect their own personal information, mentioned in Srivastava(2000). Therefore, in this paper, we suggest a methodology to recommend stores in U-market on the basis of ubiquitous environment not using personal information in order to protect individual information and privacy. The main idea behind our suggested methodology is based on Feature Matrices model (FM model, Shahabi and Banaei-Kashani, 2003) that uses clusters of customers' similar transaction data, which is similar to the Collaborative Filtering. However unlike Collaborative Filtering, this methodology overcomes the problems of personal information and privacy since it is not aware of the customer, exactly who they are, The methodology is compared with single trait model(vector model) such as visitor logs, while looking at the actual improvements of the recommendation when the context information is used. It is not easy to find real U-market data, so we experimented with factual data from a real department store with context information. The recommendation procedure of U-market proposed in this paper is divided into four major phases. First phase is collecting and preprocessing data for analysis of shopping patterns of customers. The traits of shopping patterns are expressed as feature matrices of N dimension. On second phase, the similar shopping patterns are grouped into clusters and the representative pattern of each cluster is derived. The distance between shopping patterns is calculated by Projected Pure Euclidean Distance (Shahabi and Banaei-Kashani, 2003). Third phase finds a representative pattern that is similar to a target customer, and at the same time, the shopping information of the customer is traced and saved dynamically. Fourth, the next store is recommended based on the physical distance between stores of representative patterns and the present location of target customer. In this research, we have evaluated the accuracy of recommendation method based on a factual data derived from a department store. There are technological difficulties of tracking on a real-time basis so we extracted purchasing related information and we added on context information on each transaction. As a result, recommendation based on FM model that applies purchasing and context information is more stable and accurate compared to that of vector model. Additionally, we could find more precise recommendation result as more shopping information is accumulated. Realistically, because of the limitation of ubiquitous environment realization, we were not able to reflect on all different kinds of context but more explicit analysis is expected to be attainable in the future after practical system is embodied.

• Journal of Intelligence and Information Systems
• /
• v.24 no.4
• /
• pp.111-136
• /
• 2018

In this paper, we propose a methodology to extract answer information about queries from various types of unstructured documents collected from multi-sources existing on web in order to expand knowledge base. The proposed methodology is divided into the following steps. 1) Collect relevant documents from Wikipedia, Naver encyclopedia, and Naver news sources for "subject-predicate" separated queries and classify the proper documents. 2) Determine whether the sentence is suitable for extracting information and derive the confidence. 3) Based on the predicate feature, extract the information in the proper sentence and derive the overall confidence of the information extraction result. In order to evaluate the performance of the information extraction system, we selected 400 queries from the artificial intelligence speaker of SK-Telecom. Compared with the baseline model, it is confirmed that it shows higher performance index than the existing model. The contribution of this study is that we develop a sequence tagging model based on bi-directional LSTM-CRF using the predicate feature of the query, with this we developed a robust model that can maintain high recall performance even in various types of unstructured documents collected from multiple sources. The problem of information extraction for knowledge base extension should take into account heterogeneous characteristics of source-specific document types. The proposed methodology proved to extract information effectively from various types of unstructured documents compared to the baseline model. There is a limitation in previous research that the performance is poor when extracting information about the document type that is different from the training data. In addition, this study can prevent unnecessary information extraction attempts from the documents that do not include the answer information through the process for predicting the suitability of information extraction of documents and sentences before the information extraction step. It is meaningful that we provided a method that precision performance can be maintained even in actual web environment. The information extraction problem for the knowledge base expansion has the characteristic that it can not guarantee whether the document includes the correct answer because it is aimed at the unstructured document existing in the real web. When the question answering is performed on a real web, previous machine reading comprehension studies has a limitation that it shows a low level of precision because it frequently attempts to extract an answer even in a document in which there is no correct answer. The policy that predicts the suitability of document and sentence information extraction is meaningful in that it contributes to maintaining the performance of information extraction even in real web environment. The limitations of this study and future research directions are as follows. First, it is a problem related to data preprocessing. In this study, the unit of knowledge extraction is classified through the morphological analysis based on the open source Konlpy python package, and the information extraction result can be improperly performed because morphological analysis is not performed properly. To enhance the performance of information extraction results, it is necessary to develop an advanced morpheme analyzer. Second, it is a problem of entity ambiguity. The information extraction system of this study can not distinguish the same name that has different intention. If several people with the same name appear in the news, the system may not extract information about the intended query. In future research, it is necessary to take measures to identify the person with the same name. Third, it is a problem of evaluation query data. In this study, we selected 400 of user queries collected from SK Telecom 's interactive artificial intelligent speaker to evaluate the performance of the information extraction system. n this study, we developed evaluation data set using 800 documents (400 questions * 7 articles per question (1 Wikipedia, 3 Naver encyclopedia, 3 Naver news) by judging whether a correct answer is included or not. To ensure the external validity of the study, it is desirable to use more queries to determine the performance of the system. This is a costly activity that must be done manually. Future research needs to evaluate the system for more queries. It is also necessary to develop a Korean benchmark data set of information extraction system for queries from multi-source web documents to build an environment that can evaluate the results more objectively.