• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.1
• /
• pp.42.1-42.1
• /
• 2021

In this study, we present a visual explanation of a deep learning solar flare forecast model and its relationship to physical parameters of solar active regions (ARs). For this, we use full-disk magnetograms at 00:00 UT from the Solar and Heliospheric Observatory/Michelson Doppler Imager and the Solar Dynamics Observatory/Helioseismic and Magnetic Imager, physical parameters from the Space-weather HMI Active Region Patch (SHARP), and Geostationary Operational Environmental Satellite X-ray flare data. Our deep learning flare forecast model based on the Convolutional Neural Network (CNN) predicts "Yes" or "No" for the daily occurrence of C-, M-, and X-class flares. We interpret the model using two CNN attribution methods (guided backpropagation and Gradient-weighted Class Activation Mapping [Grad-CAM]) that provide quantitative information on explaining the model. We find that our deep learning flare forecasting model is intimately related to AR physical properties that have also been distinguished in previous studies as holding significant predictive ability. Major results of this study are as follows. First, we successfully apply our deep learning models to the forecast of daily solar flare occurrence with TSS = 0.65, without any preprocessing to extract features from data. Second, using the attribution methods, we find that the polarity inversion line is an important feature for the deep learning flare forecasting model. Third, the ARs with high Grad-CAM values produce more flares than those with low Grad-CAM values. Fourth, nine SHARP parameters such as total unsigned vertical current, total unsigned current helicity, total unsigned flux, and total photospheric magnetic free energy density are well correlated with Grad-CAM values.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.12
• /
• pp.191-199
• /
• 2023

In this paper, we propose a method to extract the features of five sensor-only facilities built as infrastructure for autonomous cooperative driving, which are from point cloud data acquired by LiDAR. In the case of image acquisition sensors installed in autonomous vehicles, the acquisition data is inconsistent due to the climatic environment and camera characteristics, so LiDAR sensor was applied to replace them. In addition, high-intensity reflectors were designed and attached to each facility to make it easier to distinguish it from other existing facilities with LiDAR. From the five sensor-only facilities developed and the point cloud data acquired by the data acquisition system, feature points were extracted based on the average reflective intensity of the high-intensity reflective paper attached to the facility, clustered by the DBSCAN method, and changed to two-dimensional coordinates by a projection method. The features of the facility at each distance consist of three-dimensional point coordinates, two-dimensional projected coordinates, and reflection intensity, and will be used as training data for a model for facility recognition to be developed in the future.

• International Journal of Computer Science & Network Security
• /
• v.23 no.11
• /
• pp.59-66
• /
• 2023

Text mining (TM) is most widely used to find patterns from various text documents. Cyber-bullying is the term that is used to abuse a person online or offline platform. Nowadays cyber-bullying becomes more dangerous to people who are using social networking sites (SNS). Cyber-bullying is of many types such as text messaging, morphed images, morphed videos, etc. It is a very difficult task to prevent this type of abuse of the person in online SNS. Finding accurate text mining patterns gives better results in detecting cyber-bullying on any platform. Cyber-bullying is developed with the online SNS to send defamatory statements or orally bully other persons or by using the online platform to abuse in front of SNS users. Deep Learning (DL) is one of the significant domains which are used to extract and learn the quality features dynamically from the low-level text inclusions. In this scenario, Convolutional neural networks (CNN) are used for training the text data, images, and videos. CNN is a very powerful approach to training on these types of data and achieved better text classification. In this paper, an Ensemble model is introduced with the integration of Term Frequency (TF)-Inverse document frequency (IDF) and Deep Neural Network (DNN) with advanced feature-extracting techniques to classify the bullying text, images, and videos. The proposed approach also focused on reducing the training time and memory usage which helps the classification improvement.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.1
• /
• pp.32-43
• /
• 2009

In this paper, we improve the performance of cardiac disorder classification by continuous heart sound signals using automatic segmentation and extreme learning machine (ELM). The accuracy of the conventional cardiac disorder classification systems degrades because murmurs and click sounds contained in the abnormal heart sound signals cause incorrect or missing starting points of the first (S1) and the second heart pulses (S2) in the automatic segmentation stage, In order to reduce the performance degradation due to segmentation errors, we find the positions of the S1 and S2 pulses, modify them using the time difference of S1 or S2, and extract a single period of heart sound signals. We then obtain a feature vector consisting of the mel-scaled filter bank energy coefficients and the envelope of uniform-sized sub-segments from the single-period heart sound signals. To classify the heart disorders, we use ELM with a single hidden layer. In cardiac disorder classification experiments with 9 cardiac disorder categories, the proposed method shows the classification accuracy of 81.6% and achieves the highest classification accuracy among ELM, multi-layer perceptron (MLP), support vector machine (SVM), and hidden Markov model (HMM).

• Mass Spectrometry Letters
• /
• v.15 no.2
• /
• pp.95-106
• /
• 2024

Taraxacum coreanum, known as the native Korean white dandelion, has been historically used in traditional medicine due to its various therapeutic properties. However, the specific benefits and mechanisms of white dandelion in alleviating particular symptoms or diseases remain uncertain due to the complexity of its phytochemical profile. In this study, we aimed to elucidate the phytochemical profiles of methanolic extracts of different parts of the white dandelion (flower, leaf, stem, and root) using hybrid ion-mobility Q-TOF MS. Using the trapped ion mobility-based PASEF technique, 3715 and 2114 molecular features with MS2 fragments were obtained in positive and negative ion modes, respectively, and then a total of 360 and 156 phytochemical compounds were annotated by matching with a reference spectral library in positive and negative ion modes, respectively. Subsequent feature-based molecular networking analysis revealed the phytochemical differences across the four different parts of the white dandelion. Our findings indicated that the methanolic extracts contained various bioactive compounds, including lipids, flavonoids, phenolic acids, and sesquiterpenes. In particular, lipids such as linoleic acids, lysophosphatidylcholines, and sesquiterpenoids were predominantly present in the leaf, while flavonoid glycosides and lysophosphoethanolamines were notably enriched in the flower. An assessment of the total phenolic content (TPC) and total flavonoid content (TFC) of the methanolic extracts revealed that the majority of phytochemicals were concentrated in the flower. Interestingly, despite the root extract displaying the lowest TPC and TFC values, it exhibited the highest radical scavenging rate when normalized to TPC and TFC, suggesting a potent antioxidant effect. These findings and further investigations into the biological activities and medicinal potential of the identified compounds, particularly those exclusive to specific plant parts, may contribute to the development of novel therapeutic agents derived from white dandelion.

• Journal of Internet Computing and Services
• /
• v.25 no.4
• /
• pp.57-64
• /
• 2024

This paper aims to enhance the accuracy of music genre classification for music tracks where genre information is not provided, by utilizing machine learning to classify a large amount of music data. The paper proposes collecting and preprocessing data instead of using the commonly employed GTZAN dataset in previous research for genre classification in music. To create a dataset with superior classification performance compared to the GTZAN dataset, we extract specific segments with the highest energy level of the onset. We utilize 57 features as the main characteristics of the music data used for training, including Mel Frequency Cepstral Coefficients (MFCC). We achieved a training accuracy of 85% and a testing accuracy of 71% using the Support Vector Machine (SVM) model to classify into Classical, Jazz, Country, Disco, Soul, Rock, Metal, and Hiphop genres based on preprocessed data.

• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.1-25
• /
• 2020

In this paper, we suggest an application system architecture which provides accurate, fast and efficient automatic gasometer reading function. The system captures gasometer image using mobile device camera, transmits the image to a cloud server on top of private LTE network, and analyzes the image to extract character information of device ID and gas usage amount by selective optical character recognition based on deep learning technology. In general, there are many types of character in an image and optical character recognition technology extracts all character information in an image. But some applications need to ignore non-of-interest types of character and only have to focus on some specific types of characters. For an example of the application, automatic gasometer reading system only need to extract device ID and gas usage amount character information from gasometer images to send bill to users. Non-of-interest character strings, such as device type, manufacturer, manufacturing date, specification and etc., are not valuable information to the application. Thus, the application have to analyze point of interest region and specific types of characters to extract valuable information only. We adopted CNN (Convolutional Neural Network) based object detection and CRNN (Convolutional Recurrent Neural Network) technology for selective optical character recognition which only analyze point of interest region for selective character information extraction. We build up 3 neural networks for the application system. The first is a convolutional neural network which detects point of interest region of gas usage amount and device ID information character strings, the second is another convolutional neural network which transforms spatial information of point of interest region to spatial sequential feature vectors, and the third is bi-directional long short term memory network which converts spatial sequential information to character strings using time-series analysis mapping from feature vectors to character strings. In this research, point of interest character strings are device ID and gas usage amount. Device ID consists of 12 arabic character strings and gas usage amount consists of 4 ~ 5 arabic character strings. All system components are implemented in Amazon Web Service Cloud with Intel Zeon E5-2686 v4 CPU and NVidia TESLA V100 GPU. The system architecture adopts master-lave processing structure for efficient and fast parallel processing coping with about 700,000 requests per day. Mobile device captures gasometer image and transmits to master process in AWS cloud. Master process runs on Intel Zeon CPU and pushes reading request from mobile device to an input queue with FIFO (First In First Out) structure. Slave process consists of 3 types of deep neural networks which conduct character recognition process and runs on NVidia GPU module. Slave process is always polling the input queue to get recognition request. If there are some requests from master process in the input queue, slave process converts the image in the input queue to device ID character string, gas usage amount character string and position information of the strings, returns the information to output queue, and switch to idle mode to poll the input queue. Master process gets final information form the output queue and delivers the information to the mobile device. We used total 27,120 gasometer images for training, validation and testing of 3 types of deep neural network. 22,985 images were used for training and validation, 4,135 images were used for testing. We randomly splitted 22,985 images with 8:2 ratio for training and validation respectively for each training epoch. 4,135 test image were categorized into 5 types (Normal, noise, reflex, scale and slant). Normal data is clean image data, noise means image with noise signal, relfex means image with light reflection in gasometer region, scale means images with small object size due to long-distance capturing and slant means images which is not horizontally flat. Final character string recognition accuracies for device ID and gas usage amount of normal data are 0.960 and 0.864 respectively.

• Korean Journal of Remote Sensing
• /
• v.14 no.3
• /
• pp.296-306
• /
• 1998

Airborne MSS data is regarded as a potentially effective data source for the measurement of water quality and for the environmental change of water bodies. In this study, we measured the radiance reflectance by using multi-spectral image of low resolution camera(LRC) which will be reached in the multi-purpose satellite(KOMPSAT) to use the data in analyzing water pollution. We also investigated the possibility of extraction of water quality factors in water bodies by using high resolution remote sensing data such as Airborne MSS. Especially, we tried to extract environmental factors related with eutrophication such as chlorophyll-a, suspended sediments and turbidity, and also tried to develop the process technique and the radiance feature of reflectance related with eutrophication. Although it was difficult to explicitly correlate Airborne MSS data with water quality factors due to the insufficient number of ground truth data. The results were summarized as follows: First, the spectrum of sun's rays which reaches the surface of the earth was consistent with visible bands of 0.4${\mu}{\textrm}{m}$~0.7${\mu}{\textrm}{m}$ and about 50% of total quantity of radiation could be found. The spectrum was reached highest at around 0.5${\mu}{\textrm}{m}$ of green spectral band in visible bands. Second, as a result of the radiance reflectance Chlorophyll-a represented high mainly around 0.52${\mu}{\textrm}{m}$ of green spectral band, and suspended sediments and turbidity represented high at 0.8${\mu}{\textrm}{m}$ and at 0.57${\mu}{\textrm}{m}$, respectively. Finally, as a result of the water quality analysis by using Airborne MSS, Chlorophyll-a could have a distribution image after carrying out ratio of B3 and B5 to B7. Band 7 was useful for making the distribution image of suspended sediments. When we carried out PCA, suspended sediments and turbidity had distributions at PC 1 and PC 4 which are similar to the ground data. Above results can be changed according to the change of season and time. Therefore, in order to analyze the environmental factors of water quality by using LRC data more exactly, we need to investigate the ground data and the radiance feature of reflectance of water bodies constantly. For further studies, we will constantly analyze the radiance feature of the surface of water in wafter bodies by measuring the on-the-spot radiance reflectance and using low resolution satellite image(SeaWiFS). We will also gather the data of water quality analysis in water bodies and analyze the pattern of water pollution.

• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.205-225
• /
• 2018

Convolutional Neural Network (ConvNet) is one class of the powerful Deep Neural Network that can analyze and learn hierarchies of visual features. Originally, first neural network (Neocognitron) was introduced in the 80s. At that time, the neural network was not broadly used in both industry and academic field by cause of large-scale dataset shortage and low computational power. However, after a few decades later in 2012, Krizhevsky made a breakthrough on ILSVRC-12 visual recognition competition using Convolutional Neural Network. That breakthrough revived people interest in the neural network. The success of Convolutional Neural Network is achieved with two main factors. First of them is the emergence of advanced hardware (GPUs) for sufficient parallel computation. Second is the availability of large-scale datasets such as ImageNet (ILSVRC) dataset for training. Unfortunately, many new domains are bottlenecked by these factors. For most domains, it is difficult and requires lots of effort to gather large-scale dataset to train a ConvNet. Moreover, even if we have a large-scale dataset, training ConvNet from scratch is required expensive resource and time-consuming. These two obstacles can be solved by using transfer learning. Transfer learning is a method for transferring the knowledge from a source domain to new domain. There are two major Transfer learning cases. First one is ConvNet as fixed feature extractor, and the second one is Fine-tune the ConvNet on a new dataset. In the first case, using pre-trained ConvNet (such as on ImageNet) to compute feed-forward activations of the image into the ConvNet and extract activation features from specific layers. In the second case, replacing and retraining the ConvNet classifier on the new dataset, then fine-tune the weights of the pre-trained network with the backpropagation. In this paper, we focus on using multiple ConvNet layers as a fixed feature extractor only. However, applying features with high dimensional complexity that is directly extracted from multiple ConvNet layers is still a challenging problem. We observe that features extracted from multiple ConvNet layers address the different characteristics of the image which means better representation could be obtained by finding the optimal combination of multiple ConvNet layers. Based on that observation, we propose to employ multiple ConvNet layer representations for transfer learning instead of a single ConvNet layer representation. Overall, our primary pipeline has three steps. Firstly, images from target task are given as input to ConvNet, then that image will be feed-forwarded into pre-trained AlexNet, and the activation features from three fully connected convolutional layers are extracted. Secondly, activation features of three ConvNet layers are concatenated to obtain multiple ConvNet layers representation because it will gain more information about an image. When three fully connected layer features concatenated, the occurring image representation would have 9192 (4096+4096+1000) dimension features. However, features extracted from multiple ConvNet layers are redundant and noisy since they are extracted from the same ConvNet. Thus, a third step, we will use Principal Component Analysis (PCA) to select salient features before the training phase. When salient features are obtained, the classifier can classify image more accurately, and the performance of transfer learning can be improved. To evaluate proposed method, experiments are conducted in three standard datasets (Caltech-256, VOC07, and SUN397) to compare multiple ConvNet layer representations against single ConvNet layer representation by using PCA for feature selection and dimension reduction. Our experiments demonstrated the importance of feature selection for multiple ConvNet layer representation. Moreover, our proposed approach achieved 75.6% accuracy compared to 73.9% accuracy achieved by FC7 layer on the Caltech-256 dataset, 73.1% accuracy compared to 69.2% accuracy achieved by FC8 layer on the VOC07 dataset, 52.2% accuracy compared to 48.7% accuracy achieved by FC7 layer on the SUN397 dataset. We also showed that our proposed approach achieved superior performance, 2.8%, 2.1% and 3.1% accuracy improvement on Caltech-256, VOC07, and SUN397 dataset respectively compare to existing work.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.109-125
• /
• 2019

As interest on intelligent search engines increases, various studies have been conducted to extract and utilize the features related to products intelligencely. In particular, when users search for goods in e-commerce search engines, the 'color' of a product is an important feature that describes the product. Therefore, it is necessary to deal with the synonyms of color terms in order to produce accurate results to user's color-related queries. Previous studies have suggested dictionary-based approach to process synonyms for color features. However, the dictionary-based approach has a limitation that it cannot handle unregistered color-related terms in user queries. In order to overcome the limitation of the conventional methods, this research proposes a model which extracts RGB values from an internet search engine in real time, and outputs similar color names based on designated color information. At first, a color term dictionary was constructed which includes color names and R, G, B values of each color from Korean color standard digital palette program and the Wikipedia color list for the basic color search. The dictionary has been made more robust by adding 138 color names converted from English color names to foreign words in Korean, and with corresponding RGB values. Therefore, the fininal color dictionary includes a total of 671 color names and corresponding RGB values. The method proposed in this research starts by searching for a specific color which a user searched for. Then, the presence of the searched color in the built-in color dictionary is checked. If there exists the color in the dictionary, the RGB values of the color in the dictioanry are used as reference values of the retrieved color. If the searched color does not exist in the dictionary, the top-5 Google image search results of the searched color are crawled and average RGB values are extracted in certain middle area of each image. To extract the RGB values in images, a variety of different ways was attempted since there are limits to simply obtain the average of the RGB values of the center area of images. As a result, clustering RGB values in image's certain area and making average value of the cluster with the highest density as the reference values showed the best performance. Based on the reference RGB values of the searched color, the RGB values of all the colors in the color dictionary constructed aforetime are compared. Then a color list is created with colors within the range of ${\pm}50$ for each R value, G value, and B value. Finally, using the Euclidean distance between the above results and the reference RGB values of the searched color, the color with the highest similarity from up to five colors becomes the final outcome. In order to evaluate the usefulness of the proposed method, we performed an experiment. In the experiment, 300 color names and corresponding color RGB values by the questionnaires were obtained. They are used to compare the RGB values obtained from four different methods including the proposed method. The average euclidean distance of CIE-Lab using our method was about 13.85, which showed a relatively low distance compared to 3088 for the case using synonym dictionary only and 30.38 for the case using the dictionary with Korean synonym website WordNet. The case which didn't use clustering method of the proposed method showed 13.88 of average euclidean distance, which implies the DBSCAN clustering of the proposed method can reduce the Euclidean distance. This research suggests a new color synonym processing method based on RGB values that combines the dictionary method with the real time synonym processing method for new color names. This method enables to get rid of the limit of the dictionary-based approach which is a conventional synonym processing method. This research can contribute to improve the intelligence of e-commerce search systems especially on the color searching feature.