• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.19 no.10
• /
• pp.2028-2042
• /
• 1994

Properly defining signal and noise in a self-organizing system like ART(Adaptive Resonance Theory) neural network model raises a number of subtle issues. Pattern context must enter the definition so that input features, treated as irrelevant noise when they are embedded in a given input pattern, may be treated as informative signals when they are embedded in a different input pattern. The ATR automatically self-scales their computational units to embody context and learning dependent definitions of a signal and noise and there is no problem in categorizing input pattern that have features similar in nature. However, when we have imput patterns that have features that are different in size and nature, the use of only one vigilance parameter is not enough to differentiate a signal from noise for a good categorization. For example, if the value fo vigilance parameter is large, then noise may be processed as an informative signal and unnecessary categories are generated: and if the value of vigilance parameter is small, an informative signal may be ignored and treated as noise. Hence it is no easy to achieve a good pattern categorization. To overcome such problems, a Coupled-ART neural network capable of modularized categorization of patterns is proposed. The Coupled-ART has two layer of tightly coupled modules. the upper and the lower. The lower layer processes the global features of a pattern and the structural features, separately in parallel. The upper layer combines the categorized outputs from the lower layer and categorizes the combined output, Hence, due to the modularized categorization of patterns, the Coupled-ART classifies patterns more efficiently than the ART1 model.

• 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 the Korean Society of Environmental Restoration Technology
• /
• v.3 no.1
• /
• pp.10-19
• /
• 2000

Due to the urbanization, concentrated population, and limited land exploitation in the modern society, the environment surrounding that we live in is getting polluted more and more, and it has become hard even to let urban children experience the nature. This research was conducted to help people recognize the importance of our natural resources through the environmental education of elementary school and to use school's practical open-space for the Indigenous Plants education. The results of this study are as follows : First, the status of a plant utilization in our institutional education : There were 362 species totally of 124 species of Trees, 156 species of Herbs, 63 species of Crops, and 19 species of Hydrophytes which appear in the elementary school text book. Of all, the most frequently appearing species of tree were the Malus pumila var. dulcissima, Pinus densijlora, Citrus unshiu, Diospyros kaki. Second, the effect of plant education using the land around schools : The result of research on the open-space of the 19 elementary schools located in Mapo-gu showed that most of the species planted are the Juniperus chinensisrose, Hibiscus syriacus. Pelargonium inquinans in the order of size, and the plants appearing in text book were grown in the botanical garden organized in 7 schools. Especially most of the Indigenous Plants were being planted in botanical garden, and Pinus densijlora, Abeliophyllum distichum, Polygonatum var. plurijlorum, Liriope platyphylla and so on. Last, the result of this research on recognition of Environment, Planting education and Indigenous plants : It showed that educational necessity of students and teachers about environment and Indigenous Plants was more than 80%. The management of botanical garden was conducted by some teachers and managers. The results of this study suggested that we needed the reconstruction of curriculum, the efficient application of plant education for effectiveness of using school environment and monitoring continually and construction information sources for the better environment education in the elementary schools.

• The KIPS Transactions:PartB
• /
• v.16B no.4
• /
• pp.299-308
• /
• 2009

AAM(Active Appearance Model) is an algorithm to extract face feature points with statistical models of shape and texture information based on PCA(Principal Component Analysis). This method is widely used for face recognition, face modeling and expression recognition. However, the detection performance of AAM algorithm is sensitive to initial value and the AAM method has the problem that detection error is increased when an input image is quite different from training data. Especially, the algorithm shows high accuracy in case of closed lips but the detection error is increased in case of opened lips and deformed lips according to the facial expression of user. To solve these problems, we propose the improved AAM algorithm using lip feature points which is extracted based on a new lip detection algorithm. In this paper, we select a searching region based on the face feature points which are detected by AAM algorithm. And lip corner points are extracted by using Canny edge detection and histogram projection method in the selected searching region. Then, lip region is accurately detected by combining color and edge information of lip in the searching region which is adjusted based on the position of the detected lip corners. Based on that, the accuracy and processing speed of lip detection are improved. Experimental results showed that the RMS(Root Mean Square) error of the proposed method was reduced as much as 4.21 pixels compared to that only using AAM algorithm.

• Journal of the Korean Chemical Society
• /
• v.45 no.2
• /
• pp.162-176
• /
• 2001

This study was covered with the analysis of five kinds of the middle school science text books(chemistry part) based on the oth curriculum. Particularly, inquiry activities part was analyzed by the three dimension framework which consists of inquiry content dimension, inquiry process dimension and inquiry context dimension and the results are as follows;1. In the analysis of the contents in the middle school science textbooks(chemistry part), the average number of total pages was 197.6. The illustration and picture were contained 0.66 in number per a page, and the average number of further readings was 5.8.2. In the analysis of the inquiry content dimension of inquiry activities, the total number of themes in five kinds of textbooks was 222. And the number of imquiry activities in five kinds of textbooks was siverse : A textbook had 51, B texbook 49, C textbook 37 and E textbook 35.3. For the analysis of inquiry process dimension. it follows in the order of 'interpreing data and formulating generalizations (42.4%)','observation and measuring (38.1%)','seeing a problem and seeking ways to solve it (7.8%)' and 'building, testing and revising the theoretical model(11.7%)'.4. As for the analysis of the inquiry context dimension, the scientific context occupied 94.2%, the individual context 0.4%, the social context 2.7% and the technical context 2.7%. It shows that the proportion of STS(Science-Technology-Society) related contents in inquiry activities was only 5.8%.

• Korean Journal of Remote Sensing
• /
• v.33 no.5_1
• /
• pp.455-467
• /
• 2017

When performing target detection using hyperspectral imagery, a feature extraction process is necessary to solve the problem of redundancy of adjacent spectral bands and the problem of a large amount of calculation due to high dimensional data. This study proposes a new band selection method using the $L_{2,1}$-norm regression model to apply the feature selection technique in the machine learning field to the hyperspectral band selection. In order to analyze the performance of the proposed band selection technique, we collected the hyperspectral imagery and these were used to analyze the performance of target detection with band selection. The Adaptive Cosine Estimator (ACE) detection performance is maintained or improved when the number of bands is reduced from 164 to about 30 to 40 bands in the 350 nm to 2500 nm wavelength band. Experimental results show that the proposed band selection technique extracts bands that are effective for detection in hyperspectral images and can reduce the size of the data without reducing the performance, which can help improve the processing speed of real-time target detection system in the future.

• The Journal of the Korea Contents Association
• /
• v.10 no.8
• /
• pp.327-340
• /
• 2010

Since South Korea has problems existing in the respects of educational resources, environment, policies, etc., local universities of South Korea have been in an inferior position in the competition of overseas student enrollment. With increasingly deepening of cultural exchange between China and South Korea, due to geographic close relationship between the two countries, a great number of South Korean students go to China for study; also, Chinese students have increasingly become the important target of recruitment by all local universities in South Korea. In recent years, with rapid increase in the number of Chinese students in South Korea, the overseas student education system of South Korea has been progressed continuously. Compared with educational advantages of universities in European, American developed countries, and the capital region of South Korea, however there still exists a lot of problems in Chinese student recruitment by local universities. The major findings of the study can be summarized as follows. There exist not only some advantages but also some problems to be addressed urgently in Chinese student enrollment by South Korean local universities. How to build a distinctive recruitment environment suitable for Chinese students; and, how to develop a complete "one-stop" educational system suitable for study; and, how to make a strategy of development for Chinese students; and, how to perfect set up the overseas students' education management system; which are the remaining tasks to be solved for. To achieve a win-win for both overseas students and South Korean local universities, it is very important and urgent works to do for administrators of Chinese students in all local universities of South Korea.

• Journal of Internet Computing and Services
• /
• v.12 no.5
• /
• pp.47-61
• /
• 2011

Recently, the importance of digital forensic has been magnified because of the dramatic increase of cyber crimes and the increasing complexity of the investigation of target systems such as PCs, servers, and database systems. Moreover, some systems have to be investigated with live forensic techniques. However, even though live forensic techniques have been improved, they are still vulnerable to anti-forensic activities when the target systems are remotely accessible by criminals or their accomplices. To solve this problem, we first suggest a layer-based model and the anti-forensic scenarios which can actually be applicable to each layer. Our suggested model, the Anti-Forensic Activites layer-based model, has 5 layers - the physical layer, network layer, OS layer, database application layer and data layer. Each layer has possible anti-forensic scenarios with detailed commands. Second, we propose a fuzzy expert system for effectively detecting anti-forensic activities. Some anti-forensic activities are hardly distinguished from normal activities. So, we use fuzzy logic for handling ambiguous data. We make rule sets with extracted commands and their arguments from pre-defined scenarios and the fuzzy expert system learns the rule sets. With this system, we can detect anti-forensic activities in real time when performing live forensic.

• Korean Journal of Remote Sensing
• /
• v.36 no.6_2
• /
• pp.1579-1590
• /
• 2020

In this paper, spatially-adaptive denormalization (SPADE) based U-Net is proposed to detect changes by using high-resolution satellite images. The proposed network is to preserve spatial information using SPADE. Change detection methods using high-resolution satellite images can be used to resolve various urban problems such as city planning and forecasting. For using pixel-based change detection, which is a conventional method such as Iteratively Reweighted-Multivariate Alteration Detection (IR-MAD), unchanged areas will be detected as changing areas because changes in pixels are sensitive to the state of the environment such as seasonal changes between images. Therefore, in this paper, to precisely detect the changes of the objects that consist of the city in time-series satellite images, the semantic spatial objects that consist of the city are defined, extracted through deep learning based image segmentation, and then analyzed the changes between areas to carry out change detection. The semantic objects for analyzing changes were defined as six classes: building, road, farmland, vinyl house, forest area, and waterside area. Each network model learned with KOMPSAT-3A satellite images performs a change detection for the time-series KOMPSAT-3 satellite images. For objective assessments for change detection, we use F1-score, kappa. We found that the proposed method gives a better performance compared to U-Net and UNet++ by achieving an average F1-score of 0.77, kappa of 77.29.

• Journal of the Korean Society of Earth Science Education
• /
• v.15 no.1
• /
• pp.62-75
• /
• 2022

In this study, experiential STEAM program using 3D printer was produced focusing on the content elements of 'solar' in the 2015 revised science curriculum, and in order to find out the effectiveness of the STEAM program, analyzed creative problem solving, STEAM attitude, and STEAM satisfaction by applying it to two middle school 77 students simple random sampled. The results of this study are as follows. First, a solar tactile model was produced using a 3D printer, and a program was developed to enable students to actively learn experience-oriented activities through visual impairment experiences. Second, in the response sample t-test by the difference in pre- and post-score of STEAM attitude tests, significant statistical test results were shown in 'interest', 'consideration', 'self-concept', 'self-efficacy', and 'science and engineering career choice' sub-factors except 'consideration' and 'usefulness / value recognition' sub-factors (p<.05). Third,, the STEAM satisfaction test conducted after the application of the 3D printer-based STEAM program showed that the average value range of sub-factors were 3.66~3.97, which improved students' understanding and interest in science subjects through the 3D printer-based STEAM program.