• Proceedings of the KOSOMBE Conference
• /
• v.1991 no.05
• /
• pp.65-70
• /
• 1991

A hybrid model which uses a probabilistic model and a MLP(multi layer perceptron) model for pattern recognition of EMG(electromyogram) signals is proposed in this paper. MLP model has problems which do not guarantee global minima of error due to learning method and have different approximation grade to bayesian probabilities due to different amounts and quality of training data, the number of hidden layers and hidden nodes, etc. Especially in the case of new test data which exclude design samples, the latter problem produces quite different results. The error probability of probabilistic model is closely related to the estimation error of the parameters used in the model and fidelity of assumtion. Generally, it is impossible to introduce the bayesian classifier to the probabilistic model of EMG signals because of unknown priori probabilities and is estimated by MLE(maximum likelihood estimate). In this paper we propose the method which get the MAP(maximum a posteriori probability) in the probabilistic model by estimating the priori probability distribution which minimize the error probability using the MLP. This method minimize the error probability of the probabilistic model as long as the realization of the MLP is optimal and approximate the minimum of error probability of each class of both models selectively. Alocating the reference coordinate of EMG signal to the outside of the body make it easy to suit to the applications which it is difficult to define and seperate using internal body coordinate. Simulation results show the benefit of the proposed model compared to use the MLP and the probabilistic model seperately.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.5
• /
• pp.147-153
• /
• 2018

Beacon enables one to measure distance indoors based on low-power Bluetooth low energy (BLE) technology, while GPS (Global Positioning System) only can be used outdoors. In measuring indoor distance using Beacon, RSSI (Received Signal Strength Indication) is considered as the one of the key factors, however, it is influenced by various environmental factors so that it causes the huge gap between the estimated distance and the real. In order to handle this issue, we propose the adaptive ANN (Artificial Neural Network) based approach to measuring the exact distance using Beacon. First, we has carried out the preprocessing of the RSSI signals by applying the extended Kalman filter and the signal stabilization filter into decreasing the noise. Then, we suggest the multi-layered ANNs, each of which layer is learned by specific training data sets. The results showed an average error of 0.67m, a precision of 0.78.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.7
• /
• pp.1807-1822
• /
• 2023

Innovation and rapid increased functionality in user friendly smartphones has encouraged shutterbugs to have picturesque image macros while in work environment or during travel. Formal signboards are placed with marketing objectives and are enriched with text for attracting people. Extracting and recognition of the text from natural images is an emerging research issue and needs consideration. When compared to conventional optical character recognition (OCR), the complex background, implicit noise, lighting, and orientation of these scenic text photos make this problem more difficult. Arabic language text scene extraction and recognition adds a number of complications and difficulties. The method described in this paper uses a two-phase methodology to extract Arabic text and word boundaries awareness from scenic images with varying text orientations. The first stage uses a convolution autoencoder, and the second uses Arabic Character Segmentation (ACS), which is followed by traditional two-layer neural networks for recognition. This study presents the way that how can an Arabic training and synthetic dataset be created for exemplify the superimposed text in different scene images. For this purpose a dataset of size 10K of cropped images has been created in the detection phase wherein Arabic text was found and 127k Arabic character dataset for the recognition phase. The phase-1 labels were generated from an Arabic corpus of quotes and sentences, which consists of 15kquotes and sentences. This study ensures that Arabic Word Awareness Region Detection (AWARD) approach with high flexibility in identifying complex Arabic text scene images, such as texts that are arbitrarily oriented, curved, or deformed, is used to detect these texts. Our research after experimentations shows that the system has a 91.8% word segmentation accuracy and a 94.2% character recognition accuracy. We believe in the future that the researchers will excel in the field of image processing while treating text images to improve or reduce noise by processing scene images in any language by enhancing the functionality of VGG-16 based model using Neural Networks.

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.175-197
• /
• 2021

Recently, with the development of deep learning technology, research on unstructured data analysis is being actively conducted, and it is showing remarkable results in various fields such as classification, summary, and generation. Among various text analysis fields, text classification is the most widely used technology in academia and industry. Text classification includes binary class classification with one label among two classes, multi-class classification with one label among several classes, and multi-label classification with multiple labels among several classes. In particular, multi-label classification requires a different training method from binary class classification and multi-class classification because of the characteristic of having multiple labels. In addition, since the number of labels to be predicted increases as the number of labels and classes increases, there is a limitation in that performance improvement is difficult due to an increase in prediction difficulty. To overcome these limitations, (i) compressing the initially given high-dimensional label space into a low-dimensional latent label space, (ii) after performing training to predict the compressed label, (iii) restoring the predicted label to the high-dimensional original label space, research on label embedding is being actively conducted. Typical label embedding techniques include Principal Label Space Transformation (PLST), Multi-Label Classification via Boolean Matrix Decomposition (MLC-BMaD), and Bayesian Multi-Label Compressed Sensing (BML-CS). However, since these techniques consider only the linear relationship between labels or compress the labels by random transformation, it is difficult to understand the non-linear relationship between labels, so there is a limitation in that it is not possible to create a latent label space sufficiently containing the information of the original label. Recently, there have been increasing attempts to improve performance by applying deep learning technology to label embedding. Label embedding using an autoencoder, a deep learning model that is effective for data compression and restoration, is representative. However, the traditional autoencoder-based label embedding has a limitation in that a large amount of information loss occurs when compressing a high-dimensional label space having a myriad of classes into a low-dimensional latent label space. This can be found in the gradient loss problem that occurs in the backpropagation process of learning. To solve this problem, skip connection was devised, and by adding the input of the layer to the output to prevent gradient loss during backpropagation, efficient learning is possible even when the layer is deep. Skip connection is mainly used for image feature extraction in convolutional neural networks, but studies using skip connection in autoencoder or label embedding process are still lacking. Therefore, in this study, we propose an autoencoder-based label embedding methodology in which skip connections are added to each of the encoder and decoder to form a low-dimensional latent label space that reflects the information of the high-dimensional label space well. In addition, the proposed methodology was applied to actual paper keywords to derive the high-dimensional keyword label space and the low-dimensional latent label space. Using this, we conducted an experiment to predict the compressed keyword vector existing in the latent label space from the paper abstract and to evaluate the multi-label classification by restoring the predicted keyword vector back to the original label space. As a result, the accuracy, precision, recall, and F1 score used as performance indicators showed far superior performance in multi-label classification based on the proposed methodology compared to traditional multi-label classification methods. This can be seen that the low-dimensional latent label space derived through the proposed methodology well reflected the information of the high-dimensional label space, which ultimately led to the improvement of the performance of the multi-label classification itself. In addition, the utility of the proposed methodology was identified by comparing the performance of the proposed methodology according to the domain characteristics and the number of dimensions of the latent label space.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.29 no.1
• /
• pp.1-10
• /
• 1993

In order to investigate the environmental properties of set net grounds located in the coastal waters of Yeosu, oceanographic observations on the fishing grounds were carried out by the training ship of Yeosu Fisheries University from Jun. 1988 to Dec. 1990. The resultes obtained are summarized as follows; 1) The water mass in the fishing grounds were divided into the inner water (29.50-31.00$\textperthousand$), the mixed water (31.10-32.70$\textperthousand$) and the offshore water (32.70-34.30$\textperthousand$) according to the distribution of salinity from T-S diagram plotted all salinity data observed from Jun. 1988 to Dec. 1990. In spring the mixing water prevailed and in summer the inner and mixing water. But in autumn and winter the mixing and offshore waters prevailed. 2) The inner water which was formed by land water from the river of Somjin and the precipitation in the Yeosu district flowed southerly along the coast of Dolsando and spread south-easterly in the vicinity of Kumodo. The inner water and offshore water which supplied from the vicinity of Sorido and Yokchido formed the thermal front and halofront. 3) As the mixing water flowing from the western sea of Cheju to the southern coast of korea was low in temperature, the water mass of low temperature which appeared at the offshore bottom of Sorido in summer was considered not to be the Tsushima warm current. 4) As vertical mixing was made frequently in spring, autumn and winter, the differences in temperature and salinity between surface and bottom was respectively small. In summer, however, the mixing was not made because of the inner water expanded offshore through the space between surface and 10m layer and so a thermocline of $2.0^{\circ}C$/10m and halocline of 4.0$\textperthousand$/10m respectively in vertical gradient was formed. 5) In the vicinity of Dolsando and Kum a water low in salinity prevailed, but in the vicinity of Namhaedo and YoKchido the reverse took place. The inner and mixing waters formed at these arease was limited to the observation area not to spread widely.

• Journal of architectural history
• /
• v.9 no.2 s.23
• /
• pp.85-100
• /
• 2000

This paper is a study of the Beaux-Arts discipline of architecture, as it was established during the late nineteenth century in America. It focuses on trio particular modes of vision and representation that were at the heart of the discipline. The paper argues that Beaux Arts vision was centered on what may be called 'planar vision'; a mode of seeing through which the multiple aspects of the architectural design imbedded in the plan were read and re-interpreted. Similarly Beaux-Arts training in drawing required its student to draw within the multiple layers of historical traces; the new design being in effect a new layer placed on often unseen traces of monumental precedent. The theoretical basis of this practice was not based on history but on the concept of composition. Composition, in the French tradition was regarded more a matter of practice than theory. The Anglo-American discourse on composition, on the other hand, formed a body of theoretical literature based on formalist assumptions. There was, however, a fundamental gap between these formalist theories of composition and the 'layered' modes of vision and drawing involved in the design process. This practice leaned more on the modern romantic notion of 'intuition' for its theoretical basis, once again forming an immanent conflict with the mimetic practice of classical and historical architecture. The paper draws a picture of a discipline centered on a 'theory of the plan,' a potentially modern discipline integrated with classical forms and details. It was clearly effective as a practice. However, structured by conflicts between theory and practice, history and form, mimesis and intuition, the Beaux-Arts was unable to defend itself at the philosophical and theoretical level the modernists engaged their attacks on this system. At the same time, the paper poses the question of how different modern architecture is from this system. Is not the 'theory of plan,' in its many transformations and guises, still the central discipline of twentieth century modern architecture, and is it not structured by basically the same kind of conflicts and paradox that were immanent to the Beaux-Arts system.

• Journal of the Korea Convergence Society
• /
• v.7 no.5
• /
• pp.23-28
• /
• 2016

In this paper, smart devices using a 4 to 6 years old, infants and parents to a user layer of training and with games or studying infants by integrating them. Not losing interest and concentration, Mobile infants to learning to learn Korean, English, the estimated budget. design the app tainer. For parents of infants and a variety of media concentration and learning Korean, English, English word cards that can increase the interest of the design and, to find a picture Memory game, had provided games are various kinds such as learning to do puzzles. Also, infants and tries to help study a visual learning and auditory learning at the same time can be achieved by mothers of children is much more conveniently. Learning to guide implementation to maximize the availability, convenience and mobility.

• Journal of Forest and Environmental Science
• /
• v.32 no.1
• /
• pp.74-81
• /
• 2016

Classified into sub-tree of the growth-form, Acer pseudosieboldianum and Carpinus cordata hardly reach the uppermost canopy with limited height growth potentiality in the natural deciduous forest. However, the two species usually dominate lower stories of the forest, playing an important role in overall ecological characteristics in the forest. Based on the vegetation data of 106 $20m{\times}20m$ sample plots in Mt. Jumbong area, this study was carried out to evaluate ecological constitution by several quantitative analysis so as to understand the mechanism of the natural deciduous forest. The results indicated that individuals of A. pseudosieboldianum and C. cordata were absent or few in overstory, but emerged the most dominant species in midstory and understory, providing adequate proof of the ecological importance. The comparison of indices of succession between presented and predicted values in midstory did not make much difference, suggesting that the species composition would not change much and come close to steady state in midstory and understory. The pair combination of species association noted that A. pseudosieboldianum had significant positive association with C. cordata, Quercus mongolica, and Tilia amurensis had significant positive association with A. pseudosieboldianum, A. pictum subsp. mono and Fraxinus mandshurica but negative association with F. rhynchophylla. Being compared with other major canopy tree species in the study forest, the target species of A. pseudosieboldianum and C. cordata had strong regeneration strategies, partially characterized by large number of saplings and pole sized trees and high ratio of live crown, which indicated high shade tolerance to survive in the limited amount of light under the canopy. Even though A. pseudosieboldianum and C. cordata do not reach and occupy the canopy layer mainly due to the inherent growth form, they would have highest competitive potentiality to prosper and dominate in the midstory of the natural deciduous forest.

• Phonetics and Speech Sciences
• /
• v.11 no.4
• /
• pp.89-96
• /
• 2019

In this paper, we propose a deep learning architecture that can effectively detect speech segmentation in broadcast contents. We also propose a multi-scale time-dilated layer for learning the temporal changes of feature vectors. We implement several comparison models to verify the performance of proposed model and calculated the frame-by-frame F-score, precision, and recall. Both the proposed model and the comparison model are trained with the same training data, and we train the model using 32 hours of Korean broadcast data which is composed of various genres (drama, news, documentary, and so on). Our proposed model shows the best performance with F-score 91.7% in Korean broadcast data. The British and Spanish broadcast data also show the highest performance with F-score 87.9% and 92.6%. As a result, our proposed model can contribute to the improvement of performance of speech detection by learning the temporal changes of the feature vectors.

• Structural Monitoring and Maintenance
• /
• v.1 no.3
• /
• pp.249-271
• /
• 2014

This paper presents a unique study of Structural Health Monitoring (SHM) for the maintenance decision making about a real life movable bridge. The mechanical components of movable bridges are maintained on a scheduled basis. However, it is desired to have a condition-based maintenance by taking advantage of SHM. The main objective is to track the operation of a gearbox and a rack-pinion/open gear assembly, which are critical parts of bascule type movable bridges. Maintenance needs that may lead to major damage to these components needs to be identified and diagnosed timely since an early detection of faults may help avoid unexpected bridge closures or costly repairs. The fault prediction of the gearbox and rack-pinion/open gear is carried out using two types of Artificial Neural Networks (ANNs): 1) Multi-Layer Perceptron Neural Networks (MLP-NNs) and 2) Fuzzy Neural Networks (FNNs). Monitoring data is collected during regular opening and closing of the bridge as well as during artificially induced reversible damage conditions. Several statistical parameters are extracted from the time-domain vibration signals as characteristic features to be fed to the ANNs for constructing the MLP-NNs and FNNs independently. The required training and testing sets are obtained by processing the acceleration data for both damaged and undamaged condition of the aforementioned mechanical components. The performances of the developed ANNs are first evaluated using unseen test sets. Second, the selected networks are used for long-term condition evaluation of the rack-pinion/open gear of the movable bridge. It is shown that the vibration monitoring data with selected statistical parameters and particular network architectures give successful results to predict the undamaged and damaged condition of the bridge. It is also observed that the MLP-NNs performed better than the FNNs in the presented case. The successful results indicate that ANNs are promising tools for maintenance monitoring of movable bridge components and it is also shown that the ANN results can be employed in simple approach for day-to-day operation and maintenance of movable bridges.