• Korean Journal of Remote Sensing
• /
• v.29 no.5
• /
• pp.569-579
• /
• 2013

In this study, land-cover classification was implemented using features extracted from multi-temporal SAR data through 3D wavelet transform and the applicability of the 3D wavelet transform as a feature extraction approach was evaluated. The feature extraction stage based on 3D wavelet transform was first carried out before the classification and the extracted features were used as input for land-cover classification. For a comparison purpose, original image data without the feature extraction stage and Principal Component Analysis (PCA) based features were also classified. Multi-temporal Radarsat-1 data acquired at Dangjin, Korea was used for this experiment and five land-cover classes including paddy fields, dry fields, forest, water, and built up areas were considered for classification. According to the discrimination capability analysis, the characteristics of dry field and forest were similar, so it was very difficult to distinguish these two classes. When using wavelet-based features, classification accuracy was generally improved except built-up class. Especially the improvement of accuracy for dry field and forest classes was achieved. This improvement may be attributed to the wavelet transform procedure decomposing multi-temporal data not only temporally but also spatially. This experiment result shows that 3D wavelet transform would be an effective tool for feature extraction from multi-temporal data although this procedure should be tested to other sensors or other areas through extensive experiments.

• Journal of the Korean Geographical Society
• /
• v.38 no.5
• /
• pp.667-685
• /
• 2003

To address the requirements of gap analysis for species protection, as well as the needs of state and federal agencies for detailed digital land cover, a 43-class map at the vegetation alliance level was created for the state of Kansas using multi-temporal Thematic Mapper imagery. The mapping approach included the use of three-date multi-seasonal imagery, a two-stage classification approach that first masked out cropland areas using unsupervised classification and then mapped natural vegetation with supervised classification, visualization techniques utilizing a map of small multiples and field experts, and extensive use of ancillary data in post-hoc processing. Accuracy assessment was conducted at three levels of generalization (Anderson Level I, vegetation formation, and vegetation alliance) and three cross-tabulation approaches. Overall accuracy ranged from 51.7% to 89.4%, depending on level of generalization, while accuracy figures for individual alliance classes varied by area covered and level of sampling.

• Convergence Security Journal
• /
• v.20 no.4
• /
• pp.81-90
• /
• 2020

In the real world, new types of attacks or variants are constantly emerging, but attack traffic classification models developed through artificial neural networks and supervised learning do not properly detect new types of attacks that have not been trained. Most of the previous studies overlooked this problem and focused only on improving the structure of their artificial neural networks. As a result, a number of new attacks were frequently classified as normal traffic, and attack traffic classification performance was severly degraded. On the other hand, the softmax function, which outputs the probability that each class is correctly classified in the multi-class classification as a result, also has a significant impact on the classification performance because it fails to calculate the softmax score properly for a new type of attack traffic that has not been trained. In this paper, based on this characteristic of softmax function, we propose an efficient method to improve the classification performance against new types of attacks by classifying traffic with a probability below a certain level as attacks, and demonstrate the efficiency of our approach through experiments.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.1 no.1
• /
• pp.45-48
• /
• 1978

If two inspectors classify items in a lot into m classes, it is possible that each of them makes wrong classification in some cases, thus causing bias. Expressions have been obtained for the limits of this bias in estimating the proportion of the different classes. From the results of the classification they obtained limit for the estimates of Proportions have been worked out, based on assumption regarding the magnitudes of probabilities of misclassification. Now we suppose that $P_{ti}{\;}(t=1.2)$ is the probability that t the inspector classifies correctly an item in class $A_i$ and $q_{tji}$ is the probability that he misclassifies in $A_j$ an item actually belonging to $A_i$, therefor, $P_{ti}+ \sum\limits_{j{\neq}i}q_{tji}=1$ An estimate for the proportion $P_k$ of the class $A_k$ in the lot would be $\hat{P}_k=r_{kk}+(\frac{1}{2})\sum\limits_{j{\neq}k}r_{kj}+r_{jk}$ The % Bias in proportion $\hat{P}_k$ is $\frac{E(\hat{P}_k)-P_k}{P_k}{\times}100$

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.11
• /
• pp.563-568
• /
• 2016

The word sense disambiguation problem is to find the correct sense of an ambiguous word having multiple senses in a dictionary in a sentence. We regard this problem as a multi-class classification problem and classify the ambiguous word by using Support Vector Machines. Context words of the ambiguous word, which are extracted from Sejong sense tagged corpus, are represented to two kinds of vector space. One vector space is composed of context words vectors having binary weights. The other vector space has vectors where the context words are mapped by word embedding model. After experiments, we acquired accuracy of 87.0% with context word vectors and 86.0% with word embedding model.

• The KIPS Transactions:PartB
• /
• v.13B no.1 s.104
• /
• pp.1-6
• /
• 2006

The prediction of protein function basically make use of a protein-protein interaction map based on the concept of guilt-by-association. The method however cannot determine the functions of proteins in case that the target protein does not interact with proteins with known functions directly. This paper studies protein function prediction considering the given problem as a K-class classification problem and proposes a predictive approach utilizing a modular neural network. The proposed method uses interaction data and protein related attributes as well. The experimental results demonstrate that the proposed approach can predict the functional roles of Yeast proteins whose interaction knowledge is not known and shows better performance than the graph-based models that use protein interaction data.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.2
• /
• pp.9-17
• /
• 2023

In this paper, we propose a one-class vibration anomaly detection system for bearing defect diagnosis. In order to reduce the economic and time loss caused by bearing failure, an accurate defect diagnosis system is essential, and deep learning-based defect diagnosis systems are widely studied to solve the problem. However, it is difficult to obtain abnormal data in the actual data collection environment for deep learning learning, which causes data bias. Therefore, a one-class classification method using only normal data is used. As a general method, the characteristics of vibration data are extracted by learning the compression and restoration process through AutoEncoder. Anomaly detection is performed by learning a one-class classifier with the extracted features. However, this method cannot efficiently extract the characteristics of the vibration data because it does not consider the frequency characteristics of the vibration data. To solve this problem, we propose an AutoEncoder model that considers the frequency characteristics of vibration data. As for classification performance, accuracy 0.910, precision 1.0, recall 0.820, and f1-score 0.901 were obtained. The network design considering the vibration characteristics confirmed better performance than existing methods.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.305-307
• /
• 2021

Alzheimer's Disease (AD) is a cognitive disorder characterized by memory impairment that can be assessed at early stages based on administering clinical tests. However, the AD pathophysiological mechanism is still poorly understood due to the difficulty of distinguishing different levels of AD severity, even using a variety of brain modalities. Therefore, in this study, we present a hybrid EEG-fNIRS modalities to compensate for each other's weaknesses with the help of Machine Learning (ML) techniques for classifying four subject groups, including healthy controls (HC) and three distinguishable groups of AD levels. A concurrent EEF-fNIRS setup was used to record the data from 41 subjects during Oddball and 1-back tasks. We employed both a traditional neural network (NN) and a CNN-LSTM hybrid model for fNIRS and EEG, respectively. The final prediction was then obtained by using majority voting of those models. Classification results indicated that the hybrid EEG-fNIRS feature set achieved a higher accuracy (71.4%) by combining their complementary properties, compared to using EEG (67.9%) or fNIRS alone (68.9%). These findings demonstrate the potential of an EEG-fNIRS hybridization technique coupled with ML-based approaches for further AD studies.

• Investigative Magnetic Resonance Imaging
• /
• v.3 no.2
• /
• pp.179-187
• /
• 1999

Purpose : This paper introduces a computationally inexpensive context-dependent classification of multi-echo MRI with Bayes compound decision model. In order to produce accurate region segmentation especially in homogeneous area and along boundaries of the regions, we propose a classification method that uses contextual information of local enighborhood system in the image. Material and Methods : The performance of the context free classifier over a statistically heterogeneous image can be improved if the local stationary regions in the image are disassociated from each other through the mechanism of the interaction parameters defined at he local neighborhood level. In order to improve the classification accuracy, we use the contextual information which resolves ambiguities in the class assignment of a pattern based on the labels of the neighboring patterns in classifying the image. Since the data immediately surrounding a given pixel is intimately associated with this given pixel., then if the true nature of the surrounding pixel is known this can be used to extract the true nature of the given pixel. The proposed context-dependent compound decision model uses the compound Bayes decision rule with the contextual information. As for the contextual information in the model, the directional transition probabilities estimated from the local neighborhood system are used for the interaction parameters. Results : The context-dependent classification paradigm with compound Bayesian model for multi-echo MR images is developed. Compared to context free classification which does not consider contextual information, context-dependent classifier show improved classification results especially in homogeneous and along boundaries of regions since contextual information is used during the classification. Conclusion : We introduce a new paradigm to classify multi-echo MRI using clustering analysis and Bayesian compound decision model to improve the classification results.

• Korean Journal of Remote Sensing
• /
• v.37 no.3
• /
• pp.409-418
• /
• 2021

The land cover map is a very important data that is used as a basis for decision-making for land policy and environmental policy. The land cover map is mapped using remote sensing data, and the classification results may vary depending on the acquisition time of the data used even for the same area. In this study, to overcome the classification accuracy limit of single-period data, multi-series satellite images were used to learn the difference in the spectral reflectance characteristics of the land surface according to seasons on a U-Net model, one of the deep learning algorithms, to improve classification accuracy. In addition, the degree of improvement in classification accuracy is compared by comparing the accuracy of single-period data. Seoul, which consists of various land covers including 30% of green space and the Han River within the area, was set as the research target and quarterly Sentinel-2 satellite images for 2020 were aquired. The U-Net model was trained using the sub-class land cover map mapped by the Korean Ministry of Environment. As a result of learning and classifying the model into single-period, double-series, triple-series, and quadruple-series through the learned U-Net model, it showed an accuracy of 81%, 82% and 79%, which exceeds the standard for securing land cover classification accuracy of 75%, except for a single-period. Through this, it was confirmed that classification accuracy can be improved through multi-series classification.