• IE interfaces
• /
• v.20 no.1
• /
• pp.41-48
• /
• 2007

The main purpose of this paper is cost reduction in absurd pilot positive expense and human accident prevention which is caused by in the pilot selection process. We use classification models such as logistic regression, decision tree, and neural network based on aptitude test results of 505 ROK Air Force applicants in 2001~2004. First, we determine the reliability and propriety against the aptitude test system which has been improved. Based on this conference flight simulator test item was compared to the new aptitude test item in order to make additional yes or no decision from different models in terms of classification accuracy, ROC and Response Threshold side. Decision tree was selected as the most efficient for each sequential flight training result and the last flight training results predict excellent. Therefore, we propose that the standard of pilot selection be adopted by the decision tree and it presents in the aptitude test item which is new a conference flight simulator test.

• Smart Structures and Systems
• /
• v.33 no.1
• /
• pp.55-64
• /
• 2024

This paper proposes a fusion imaging-based coating-defect classification method for steel structures that uses zero-shot learning. In the proposed method, a halogen lamp generates heat energy on the coating surface of a steel structure, and the resulting heat responses are measured by an infrared (IR) camera, while photos of the coating surface are captured by a charge-coupled device (CCD) camera. The measured heat responses and visual images are then analyzed using zero-shot learning to classify the coating defects, and the estimated coating defects are visualized throughout the inspection surface of the steel structure. In contrast to older approaches to coating-defect classification that relied on visual inspection and were limited to surface defects, and older artificial neural network (ANN)-based methods that required large amounts of data for training and validation, the proposed method accurately classifies both internal and external defects and can classify coating defects for unobserved classes that are not included in the training. Additionally, the proposed model easily learns about additional classifying conditions, making it simple to add classes for problems of interest and field application. Based on the results of validation via field testing, the defect-type classification performance is improved 22.7% of accuracy by fusing visual and thermal imaging compared to using only a visual dataset. Furthermore, the classification accuracy of the proposed method on a test dataset with only trained classes is validated to be 100%. With word-embedding vectors for the labels of untrained classes, the classification accuracy of the proposed method is 86.4%.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.2226-2229
• /
• 2003

A new method is presented to construct a classifier. This was accomplished by combining a generalized regression neural network (GRNN) and a genetic algorithm (GA). The classifier constructed in this way is referred to as a GA-GRNN. The GA played a role of controlling training factors simultaneously. In GA optimization, neuron spreads were represented in a chromosome. The proposed optimization method was applied to a data set, consisted of 4 different promoter sequences. The training and test data were composed of 115 and 58 sequence patterns, respectively. The range of neuron spreads was experimentally varied from 0.4 to 1.4 with an increment of 0.1. The GA-GRNN was compared to a conventional GRNN. The classifier performance was investigated in terms of the classification sensitivity and prediction accuracy. The GA-GRNN significantly improved the total classification sensitivity compared to the conventional GRNN. Also, the GA-GRNN demonstrated an improvement of about 10.1% in the total prediction accuracy. As a result, the proposed GA-GRNN illustrated improved classification sensitivity and prediction accuracy over the conventional GRNN.

• Journal of KIISE:Software and Applications
• /
• v.33 no.5
• /
• pp.489-498
• /
• 2006

In this paper, we suggested a vehicle classification algorithm using pattern recognition method. At present, Inductive Loop Detector is rarely used for vehicle classification because of its low accuracy. To improve the accuracy, we suggest a new algorithm for Loop Detector using neural networks. In the developed algorithm, the inputs to the neural networks are the variation rate of frequency and occupancy-time. The output is classified vehicles. The developed algorithm was assessed at test sites and the recognition rate was 91.3percent. The results verified that the proposed algorithm improves the vehicle classification accuracy compared to the conventional method based on Loop Detector.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.24 no.3
• /
• pp.1-9
• /
• 2021

In this study, we evaluated the accuracy of deep learning-based tree species classification model trained by using high-resolution images. We selected five species classed, i.e., pine, birch, larch, korean pine, mongolian oak for classification. We created 5,000 datasets using high-resolution orthophoto and forest type map. CNN deep learning model is used to tree species classification. We divided training data, verification data, and test data by a 5:3:2 ratio of the datasets and used it for the learning and evaluation of the model. The overall accuracy of the model was 89%. The accuracy of each species were pine 95%, birch 89%, larch 80%, korean pine 86% and mongolian oak 98%.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.1
• /
• pp.117-125
• /
• 2019

Deep Learning, which is being used in various fields recently, is being threatened with Adversarial Attack. In this paper, we experimentally verify that the classification accuracy is lowered by adversarial samples generated by malicious attackers in image classification models. We used MNIST dataset and measured the detection accuracy by injecting adversarial samples into the Autoencoder classification model and the CNN (Convolution neural network) classification model, which are created using the Tensorflow library and the Pytorch library. Adversarial samples were generated by transforming MNIST test dataset with JSMA(Jacobian-based Saliency Map Attack) and FGSM(Fast Gradient Sign Method). When injected into the classification model, detection accuracy decreased by at least 21.82% up to 39.08%.

• Journal of Korean Society for Geospatial Information Science
• /
• v.22 no.1
• /
• pp.81-88
• /
• 2014

The researcher of this study applied PCA on aerial hyper-spectral sensor and selectively combined bands which contain high amount of information, creating five types of PCA images. By applying Spectral Angle Mapping-supervised classification technique on each type of image, classification process was carried out and accuracy was evaluated. The test result showed that the amount of information contained in the first band of PCA-transformation image was 76.74% and the second accumulated band contained 98.40%, suggesting that most of information were contained in the first and the second PCA components. Quantitative classification accuracy evaluation of each type of image showed that total accuracy, producer's accuracy and user's accuracy had similar patterns. What drew the researcher's attention was the fact that the first and the second bands of the PCA-transformation image had the highest accuracy according to the classification accuracy although it was believed that more than four bands of PCA-transformation image should be contained in order to secure accuracy when doing the qualitative classification accuracy.

• Proceedings of the KSRS Conference
• /
• 2008.10a
• /
• pp.169-172
• /
• 2008

Land-use or land-cover classification of satellite images is one of the important tasks in remote sensing application and many researchers have been tried to enhance classification accuracy. Previous studies show that the classification technique based on wavelet transform is more effective than that of traditional techniques based on original pixel values, especially in complicated imagery. Various wavelets can be used in wavelet transform. Wavelets are used as basis functions in representing other functions, like sinusoidal function in Fourier analysis. In these days, some basis functions such as Haar, Daubechies, Coiflets and Symlets are mainly used in 2D image processing. Selecting adequate wavelet is very important because different results could be obtained according to the type of basis function in classification. However, it is not easy to choose the basis function which is effective to improve classification accuracy. In this study, we computed the wavelet coefficients of satellite image using 10 different basis functions, and then classified test image. After evaluating classification results, we tried to ascertain which basis function is the most effective for image classification. We also tried to see if the optimum basis function is decided by energy parameter before classifying the image using all basis function. The energy parameter of signal is the sum of the squares of wavelet coefficients. The energy parameter is calculated by sub-bands after the wavelet decomposition and the energy parameter of each sub-band can be a favorable feature of texture. The decision of optimum basis function using energy parameter in the wavelet based image classification is expected to be helpful for saving time and improving classification accuracy effectively.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.47 no.2
• /
• pp.1-9
• /
• 2024

New motor development requires high-speed load testing using dynamo equipment to calculate the efficiency of the motor. Abnormal noise and vibration may occur in the test equipment rotating at high speed due to misalignment of the connecting shaft or looseness of the fixation, which may lead to safety accidents. In this study, three single-axis vibration sensors for X, Y, and Z axes were attached on the surface of the test motor to measure the vibration value of vibration. Analog data collected from these sensors was used in classification models for anomaly detection. Since the classification accuracy was around only 93%, commonly used hyperparameter optimization techniques such as Grid search, Random search, and Bayesian Optimization were applied to increase accuracy. In addition, Response Surface Method based on Design of Experiment was also used for hyperparameter optimization. However, it was found that there were limits to improving accuracy with these methods. The reason is that the sampling data from an analog signal does not reflect the patterns hidden in the signal. Therefore, in order to find pattern information of the sampling data, we obtained descriptive statistics such as mean, variance, skewness, kurtosis, and percentiles of the analog data, and applied them to the classification models. Classification models using descriptive statistics showed excellent performance improvement. The developed model can be used as a monitoring system that detects abnormal conditions of the motor test.

• Journal of Korea Multimedia Society
• /
• v.19 no.5
• /
• pp.874-880
• /
• 2016

The gesture classification has been applied to many fields. But it is not efficient in the environment for game interface with low specification devices such as mobile and tablet, In this paper, we propose a effective way for realistic game interface using k-nearest neighbors algorithm for gesture classification. It is time consuming by realtime rendering process in game interface. To reduce the process time while preserving the accuracy, a reconstruction method to minimize error between training and test data sets is also proposed. The experimental results show that the proposed method is better than the conventional methods in both accuracy and time.