• Korean Journal of Remote Sensing
• /
• 제38권4호
• /
• pp.327-341
• /
• 2022

Cloud removal is often required to construct time-series sets of optical images for environmental monitoring. In regression-based cloud removal, the selection of an appropriate regression model and the impact analysis of the input images significantly affect the prediction performance. This study evaluates the potential of Gaussian process (GP) regression for cloud removal and also analyzes the effects of cloud-free optical images and spectral bands on prediction performance. Unlike other machine learning-based regression models, GP regression provides uncertainty information and automatically optimizes hyperparameters. An experiment using Sentinel-2 multi-spectral images was conducted for cloud removal in the two agricultural regions. The prediction performance of GP regression was compared with that of random forest (RF) regression. Various combinations of input images and multi-spectral bands were considered for quantitative evaluations. The experimental results showed that using multi-temporal images with multi-spectral bands as inputs achieved the best prediction accuracy. Highly correlated adjacent multi-spectral bands and temporally correlated multi-temporal images resulted in an improved prediction accuracy. The prediction performance of GP regression was significantly improved in predicting the near-infrared band compared to that of RF regression. Estimating the distribution function of input data in GP regression could reflect the variations in the considered spectral band with a broader range. In particular, GP regression was superior to RF regression for reproducing structural patterns at both sites in terms of structural similarity. In addition, uncertainty information provided by GP regression showed a reasonable similarity to prediction errors for some sub-areas, indicating that uncertainty estimates may be used to measure the prediction result quality. These findings suggest that GP regression could be beneficial for cloud removal and optical image reconstruction. In addition, the impact analysis results of the input images provide guidelines for selecting optimal images for regression-based cloud removal.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• 제24권8호
• /
• pp.1077-1084
• /
• 2000

In this work, a modified stage-stacking method for the performance prediction of multi-stage axial flow compressors is proposed. The method is based on a simultaneous calculation of all interstage variables (temperature, pressure, flow velocity) instead of the conventional sequential stage-by-stage scheme. The method is also very useful in simulating the effect of changing angles of the inlet guide vane and stator vanes on the compressor operating characteristics. Generalized stage performance curves are used in presenting the performance characteristics of each stage. General assumptions enable determination of flow path data and stage design performance. Performance of various real compressors is predicted and comparison between prediction and field data validates the usefulness of the present method.

• Journal of Mechanical Science and Technology
• /
• 제15권9호
• /
• pp.1292-1301
• /
• 2001

The mean-line method using empirical models is the most practical method of predicting off-design performance. To gain insight into the empirical models, the influence of empirical models on the performance prediction results is investigated. We found that, in the two-zone model, the secondary flow mass fraction has a considerable effect at high mass flow-rates on the performance prediction curves. In the TEIS model, the first element changes the slope of the performance curves as well as the stable operating range. The second element makes the performance curves move up and down as it increases or decreases. It is also discovered that the slip factor affects pressure ratio, but it has little effect on efficiency. Finally, this study reveals that the skin friction coefficient has significant effect on both the pressure ratio curve and the efficiency curve. These results show the limitations of the present empirical models, and more resonable empirical models are reeded.

• Proceedings of the KSR Conference
• /
• 한국철도학회 2011년도 정기총회 및 추계학술대회 논문집
• /
• pp.2844-2851
• /
• 2011

The purpose of this study is to set up an acoustic prediction technique and to perform the IL test of scale down noise reducing device for the development of the noise reducing device as the development of 400km/h class high speed train. First of all, the IL prediction of noise reducing device was performed with the 2D BEM method. And the noise test of scale down noise reducing device in anechoic chamber was performed for the verification of acoustic prediction technique and IL performance evaluation. As the results, the acoustic prediction technique for the development of noise reducing device was verified because the averaged IL difference between prediction and test is in 2dB(A). And the measured IL value of noise reducing device is less than 2dB(A), and additional IL with polyester absorption material is increased about 0.5dB(A).

• Journal of the Korea Society of Computer and Information
• /
• 제21권4호
• /
• pp.73-80
• /
• 2016

Most fault prediction models have class imbalance problems because training data usually contains much more non-fault class modules than fault class ones. This imbalanced distribution makes it difficult for the models to learn the minor class module data. Data imbalance is much higher when severity-based fault prediction is used. This is because high severity fault modules is a smaller subset of the fault modules. In this paper, we propose severity-based models to solve these problems using the three sampling methods, Resample, SpreadSubSample and SMOTE. Empirical results show that Resample method has typical over-fit problems, and SpreadSubSample method cannot enhance the prediction performance of the models. Unlike two methods, SMOTE method shows good performance in terms of AUC and FNR values. Especially J48 decision tree model using SMOTE outperforms other prediction models.

• Journal of Institute of Control, Robotics and Systems
• /
• 제15권9호
• /
• pp.906-908
• /
• 2009

AVI (Automatic Vision Inspection) systems automatically detect defect features and measure their sizes via camera vision. It is important to predict the performance of an AVI to meet customer's specification in advance. Also the prediction can indicate the level of current performance of an AVI system. In this paper, we propose a statistical method for prediction of false alarm rate regarding inconsistency of defect size measurement process. For this purpose, only simple experiments are needed to measure the defect sizes for certain number of times. The statistical features from the experiment are utilized in the prediction process. Therefore, the proposed method is swift and easy to implement and use. The experiment shows a close prediction compared to manual inspection results.

• Geomechanics and Engineering
• /
• 제18권4호
• /
• pp.373-389
• /
• 2019

It is important to study the problem of durability for tunnel structures. As a main influence on the durability of tunnel structures, carbonation-induced corrosion is studied. For the complicated environment of tunnel structures, based on the data samples from real engineering examples, the intelligent method (genetic programming) is used to construct the service life prediction model of tunnel structures. Based on the model, the prediction of service life for tunnel structures in carbonation environments is studied. Using the data samples from some tunnel engineering examples in China under carbonation environment, the proposed method is verified. In addition, the performance of the proposed prediction model is compared with that of the artificial neural network method. Finally, the effect of two main controlling parameters, the population size and sample size, on the performance of the prediction model by genetic programming is analyzed in detail.

• ETRI Journal
• /
• 제44권2호
• /
• pp.208-219
• /
• 2022

Crime amount prediction is crucial for optimizing the police patrols' arrangement in each region of a city. First, we analyzed spatiotemporal correlations of the crime data and the relationships between crime and related auxiliary data, including points-of-interest (POI), public service complaints, and demographics. Then, we proposed a crime amount prediction model based on 2D convolution and long short-term memory neural network (2DCONV-LSTM). The proposed model captures the spatiotemporal correlations in the crime data, and the crime-related auxiliary data are used to enhance the regional spatial features. Extensive experiments on real-world datasets are conducted. Results demonstrated that capturing both temporal and spatial correlations in crime data and using auxiliary data to extract regional spatial features improve the prediction performance. In the best case scenario, the proposed model reduces the prediction error by at least 17.8% and 8.2% compared with support vector regression (SVR) and LSTM, respectively. Moreover, excessive auxiliary data reduce model performance because of the presence of redundant information.

• Journal of Broadcast Engineering
• /
• 제19권2호
• /
• pp.227-239
• /
• 2014

This paper proposes an enhance prediction for conventional near-lossless coder to effectively lower external memory bandwidth in image processing SoC. First, we utilize an already reconstructed green component as a base of predictor of the other color component because high correlation between RGB color components usually exists. Next, we can improve prediction performance by applying variable block size prediction. Lastly, we use minimum internal memory and improve a temporal prediction performance by using a template dictionary that is sampled in previous frame. Experimental results show that the proposed algorithm shows better performance than the previous works. Natural images have approximately 30% improvement in coding efficiency and CG images have 60% improvement on average.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• 제30권11B호
• /
• pp.744-749
• /
• 2005

This paper proposes the prediction based QoS routing algorithm, PSS(Prediction Safety-Shortest) algorithm that minimizes network state information overhead and presumes more accurate knowledge of the present state of all the links within the network. We apply time series model to the available bandwidth prediction to overcome inaccurate information of the existing QoS routing algorithms. We have evaluated the performance of the proposed model and the existing algorithms on MCI networks, it thus appears that we have verified the performance of this algorithm.