• Journal of the Korean Society of Industry Convergence
• /
• 제27권5호
• /
• pp.1243-1251
• /
• 2024

The Integrated Condition Assessment System (ICAS) is a system that supports Condition Based Maintenance (CBM) by diagnosing the status of major onboard equipment on a naval ship in real time and allowing maintenance personnel to immediately perform maintenance when an abnormal condition occurs to maintain the operational performance of the on-board equipment. This study introduces the necessity of data preprocessing collected from naval ship, and compare and review baselines generated through statistical and designed machine learning algorithms using the same data preprocessing. Through these, this paper analyzes and proposes the suitability of a baseline algorithm, a machine learning methods that has not been applied to the condition based maintenance of naval ship equipment.

• The Journal of the Korea Contents Association
• /
• 제9권1호
• /
• pp.159-166
• /
• 2009

In this paper, we analyse the noise robustness of MLPs(Multilayer perceptrons) through deriving the probability density function(p.d.f.) of output nodes with additive input noises and the misclassification ratio with the integral form of the p.d.f. functions. Also, we propose linear preprocessing methods to improve the noise robustness. As a preprocessing stage of MLPs, we consider ICA(independent component analysis) and PCA(principle component analysis). After analyzing the noise reduction effect using PCA or ICA in the viewpoints of SNR(Singal-to-Noise Ratio), we verify the preprocessing effects through the simulations of handwritten-digit recognition problems.

• Journal of Advanced Navigation Technology
• /
• 제10권1호
• /
• pp.48-56
• /
• 2006

This paper proposes an efficient preprocessing technique for a binary centroid tracker in correlated image sequences. It is known that the following factors determine the performance of the binary centroid target tracker: (1) an efficient real-time preprocessing technique, (2) an exact target segmentation from cluttered background images and (3) an intelligent tracking window sizing, and etc. The proposed centroid tracker consists of an adaptive segmentation method based on novel distance features and an efficient real-time preprocessing technique in order to enhance the distinction between the objects of interest and their local background. Various tracking experiments using synthetic images as well as real Forward-Looking InfraRed (FLIR) images are performed to show the usefulness of the proposed methods.

• Journal of KIISE:Software and Applications
• /
• 제34권5호
• /
• pp.487-501
• /
• 2007

To achieve high resource utilization for multi-issue DSPs, production compiler commonly includes variants of iterative modulo scheduling algorithm. However, excessive cyclic data dependences, which exist in communication and media processing loops, unduly restrict modulo scheduling freedom. As a result, replicated functional units in multi-issue DSPs are often under-utilized. To address this resource under-utilization problem, our paper describes a novel compiler preprocessing strategy for effective modulo scheduling. The preprocessing strategy proposed capitalizes on two new transformations, which are referred to as cloning and dismantling. Our preprocessing strategy has been validated by an implementation for StarCore SC140 DSP compiler.

• Journal of Korea Society of Industrial Information Systems
• /
• 제23권4호
• /
• pp.41-53
• /
• 2018

Recently, the development of machine learning technology has led to the application of deep learning technology to existing image based application systems. In this context, some researches have been made to apply CNN (Convolutional Neural Network) to the field of fire detection. To verify the effects of existing preprocessing and feature extraction methods on fire detection when combined with CNN, in this paper, the recognition performance and learning time are evaluated by changing the VGG19 CNN structure while gradually increasing the convolution layer. In general, the accuracy is better when the image is not preprocessed. Also it's shown that the preprocessing method and the feature extraction method have many benefits in terms of learning speed.

• Journal of Biosystems Engineering
• /
• 제35권6호
• /
• pp.450-457
• /
• 2010

Although cherry tomato is one of major vegetables consumed in fresh vegetable market, the quality grading method is mostly dependant on size measurement using drum shape sorting machines. Using Visible/Near-infrared spectroscopy, apparatus to be able to acquire transmittance spectrum data was made and used to estimate firmness, sugar content, and acidity of cherry tomatoes grown at hydroponic and soil culture. Partial least square (PLS) models were performed to predict firmness, sugar content, and acidity for the acquired transmittance spectra. To enhance accuracy of the PLS models, several preprocessing methods were carried out, such as normalization, multiplicative scatter correction (MSC), standard normal variate (SNV), and derivatives, etc. The coefficient of determination ($R^2_p$) and standard error of prediction (SEP) for the prediction of firmness, sugar, and acidity of cherry tomatoes from green to red ripening stages were 0.859 and 1.899 kgf, with a preprocessing of normalization, 0.790 and $0.434^{\circ}Brix$ with a preprocessing of the 1st derivative of Savitzky Golay, and 0.518 and 0.229% with a preprocessing normalization, respectively.

• Journal of the Korean Society for Nondestructive Testing
• /
• 제22권4호
• /
• pp.361-368
• /
• 2002

The aim of this study was to investigate the effect of preprocessing the transmitted energy spectrum data on development of a robust model to predict the sugar content in intact apples. The spectrum data were measured from 120 Fuji apple samples conveying at the speed of 2 apples per second. Computer algorithms of preprocessing methods such as MSC, SNV, first derivative, OSC and their combinations were developed and applied to the raw spectrum data set. The results indicated that correlation coefficients between the transmitted energy values at each wavelength and sugar contents of apples were significantly improved by the preprocessing of MSC and SNV in particular as compared with those of no-preprocessing. SEPs of the prediction models showed great difference depending on the preprocessing method of the raw spectrum data, the largest of 1.265%brix and the smallest of 0.507% brix. Such a result means that an appropriate preprocessing method corresponding to the characteristics of the spectrum data set should be found or developed for minimizing the prediction errors. It was observed that MSC and SNV are closely related to prediction accuracy, OSC is to number of PLS factors and the first derivative resulted in decrease of the prediction accuracy. A robust calibration model could be d3eveloped by the combined preprocessing of MSC and OSC, which showed that SEP=0.507%brix, bias=0.0327 and R2=0.8823.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• 제26권1호
• /
• pp.41-50
• /
• 2008

Recent progress in state-of-the-art geospatial information technologies such as digital mapping, LiDAR(Light Detection And Ranging), and high-resolution satellite imagery provides various data sources fer Digital Elevation Model(DEM). DEMs are major source to extract elements of the hydrological terrain property that are necessary for efficient watershed management. Especially, watersheds extracted from DEM are important geospatial database to identify physical boundaries that are utilized in water resource management plan including water environmental survey, pollutant investigation, polluted/wasteload/pollution load allocation estimation, and water quality modeling. Most of the previous studies related with watershed extraction using DEM are mainly focused on the hydrological elements analysis and preprocessing without considering grid size of the DEMs. This study aims to analyze accuracy of the watersheds extracted from DEMs with various grid sizes generated by LiDAR data and digital map, and appropriate preprocessing methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권7호
• /
• pp.1773-1793
• /
• 2023

Automated machine learning, often referred to as "AutoML," is the process of automating the time-consuming and iterative procedures that are associated with the building of machine learning models. There have been significant contributions in this area across a number of different stages of accomplishing a data-mining task, including model selection, hyper-parameter optimization, and preprocessing method selection. Among them, preprocessing method selection is a relatively new and fast growing research area. The current work is focused on the recommendation of preprocessing methods, i.e., feature subset selection (FSS) algorithms. One limitation in the existing studies regarding FSS algorithm recommendation is the use of a single learner for meta-modeling, which restricts its capabilities in the metamodeling. Moreover, the meta-modeling in the existing studies is typically based on a single group of data characterization measures (DCMs). Nonetheless, there are a number of complementary DCM groups, and their combination will allow them to leverage their diversity, resulting in improved meta-modeling. This study aims to address these limitations by proposing an architecture for preprocess method selection that uses ensemble learning for meta-modeling, namely AutoFE-Sel. To evaluate the proposed method, we performed an extensive experimental evaluation involving 8 FSS algorithms, 3 groups of DCMs, and 125 datasets. Results show that the proposed method achieves better performance compared to three baseline methods. The proposed architecture can also be easily extended to other preprocessing method selections, e.g., noise-filter selection and imbalance handling method selection.

• Fashion & Textile Research Journal
• /
• 제17권5호
• /
• pp.838-843
• /
• 2015

In this study, a cotton knit was dyed with elm bark extract; subsequently, the dyed fabric was measured according to the types of mordants and the preprocessing cationizers used. Additionally, antibiosis against super bacteria was examined. The results follow. First, the color of the dyed cotton knit appeared reddish and yellowish for fabrics treated with non-mordants and mordants. When preprocessing with a cationizer was conducted, the dyeing properties were the best. Second. even when mordants were not used for dyeing, color fastness after washing, sweating, and rubbing was generally good Grade 4 and 5. Color fastness after exposure to sunlight was the best Grade 4 for fabric prepared with ferrous sulfate as the mordant. Third. as for antimicrobial properties, or resistance to super bacteria, the growth of bacteria was suppressed in a meaningful way for fabrics treated with non-mordants and mordants, compared to the control group fabric. The dyeing methods with the most powerful antimicrobial effects were dyeing after preprocessing with a cationizer and preparing fabric with copper sulfate as the mordant. The results stated above show that in case of dyeing with elm bark extract, preprocessing of the cotton knit with a cationizer and dying with copper mordant displayed high levels of antimicrobial properties that were useful for resisting super bacteria. Of these the dyeing properties were the best when preprocessing with a cationizer.