• The Journal of the Acoustical Society of Korea
• /
• v.29 no.3
• /
• pp.216-221
• /
• 2010

In this paper, we propose speaker verification method using Support Vector Machine (SVM) kernel with Gaussian Mixture Model (GMM)-supervector based on the Mahalanobis distance. The proposed GMM-supervector SVM kernel method is combined GMM with SVM. The GMM-supervectors are generated by GMM parameters of speaker and other speaker utterances. A speaker verification threshold of GMM-supervectors is decided by SVM kernel based on Mahalanobis distance to improve speaker verification accuracy. The experimental results for text-independent speaker verification using 20 speakers demonstrates the performance of the proposed method compared to GMM, SVM, GMM-supervector SVM kernel based on Kullback-Leibler (KL) divergence, and GMM-supervector SVM kernel based on Bhattacharyya distance.

• KEPCO Journal on Electric Power and Energy
• /
• v.7 no.1
• /
• pp.115-123
• /
• 2021

Through machine learning-based load prediction, it is possible to prevent excessive power generation or unnecessary economic investment by estimating the appropriate amount of facility investment in consideration of the load that will increase in the future or providing basic data for policy establishment to distribute the maximum load. However, in order to secure the reliability of the developed load prediction model in the field, the performance comparison verification between the distribution line load prediction models must be preceded, but a comparative performance verification system between the distribution line load prediction models has not yet been established. As a result, it is not possible to accurately determine the performance excellence of the load prediction model because it is not possible to easily determine the likelihood between the load prediction models. In this paper, we developed a reliability verification system for load prediction models including a method of comparing and verifying the performance reliability between machine learning-based load prediction models that were not previously considered, verification process, and verification result visualization methods. Through the developed load prediction model reliability verification system, the objectivity of the load prediction model performance verification can be improved, and the field application utilization of an excellent load prediction model can be increased.

• Journal of Information Processing Systems
• /
• v.12 no.3
• /
• pp.468-488
• /
• 2016

The paper proposes a novel framework for 3D face verification using dimensionality reduction based on highly distinctive local features in the presence of illumination and expression variations. The histograms of efficient local descriptors are used to represent distinctively the facial images. For this purpose, different local descriptors are evaluated, Local Binary Patterns (LBP), Three-Patch Local Binary Patterns (TPLBP), Four-Patch Local Binary Patterns (FPLBP), Binarized Statistical Image Features (BSIF) and Local Phase Quantization (LPQ). Furthermore, experiments on the combinations of the four local descriptors at feature level using simply histograms concatenation are provided. The performance of the proposed approach is evaluated with different dimensionality reduction algorithms: Principal Component Analysis (PCA), Orthogonal Locality Preserving Projection (OLPP) and the combined PCA+EFM (Enhanced Fisher linear discriminate Model). Finally, multi-class Support Vector Machine (SVM) is used as a classifier to carry out the verification between imposters and customers. The proposed method has been tested on CASIA-3D face database and the experimental results show that our method achieves a high verification performance.

• Journal of Platform Technology
• /
• v.9 no.1
• /
• pp.32-39
• /
• 2021

Existing environmental measuring devices mainly focus on electromagnetic wave and eco-friendly product certification and durability test, and sensor reliability verification and verification of measurement data are conducted mainly through sensor performance evaluation through type approval and registration, acceptance test, initial calibration, and periodic test. This platform has established an ICT-based environmental monitoring sensor reliability verification system that supports not only performance evaluation for each target sensor, but also a verification system for sensor data reliability. A sensor board to collect sensor data for environmental information was produced, and a sensor and data reliability evaluation and verification service system was standardized. In addition, to evaluate and verify the reliability of sensor data based on ICT, a sensor data platform monitoring prototype using LoRa communication was produced, and the test was conducted in smart cities. To analyze the data received through the system, an optimization algorithm was developed using machine learning. Through this, a sensor big data analysis system is established for reliability verification, and the foundation for an integrated evaluation and verification system is provide.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.8 no.3
• /
• pp.57-67
• /
• 2018

The ensemble is a unified approach used for getting better performance by using multiple algorithms in machine learning. In this paper, we introduce boosting and bagging, which have been widely used in ensemble techniques, and design a method using support vector regression, radial basis function network, Gaussian process, and multilayer perceptron. In addition, our experiment was performed by adding a recurrent neural network and MOHID numerical model. The drifter data used for our experimental verification consist of 683 observations in seven regions. The performance of our ensemble technique is verified by comparison with four algorithms each. As verification, mean absolute error was adapted. The presented methods are based on ensemble models using bagging, boosting, and machine learning. The error rate was calculated by assigning the equal weight value and different weight value to each unit model in ensemble. The ensemble model using machine learning showed 61.7% improvement compared to the average of four machine learning technique.

• Proceedings of the IEEK Conference
• /
• 2000.11c
• /
• pp.37-40
• /
• 2000

A fast method of Hangul address word verification is presented in this Paper. Pre-segmentation and recognition by DP matching is adopted in this paper. An address line image is over-segmented by analyzing the topology of connected components and the projection profile. A fast individual Hangul character verifier was developed by applying SVM (Support Vector Machine). The segmentation hypothesis was represented by lattice structure, and a best path search by dynamic programming generates the most probable segmentation path and the final verification score. The word verifier was tested on 310 address image DB, and it show the possibility of improvements of this method.

• Journal of Platform Technology
• /
• v.9 no.1
• /
• pp.23-31
• /
• 2021

In recent years, in the domestic industry, personal damage has occurred due to sensor malfunction and the emission of harmful gases. But there is a limit to the reliability verification of sensor data because the evaluation of environmental sensors is focused on durability and risk tests. This platform designed a sensor board that measures 10 major substances and a performance verification system for each sensor. In addition, the data collected by the sensor board was transferred to the server for data reliability evaluation and verification using LoRa communication, and a prototype of the sensor data platform was produced to monitor the transferred data. And the collected data is analyzed and predicted by using machine learning techniques.

• Design & Manufacturing
• /
• v.16 no.1
• /
• pp.21-26
• /
• 2022

In the CNC machining process, problems such as lowering of machine operation rate, setting errors, and machining precision occur due to the increase in setting time and preparation time. These machining errors cause delays in delivery and increase in cost due to an increase in the number of mounting and dismounting of the workpiece, an increase in measurement and reprocessing time, and an increase in the finishing time in the assembly process. Therefore, in this study, by automating the setting of the work piece using OMV (On Machine Verification), which is a meteorological measurement system, the preparation time for machining the work piece and the setting accuracy were improved, the rework rate was reduced, and the mold manufacturing process was shortened. Through the advancement, standardzation, and automation of the mold part manufacturing process, we have improved productivity by minimizing low-value-added repetitive tasks. In addition, the measurement time was reduced by more than 50% and the machining measurement rate was improved by more than 20%, eliminating repetitive work for correcting machining defects, and reducing the work preparation time by more than 15% through automatic setting.

• Journal of the Korean Society for Precision Engineering
• /
• v.18 no.2
• /
• pp.79-83
• /
• 2001

A verification of multi-spindles drilling NC data is presented. The drilling of multi-spindles can offer productivity over three times as fast as that of single spindle. The most important things in machining tube sheet are precision of hole position and machining time. The drilling of multi-spindles has difficulties in controlling many motors to drive spindles and assign a correspondent number to each spindles. Multi-spindles drilling has different codes from CNC milling ; many subroutines, assignment of spindle, and so on. The conventional method, which inspects the NC code of the drilling, is to drill holes on a thin plate or tube sheet previously. The method results in low productivity because it consumed long machining time and welding for correction. This paper describes details of multi-spindles NC code and operation of multi-spindles drilling machine. A verification software of the multi-spindles drilling NC code is developed on the details.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 1996.03a
• /
• pp.183-187
• /
• 1996

On the machine measuring system composed of touch trigger probes, a DNC module, a CMM module, an analysis module and a man-machine interface unit was developed. Measuring accuracy is affected by working accuracy of the on the machine measuring system. The working accuracy of the system is due to geometric errors of th machine tool, servo errors of feed drives and positioning errors of probes. In order to compensate for the measuring errors due to the working accuracy, a calibration module was developed. The measuring automation system was realized with the on the machine measuring system and an IBM-PC on the machine center through a RS-232C. It turns the machining machine (CMM). The system is used for dimensional checking of machined components. initial job setup, part identification, identification of machining errors due to deflection and wear of tools. cutter run out, and calibration of machine tools. A horizontal machining center equipped with FANUC OMC wre used for verification of the system. The validity and reliability of the system. The validity and reliability of the system were confirmed through a series of experiments with gage blocks, ring gages, comparison measurement with a commercial CMM, and so on.