• The Journal of the Acoustical Society of Korea
• /
• v.32 no.5
• /
• pp.438-445
• /
• 2013

In this paper, we propose an enhancement of Japanese acoustic model which is trained with Korean speech database by using several combination strategies. We describe the strategies for training more than two language combination, which are Cross-Language Transfer, Cross-Language Adaptation, and Data Pooling Approach. We simulated those strategies and found a proper method for our current Japanese database. Existing combination strategies are generally verified for under-resourced Language environments, but when the speech database is not fully under-resourced, those strategies have been confirmed inappropriate. We made tyied-list with only object-language on Data Pooling Approach training process. As the result, we found the ERR of the acoustic model to be 12.8 %.

• International Journal of Railway
• /
• v.2 no.2
• /
• pp.50-59
• /
• 2009

Numerous train operating companies provide training by means of driving simulators. A detailed analysis in the course of the rail research project 2TRAIN has shown that the simulation technology, the purposes of training and the overall concept of simulator-based training are rather diverse (Schmitz & Maag, 2008). A joint factor however are weak assessment capabilities and the fact that the simulator training is often not embedded into the overall competence management. This fact hinders an optimal use of the simulators. Therefore, 2TRAIN aims at the development of enhanced training and assessment tools. Taking into account that several simulators are already in use, the focus lays on the extension of existing simulation technology instead of developing entirely new systems. This extension comprises (1) a common data simulation interface (CDSI), (2) a rule-based expert system (ExSys), (3) a virtual instructor (VI), and (4) an _assessment database (AssDB). The foundation of this technical development is an assessment concept (PERMA concept) that is based on performance markers. The first part of the paper presents this assessment concept and a process model for the two major steps of driver performance assessment, i.e. (1) the specification of exercise and assessment and (2) the assessment algorithm and execution of the assessment. The second part describes the rationale and the functionalities of the simulator add-on tools. Finally, recommendations for further technical improvement and appropriate usage are given. based on the results of a pilot study.

• Journal of Korea Multimedia Society
• /
• v.22 no.1
• /
• pp.44-54
• /
• 2019

Recently, image-based object detection has made great progress with the introduction of Convolutional Neural Network (CNN). Many trials such as Region-based CNN, Fast R-CNN, and Faster R-CNN, have been proposed for achieving better performance in object detection. YOLO has showed the best performance under consideration of both accuracy and computational complexity. However, these data-driven detection methods including YOLO have the fundamental problem is that they can not guarantee the good performance without a large number of training database. In this paper, we propose a data sampling method using CycleGAN to solve this problem, which can convert styles while retaining the characteristics of a given input image. We will generate the insufficient data samples for training more robust object detection without efforts of collecting more database. We make extensive experimental results using the day-time and night-time road images and we validate the proposed method can improve the object detection accuracy of the night-time without training night-time object databases, because we converts the day-time training images into the synthesized night-time images and we train the detection model with the real day-time images and the synthesized night-time images.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.1
• /
• pp.63-74
• /
• 2023

In order to ensure safety to invasion of privacy, Federated Learning(FL) that learns using parameters is emerging. However a paper that leaks training data using gradients was recently published. Our paper implements an experiment to leak training data using gradients in a federated learning environment, and proposes a method to improve reconstruction performance by improving existing attacks that leak training data. Experiments using Yale face database B, MNIST dataset on the proposed method show that federated learning is not safe from invasion of privacy by reconstructing up to 100 data out of 100 training data when performance of federated learning is high at accuracy=99~100%. In addition, by comparing the performance (MSE, PSNR, SSIM) of pixels and the performance of identification by Human Test, we want to emphasize the importance of the performance of identification rather than the performance of pixels.

• Proceedings of the KSRS Conference
• /
• v.2
• /
• pp.635-638
• /
• 2006

Regression of conventional prediction techniques in data mining uses the model which is generated from the training step. This model is applied to new input data without any change. If this model is applied directly to time series, the rate of prediction accuracy will be decreased. This paper proposes an incremental regression for time series prediction like typhoon track prediction. This technique considers the characteristic of time series which may be changed over time. It is composed of two steps. The first step executes a fractional process for applying input data to the regression model. The second step updates the model by using its information as new data. Additionally, the model is maintained by only recent data in a queue. This approach has the following two advantages. It maintains the minimum information of the model by using a matrix, so space complexity is reduced. Moreover, it prevents the increment of error rate by updating the model over time. Accuracy rate of the proposed method is measured by RME(Relative Mean Error) and RMSE(Root Mean Square Error). The results of typhoon track prediction experiment are performed by the proposed technique IMLR(Incremental Multiple Linear Regression) is more efficient than those of MLR(Multiple Linear Regression) and SVR(Support Vector Regression).

• Journal of Environmental Health Sciences
• /
• v.46 no.4
• /
• pp.410-422
• /
• 2020

Objectives: According to the Act on Registration, Evaluation, Etc. of Chemicals, new and existing chemicals must be registered by 2030. In addition, industries need to submit hazard data as an attachment during the registration process. Therefore, we constructed a nationwide chemical database to support small industry by providing hazard data and original sources. During the process, we developed a new standard procedure for minimizing errors and increasing reliability. Methods: We analyzed the categories of errors and the cause of the errors through the verification results of the 2019 project. We present an improved database construction methodology and system. Results: Errors are categorized according to their causative factors into simple, technical, and structural type errors. Simple errors arise simply because of decreased concentration or negligence in following the instructions. Technical errors are caused by a discrepancy between the professional field and the type of data. Structural errors indicate systemic errors such as incomplete forms on the excel database or ambiguity in the guidelines. Lessons from the errors collected in the 2019 project are used to update the procedures for database authorization and technical guidelines. The main update points are as follows; 'supplementation of review process', 'giving regular training to external reviewers', 'giving additional information to authors, like physico-chemical properties of substances, degradability, etc.', 'amendment of excel form', and 'guideline upgrades'. Conclusions: We conducted this study with the aim of improving the accuracy and reliability of the database of hazard information for chemical substances. The new procedures and guidelines are now being used in the 2020 project for construction of a hazard information database for Korea.

• International Journal of Knowledge Content Development & Technology
• /
• v.6 no.1
• /
• pp.41-68
• /
• 2016

This paper aims to analyze the function of database management systems from the perspective of librarians working in engineering institutions in Andhra Pradesh. Ninety-eight librarians from one hundred thirty engineering institutions participated in the study. The paper reveals that training by computer suppliers and software packages are the significant mode of acquiring DBMS skills by librarians; three-fourths of the librarians are postgraduate degree holders. Most colleges use database applications for automation purposes and content value. Electrical problems and untrained staff seem to be major constraints faced by respondents for managing library databases.

• Journal of Information Processing Systems
• /
• v.12 no.3
• /
• pp.381-391
• /
• 2016

The Joint Bayesian (JB) method has been used in most state-of-the-art methods for face verification. However, since the publication of the original JB method in 2012, no improved verification method has been proposed. A lot of studies on face verification have been focused on extracting good features to improve the performance in the challenging Labeled Faces in the Wild (LFW) database. In this paper, we propose an improved version of the JB method, called the two-dimensional Joint Bayesian (2D-JB) method. It is very simple but effective in both the training and test phases. We separated two symmetric terms from the three terms of the JB log likelihood ratio function. Using the two terms as a two-dimensional vector, we learned a decision line to classify same and not-same cases. Our experimental results show that the proposed 2D-JB method significantly outperforms the original JB method by more than 1% in the LFW database.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.1
• /
• pp.485-492
• /
• 2018

In this paper, we propose facial expression recognition using CNN (Convolutional Neural Network), one of the deep learning technologies. The proposed structure has general classification performance for any environment or subject. For this purpose, we collect a variety of databases and organize the database into six expression classes such as 'expressionless', 'happy', 'sad', 'angry', 'surprised' and 'disgusted'. Pre-processing and data augmentation techniques are applied to improve training efficiency and classification performance. In the existing CNN structure, the optimal structure that best expresses the features of six facial expressions is found by adjusting the number of feature maps of the convolutional layer and the number of nodes of fully-connected layer. The experimental results show good classification performance compared to the state-of-the-arts in experiments of the cross validation and the cross database. Also, compared to other conventional models, it is confirmed that the proposed structure is superior in classification performance with less execution time.

• Journal of Information Processing Systems
• /
• v.14 no.4
• /
• pp.807-823
• /
• 2018

Speaker verification system performance depends on the utterance of each speaker. To verify the speaker, important information has to be captured from the utterance. Nowadays under the constraints of limited data, speaker verification has become a challenging task. The testing and training data are in terms of few seconds in limited data. The feature vectors extracted from single frame size and rate (SFSR) analysis is not sufficient for training and testing speakers in speaker verification. This leads to poor speaker modeling during training and may not provide good decision during testing. The problem is to be resolved by increasing feature vectors of training and testing data to the same duration. For that we are using multiple frame size (MFS), multiple frame rate (MFR), and multiple frame size and rate (MFSR) analysis techniques for speaker verification under limited data condition. These analysis techniques relatively extract more feature vector during training and testing and develop improved modeling and testing for limited data. To demonstrate this we have used mel-frequency cepstral coefficients (MFCC) and linear prediction cepstral coefficients (LPCC) as feature. Gaussian mixture model (GMM) and GMM-universal background model (GMM-UBM) are used for modeling the speaker. The database used is NIST-2003. The experimental results indicate that, improved performance of MFS, MFR, and MFSR analysis radically better compared with SFSR analysis. The experimental results show that LPCC based MFSR analysis perform better compared to other analysis techniques and feature extraction techniques.