Masaru Teranishi

Continuous fatigue level estimation for the classification of fatigued bills based on an acoustic signal feature by a supervised SOM

Fatigued bills have a harmful influence on the daily operation of automated teller machines (ATMs). To make the classification of fatigued bills more efficient, the development of an automatic fatigued bill classification method with a continu ous fatigue level is desirable. We propose a new method to estimate the bending rigidity of bills using the acoustic signal feature of banking machines. The estimated bending rigidities are used as the continuous fatigue level for the classification of fatigued bills. By using a supervised self-organizing map (SOM), we effectively estimate the bending rigidity using only the acoustic energy pattern. The experimental results with real bill samples show the effectiveness of the proposed method.

Fatigue level estimation of bill based on acoustic energy features by Supervised SOM

Fatigued bills have harmful influence on daily operation of Automated Teller Machine (ATM). To make the fatigued bills classification more efficient, development of an automatic fatigued bill classification method is desired. We proposed a new method to estimate fatigue levels of bills from acoustic energy feature of banking machines by using the supervised SOM in the previous study. Though the proposed method has achieved to estimate the bending rigidity from only the acoustic energy pattern effectively, there were some problems about estimation performances. In this paper, we try to improve estimation performance by reconsidering configurations of the Supervised SOM. Furthermore, we show effectiveness of the proposed method by comparing it with other estimation methods. The experimental results with real bill samples show the improved estimation accuracy, and effectiveness of the proposed method.

Fatigue Level Estimation of Bill by Using Supervised SOM Based on Feature-Selected Acoustic Energy Pattern

Fatigued bills have harmful influence on daily operation of automated teller machine (ATM). To make the fatigued bills classification more efficient, development of an automatic fatigued bill classification method is desired. In this paper, we propose a new method to estimate fatigue levels of bills from feature-selected acoustic energy pattern of banking machines by using the supervised SOM. The proposed method also selects feature components of an acoustic energy pattern based on correlation between acoustic energy features and fatigue level of bill to let the supervised SOM work effectively. The experimental results with real bill samples show the effectiveness of the proposed method. Furthermore, we show an advantage of the proposed method by comparing it with another estimation method.

Reconstruction of 3-D Radiation Profile From 2-D Images Viewed by IR Imaging Video Bolometers With SVD

Radiation is one of the major paths of power loss from fusion plasmas. Since the radiation distribution has a complex 3-D structure in helical devices, such as a large helical device (LHD), 3-D radiation measurements, are required for understanding the power balance of fusion plasmas. IR imaging video bolometers (IRVBs) installed in the LHD were examined as one system of 3-D imaging using a projection matrix that is obtained under the assumption of plasma periodicity in the toroidal direction, and using the phantoms obtained numerically with the EMC3-EIRENE code as the impurity distribution. The reconstructed profiles indicate the capability of the current system of IRVBs for imaging the 3-D radiation distribution in LHD.

Fatigue level estimation of bill based on feature-selected acoustic energy pattern by using supervised SOM

Fatigued bills have harmful influence on daily operation of Automated Teller Machine (ATM). To make the fatigued bills classification more efficient, development of an automatic fatigued bill classification method is desired. In this paper, we propose a new method to estimate fatigue levels of bills from feature-selected acoustic energy pattern of banking machines by using the Supervised SOM. The proposed method also selects feature components of an acoustic energy pattern based on correlation between acoustic energy features and fatigue level of bill to let the supervised SOM work effectively. The experimental results with real bill samples show the effectiveness of the proposed method. Furthermore, we show an advantage of the proposed method by comparing it with another estimation method.

Personal Peculiarity Classification of Flat Finishing Skill Training by using Torus type Self-Organizing Maps

The paper proposes an unsupervised classification method for peculiarities of flat finishing motion with an iron file, measured by a 3D stylus. The classified personal peculiarities are used to correct learner’s finishing motions effectively for skill training. In the case of such skill training, the number of classes of peculiarity is unknown. A torus type Self-Organizing Maps is effectively used to classify such unknown number of classes of peculiarity patterns.

Peculiarity Classification of Flat Finishing Motion Based on Tool Trajectory by Using Self-organizing Maps

The paper proposes an unsupervised classification method for peculiarities of flat finishing motion with an iron file, measured by a 3D stylus. The proposed method extract personal peculiarities based on trajectory of an iron file. The classified peculiarities are used to correct learner’s finishing motions effectively for skill training. In the case of such skill training, the number of classes of peculiarity is unknown. A torus type Self-Organizing Maps is effectively used to classify such unknown number of classes of peculiarity patterns.

Examining efficient instructional methods for computer-aided brush coating skill training system in elementary and secondary education

In recent Japan, many elementary and secondary school children strongly feel an awareness of resistance toward manufacturing class. It is considered as one of the aversion causes to science and engineering. We suspect that the shortage of teacher’s manufacturing experience is related to this background. So, the authors previously have developed a brush coating skill training system with a haptic device to improve the ability. In the skill training, historically, the instruction method while reproducing the video of trainee’s action has been considered to be effective. Thus, our developed system includes a software system to visualize the brush coating motion. This software system can record trainee’s brush coating motion and visualize the data in virtual three-dimensional space, so the advisor and the trainee can share his/her skill level and the wrong motion such as extra shaking and tilt by reproducing his/her past training data on a computer screen. In this paper, to improve the learning effectiveness of our brush coating skill training system, some kinds of instructional methods are designed, and the difference of the instructional methods is analyzed based on the proposed criteria. This paper performed some experiments with 10 college students who have less experience of brush coating. From the analysis results, the most effective instruction method was clarified.

Personal Peculiarity Classification of Flat Finishing Motion for Skill Training by Using Expanding Self-Organizing Maps

The paper proposes an unsupervised classification method for peculiarities of flat finishing motion with an iron file, measured by a 3D stylus. The classified personal peculiarities are used to correct learner’s finishing motions effectively for skill training. In the case of such skill training, the number of classes of peculiarity is unknown. An expanding Self-Organizing Maps is effectively used to classify such unknown number of classes of peculiarity patterns.

Improved neural network tomography by initial learning with coarse reconstructed image

This paper presents a stable learning method of the neural network tomography, in case of asymmetrical few view projection. The neural network collocation method (NNCM) is one of effective reconstruction tools for symmetrical few view tomography. But in cases of asymmetrical few view, the learning process of NNCM tends to be unstable and fails to reconstruct appropriate tomographic images. We solve the unstable learning problem of NNCM by introducing a coarse reconstructed image in the initial learning stage of NNCM. The numerical simulation with an assumed tomographic image shows the effectiveness of the proposed method.