This paper describes a newly developed image processing program for grain boundary extraction with some experimental results. It is available on an MS-DOS personal computer without any special equipment. The program deals with image data directly, and extracts a thinned image of grain boundaries from an input original gray image. Its robustness comes from refinement process inserted between binarization and thinning. The refinement process is carefully designed so as to be suitable for the property of a grain images, so that the program is applicable to many kinds of grain images. Since the operators provided in the program are of general purpose inherent, it is also applicable to any kinds of digital images by combining the operators as you like.

抄録全体を表示

We have constructed a general framework for integrating application programs with control through a local Web browser. This method is based on a simple inter-process message function from an external process to application programs. Commands to a target program areprepared in a script file, which is parsed by a message dispatcher program. When it is used as ahelper application to a Web browser, these messages will be sent from the browser by clicking a hyper-link in a Web document. Our framework also supports pluggable extension-modules for application programs by means of dynamic linking. A prototype system is implemented on our molecular structure-viewer program, MOSBY. It successfully featured a function to load an extension-module required for the docking study of molecular fragments from a Web page. Our simple framework facilitates the concise configuration of Web softwares without complicated knowledge on network computation and security issues. It is also applicable for a wide range of network computations processing private data using a Web browser.

抄録全体を表示

This paper presents a new computational method in the modeling and simulation of gene expression by introducing the artificial chemical system. The artificial chemical system is specified by its four items:(1) components (five kinds of particles and DNA with Genetic Switches);(2) space (2-dimensional polar grids);(3) simple reaction rules (construction and destruction of molecules, etc.);(4) simple behavioral rules (stochastic movements and stochastic collisions, etc.). The simulation demonstrates the capability of the system to exhibit emergent behavior: that is, global order of the system (regular rhythms, i.e. regular oscillations in the amounts of some gene products, in this case) emerges out of the randomness (through stochastic movements and collisions) of the components.

抄録全体を表示

This paper describes a model of the artificial chemical world and its computer simulation, in which rhythms emerge. The model specifies four items of the artificial chemical world:(1) components (five kinds of particles and DNA having Genetic Switches);(2) space (2-dimensional polar grids);(3) simple reaction rules (construction and destruction of molecules, etc.);(4) simple behavioral rules (stochastic movements and stochastic collisions, etc.); The simulation demonstrates the capability of the system to exhibit emergent behavior: that is, global order of the system (regular rhythms in this case) emerges out of randomness (thorough stochastic movements and collisions) of its components.

抄録全体を表示

To improve the performance of sparse representation based space-time adaptive processing methods, the multiple measurement vectors joint sparse representation (MMV-JSR) model of MIMO radar received signal of testing range cell is established. Then, MMV FOCal Underdetermined System Solver is applied to estimate clutter-target space-time spectrum. At last, clutter plus noise covariance matrix and the weight vector of STAP processor are calculated based on the clutter distribution obtained by using the rough priori knowledge of target. Numerical results demonstrate that our method has great performance of clutter suppression and target detection.

抄録全体を表示

In this paper, we propose the Active Net model and its application to region extraction. Active Net is a simulated two dimensional elastic network model which minimizes the energy functional and therefore shows both active and dynamic behavior. Its energy functional consists of the internal strain energy of the net and the image energy which attracts the net to features in a target region. The net deforms to wrap the region as the energy functional is minimized. We can apply Active Net to the extraction of a textured region. Some experimental results are also described.

抄録全体を表示

Plastic pollution is a pressing global environmental challenge. Encouraging sustainable behavior among consumers is essential for solving this problem. This study investigated the role of faith in shaping behaviors to mitigate plastic pollution in Japan and Malaysia. In this study, faith was translated into perceived faith connections with the environment, culture, cleanliness, and solutions to understand its universal role in pro-environmental behaviors regardless of the type and depth of religious beliefs. An online questionnaire survey was administered to university students (Japan, N = 157; Malaysia, N = 192), and the psychological processes involved in plastic pollution mitigation were examined based on the theory of planned behavior. A factor analysis of faith connections revealed the relevance of faith as one of three factors. The results of the structural equation modeling showed that the relevance of faith was positively associated with behavior in both countries, while the primary significant paths were distinctive. Positive attitudes toward behavioral change and perceived behavioral control (i.e., opportunities) are critical psychological factors encouraging behaviors in Japan and Malaysia. These findings suggest that faith plays a pivotal role in shaping pro-environmental intentions and behaviors and that structural and psychological barriers must be overcome.

抄録全体を表示

In this paper, we presented a novel method for automatic pronunciation quality assessment. Unlike the popular “Goodness of Pronunciation” (GOP) method, this method does not map the decoding confidence into pronunciation quality score, but differentiates the different pronunciation quality utterances directly. In this method, the student's utterance need to be decoded for two times. The first-time decoding was for getting the time points of each phone of the utterance by a forced alignment using a conventional trained acoustic model (AM). The second-time decoding was for differentiating the pronunciation quality for each triphone using a specially trained AM, where the triphones in different pronunciation qualities were trained as different units, and the model was trained in discriminative method to ensure the model has the best discrimination among the triphones whose names were same but pronunciation quality scores were different. The decoding network in the second-time decoding included different pronunciation quality triphones, so the phone-level scores can be obtained from the decoding result directly. The phone-level scores were combined into the sentence-level scores using maximum entropy criterion. The experimental results shows that the scoring performance was increased significantly compared to the GOP method, especially in sentence-level.

抄録全体を表示

This paper presents our investigation into improving the performance of our previous automatic reading quality assessment system. The method of the baseline system is calculating the average value of the Phone Log-Posterior Probability (PLPP) of all phones in the voice to be assessed, and the average value is used as the reading quality assessment feature. In this paper, we presents three improvements. First, we cluster the triphones, and then calculate the average value of the normalized PLPP for each classification separately, and use this average values as the multi-dimensional assessment features instead of the original one-dimensional assessment feature. This method is simple but effective, which made the score difference of the machine scoring and manual scoring decrease by 30.2% relatively. Second, in order to assess the reading rhythm, we train Gaussian Mixture Models (GMM), which contain the information of each triphone's relative duration under standard pronunciation. Using the GMM, we can calculate the probability that the relative duration of each phone is conform to the standard pronunciation, and the average value of the probabilities is added to the assessment feature vector as a dimension of feature, which decreased the score difference between the machine scoring and manual scoring by 9.7% relatively. Third, we detect Filled Pauses (FP) by analyzing the formant curve, and then calculate the relative duration of FP, and add the relative duration of FP to the assessment feature vector as a dimension of feature. This method made the score difference between the machine scoring and manual scoring be further decreased by 10.2% relatively. Finally, when the feature vector extracted by the three methods are used together, the score difference between the machine scoring and manual scoring was decreased by 43.9% relatively compared to the baseline system.

抄録全体を表示

Deep level emotion and its visualization method are proposed for multi-agent communication such as humans-robots communication, where customized learning knowledge of an observed agent is used with the observed input information from Kinect. It aims to realize agent dependent emotion by utilizing special customized knowledge of the agent, rather than ordinary surface level emotion by using visual/acoustic/distance information without any customized knowledge and visualizing the final emotion in shape-color-size diagram. In the experiment employing special demonstration scenarios where a company employee's emotion is understood by a secretary eye robot equipped with Kinect, it is confirmed that the proposed method provides deep level emotion different from ordinary surface level emotion . The proposal is being planned to be applied to a part of the emotion module in the demonstration experiments of an ongoing robotics research project entitled "Multi-Agent Fuzzy Atmosfield".

抄録全体を表示

This letter proposes a reliable depth-image-based rendering (DIBR) technique, which can provide users with reliable 3D services even if contents contain unknown or inappropriate values in some parameters. For the reliable DIBR, this letter presents how to set rendering parameters such as far/near clipping plane, convergence distance, and scale factor by taking mobile display size into consideration. Various experiments show that the proposed method can be effectively employed for 3D services based on DIBR over terrestrial-digital multimedia broadcasting (T-DMB).

抄録全体を表示

Sparse representation has been studied within the field of signal processing as a means of providing a compact form of signal representation. This paper introduces a sparse representation based framework named Sparse Probabilistic Linear Discriminant Analysis in speaker recognition. In this latent variable model, probabilistic linear discriminant analysis is modified to obtain an algorithm for learning overcomplete sparse representations by replacing the Gaussian prior on the factors with Laplace prior that encourages sparseness. For a given speaker signal, the dictionary obtained from this model has good representational power while supporting optimal discrimination of the classes. An expectation-maximization algorithm is derived to train the model with a variational approximation to a range of heavy-tailed distributions whose limit is the Laplace. The variational approximation is also used to compute the likelihood ratio score of all trials of speakers. This approach performed well on the core-extended conditions of the NIST 2010 Speaker Recognition Evaluation, and is competitive compared to the Gaussian Probabilistic Linear Discriminant Analysis, in terms of normalized Decision Cost Function and Equal Error Rate.

抄録全体を表示

In this paper, a logarithmic adaptive quantization projection (LAQP) algorithm for digital watermarking is proposed. Conventional quantization index modulation uses a fixed quantization step in the watermarking embedding procedure, which leads to poor fidelity. Moreover, the conventional methods are sensitive to value-metric scaling attack. The LAQP method combines the quantization projection scheme with a perceptual model. In comparison to some conventional quantization methods with a perceptual model, the LAQP only needs to calculate the perceptual model in the embedding procedure, avoiding the decoding errors introduced by the difference of the perceptual model used in the embedding and decoding procedure. Experimental results show that the proposed watermarking scheme keeps a better fidelity and is robust against the common signal processing attack. More importantly, the proposed scheme is invariant to value-metric scaling attack.

抄録全体を表示

Mel Frequency Cepstral Coefficients (MFCC) are the most popular acoustic features used in automatic speech recognition (ASR), mainly because the coefficients capture the most useful information of the speech and fit well with the assumptions used in hidden Markov models. As is well known, MFCCs already employ several principles which have known counterparts in the peripheral properties of human hearing: decoupling across frequency, mel-warping of the frequency axis, log-compression of energy, etc. It is natural to introduce more mechanisms in the auditory periphery to improve the noise robustness of MFCC. In this paper, a k-nearest neighbors based frequency masking filter is proposed to reduce the audibility of spectra valleys which are sensitive to noise. Besides, Moore and Glasberg's critical band equivalent rectangular bandwidth (ERB) expression is utilized to determine the filter bandwidth. Furthermore, a new bandpass infinite impulse response (IIR) filter is proposed to imitate the temporal masking phenomenon of the human auditory system. These three auditory perceptual mechanisms are combined with the standard MFCC algorithm in order to investigate their effects on ASR performance, and a revised MFCC extraction scheme is presented. Recognition performances with the standard MFCC, RASTA perceptual linear prediction (RASTA-PLP) and the proposed feature extraction scheme are evaluated on a medium-vocabulary isolated-word recognition task and a more complex large vocabulary continuous speech recognition (LVCSR) task. Experimental results show that consistent robustness against background noise is achieved on these two tasks, and the proposed method outperforms both the standard MFCC and RASTA-PLP.

抄録全体を表示

In this letter, we explored the usage of spatio-temporal information in one unified framework to improve the performance of multichannel speech recognition. Generalized cross correlation (GCC) is served as spatial feature compensation, and an attention mechanism across time is embedded within long short-term memory (LSTM) neural networks. Experiments on the AMI meeting corpus show that the proposed method provides a 8.2% relative improvement in word error rate (WER) over the model trained directly on the concatenation of multiple microphone outputs.

抄録全体を表示

Recently, there has been a growing focus on wellbeing. However, what constitutes wellbeing varies from person to person. In a diverse society, it’s important for everyone to consider their own and others’ wellbeing and show mutual respect. The authors have developed a tool called “Cards for Collective Wellbeing” to facilitate awareness and sharing of values related to wellbeing. This paper reports an experiment comparing the effects of using the cards and free recall in a workshop in which participants identified their values and shared them with others, including the results of the effects on self-

and of others.

抄録全体を表示