We propose a new method of direct mode motion compensation for bi-directionally predicted pictures (B-pictures). The proposed direct mode system is utilized in extended multiple picture prediction, in which blocks in B-pictures are encoded by referring to previous B-pictures in addition to I- or P-pictures as forward reference pictures in a multiple picture prediction framework. The proposed direct mode is suitable for extended multiple picture prediction, since it always uses the immediately previous picture as the forward reference picture. In the simulation, our proposed method is implemented in the H.26L codec. The simulation results show that the extended multiple picture prediction employing the proposed direct mode can reduce the bit rate of B-pictures by up to nearly 13% compared to conventional multiple picture prediction under typical encoding conditions. We also evaluate the performance of the proposed direct mode with the extended multiple picture prediction under several different encoding conditions.

We propose a fully automatic method for detecting the carotid artery from volumetric ultrasound images as a preprocessing stage for building three-dimensional images of the structure of the carotid artery. The proposed detector utilizes support vector machine classifiers to discriminate between carotid artery images and non-carotid artery images using two kinds of LBP-based features. The detector switches between these features depending on the anatomical position along the carotid artery. We evaluate our proposed method using actual clinical cases. Accuracies of detection are 100%, 87.5% and 68.8% for the common carotid artery, internal carotid artery, and external carotid artery sections, respectively.

This study proposes fuzzy matrix quantization (FMQ) which is a new coding technique developed to obtain discrete symbols employed for hidden Markov models (HMM's). FMQ is a coding technique combining fuzzy vector quantization with matrix quantization. The validity of FMQ is evaluated by a speaker-independent isolated word recognition task. First, the effect of FMQ is examined when FMQ is applied to the training phase and/or recognition phase. The effects of number of training data, codebook size and codeword matrix size for recognition accuracy are investigated. And the results of the speech recognition based on HMM recognizer using FMQ technique is compared with HMM recognizers using conventional quantization methods, vector quantization and fuzzy vector quantization. As a result, FMQ is the effective coding technique for isolated word recognition on condition that codebook size is large, above all, when FMQ is applied to the training phase and training data set is small.

In this paper, we propose a fully automatic method for extracting carotid artery contours from ultrasound images based on an active contour approach. Several contour extraction techniques have been proposed to measure carotid artery walls for early detection of atherosclerotic disease. However, the majority of these techniques require a certain degree of user interaction that demands time and effort. Our proposal automatically detects the position of the carotid artery by identifying blood flow information related to the carotid artery, and an active contour model is employed that uses initial contours placed in the detected position. Our method also applies a global energy minimization scheme to the active contour model. Experiments on clinical cases show that the proposed method automatically extracts the carotid artery contours at an accuracy close to that achieved by manual extraction.

One of the major applications of contrast-enhanced ultrasound (CEUS) is lesion classification. After contrast agents are administered, it is possible to identify a lesion type from its enhancement pattern. However, CEUS image reading is not easy because there are various types of enhancement patterns even for the same type of lesion, and clear classification criteria have not yet been defined. Some studies have used conventional time intensity curves (TICs), which show the vessel dynamics of a lesion. It is possible to predict lesion type from the TIC parameters, such as the coefficients obtained by curve fitting, peak intensity, flow rate and time to peak. However, these parameters are not always provide sufficient accuracy. In this paper, we prepare 1D Haar-like features which describe intensity changes in a TIC and adopt the Adaboost machine learning technique, which eases understanding of which features are useful. Hyperparameters of weak classifiers, e.g., the step size of a Haar-like filter length and threshold for output of the filter, are optimized by searching for those parameters that give the best accuracy. We evaluate the proposed method using 36 focal splenic lesions in canines 16 of which were benign and 20 malignant. The accuracies were 91.7% (33/36) when inspected by an experienced veterinarian, 75.0% (27/36) by linear discriminant analysis (LDA) using conventional three TIC parameters: time to peak, area under curve and peak intensity, and 91.7% (33/36) using our proposed method. McNemar testing shows the p-value to be less than 0.05 between the proposed method and LDA. This result shows the statistical significance of differences between the proposed method and the conventional TIC analysis method using LDA.

The authors propose a model of the top down parser for continuous speech recognition. It utilizes a subject of an input sentence for its top down process and a preceding transition among subjects for the determination of a new subject. A task, a washing machine operation, which has five subjects are examined.

This study proposes a new training method for hidden Markov model with separate vector quantization (SVQ-HMM) in speech recognition. The proposed method uses the correlation of two different kinds of features: cepstrum and delta-cepstrum. The correlation is used to decrease the number of reestimation for two features thus the total computation time for training models decreases. The proposed method is applied to Japanese language isolated dgit recognition.

In this paper, we propose an image completion algorithm which takes advantage of the countless number of images available on Internet photo sharing sites to replace occlusions in an input image. The algorithm 1) automatically selects the most suitable images from a database of downloaded images and 2) seamlessly completes the input image using the selected images with minimal user intervention. Experimental results on input images captured at various locations and scene conditions demonstrate the effectiveness of the proposed technique in seamlessly reconstructing user-defined occlusions.