A Two-Channel Picture Coding System: I--Real-Time Implementation

The two-channel system previously reported has been implemented in hardware using system parameters appropriate to consumer television. The basic system divides the spectrum into lowand high-frequency spatial components. The lows are coarsely sampled and finely quantized and the highs finely sampled and coarsely quantized using a companded, randomized quantizer. The purpose of this experiment was to demonstrate the potential of the system for low-cost practical application, to study the effect of the character of the randomizing noise, and to ascertain that there were no deleterious effects due to interlace, motion, or input noise. Theoretical noise calculations were qualitatively confirmed.     

Robustifying AdaBoost by adding the naive error rate

AdaBoost can be derived by sequential minimization of the exponential loss function. It implements the learning process by exponentially reweighting examples according to classification results. However, weights are often too sharply tuned, so that AdaBoost suffers from the nonrobustness and overlearning. Wepropose a new boosting method that is a slight modification of AdaBoost. The loss function is defined by a mixture of the exponential loss and naive error loss functions. As a result, the proposed method incorporates the effect of forgetfulness into AdaBoost. The statistical significance of our method is discussed, and simulations are presented for confirmation.     

Trading variance reduction with unbiasedness: the regularized subspace information criterion for robust model selection in kernel regression

A well-known result by Stein (1956) shows that in particular situations, biased estimators can yield better parameter estimates than their generally preferred unbiased counterparts. This letter follows the same spirit, as we will stabilize the unbiased generalization error estimates by regularization and finally obtain more robust model selection criteria for learning. We trade a small bias against a larger variance reduction, which has the beneficial effect of being more precise on a single training set. We focus on the subspace information criterion (SIC), which is an unbiased estimator of the expected generalization error measured by the reproducing kernel Hilbert space norm. SIC can be applied to the kernel regression, and it was shown in earlier experiments that a small regularization of SIC has a stabilization effect. However, it remained open how to appropriately determine the degree of regularization in SIC. In this article, we derive an unbiased estimator of the expected squared error, between SIC and the expected generalization error and propose determining the degree of regularization of SIC such that the estimator of the expected squared error is minimized. Computer simulations with artificial and real data sets illustrate that the proposed method works effectively for improving the precision of SIC, especially in the high-noise-level cases. We furthermore compare the proposed method to the original SIC, the cross-validation, and an empirical Bayesian method in ridge parameter selection, with good results.     

Wind-flow measurement over the Subaru Telescope

Wind flows over the 8.2-m Subaru Telescope at Mauna Kea in Hawaii were analyzed with a correlation method. Three or four wind flows were detected from our measurements. Spatial and temporal resolution of the wind-flow analysis across the 8.2 m pupil were investigated experimentally. A three-dimensional spatiotemporal-frequency analysis was also applied to the wind-flow data.     

Polymer film-type channel electron multipliers: Their characteristics and comparison with bulk-type flexible multipliers

Polymer film-type channel electron multipliers (CEMs) were developed and their characteristics investigated in comparison with bulk-type flexible CEM. As film-type CEMs, a paint-coating-type CEM and a solvent-etching-type CEM were fabricated. The paint-coating-type CEM is made by coating a “paint” which is prepared by dissolving in a solvent an electronically conductive polymeric composition having a high secondary electron emission yield. Up to the present study the mean gain of the film-type CEM, in spite of the lower resistivity of the dynode, seems to be slightly more dependent on the count rate than that of the bulk-type CEM made of the same material. However, the convenience of fabricating the dynode film by coating the dynode-forming paint on the internal surface of any shaped tube or a flexible polymer tube would overcome the slight disadvantage mentioned above.     

Rate of emulsion polymerization of styrene

In emulsion polymerization, the Smith and Ewart theory gives about two or three times the number of polymer particles obtained by experiment. In this paper, a reaction model is proposed which, from the standpoint of reactor design, can give an adequate explanation of the whole course of an emulsion polymerization of monomer highly insoluble in water. Among other things, the generating process of polymer particles is examined in detail. It is demonstrated experimentally that a new parameter proposed here, which represents the degree of difficulty of monomer initiation in micelles, is indispensable in explaining that process. Also confirmed is that monomer initiation takes place more easily in polymer particles than in micelles. According to the new model, the progress of polymerization, i.e., monomer conversion, the number of the polymer particles, and properties of polymer thus produced can be estimated with satisfactory accuracy. Furthermore, approximate equations are derived for easier estimation.     

Synthesis of TiO2/CdS nanocomposite via TiO2 coating on CdS nanoparticles by compartmentalized hydrolysis of Ti alkoxide

Hydrolysis of Ti alkoxide in the presence of inverse micelles containing CdS nanoparticles in their hydrophilic interior results in formation of TiO₂/CdS nanocomposites in which the CdS nanoparticles are embedded in a TiO₂ matrix with a thickness of 10 nm at the surface of the particles. The primary hydroxyl groups introduced by 2-mercaptoethanol as a capping agent used for preparation of the CdS nanoparticles are considered to play an important role for successful adhesion and growth of the TiO₂ layer on the CdS surfaces. TEM observation strongly supports formation of semiconductor-in-semiconductor heteronanostructure by compartmentalized hydrolysis of Ti alkoxide within the inverse micelles in which the surface-capped CdS nanoparticles coexist.     

Real-time exploration of a multi-robot rescue system in disaster areas

A system procedure is proposed for a multi-robot rescue system that performs real-time exploration over disaster areas. Real-time exploration means that every robot exploring the area always has a communication path to human operators standing by at a base station and that the communication path is configured by ad hoc wireless networking. Real-time exploration is essential in multi-robot systems for USAR (urban search and rescue) because operators must communicate with every robot to support the victim detection process and ad hoc networking is suitable to configure a communication path among obstacles. The proposed system procedure consists of the autonomous classification of robots into search and relay types and behavior algorithms for each class of robot. Search robots explore the areas and relay robots act as relay terminals between search robots and the base station. The rule of the classification and the behavior algorithm refer to the forwarding table of each robot constructed for ad hoc networking. The table construction is based on DSDV (destination-sequenced distance vector) routing that informs each robot of its topological position in the network and other essentials. Computer simulations are executed with a specific exploration strategy of search robots. The results show that a multi-robot rescue system can perform real-time exploration with the proposed system procedure and reduce exploration time in comparison with the case where the proposed scheme is not adopted.     

Density estimation with minimization of U-divergence

This paper is concerned with density estimation based on the stagewise minimization of the U-divergence. The U-divergence is a general divergence measure involving a convex function U which includes the Kullback-Leibler divergence and the L ₂ norm as special cases. The algorithm to yield the density estimator is closely related to the boosting algorithm and it is shown that the usual kernel density estimator can also be seen as a special case of the proposed estimator. Non-asymptotic error bounds of the proposed estimators are developed and numerical experiments show that the proposed estimators often perform better than several existing methods for density estimation.