Most learning-based super-resolution methods neglect the illumination problem. In this paper we propose a novel method to combine blind single-frame super-resolution and shadow removal into a single operation. Firstly, from the pattern recognition viewpoint, blur identification is considered as a classification problem. We describe three methods which are respectively based on Vector Quantization (VQ), Hidden Markov Model (HMM) and Support Vector Machines (SVM) to identify the blur parameter of the acquisition system from the compressed/uncompressed low-resolution image. Secondly, after blur identification, a super-resolution image is reconstructed by a learning-based method. In this method, Logarithmic-wavelet transform is defined for illumination-free feature extraction. Then an initial estimation is obtained based on the assumption that small patches in low-resolution space and patches in high-resolution space share a similar local manifold structure. The unknown high-resolution image is reconstructed by projecting the intermediate result into general reconstruction constraints. The proposed method simultaneously achieves blind single-frame super-resolution and image enhancement especially shadow removal. Experimental results demonstrate the effectiveness and robustness of our method.

We have built a CLQA (Cross Language Question Answering) system for a source language with limited data resources (e.g. Indonesian) using a machine learning approach. The CLQA system consists of four modules: question analyzer, keyword translator, passage retriever and answer finder. We used machine learning in two modules, the question classifier (part of the question analyzer) and the answer finder. In the question classifier, we classify the EAT (Expected Answer Type) of a question by using SVM (Support Vector Machine) method. Features for the classification module are basically the output of our shallow question parsing module. To improve the classification score, we use statistical information extracted from our Indonesian corpus. In the answer finder module, using an approach different from the common approach in which answer is located by matching the named entity of the word corpus with the EAT of question, we locate the answer by text chunking the word corpus. The features for the SVM based text chunking process consist of question features, word corpus features and similarity scores between the word corpus and the question keyword. In this way, we eliminate the named entity tagging process for the target document. As for the keyword translator module, we use an Indonesian-English dictionary to translate Indonesian keywords into English. We also use some simple patterns to transform some borrowed English words. The keywords are then combined in boolean queries in order to retrieve relevant passages using IDF scores. We first conducted an experiment using 2,837 questions (about 10% are used as the test data) obtained from 18 Indonesian college students. We next conducted a similar experiment using the NTCIR (NII Test Collection for IR Systems) 2005 CLQA task by translating the English questions into Indonesian. Compared to the Japanese-English and Chinese-English CLQA results in the NTCIR 2005, we found that our system is superior to others except for one system that uses a high data resource employing 3 dictionaries. Further, a rough comparison with two other Indonesian-English CLQA systems revealed that our system achieved higher accuracy score.

This paper discusses design of Generalized Predictive Control (GPC) scheme. GPC is designed in two cases; the first is a dual-rate (DR) system, where the sampling interval of a plant output is an integer multiple of the holding interval of a control input, and the second is a fast-rate single-rate (FR-SR) system, where both the holding and sampling intervals are equal to the holding interval of the DR system. Furthermore, the relation between them is investigated, and this study gives the conditions that FR-SR and DR GPC become equivalent. To this end, a future reference trajectory of DR GPC is rewritten, and a future predictive output of the FR-SR GPC is rearranged.

The objective of this paper is to analyze a pull type multi-product, multi-line and multi-stage flexible manufacturing system whose resources are subject to planned and unplanned breakdown conditions. To ensure a continual supply of the finished products, under breakdown conditions, parts/materials flow through alternate routes exhibiting routing flexibility. The machine resources are flexible in this study and are capable of producing more than one item. Every machining and assembly station has been equipped with automated inspection units to ensure the quality of the products. The system is modelled through coloured Petri net methodology and the impact of input factors have been shown on the performance of the system. The study has been extended to explore near-optimal conditions of the system using design of experiment and response surface methods.

This paper presents a middleware system for multi-agents on a distributed system as a general test-bed for bio-inspired approaches. The middleware is unique to other approaches, including distributed object systems, because it can maintain and migrate a dynamic federation of multiple agents on different computers. It enables each agent to explicitly define its own deployment policy as a relocation between the agent and another agent. This paper describes a prototype implementation of the middleware built on a Java-based mobile agent system and its practical applications that illustrates the utility and effectiveness of the approach in real distributed systems.

The support vector machine has received wide acceptance for its high generalization ability in real world classification applications. But a drawback is that it uniquely classifies each pattern to one class or none. This is not appropriate to be applied in classification problem involves overlapping patterns. In this paper, a novel multi-model classifier (DR-SVM) which combines SVM classifier with kNN algorithm under rough set technique is proposed. Instead of classifying the patterns directly, patterns lying in the overlapped region are extracted firstly. Then, upper and lower approximations of each class are defined on the basis of rough set technique. The classification operation is carried out on these new sets. Simulation results on synthetic data set and benchmark data sets indicate that, compared with conventional classifiers, more reasonable and accurate information about the pattern's category could be obtained by use of DR-SVM.

This work is concerned with the problem of robot localization using standard RFID tags as landmarks and an RFID reader as a landmark sensor. A main advantage of such an RFID-based localization system is the availability of landmark ID measurement, which trivially solves the data association problem. While the main drawback of an RFID system is its low spatial accuracy. The result in this paper is an improvement of the localization accuracy for a standard short-range RFID sensor. One of the main contributions is a proposal of a machine learning approach in which multiple classifiers are trained to distinguish RFID-signal features of each location. Another contribution is a design tool for tag arrangement by which the tag configuration needs not be manually designed by the user, but can be automatically recommended by the system. The effectiveness of the proposed technique is evaluated experimentally with a real mobile robot and an RFID system.

The presented unified data/instruction cache design uses multiple banks and features 4 ports, distributed crossbar, different word-length for data and instruction ports, interleaved cache-line words and synchronous access with hidden precharge. A 20.5 KByte storage capacity is integrated in 5-metal-layer CMOS logic technology with 200 nm minimum gate length and a 3.4 ns access-cycle time is achieved. The access bandwidth corresponds to 10 ports with standard word-length, while the cost in increased Si-area is only 25% in comparison to a 1-port cache.

Content delivery network improves end-user performance by replicating Web contents on a group of geographically distributed sites interconnected over the Internet. However, with the development whereby content distribution systems can manage dynamically changing files, an important issue to be resolved is consistency management, which means the cached replicas on different sites must be updated if the originals change. In this paper, based on the analytical formulation of object freshness, web access distribution and network topology, we derive a novel algorithm as follows: (1) For a given content which has been changed on its original server, only a limited number of its replicas instead of all replicas are updated. (2) After a replica has been selected for update, the latest version will be sent from an algorithm-decided site instead of from its original server. Simulation results verify that the proposed algorithm provides better consistency management than conventional methods with the reduced the old hit ratio and network traffic.

Broadcast and synchronization techniques are used for cache coherence control in conventional larger scale snoop-based SMP systems. The penalty for synchronization is directly proportional to system size. Meanwhile, advances in LSI technology now enable placing a memory controller on a CPU die. The latency to access directly linked memory is drastically reduced by an on-die controller. Developing an enterprise server system with these CPUs allows us an opportunity to achieve higher performance. Though the penalty of synchronization is counted whenever a cache miss occurs, it is necessary to improve the coherence method to receive the full benefit of this effect. In this paper, we demonstrate a coherence directory organization that fits into DSM enterprise server systems. Originally, a directory-based method was adopted in high performance computing systems because of its huge scalability in comparison with snoop-based method. Though directory capacity miss and long directory access latency are the major problems of this method, the relaxed scalability requirement of enterprise servers is advantageous to us to solve these problems along with an advanced LSI technology. Our proposed directory solves both problems by implementing a full bit vector level map of the coherence directory on an LSI chip. Our experimental results validate that a system controlled by our proposed directory can surpass a snoop-based system in performance even without applying data localization optimization to an online transaction processing (OLTP) workload.

Petri nets are known as a modeling language for concurrent and distributed systems. In recent years, various object-oriented Petri nets were proposed, and we are proposing a kind of object-oriented Petri nets, called multi agent nets (MANs). In this letter, we consider the reachability analysis of MANs. We propose an algorithm for generating an abstract state space of a multi agent net, and report results of computational experiments.

The support vector machines (SVMs) are one of the most effective classification techniques in several knowledge discovery and data mining applications. However, a SVM requires the user to set the form of its kernel function and parameters in the function, both of which directly affect to the performance of the classifier. This paper proposes a novel method, named a kernel-tree, the function of which is composed of multiple kernels in the form of a tree structure. The optimal kernel tree structure and its parameters is determined by genetic programming (GP). To perform a fine setting of kernel parameters, the gradient descent method is used. To evaluate the proposed method, benchmark datasets from UCI and dataset of text classification are applied. The result indicates that the method can find a better optimal solution than the grid search and the gradient search.

In mobile computing environments, cache invalidation techiniques are widely used. However, theses techniques require a large-sized invalidation report and show low cache utilization under high server update rate. In this paper, we propose a new cache-level cache invalidation technique called TTCI (Timestamp Tree-based Cache Invalidation technique) to overcome the above two problems. TTCI also supports selective tuning for a cache-level cache invalidation. We show in our experiment that our technique requires much smaller size of cache invalidation report and improves cache utilization.

Machine learning techniques have long been applied in many fields and have gained a lot of success. The purpose of learning processes is generally to obtain a set of parameters based on a given data set by minimizing a certain objective function which can explain the data set in a maximum likelihood or minimum estimation error sense. However, most of the learned parameters are highly data dependent and rarely reflect the true physical mechanism that is involved in the observation data. In order to obtain the inherent knowledge involved in the observed data, it is necessary to combine physical models with learning process rather than only fitting the observations with a black box model. To reveal underlying properties of human speech production, we proposed a learning process based on a physiological articulatory model and a coarticulation model, where both of the models are derived from human mechanisms. A two-layer learning framework was designed to learn the parameters concerned with physiological level using the physiological articulatory model and the parameters in the motor planning level using the coarticulation model. The learning process was carried out on an articulatory database of human speech production. The learned parameters were evaluated by numerical experiments and listening tests. The phonetic targets obtained in the planning stage provided an evidence for understanding the virtual targets of human speech production. As a result, the model based learning process reveals the inherent mechanism of the human speech via the learned parameters with certain physical meaning.

Several approaches have been studied to cope with the exceptional features of non-segmented languages. When there is no explicit information about the boundary of a word, segmenting an input text is a formidable task in language processing. Not only the contemporary word list, but also usages of the words have to be maintained to cover the use in the current texts. The accuracy and efficiency in higher processing do heavily rely on this word boundary identification task. In this paper, we introduce some statistical based approaches to tackle the problem due to the ambiguity in word segmentation. The word boundary identification problem is then defined as a part of others for performing the unified language processing in total. To exhibit the ability in conducting the unified language processing, we selectively study the tasks of language identification, word extraction, and dictionary-less search engine.

The design aspects of the bulk InGaAsP semiconductor optical amplifier integrated Mach-Zehnder interferometer (SOA-MZI) optimized for 40 Gbps-NRZ all optical wavelength conversion are described. The dimensions of the SOA active waveguide have been optimized for fast gain recovery by maximizing the gain and adjusting the wavelength-converted NRZ waveforms. Submicron-width buried heterostructure (BH) SOA waveguides were fabricated successfully and showed little leakage current. The experimental wavelength-converted optical waveform agreed well to the numerical simulations, and mask-compliant 40 G-NRZ wavelength-converted waveform was obtained by the optimized SOA-MZI. 40 G-NRZ full C-band operation and polarization-insensitive operation of SOA-MZI were also achieved.

This paper describes a new criterion for speech recognition using an integrated confidence measure to minimize the word error rate (WER). The conventional criteria for WER minimization obtain the expected WER of a sentence hypothesis merely by comparing it with other hypotheses in an n-best list. The proposed criterion estimates the expected WER by using an integrated confidence measure with word posterior probabilities for a given acoustic input. The integrated confidence measure, which is implemented as a classifier based on maximum entropy (ME) modeling or support vector machines (SVMs), is used to acquire probabilities reflecting whether the word hypotheses are correct. The classifier is comprised of a variety of confidence measures and can deal with a temporal sequence of them to attain a more reliable confidence. Our proposed criterion for minimizing WER achieved a WER of 9.8% and a 3.9% reduction, relative to conventional n-best rescoring methods in transcribing Japanese broadcast news in various environments such as under noisy field and spontaneous speech conditions.

The watershed transform has been used as a powerful morphological segmentation tool in a variety of image processing applications. This is because it gives a good segmentation result if a topographical relief and markers are suitably chosen for different type of images. This paper proposes a parallel implementation of the watershed transform on the cellular neural network (CNN) universal machine, called cellular watersheds. Owing to its fine grain architecture, the watershed transform can be parallelized using local information. Our parallel implementation is based on a simulated immersion process. To evaluate our implementation, we have experimented on the CNN universal chip, ACE16k, for synthetic and real images.

This paper presents a dual data cache system structure, called a cooperative cache system, that is designed as a low power cache structure for embedded processors. The cooperative cache system consists of two caches, i.e., a direct-mapped temporal oriented cache (TOC) and a four-way set-associative spatial oriented cache (SOC). The cooperative cache system achieves improvement in performance and reduction in power consumption by virtue of the structural characteristics of the two caches designed inherently to help each other. An evaluation chip of an embedded processor having the cooperative cache system is manufactured by Samsung Electronics Co. with 0.25 µm 4-metal process technology.

To control various access privileges in group-oriented applications having multiple data streams, we present a novel reactive key management scheme where each member can obtain the key of a data stream from public parameters only when necessary. Compared with the previous schemes, this scheme significantly reduces the amount of rekey messages for dynamic membership change due to its reactive nature.