Recognition of consonant-vowel (CV) utterances in Indian languages is a challenging task because of the large number of classes and the high confusability among several classes. Modular approach based on artificial ne...
详细信息
Recognition of consonant-vowel (CV) utterances in Indian languages is a challenging task because of the large number of classes and the high confusability among several classes. Modular approach based on artificial neural network models is considered for recognition of CV utterances. In this approach, the large number of classes is divided into sub-groups and a separate network is trained for each subgroup. Three different grouping criteria are considered and the performance of modular networks based on these criteria is studied. An improved performance is obtained by combining evidence from the three modular networks. Because of similarities among several classes, the class of a test utterance may not always have the strongest evidence. However, it may be among a small set of alternative classes with strong evidence. We propose to train another neural network to further discriminate among these classes and reorder the alternatives. A significant increase in the performance is obtained by using the reordering network as a postprocessor for recognition of isolated utterances of 65 CV classes in Indian languages.
Data abstraction, information hiding, inheritance, and dynamic binding are four basic elements used in designing object-oriented (00) programs. These four mechanisms enable programmers to write object-oriented program...
Data abstraction, information hiding, inheritance, and dynamic binding are four basic elements used in designing object-oriented (00) programs. These four mechanisms enable programmers to write object-oriented programs that are easily extended, reused and maintained. Software implemented by 00 methodologies has increased software quality and productivity.1 However, this cannot be guaranteed without an appropriate software testing approach. Traditional testing approaches cannot be applied directly to the testing of 00 programs due to the 00 characteristics used in these programs. 00 program testing strategies and methods are proposed continually, but fewer 00 testing tools with flexibility are discussed.
Current audio coding standards, such as MPEG-I layers 1-3, MPEG-II layers 1-4, MPEG-IV, and AC-3, have adopted for comp.ession various forms of the cosine modulated filter banks (CMFBs). It is demonstrated that all mo...
详细信息
Current audio coding standards, such as MPEG-I layers 1-3, MPEG-II layers 1-4, MPEG-IV, and AC-3, have adopted for comp.ession various forms of the cosine modulated filter banks (CMFBs). It is demonstrated that all modulated cosine transforms can be divided into two modules - the permutation and discrete cosine transform. The derived discrete cosine transforms are either type II, type III, or type IV. A fast comp.ting algorithm is proposed that will comp.te all three types of transforms uniformly. The new fast algorithm has good features with regard to regularity, comp.exity, and general applicability.
Most of public-key cryptosystems rely on one-way functions, which can be used to encrypt and sign messages. Their encryption and signature operations are based on the comp.tation of exponentiation. Recently, some publ...
详细信息
Most of public-key cryptosystems rely on one-way functions, which can be used to encrypt and sign messages. Their encryption and signature operations are based on the comp.tation of exponentiation. Recently, some public-key cryptosystems are proposed and based on Lucas functions, and the Lucas sequences are performed as S = V(d)modN. In this paper, we will transform the concept of addition chains for comp.ting the exponentiation evaluations to the Lucas chains for comp.ting the Lucas sequences. Theoretically, the shorter Lucas chain for d is generated, the less comp.tation time for evaluating the value V(d) is required. Therefore, we proposed a heuristic algorithm for evaluating a shorter Lucas chain and then use it to comp.te the Lucas sequence with less modular multiplications.
Recent work on optical microcavities is concisely reviewed in this article. Optical microcavities can be fabricated using III-V comp.und semiconductors by introducing a large periodic refractive-index variation along ...
详细信息
Recent work on optical microcavities is concisely reviewed in this article. Optical microcavities can be fabricated using III-V comp.und semiconductors by introducing a large periodic refractive-index variation along the axis of GaAs waveguides. Strong optical confinement in the microcavities is achieved by oxidizing the underlying III-V material and further optimized by preferentially etching away the oxide to form an air-bridge structure. The III-V thermal oxide is obtained from the wet thermal oxidation of high Al-content AlxGa1-xAs material. The critical role of the thermal oxide in the design and fabrication of the waveguide microcavities is outlined and the characterization of the properties of the thermal oxide salient to the fabrication of the devices is described. Finally, preliminary transmission data demonstrating the successful realization of these microcavities are presented.
"Digital Museum/Library" is an ongoing project sponsored by the National sci.nce Council of Taiwan. The content of this paper is partial result of one of its subprojects called "Image content-based retr...
详细信息
"Digital Museum/Library" is an ongoing project sponsored by the National sci.nce Council of Taiwan. The content of this paper is partial result of one of its subprojects called "Image content-based retrieval on a fish database of Taiwan." The objective of this project is to develop an image content-based search engine which can perform identity check of a fish. It is well known that conventional fish databases can only be retrieved by text-based query. In this study we shall use shape, color, and other features extracted from a fish to search the fish database of Taiwan. The developed technique is able to perform scale, translation, and rotation invariant matching between two fishes. Currently, the database contains several hundreds of fishes. In the future, we shall enhance the capability of the search engine to deal with more than 2,000 fish species, which is the total amount of fish species along the coast of Taiwan.
Due to the non-deterministic nature of process sequence design for multi-stage design for multi-stage cold forging, various process designs are available depending on the initial billet geometry and the order of basic...
详细信息
Due to the non-deterministic nature of process sequence design for multi-stage design for multi-stage cold forging, various process designs are available depending on the initial billet geometry and the order of basic processes such as forward/backward extrusion, upsetting and trimming process. Therefore, various process sequences should be determined and comp.red to obtain an optimal solution. For this purpose, a depth-first search, a searching technique used in artificial intelligence, has been introduced in developing an expert system for multi-stage cold forging process design. As a result, process designers can select the optimal process sequence from the searched feasible solutions by estimating the values of evaluation functions that are introduced to represent the important design characteristics. In the present investigation, the distributions of the global effective strains in the final product and the forming loads required at each forging stage were selected to be controlled. In general, a more realistic process sequence should be determined by taking into account manufacturing conditions such as the number of forming stages, the forming loads, the shearing diameter of the coil, the open upsetting diameter, and the knock-out lengths of the die and punch. In this paper, a methodology of applying the searching technique for process sequence design is discussed, and the flexibility of the introduced searching technique is evaluated by generating design examples of a shaft part, a wrench and hexagonal bolts of AISI 1045.
暂无评论