Abstract: A leak current is prevented from flowing during a power off state in a power supply unit having a remote power on/off function by using the momentary switch. In a state of power off, i.e. a state in which a switch SW2 is turned off, power is not applied to a latch circuit 30, and hence power is not consumed. A switch drive circuit 40 does not consume power. When a user depresses a momentary switch SW1 during a state that the switch SW2 is turned off, the switch drive circuit 40 turns the switch SW2 on. As a result, the latch circuit 30 can receive a drive voltage VCC from a node located down stream of the switch SW2, so that it can store or latch the turn on state of the momentary switch SW1. After that, when the user releases the momentary switch SW1, it returns to an original position and operation of the switch drive circuit 40 which turns the switch SW2 on is terminated.

Abstract: A user's burden is decreased when transferring a file via a network by simplifying the directory changing operation at each node. FTP-C-S/W 114 displays a current directory and a directory/file included therein of a local node and a remote node in a GUI panel. When the user clicks any one of the displayed directories with a mouse, the FTP-C-S/W 114 accepts the clicked directory as a directory specified by the user and selects it as a new current directory in place of the existing current directory. When a current directory of the local node is newly selected, the FTP-C-S/W 114 automatically selects a directory of the remote node corresponding to the new current directory of the local node and makes it a new current directory of the remote node to have both current directories coincide.

Abstract: Data hidden in a plurality of contents is protected from illegal access by a third party. An embedding process is performed by using a frame pattern (mark) that is composed of multiple embedded block patterns comprising several different types and wherein the embedding process is varied in accordance with the time and the frame. One embedding method mixes at random several types of frame patterns having different arrangements of block patterns. The embedding method may change at random a range in which a block pattern is embedded. An embedded mark is detected by processing in parallel a plurality of patterns that are to be detected in contents, and if as a result of detection there is a reaction, it is ascertained that a mark is present. An illegal process may be traced by using an electronic fingerprint obtained by entering a false detection mark in a detector or by using an electronic fingerprint using a random sync signal for an embedding detector.

Abstract: An icon/window which a user is selecting is precisely predicted to improve the operability of a GUI. This invention displays a GUI picture, including an icon/window and a pointer of a mouse or the like in a display screen of a computer, predicts an icon which the user is selecting based on the moving speed and the direction of movement of the pointer, and displays a tool tip or the like which indicate the function of the icon. This invention also predicts an operation which the user is effecting on the window based on the moving speed of the pointer and a distance between the pointer and a window, changes the display of the pointer from a conventional arrow head during the movement to a bidirectional arrow head appearing in changing the size of the window, for example, depending on the prediction, and further assists an operation to change the size of the window in response to clicking by the user.

Abstract: To select a graphic primitive hidden behind a graphic primitive displayed in foreground, a point on the display screen is pointed to by a pointing device. A CPU detects graphic primitives that include the specified point based on the position and the size indicated by the graphic primitive information stored in the system memory and further detects the smallest graphic primitive completely included in one or more graphic primitives among the detected graphic primitives. The detected smallest graphic primitive is thereafter processed as the selected graphic primitive.

Abstract: An efficient method and system is provided for searching for the shortest paths between a source and multiple destinations, and between multiple sources and multiple destinations. The speed of the conventional Dijkstra method, which is the basic calculation method, is increased by employing information concerning the relationship between a node and a set of destinations on a graph. The relationship information is constituted by the estimate function h(v) concerning a specific node v and a set T of destinations, where h(v) is a lower bound of all the shortest path lengths extending from node v to each of the destination sets T. Employment of the estimate function can increase the speed of the Dijkstra method.

Abstract: A method for determining optimal database materializations utilizing a query optimizer in a database management system. The method takes one or more queries as inputs and using the query optimizer in the database management system generates a series of virtual materializations by materializing some subsets of the database. The virtual materializations are used to consider the relative performance benefits, i.e. cost-benefits, for the queries based on the various virtual materializations. If the query optimizer decides to use any of the materializations in its plan, then those materializations are recommended to the user, or created automatically for the user.

Abstract: In order to make sure that a user does not attempt to build an invalid database query in which tables are incorrectly joined on fields that cannot be joined due to an incompatibility of data types, after the user selects a source data field the incompatible target data fields which cannot be joined with the selected source data field are determined automatically and displayed distinctly differently in a target list box of possible target data field names, such as in a different color.

Abstract: The improved neural network of the present invention results from the combination of a dedicated logic block with a conventional neural network based upon a mapping of the input space usually employed to classify an input data by computing the distance between said input data and prototypes memorized therein. The improved neural network is able to classify an input data, for instance, represented by a vector A even when some of its components are noisy or unknown during either the learning or the recognition phase. To that end, influence fields of various and different shapes are created for each neuron of the conventional neural network. The logic block transforms at least some of the n components (A1, . . . , An) of the input vector A into the m components (V1, . . . , Vm) of a network input vector V according to a linear or non-linear transform function F. In turn, vector V is applied as the input data to said conventional neural network.

Abstract: The invention provides an imaging system including a camera with an image sensor sensitive to light and a light emitter with the camera to transmit image data from the camera. It includes a controller to control the light emitter to illuminate a subject in the field of view of the camera.

Abstract: An apparatus for automatically generating punctuation marks in a continuous speech recognition system, comprises means (1,2,3,5) for recognizing user speech and converting the user speech into words, characterized in that means (1,2,3,5) for recognizing user speech is further used to recognize pseudo noises in the user speech; and the apparatus characterized by further comprising: means (9) for marking pseudo noises in output results of means (1,2,3,5) for recognizing user speech; means (10,14,13) for generating punctuation marks by finding most likely pseudo punctuation marks at locations of pseudo noises marked by the means (9) for marking pseudo noises based on a language model containing pseudo punctuation marks.

Abstract: A method is provided for generating a quadrilateral mesh and a hexahedral mesh. First, a geometric model, which is an object of meshing, is input. The model is either two-dimensional or three-dimensional. Multiple bubbles of a plurality of types are generated in a region of the geometric model. The bubbles are moved according to an interbubble force defined by a predetermined rule, and the bubble count is controlled to adjust a neighboring relationship between the bubbles, so that the stable allocation of the bubbles are determined. By connecting the centers of bubbles of a specific type, a mesh is generated.

Abstract: A client computer system (10) is adapted to connect to a server (16) across a network (20). The client system includes memory (25 or a storage medium) into which an operating system is loadable. The client includes a network card (12) having a PROM (14) adapted to download a bootstrap program (17) from the server. The bootstrap program (17) is in turn adapted to load an operating system image (18) over the network from the server into a location in the memory. The client then boots from the operating system image in memory, with the client system being adapted to subsequently modify the operating system image in memory, without reloading the image over the network, and re-boot the client system.

Abstract: In order to efficiently thread n chronologically ordered documents, a similarity among the n documents is calculated and the similarity is employed to create a similarity matrix using time constraints, and is converted into an adjacency matrix for identifying a relationship among the n documents. By applying this threading method that employs the time constraints, a large quantity of article data can be efficiently threaded in the O(n) order. Users can easily access a large quantity of data and can understand the contents.

Abstract: To provide a numerical integration system and method, suitable for a high-order dimensional Monte Carlo calculation and having high convergence speed without incurring a calculation time penalty, a multi-dimensional sequence is first converted into a one-dimensional value. Then the conversion is optimized in accordance with characteristics of a model. A summation calculation is corrected by using an averaged weight, instead of correcting the values of a random sequence. Then a histogram is prepared to correct the overall shape of the distribution of a random number sequence. DA sum is calculated for each group obtained when the histogram is prepared. As a result, the convergence speed is increased without sacrificing the calculation time.

Abstract: A data hiding method enables embedded information to be extracted without providing original media data as information and enables the information to be extracted with relatively low calculation cost.

Abstract: A method of manufacturing a cantilever with tip comprising the steps of forming a tip-like indent in a substrate, depositing a photoresist layer which fills said tip-like indent and covers at least a part of said substrate, and photolithographically structuring said photoresist layer to form said cantilever with tip out of said photoresist.

Abstract: To provide power management for a computer, the computer includes: a resource manager for examining a running application and a usage status of an examination object device; and a policy manger for receiving, from the resource manager, information about the running application and the usage status of the examination object device, and for selecting a job type corresponding to the information, in a plurality of job types predetermined for the running applications. The resource manager receives, from the policy manager, the selected job type, and sets a parameter of a control object device thereinto. The parameter of the control object device is predetermined for the selected job type. A job type reported by the policy manager may be a job type name or a set of operating modes of the control object device.

Abstract: An output bit sequences is derived from an initial bit sequence and this output bit sequence is used to encrypt an input bit sequence in a first mode of operation or not to so encrypt the input bit sequence in a second mode of operation. The mode of operation is switched automatically whenever the output bit sequence contains a predetermined trap bit sequence. As a result of this automatic switching between such encryption and no such encryption, unauthorized determination of secret codes is thwarted.

Abstract: An integrated manual control input device is provided for use in a computer system having a graphical user interface. The device comprises a pointing actuator mounted to a computer keyboard between preselected ones of the keys for engagement by a user's index finger. The pointing actuator includes a coil in the form of a cylinder that, when carrying a current, produces a magnetic field to drive a ferromagnetic slug to provide tactile feedback to a user of the control input device.