Abstract: A laser apparatus having a laser that generates at least one laser beam of a fundamental wavelength, a diffractive optical element that converts the laser beam into a plurality of light beams at the fundamental wavelength, and a harmonic crystal that converts the plurality of light beams at the fundamental wavelength into respective light beams having respective wavelengths which are harmonically related to the fundamental wavelength. By splitting the laser beam into a plurality of beams before converting to harmonic wavelengths using the harmonic crystal, the crystal volume is more efficiently used and the thermal stress on the crystal is reduced. In an alternative embodiment, the invention provides for the use of a plurality of harmonic crystals in series to convert the wavelength to shorter harmonics.

Abstract: Apparatus and method for soldering a first material to a second material uses a laser and involves applying solder paste over the first material, placing the second material on the solder paste in alignment with the first material to form a sample, and focusing a laser beam onto a predetermined location on the sample to segment the solder paste and form a solder joint between the first material and the second material. During the soldering, optical images of at least one region-of-interest in the solder paste adjacent the solder joint are captured. Formation of the solder joint is monitored by repeatedly calculating pixel value sums in the region-of-interest from the optical images to determine whether the solder paste in the region-of-interest has segmented. Control signals may be generated based on the pixel value sums and used to adjust at least one of the intensity and duration of the laser beam.

Abstract: A circuit which restores portions of an input signal lost due to clipping distortion includes a comparator which locates the portions of the input signal which have been subject to clipping distortion and generates a pulse signal where each pulse corresponds to a clipped portion of the input signal. This pulse signal is used to generate parabolic pulses which are combined with the input signal to generate a modified signal that approximates the input signal without the clipping distortion. One circuit generates the pulse signals by controlling an oscillator to produce a parabolic pulse train which changes in frequency and amplitude to match each pulse of the pulse signal and then selecting one of the parabolic pulses for each pulse of the pulse signal. Another circuit generates the parabolic pulses by doubly integrating the pulses of the pulse signal.

Abstract: An apparatus for distributing control menus over a multi-channel distribution network such as a CEBus throughout a home is disclosed. The apparatus includes a controller such as a personal computer for centrally generating control menus as video information on available channels of the multi-channel distribution network. The controller also executes commands received over the distribution network. The apparatus further includes a display device such as a TV for receiving and displaying the control menu on the channel selected by the control means and a receiver, located proximate to the display device, for receiving user commands related to the displayed menu and for delivering the user commands, via the distribution network, to the control processor. User commands can be given using, for example, infra-red transmission or voice recognition circuitry.

Abstract: A method for managing the power consumed by a disk drive in a portable laptop computer which includes spinning the disk up during period when the computer apparatus is in an active or idle mode, creating a threshold to determine when to spin down the disk as a function of a period of disk inactivity, spinning the disk down when the threshold is exceeded in order to reduce the power consumption of the disk, automatically increasing the threshold when an undesirable spin up of the disk has occurred and automatically decreasing the threshold when an acceptable spin up of the disk occurs. By virtue of this method, the threshold for disk inactivity is continually monitored and adjusted to maintain a balance between energy consumption and undesirable disk spin down.

Abstract: A video camera generates an approximation of the dark current or black shading distortion signals based on scan position in the image, and subtracts the approximation signal from the signal generated by the camera during normal image-sensing operation to reduce the black shading distortion. In a particular embodiment of the invention, counters generate location signals representative of the location of pixel currently being read, and the location signals are processed by functions such as squaring or raising to a constant power, and by weighting, to produce components of the approximation of the dark signal. The components are subtracted from the imager signal, to reduce the dark shading. In one embodiment of the invention, the functions are automatically selected from among a plurality of preselected functions, and implemented by look-up tables.

Abstract: A Node "0" device for a Consumer Electronics Coaxial Bus (CXBUS) network is provided. The network has a plurality of internal cables coupled to a source of up-link signals located within a first frequency band. The up-link signals include first audio/video (A/V) signals (which may be digital or analog) and control signals. External cables provide second A/V signals located within a second frequency band. An active input section is coupled to the internal cables. The input section individually filters and amplifies the up-link signals and combines the amplified signals. The input section provides a down-link signal path. A block converter coupled to the input section converts the first A/V signal from the first frequency band to the second frequency band, and provides the converted signals at its output port. A control channel regenerator coupled to the input section receives the control signal, generates a down-link control signal, and transmits the down-link control signal to the input section.

Abstract: A method for managing the power consumed by a disk drive in a portable laptop computer which includes quantizing predetermined periods of disk inactivity into states which are stored in a state table in memory. Based upon a history of disk accesses by a user, the number of transitions between each pair of states is counted and stored in memory. In view of this history, a future period of disk inactivity can be predicted and said prediction is compared with a threshold value. If the predicted period of disk inactivity is greater than the threshold value, the computer is automatically placed in a low power mode by spinning down the disk. If not, the disk continues to spin.

Abstract: A CCD imager is used in a television camera which generates signals for use with a particular television standard, such as the proposed U.S. HDTV production standard. The imager includes a light-sensitive A register which generates signal in the form of charge during an integrating interval, a B charge storage register into which the charge is transferred during a pull-down interval, and a horizontal shift register coupled to the last row of the B register, into which the charges are simultaneously coupled, and serially read. The video signals produced by the imager are subject to distortion caused by incomplete charge transfer in the pull-down interval and crosstalk noise resulting from gating the clock signals to the horizontal shift register on and off. Vertical transfer distortion is reduced by reading the horizontal lines from the horizontal shift register with reduced intervals between the active portions, which allows the pull-down interval to be increased.

Abstract: An apparatus for transmitting data from a first device to a second device is provided. The invention is suitable for use in a system having first and second communications media. The apparatus includes a network interface coupled to the first medium for receiving and examining data from the first device. A clustering facility is provided for coupling the first device directly to the network interface and for selectively coupling the second device to the network interface. The network interface identifies the second device as being coupled to either the clustering facility or the second medium. The apparatus transmits the data from the network interface to the second device by way of the clustering facility, if the network interface identifies the second device as being coupled to the clustering facility. The apparatus routes the data from the network interface to the second device by way of the second medium if the network interface identifies the second device as being coupled to the second medium.

Abstract: A system and method for the transmission and recovery of asynchronous data. For instantaneous synchronization, the receiver is equipped with a high frequency timing base which has a far higher frequency than either the data-generating or transmitting rate. The receiver clock is instantly synchronized upon detection of the first transition of the incoming data packet. Data packet verification can be conducted and data processed with minimal loss of data.

Abstract: A content addressable memory (CAM) system, is disclosed for writing data including a key portion and an extended portion, and for reading the extended portion in response to the match of a search key a key portion of the data. The system includes a conventionally addressed RAM, organized into segments, each segment of which is identified by a starting address. It also includes a CAM, also organized into words. Each of the words of the CAM is identified by a CAM address. Data to be written into the memory includes a key portion and an extended data portion. The key portion of the data is written into the CAM at a selected address, and the extended data portion is written into a segment of the random access memory which begins at an address determined from the selected address. Data in the random access memory is read by searching the CAM for a search key and then using address of matching entries found in the CAM to generate address values for the random access memory.

Abstract: An edge finder for an image processing system locates edges in a digitized image over several different ranges of pixel values and assigns, to each pixel value an edge characteristic value which represents a minimum edge value for the pixel in the image. The exemplary edge finder that is able to resolve transitions occurring between two pixels or over groups of four or eight pixels to determine whether an edge should be defined between the original two pixels. The edge finder simultaneously examines four edges in two dimensions to determine the edgeness value for each pixel in the digitized image.

Abstract: An apparatus is provided for processing the output signals of a charge transfer device having at least one row of photoelectric elements. The charge transfer device produces output signals with charge levels that are provided to a shift register having a plurality of elements. The successive charge levels from each photoelectric element are provided to respective shift register elements, and the charge levels are transferred within the shift register to a readout terminal. Some of the charge is retained by each element as the charge is transferred. The retained charge is added to the next charge transferred into the element. This adding of residual charges averages the charges transferred, producing undesirable noise. A sample and hold device removes sampling artifacts inherent in the signal provided by the CCD, to produce a signal representing successive image pixel values.

Abstract: A clocking circuit and method for phasing a signal with an unknown phase to a clock signal includes a time requirement duration (TRD) signal which has two states and the duration of one of the states is indicative of a combined setup and hold time requirement for the storage device intended to capture the unknown phase signal. The TRD signal is then clocked as a function of the unknown phase signal and called a toggle clock phase (TCP) signal. The TCP signal is then used to modify the phase of a clock enable for the storage device signal to ensure adequate setup and hold time for the clocking of the unknown phase signal.

Abstract: A noise suppression, signal recovery circuit for use with an array of charge coupled device having a plurality of phase readouts. The video information present in the multiple phases are combined into a single video output signal by using a minimal number of parts while providing for noise suppression designed to suppress typical noise present in charge coupled device arrays.

Abstract: A method for polishing a substrate having at least one micro-sized structure. The method includes identifying a first region of the substrate on which a micro-sized structure is to be located. The first region is the region in which polishing is desired. A second region of the substrate, in which polishing is not desired, is also identified. An adhesion promoter is optionally applied to the substrate. The second region of the substrate is coated with a selected coating material that does not degrade substantially when exposed to a selected electrolyte. Material is removed from the first region, exposing a micro-sized structure. The coating material may be removed by the same machining process that forms the micro-sized structure. The substrate is submerged in the selected electrolyte so that the first region is exposed to the electrolyte. The first region of the substrate is electropolished. The coating is then optionally removed.

Abstract: A signal processing system for a video camera uses a single neural network to implement multiple nonlinear signal processing functions. In one example, the neural network implements gamma correction and contrast compression, in another example, color correction and aperture correction are added to the combined function emulated by the network. The network is trained using back propagation to emulate one function then a combination of two functions, then a combination of three functions, and so on. The programmed neural network replaces multiple pipelined signal processors in the video camera. The use of a single neural network in place of the multiple dedicated processing functions reduces engineering effort to develop the product and may reduce the cost of the total system.

Abstract: A signal processing system for a video camera uses a single neural network to implement multiple nonlinear signal processing functions. In one example, the neural network implements gamma correction, contrast compression, color correction, high pass filtering and aperture correction as a combined function which is emulated by the network. The network is trained off-line using back propagation to emulate the entire composite function for a set of parameters which results in multiple sets of weighting factors. Then, using the stored multiple sets of weighting factors as initial values, the neural network is "re-trained" on-line for each new parameter setting. The use of a single neural network in place of the multiple dedicated processing functions reduces engineering effort to develop the product and may reduce the cost of the total system.

Abstract: Apparatus and a method are described for modifying the magnification and registration factors of an image using digital signal processing and interpolation. The apparatus is used in a camera system which samples an image to generate sampled data signals. The sampled data signals are first interpolated across the either the rows or columns of the image and then across the columns or rows of the image to change the magnification and registration factors of the images represented by each of the sampled data signals to some common desired magnification and registration factors. The system electronically corrects for chromatic aberration and geometric distortion in the image caused by the lens system and skew among the images caused by misalignment of the respective color image sensors.