摘要:
In a light-responsive semiconductor device in which the laser oscillation state is controlled by incident light, a layer of material having a conductivity modulatable by light is integrated in a semiconductor laser structure and is responsive to the incident light for the control of the laser oscillation state of the device. In one embodiment constituting a light logic element, a plurality of semiconductor lasers are integrated on a substrate with each respective semiconductor laser having a grating along the entirety or a portion thereof and provided with a light waveguide connecting with a waveguide of an adjacent semiconductor laser. In a bistable semiconductor laser there are provided two different conductivity type light confinement layers, an active layer comprising a multiple quantum well structure which is between the two light confinement layers, a grating provided close to the quantum well active layer. The oscillation wavelength established by the grating is shorter than the peak gain wavelength produced by the quantum well active layer.
摘要:
A surface-emitting semiconductor laser device comprises a first electrode, a first-conductivity-type clad layer provided on the first electrode, an active layer provided on the first-conductivity-type clad layer, a guide layer provided on the active layer and having an even-numbered-order diffraction grating opposite to the active layer, a second-conductivity-type clad layer provided on the guide layer, contact layers provided on portions of the second-conductivity-type clad layer, and second electrodes provided on the contact layers.
摘要:
A learning method for a neural network, in which at least a portion of the interconnection strength between neurons takes discrete values, includes the steps of updating an imaginary interconnection strength taking continuous values by using the quantity of update of the interconnection strength which has been calculated by using the discrete interconnection strength, and discretizing the imaginary interconnection strength so as to provide a novel interconnection strength.
摘要:
An optical logic element includes an optical bistable npnp element for switching from a high resistance state to a low resistance state in response to electrical bias and incident light energy in which the switching time becomes shorter as the incident light energy becomes larger. The elements emit light in the low resistance on state. The optical logic element is designed for analog threshold processing of light energy. A plurality of optical bistable elements connected in parallel differentially threshold process incident light energy. An opto-electronic conversion apparatus includes linear arrays of light emitting elements, a two-dimensional array of optical memories, i.e., optical bistable elements, and linear light receiving element arrays arranged transverse to the light emitting element arrays, all integrated with each other. Corresponding light emitting elements, optical memories, and light receiving elements permit arbitrary transfers of signals. Periodic refresh light pulses maintain each optical memory in an established state.
摘要:
An optical neural network which imitates a biological neural network, to provide an associative and/or pattern recognition function, is made of light emitting elements to represent an input neuron state vector, a correlation matrix which modulates light according to stored vector information, light receiving elements, an accumulator and a comparator to perform a threshold function. A stored vector closest to an input vector can be found from a large amount of information without increasing the system size by dividing the correlation matrix and the input neuron state vector with time division techniques, frequency modulation or phase modulation techniques. Positive and negative valves can also be provided with similar techniques.
摘要:
A video display apparatus generates luminance information for individual frames of a video signal from histograms of the luminance component of the video signal, and classifies the content of the video signal on the basis of this information. Color saturation information is also generated from color saturation histograms, and scene changes are detected. Video correction parameters and display control parameters are derived from the content classification and color saturation information. The video signal is corrected according to the video correction parameters, and displayed according to the display control parameters. Display characteristics suitable for the video content are thereby obtained. The parameters are changed when a scene change is detected, so that the viewer is not disturbed by the change in video display characteristics.
摘要:
An apparatus for sensing and processing images is provided with a photo detector array arranged in a matrix form, a control circuit for feeding a row of the array with voltage for sensitivity control, and a neural network for processing current flowing from a column of the array to the ground in order to obtain an apparatus for sensing and processing images having a simple configuration, a high frame speed, the capability of forming a focus of attention, and high throughput of data and possibility of connecting to the neural network.
摘要:
An input pattern or two-dimensional pattern is associated and stored with use of associative matrices having the same size as that of an input pattern without converting it into one-dimensional pattern, wherein the associative matrices are generated from the input pattern so as to maximize the Hamming distance between state invariants which correspond to each different storage pattern, and thereby a power of categorization that corresponds to each storage pattern can be enhanced, thus increasing the storage capability and robustness.
摘要:
The part arrangement optimizing method of the present invention is a part arranging method capable of obtaining part arrangements having smaller evaluation function values rapidly, and the method improves block arrangements so that evaluation function values such as total route lengths become small as much as possible by replacing plural blocks or block sets to determine the arrangements of parts on the basis of the arrangements of the blocks. Furthermore, as aforementioned improving method of the blocks, such a method as assigns expected positional coordinate to each part and arranges a block to a block arranging position near to the expected positional coordinate while renewing the expected positional coordinate so as to be located at the position where the evaluation function value becomes smaller.
摘要:
An information retrieval system allowing the user to build a database or to retrieve data therefrom based on vague retrieval-designating data without becoming aware of the database structure. The system comprises a neural network, a memory, an interface part and a CRT. The neural network stores data designating electronic still pictures contained in the memory. When data for designating retrieval are input, the interface part groups the data into such categories as the place where the desired picture was taken and the date on which it was taken, and supplies the neural network with the categorized input data. In turn, the neural network outputs by association the data corresponding to the input data. Given the data from the network, the memory outputs the relevant electronic still picture to the CRT for display.