Abstract: A method and apparatus under software control for pattern recognition utilizes a neural network implementation to recognize two dimensional input images which are sufficiently similar to a database of previously stored two dimensional images. Images are first image processed and subjected to a Fourier transform which yields a power spectrum. An in-class to out-of-class study is performed on a typical collection of images in order to determine the most discriminatory regions of the Fourier transform. A feature vector consisting of the highest order (most discriminatory) magnitude information from the power spectrum of the Fourier transform of the image is formed. Feature vectors are input to a neural network having preferably two hidden layers, input dimensionality of the number of elements in the feature vector and output dimensionality of the number of data elements stored in the database. Unique identifier numbers are preferably stored along with the feature vector.

Abstract: A pattern recognition method and apparatus utilizes a neural network to recognize input images which are sufficiently similar to a database of previously stored images. Images are first processed and subjected to a Fourier transform which yields a power spectrum. An in-class to out-of-class study is performed on a typical collection of images in order to determine the most discriminatory regions of the Fourier transform. A feature vector consisting of the (most discriminatory) information from the power spectrum of the Fourier transform of the image is formed. Feature vectors are input to a neural network having preferably two hidden layers, input dimensionality of the number of elements in the feature vector and output dimensionality of the number of data elements stored in the database. Unique identifier numbers are preferably stored along with the feature vector. Application of a query feature vector to the neural network results in an output vector.

Abstract: A method and apparatus under software control for pattern recognition utilizes a neural network implementation to recognize two dimensional input images which are sufficiently similar to a database of previously stored two dimensional images. Images are first image processed and subjected to a Fourier transform which yields a power spectrum. An in-class to out-of-class study is performed on a typical collection of images in order to determine the most discriminatory regions of the Fourier transform. A feature vector consisting of the highest order (most discriminatory) magnitude information from the power spectrum of the Fourier transform of the image is formed. Feature vectors are input to a neural network having preferably two hidden layers, input dimensionality of the number of elements in the feature vector and output dimensionality of the number of data elements stored in the database. Unique identifier numbers are preferably stored along with the feature vector.

Abstract: A system and apparatus is disclosed for retrieving analog or digital information stored on an optically sensed disc. In a preferred embodiment, a video disc is disposed between a flooding light source and a matrix array of photosensitive elements, there being provided optical focusing means so that the illuminated data tracks of the video disc are imaged onto the photosensitive array. Circuitry is provided for sensing output signals from selected elements of the array and in response thereto selecting those elements of the array onto which the selected track (or tracks) is best imaged. In this manner the apparatus of the present invention can provide data track following, conventional data retrieval and parallel simultaneous and sequential data retrieval. Alternative circuitry provides signal stream manipulation to compensate for fluctuations in rotational velocity of the data disc and to provide improved signal-to-noise ratios as well as frequency response enhancement.

Abstract: A photovoltaic device composed of a luminescent sheet having a pair of flat, opposed semicircularly shaped sides connected by a thin edge face having a curved, semicylindrically shaped section and a straight section and at least one photovoltaic cell carried on the straight section of said thin edge face.

Abstract: A luminescent solar collector having a luminescent member with at least two types of photovoltaic cells, each type of cell operating efficiently for the generation of electricity over a wavelength range which is different from the efficient wavelength range of the other types of photovoltaic cells present on the luminescent member, each type of photovoltaic cell carrying intermediate to it and the luminescent member a filter means which allows only light within the efficient wavelength range for that type of photovoltaic cell to pass from the luminescent member into the photovoltaic cell.

Abstract: A luminescent solar collector comprising a light transmitting member which carries at least one photovoltaic means, a luminescent member removably attached to the light transmitting member, and interface material between the two members assuring the transmission of light between said members. A method for increasing the efficiency of a luminescent photovoltaic device which contains a luminescent member of reduced efficiency comprising attaching to the low efficiency luminescent member a higher efficiency luminescent member with an interface material, the interface material assuring transmission of light from one luminescent member to the other.

Abstract: A system for sensing information located in tracks in various layers or depths of a recording medium. The desired information is electro-optically tracked in three dimensions to provide sensing of the information contained in the multi-layered recording medium.