摘要:
A large number of stable local regions can be set with low calculation cost. In a face recognition apparatus which discriminates similar face images using feature amounts extracted from local regions included in an image to be discriminated, a moving destination of a feature point extracted from the image to be discriminated, and the size of an image to be clipped at the moving destination are calculated based on a table which defines information required to designate a moving destination of each feature point, and information required to designate the size of an image to be clipped at the moving destination, and an image with the calculated size is clipped at the calculated moving destination as the local region.
摘要:
A large number of stable local regions can be set with low calculation cost. In a face recognition apparatus which discriminates similar face images using feature amounts extracted from local regions included in an image to be discriminated, a moving destination of a feature point extracted from the image to be discriminated, and the size of an image to be clipped at the moving destination are calculated based on a table which defines information required to designate a moving destination of each feature point, and information required to designate the size of an image to be clipped at the moving destination, and an image with the calculated size is clipped at the calculated moving destination as the local region.
摘要:
An object identification apparatus includes an image data input unit configured to input captured image data including an object, an object identification data generation unit configured to generate data for identifying the object by extracting a feature vector from a partial area of the input image data to convert the feature vector according to the partial area, an object dictionary data storage unit configured to store object dictionary data generated from previously recorded image data, and an object identification unit configured to identify a class to which the object belongs, which is included in the image data input by the image data input unit, based on the data for identifying the object and the object dictionary data.
摘要:
A pattern recognition apparatus that is lightweight for mounting, and reduces the effects of registration conditions or check conditions on recognition accuracy. Similarity sets for respective local features are calculated from a local feature of input data and local features of a plurality of pieces of dictionary data corresponding to the local feature of the input data. Integrated similarities are calculated by integrating a plurality of similarity sets in the local features according to a registration condition or a check condition. Dictionary data corresponding to the input data is identified based on the calculated integrated similarities.
摘要:
A pattern recognition apparatus that is lightweight for mounting, and reduces the effects of registration conditions or check conditions on recognition accuracy. Similarity sets for respective local features are calculated from a local feature of input data and local features of a plurality of pieces of dictionary data corresponding to the local feature of the input data. Integrated similarities are calculated by integrating a plurality of similarity sets in the local features according to a registration condition or a check condition. Dictionary data corresponding to the input data is identified based on the calculated integrated similarities.
摘要:
An object identification apparatus includes an image data input unit configured to input captured image data including an object, an object identification data generation unit configured to generate data for identifying the object by extracting a feature vector from a partial area of the input image data to convert the feature vector according to the partial area, an object dictionary data storage unit configured to store object dictionary data generated from previously recorded image data, and an object identification unit configured to identify a class to which the object belongs, which is included in the image data input by the image data input unit, based on the data for identifying the object and the object dictionary data.
摘要:
There is provided a sintered bearing having high rotational accuracy and low rotational fluctuation. This bearing includes a bearing surface (4a), and is made of a sintered compact (4″) produced by molding and sintering a raw material powder (10) containing a partially diffusion-alloyed powder (11) in which a copper powder (13) is partially diffused on a surface of an iron powder (12), a tin powder (14) as a low-melting-point metal powder, and a graphite powder as a solid lubricant powder. The sintered bearing has a radial crushing strength greater than or equal to 300 MPa.
摘要:
When a cover 18 closes an opening end surface 11d of a boss 11, an annular piece 18a moves close to an annular groove 20 to form a labyrinth structure, and thus a flow path for turbocharged air becomes narrower to produce air resistance, which causes the cover 18 to be hardly closed. Thus, chattering is prevented from occurring, and a relief valve 10 can be prevented from being damaged.
摘要:
Input information of a multidimensional array is divided into a plurality of divided areas, accumulated information is generated by calculating accumulated values at respective element positions of the input information from a corresponding reference location for each of the plurality of divided areas, and the generated accumulated information is held in a memory for each divided area. Calculation using the accumulated information is executed for a predetermined processing range. The input information is divided into the plurality of divided areas so that two neighboring divided areas have an overlapping area, and the overlapping area has a size at least in which the whole processing range fits.
摘要:
An anti-reflection sheet includes a transparent substrate; and a resin layer formed on a surface of a transparent substrate. The resin layer includes a first surface that faces the surface of the transparent substrate; a second surface that faces away from the surface of the transparent substrate; and a plurality of projections arranged on the second surface of the resin layer so as to form a projection layer. Each of the plurality of projections has a cross section that includes an entire center axis perpendicular to the surface of the transparent substrate. Each of the plurality of projections has a cross section that is in a shape of quadratic function. An arranged pitch of each of the plurality of projections is not more than half of a wavelength of incident light. An aspect ratio of each of the plurality of projections is 1 or less. A difference between a refractive index n1 of the transparent substrate and a refractive index n2 of the resin layer satisfies the following expression: |n1−n2|≦0.05. A thickness of the resin layer is 11 μm or more. The plurality of projections are shaped and arranged so as to satisfy a linear relationship between an effective refractive index of the projection layer and a height of the projection layer.