摘要:
The image signature extraction device includes an extraction unit and a generation unit. The extraction unit extracts region features from respective sub-regions in an image in accordance with a plurality of pairs of sub-regions in the image, the pairs of sub-regions including at least one pair of sub-regions in which both a combination of shapes of two sub-regions of the pair and a relative position between the two sub-regions of the pair differ from those of at least one of other pairs of sub-regions. The generation unit generates, based on the extracted region features of the respective sub-regions, an image signature to be used for identifying the image.
摘要:
Provided is a content reproduction control system equipped with: a signal characterizing quantity sequence extraction means that, for each content item stored in a content group storage unit that stores at least more than one content item, extracts a signal characterizing quantity sequence which is the sequence of the signal characterizing quantities associated with the content positions on the time axis; a common section group detection means that detects, as a common section group in the signal characterizing quantity sequence for each content item, a common section group for which the signal characterizing quantities for different content items are similar to each other; a content reproduction quality information collection means that, for each common section group, collects content reproduction quality information which indicates the reproduction quality of the content to which each common section of a common section group belongs; and a content reproduction control means that, when the content contained in the content group storage unit is reproduced, switches the content reproduced for the content to be reproduced for the sections included in the common section group and performs reproduction such that the reproduction quality of the content will be higher, based on the content reproduction quality information for the content to which each common section of a common section group belongs.
摘要:
This image identity scale calculation system can calculate an identity scale representing a degree of identity of two images in consideration of identification capability and robustness. An image feature comparison unit is supplied with hierarchical quantization index codes, which are encodings allowing unique specification of quantization indexes of a plurality of hierarchies calculated by hierarchical quantization for each quantization target region of the two images, and selects a quantization index set used for comparison as a comparing quantization index set based on additionally supplied information in accordance with a previously defined hierarchical quantization method. Then, the image feature comparison unit compares the hierarchical quantization index codes of the two images by using the comparing quantization index set, and calculates an identity scale of the two images.
摘要:
The feature selection device includes a feature extraction unit that extracts M types of features from each of a plurality of original images and each of a plurality of altered images obtained by applying an alteration process to the plurality of original images; and a feature selection unit that handles an original image and an altered image of the original image as identical images and handles altered images of the same original image as identical images, while handles other images as different images, and with use of discrimination capability which is a degree of discriminating different images and robustness which is a degree that a value of a feature does not vary due to the alteration process applied to an image as evaluation criteria, evaluates the M types of features extracted from the respective images, and selects a collection of N types of features, the N types being smaller in number than that of the M types, from the M types of features as features for discriminating images.
摘要:
A person's region is detected from input video of a surveillance camera; a person's direction in the person's region is determined; the separability of person's clothes is determined to generate clothing segment separation information; furthermore, clothing features representing visual features of person's clothes in the person's region are extracted in consideration of the person's direction and the clothing segment separation information. The person's direction is determined based on a person's face direction, person's motion, and clothing symmetry. The clothing segment separation information is generated based on analysis information regarding a geometrical shape of the person's region and visual segment information representing person's clothing segments which are visible based on the person's region and background prior information. A person is searched out based on a result of matching between a clothing query text, representing a type and a color of person's clothes, and the extracted person's clothing features.
摘要:
The video descriptor generation device includes a first extraction unit, a second extraction unit, and a feature combining unit. The first extraction unit extracts a first feature for each picture which is a frame or a field of a video. The second extraction unit extracts a second feature from a region defined by an edge of an image included in the video. The feature combining unit combines the first feature and the second feature to generate a video descriptor.
摘要:
This image feature extraction device extracts, from an image, an image feature that makes it possible to adjust the balance between identification capability and robustness, which are the scales of the capability of determining identity of images. This image feature extraction device executes hierarchical quantization to calculate quantization indexes of a plurality of hierarchies in accordance with a previously defined hierarchical quantization method for each quantization target region of an image, and outputs a hierarchical quantization index code, which is an encoding allowing unique specification of the quantization indexes of the respective hierarchies of each quantization target region.
摘要:
A reversible eight-element DCT providing values near to transform values of the original DCT. In a 4×4 matrix, coefficients (X1, X7, X3, X5) are separated into (X1, X7) and (X3, X5. If (X1, X7) are determined, then (X3, X5) values are limited. (X1, X7) are linearly quantized with steps k1, k7 to obtain quantization values (Xq1, Xq7). (X3, X5) are divided into global and local signals. The global signals are linearly quantized with steps L3, L5, and local signals are determined using a table. The quantization values of (X3, X5) are added to obtain quantization values (Xq3, Xq5). The reversible 4×4 matrix and conventional 2×2 matrices are combined to construct a reversible eight-element DCT.
摘要:
The invention provides a reversible coding system for an interlaced scanned moving picture based on reversible discrete cosine transform which controls selection of frame/field discrete cosine transform so as to assure a high degree of compatibility with an ordinary discrete cosine transform coding system. A motion compensated predictive error is detected from an input signal, and a frame-scan block and a field-scan block are produced for each block of the input signal by a frame/field scan production circuit. One of the blocks is selected by a switch, and reversible discrete cosine transform is performed for the selected block. Upon the selection, first and second estimated error calculation circuits estimate error amounts when reversible discrete cosine transform is performed for the frame- and field-scan blocks and then inverse discrete cosine transform is performed for the resultant blocks. The error amounts are compared with each other, and the lower one of them is selected by a comparator.
摘要:
Provided is an equipment control device capable of performing more appropriate equipment control in accordance with detection by a user.The present invention includes: an equipment layout information storage unit for storing equipment layout information including positional information of control-target equipment to be controlled within a managed space; related ID information storage unit for storing related ID information representing the correspondence relationship between the control-target equipment and IDs given to users of the control-target equipment; an ID-affixed motion line information acquisition unit for acquiring ID-affixed motion line information, in which the ID of a user is associated with the time series of the positional information of the user in the managed space; and an equipment controller for acquiring, on the basis of the related ID information, the control-target equipment associated with the ID in the ID-affixed motion line information, generating control information for controlling, on the basis of the ID-affixed motion line information and equipment layout information, the state of the control-target equipment, and outputting the control information to the control-target equipment.