Abstract: According to one embodiment, a feature extraction device includes an obtaining unit that obtains image data having a plurality of pixels. The device includes a pixel feature calculation unit that calculates first pixel features and second pixel features of each of the pixels, which are different from each other, and a classification unit that classifies a pair of pixels by using the first features for at least some of the plurality of pixels. The device includes a co-occurrence frequency calculation unit that calculates a co-occurrence frequency representing a frequency of co-occurrence of the second pixel features of the first pixel and the second pixel features of the second pixel for the set for which a result of the classification by the classification unit is consistent.

Abstract: According to an embodiment, a recognition device includes a generation unit to select, plural times, groups each including learning samples from a storage unit, learn a classification metric for classifying the groups selected in each selection, and generate an evaluation metric including the classification metrics; a transformation unit to transform a first feature value of an image including an object into a second feature value using the evaluation metric; a calculation unit to calculate similarities of the object to categories in a table using the second feature value and reference feature values; and a registration unit to register the second feature value as the reference feature value in the table associated with the category of the object and register the first feature value as the learning sample belonging to the category of the object in the storage unit. The generation unit performs the generation again.

Abstract: According to an embodiment, a control apparatus controls a touch panel includes a first resistive film includes a first terminal and a second terminal and a second resistive film includes a third terminal and a fourth terminal, the first and second resistive films being laid on top of each other with a gap between the resistive films. The control apparatus comprises an application unit, a measurement unit and a calculation unit. The application unit is configured to apply a voltage between the first terminal and the second terminal at a first timing. The measurement unit is configured to measure a voltage of each of the third and fourth terminals at the first timing. The calculation unit is configured to derive a positional relationship between two points at which the touch panel has been touched based on the voltages of the third and fourth terminals measured at the first timing.

Abstract: According to an embodiment, a recognition device includes a storage unit, an acquiring unit, a first calculator, a second calculator, a determining unit, and an output unit. The storage unit stores multiple training patterns each belonging to any one of multiple categories. The acquiring unit acquires a recognition target pattern to be recognized. The first calculator calculates, for each of the categories, a distance histogram representing distribution of the number of training patterns belonging to the category with respect to distances between the recognition target pattern and the training patterns belonging to the category. The second calculator analyzes the distance histogram of each of the categories to calculate confidence of the category. The determining unit determines a category of the recognition target pattern from the multiple categories by using the confidences. The output unit outputs the category of the recognition target pattern.

Abstract: According to an embodiment, an image processing device includes a generator and a processor. The generator is configured to generate, from a plurality of unit images in which points on an object are imaged by an imaging unit at different positions according to distances between the imaging unit and the positions of the points on the object, a refocused image focused at a predetermined distance. The processor is configured to perform blurring processing on each pixel of the refocused image according to an intensity corresponding to a focusing degree of the pixel of the refocused image.

Abstract: According to one embodiment, a three-dimensional object determining apparatus includes: a detecting unit configured to detect a plurality of feature points of an object included in an image data that is acquired; a pattern normalizing unit configured to generate a normalized pattern that is normalized by a three-dimensional model from the image data using the plurality of feature points; an estimating unit configured to estimate an illumination direction in which light is emitted to the object in the image data from the three-dimensional model and the normalized pattern; and a determining unit configured to determine whether or not the object in the image data is a three-dimensional object on the basis of the illumination direction.

Abstract: According to an embodiment, an image processing apparatus includes a feature data calculator, a generating unit, and an adding unit. The feature data calculator calculates feature data representing changes in pixel values within a first range of an input image. The generating unit obtains a weight of a predetermined image pattern on the basis of a probability distribution and the feature data. The weight represents a pattern of changes in the pixel values. The probability distribution represents a distribution of relative values of feature data of a learning image containing a high-frequency component with respect to feature data of a learning image. The generating unit weights the predetermined image pattern with the weight so as to generate a high-frequency component with respect to the input image. The adding unit adds the high-frequency component to the input image.

Abstract: It is an object of the present invention to uniformly disperse fibrous material such as short glass fibers so as to be distributed on a collection conveyor, without using compressed air for dispersion of the fibrous material. The present invention provides a method for collecting fibrous material, wherein the fibrous material fiberized by a spinner of a fiberizing unit is dispersed by a hollow bucket disposed just under the spinner, so as to be collected on a collection conveyor disposed below the hollow bucket, comprising: forming said hollow bucket by connecting a blasting section having an oval opening at its lower end, with a waistline section as a lower end of a hopper section having a circular shape in cross section, and deforming the inner surface of the blasting section toward said oval opening, thereby dispersing the fibrous material dropped in the hollow bucket in a width direction of the collection conveyor from the blasting section, so as to be collected on the collection conveyor.

Abstract: According to an embodiment, a recognition device includes a generation unit to select, plural times, groups each including learning samples from a storage unit, learn a classification metric for classifying the groups selected in each selection, and generate an evaluation metric including the classification metrics; a transformation unit to transform a first feature value of an image including an object into a second feature value using the evaluation metric; a calculation unit to calculate similarities of the object to categories in a table using the second feature value and reference feature values; and a registration unit to register the second feature value as the reference feature value in the table associated with the category of the object and register the first feature value as the learning sample belonging to the category of the object in the storage unit. The generation unit performs the generation again.

Abstract: According to one embodiment, a feature extraction device includes an obtaining unit that obtains image data having a plurality of pixels. The device includes a pixel feature calculation unit that calculates first pixel features and second pixel features of each of the pixels, which are different from each other, and a classification unit that classifies a pair of pixels by using the first features for at least some of the plurality of pixels. The device includes a co-occurrence frequency calculation unit that calculates a co-occurrence frequency representing a frequency of co-occurrence of the second pixel features of the first pixel and the second pixel features of the second pixel for the set for which a result of the classification by the classification unit is consistent.

Abstract: According to an embodiment, a control apparatus controls a touch panel includes a first resistive film includes a first terminal and a second terminal and a second resistive film includes a third terminal and a fourth terminal, the first and second resistive films being laid on top of each other with a gap between the resistive films. The control apparatus comprises an application unit, a measurement unit and a calculation unit. The application unit is configured to apply a voltage between the first terminal and the second terminal at a first timing. The measurement unit is configured to measure a voltage of each of the third and fourth terminals at the first timing. The calculation unit is configured to derive a positional relationship between two points at which the touch panel has been touched based on the voltages of the third and fourth terminals measured at the first timing.

Abstract: A linear transformation matrix calculating apparatus linearly transforms a plurality of dictionary subspaces which belong to respective categories by a linear transformation matrix respectively, selects a plurality of sets of two dictionary subspaces from the plurality of linearly transformed dictionary subspaces, calculates a loss function using similarities among the selected sets of dictionary subspaces respectively, calculates a differential parameter obtained by differentiating the loss function by the linear transformation matrix, calculates a new linear transformation matrix from the differential parameter and the linear transformation matrix by Deepest Descent Method, and updates the new linear transformation matrix as the linear transformation matrix used in the linear transformation unit.

Abstract: According to one embodiment, a three-dimensional object determining apparatus includes: a detecting unit configured to detect a plurality of feature points of an object included in an image data that is acquired; a pattern normalizing unit configured to generate a normalized pattern that is normalized by a three-dimensional model from the image data using the plurality of feature points; an estimating unit configured to estimate an illumination direction in which light is emitted to the object in the image data from the three-dimensional model and the normalized pattern; and a determining unit configured to determine whether or not the object in the image data is a three-dimensional object on the basis of the illumination direction.

Abstract: A disparity profile indicating a relation between a perpendicular position on time series images and a disparity on a target monitoring area based on an arrangement of a camera is calculated. Processing areas are set, by setting a height of each of the processing areas using a length at the bottom of the image obtained by converting a reference value of a height of an object according to the profile, while setting a position of each bottom of processing areas on the image. An object having a height higher than a certain height with respect to the monitoring area, unify an object detection result in each processing area according to the disparity of the object, and detect the object of the whole monitoring area from each processing area is detected. Position and speed for the object detected by the object primary detection unit are estimated.

Abstract: It is an object of the present invention to uniformly disperse fibrous material such as short glass fibers so as to be distributed on a collection conveyor, without using compressed air for dispersion of the fibrous material. The present invention provides a method for collecting fibrous material, wherein the fibrous material fiberized by a spinner of a fiberizing unit is dispersed by a hollow bucket disposed just under the spinner, so as to be collected on a collection conveyor disposed below the hollow bucket, comprising: forming said hollow bucket by connecting a blasting section having an oval opening at its lower end, with a waistline section as a lower end of a hopper section having a circular shape in cross section, and deforming the inner surface of the blasting section toward said oval opening, thereby dispersing the fibrous material dropped in the hollow bucket in a width direction of the collection conveyor from the blasting section, so as to be collected on the collection conveyor.

Abstract: A linear transformation matrix calculating apparatus linearly transforms a plurality of dictionary subspaces which belong to respective categories by a linear transformation matrix respectively, selects a plurality of sets of two dictionary subspaces from the plurality of linearly transformed dictionary subspaces, calculates a loss function using similarities among the selected sets of dictionary subspaces respectively, calculates a differential parameter obtained by differentiating the loss function by the linear transformation matrix, calculates a new linear transformation matrix from the differential parameter and the linear transformation matrix by Deepest Descent Method, and updates the new linear transformation matrix as the linear transformation matrix used in the linear transformation unit.

Abstract: An image processing apparatus includes a classification unit configured to extract N features from an input image using pre-generated N feature extraction units and calculate confidence value which represents object-likelihood based on the extracted N features, an object detection unit configured to detect an object included in the input image based on the confidence value, a feature selection unit configured to select M feature extraction units from the N feature extraction units such that separability between the confidence value of the object and that of background thereof becomes greater than a case where the N feature extraction units are used, the M being a positive integer smaller than N, and an object tracking unit configured to extract M features from the input image and tracks the object using the M features selected by the feature selection unit.

Abstract: An apparatus of detecting an object on a road surface includes a stereo set of video cameras mounted on a vehicle to produce right and left images, a storage to store the right and left images, a parameter computation unit to compute a parameter representing road planarity constraint based on the images of the storage, a corresponding point computation unit to compute correspondence between a first point on one of the right and left images and a second point on the other, which corresponds to the first point, based on the parameter, an image transformation unit to produce a transformed image from the one image using the correspondence, and a detector to detect an object having a dimension larger than a given value in a vertical direction with respect to the road surface, using the correspondence and the transformed image.

Abstract: A disparity profile indicating a relation between a perpendicular position on time series images and a disparity on a target monitoring area based on an arrangement of a camera is calculated. Processing areas are set, by setting a height of each of the processing areas using a length at the bottom of the image obtained by converting a reference value of a height of an object according to the profile, while setting a position of each bottom of processing areas on the image. An object having a height higher than a certain height with respect to the monitoring area, unify an object detection result in each processing area according to the disparity of the object, and detect the object of the whole monitoring area from each processing area is detected. Position and speed for the object detected by the object primary detection unit are estimated.

Abstract: An apparatus of detecting an object on a road surface includes a stereo set of video cameras mounted on a vehicle to produce right and left images, a storage to store the right and left images, a parameter computation unit to compute a parameter representing road planarity constraint based on the images of the storage, a corresponding point computation unit to compute correspondence between a first point on one of the right and left images and a second point on the other, which corresponds to the first point, based on the parameter, an image transformation unit to produce a transformed image from the one image using the correspondence, and a detector to detect an object having a dimension larger than a given value in a vertical direction with respect to the road surface, using the correspondence and the transformed image.