Konica-minolta Holdings, Inc.

Minolta Company, Ltd.

Konica Minolta, Inc.

Hironori Sumitomo

Mutsuhiro Yamanaka

Koji Fujiwara

Osamu Toyama

Yuichi Kawakami

Yoshihisa Abe

Daisaku Horie

Yuusuke Nakano

An image processing apparatus for searching for corresponding points between a first image and a second image is provided. The image processing apparatus includes a window setting portion, a phase information extraction portion that extracts phase information by subjecting a representation included in each set window to frequency decomposition, a corresponding points determination portion that determines the corresponding points based on comparison of the phase information extracted from each window, and a window size control portion that causes the window setting portion to set a window different in size in accordance with a frequency used for calculation of the frequency decomposition. The window size control portion causes the window setting portion to set a window having a first size in response to a first frequency and to set a window having a second size smaller than the first size in response to a second frequency higher than the first frequency.

Image processing apparatus, image processing method, and image processing program

An image integration unit includes: an imaging section which is installed in a moving body and which images a plurality of time-series images at different times; a three-dimensional image information calculating section which calculates three-dimensional image information in each of the time-series images based on the time-series images imaged by the imaging section; a stationary body area extracting section which extracts stationary body areas in each of the time-series images based on the three-dimensional image information; and an integrating section which calculates the corresponding stationary body areas between the time-series images from each of the stationary body areas extracted in each of the time-series images, and matches the corresponding stationary body areas to integrate the time-series images.

Image integration unit and image integration method

In a motion vector generation apparatusaccording to the present invention, when motion vectors are sequentially generated for a plurality of images, a target point is determined at a sub-pixel level in any of the plurality of images, a corresponding point corresponding to the target point is searched at the sub-pixel level in another image different from this image, and a motion vector is calculated on the basis of the target point and the corresponding point. Then, the corresponding point is defined as a new target point, and the aforementioned process is repeated to sequentially generate the motion vectors. Thus, the motion vector generation apparatusof the present invention can compute consecutive motion vectors by performing correspondence search at the sub-pixel level in the plurality of images.

Motion vector generation apparatus and motion vector generation method

A corresponding point candidate determiner () determines whether plural correlation peaks appear, based on a correlation value calculated by a corresponding point determiner (). In the case where the corresponding point candidate determiner () determines that plural correlation peaks appear, the corresponding point determiner () calculates a ratio between the correlation values as represented by the correlation peaks, determines one or more corresponding points, based on the calculated ratio, and notifies the determination result to an initial position setter (). In the case where the corresponding point determiner () searches plural corresponding points, the initial position setter () sets an initial search position with respect to each of the corresponding points in a reference image of a layer immediately higher than a target layer.

Corresponding point searching apparatus

A converted reference image generated from a reference image is searched for a first corresponding point corresponding to a first standard point included in a converted standard image generated from a standard image. Based on a position of the first corresponding point in the converted reference image, a second search standard point is determined on the reference image. Further, based on phase information on each frequency component regarding a second standard area including a second standard point corresponding to the first standard point of the standard image, and based on phase information on each frequency component regarding a second reference area including the second search standard point of the reference image, the reference image is searched for a second corresponding point corresponding to the second standard point. The information on a frequency component obtained in the computation on another stage is reused in the search on the two stages above.

Image processing apparatus, image processing method, and non-transitory computer readable recording medium

An objective of the present invention is to provide a technique capable of generating a virtual viewpoint image without causing any visually uncomfortable feeling. In order to achieve this objective, a first image obtained by being captured from a first viewpoint at a first image capture time, and a second image obtained by being captured at a second image capture time different from the first image capture time are acquired. To each of pixels in a non-image capture area corresponding to a portion of a subject not captured in the first image of a first virtual viewpoint image that is generated in a pseudo manner based upon the first image and can be acquired by being captured from a first virtual viewpoint different from the first viewpoint, a pixel value is added in accordance with the second image.

System, method and program for capturing images from a virtual viewpoint

It is an object to reduce a computation related to a detection of a corresponding point intended for a compressed dynamic image. In order to attain the object, there are acquired first and second compressed dynamic images including a reference frame and a prediction frame with each pixel indicated by motion information based on the other frame respectively. Moreover, there is executed a detection processing for detecting a corresponding point which corresponds to each reference point of one frame contained in the first compressed dynamic image from one frame contained in the second compressed dynamic image by causing each set of frames between said first compressed dynamic image and said second compressed dynamic image to be a target. The detection processing intended for a set of prediction frames is executed by using the motion information indicative of the set of prediction frames.

Image processing system, image processing method, and program

A technique is provided which can improve the precision of a matching point search with a plurality of images taking the same object where distant and near views coexist. A plurality of first images obtained by time-sequentially imaging an object from a first viewpoint, and a plurality of second images obtained by time-sequentially imaging the object from a second viewpoint, are obtained. Reference regions including a reference point are set respectively in the first images with the same arrangement, and comparison regions corresponding to the form of the reference regions are set respectively in the second images with the same arrangement. One reference distribution of pixel values about two-or-more-dimensional space is generated from the distributions of pixel values about the plurality of reference regions, and one comparison distribution of pixel values about two-or-more-dimensional space is generated from the distributions of pixel values about the plurality of comparison regions. Then, a matching point in the plurality of second images that corresponds to the reference point is detected by using the reference distribution of pixel values and the comparison distribution of pixel values.

Image processing apparatus, information processing system, and image processing method

An information processing system for performing processes on first image and second image captured from different viewpoints, comprising: a first specifying part for specifying a first corresponding point on the second image, corresponding to a designation point designated on the first image, by searching on a line along a first basis direction corresponding to a predetermined direction and passing through a position corresponding to the designation point in the second image; a second specifying part for specifying a second corresponding point on the second image, corresponding to the designation point, by searching on a line passing through the first corresponding point in the second image and along a second basis direction almost perpendicular to the first basis direction; and a third specifying part for specifying a third corresponding point on the second image, corresponding to the designation point, by searching on a line passing through the second corresponding point in the second image and along the first basis direction.

Information processing system

A monitoring system according to the present invention has a plurality of cameras respectively monitoring a monitor area and having a changeable imaging magnification, monitoring section detecting an object to be monitored in the monitor area, and control section selecting at least one camera capable of obtaining an enlarged image to contribute to identification of the object to be monitored as camera for enlargement from the plural cameras when the object to be monitored is detected and selecting at least one camera other than the camera for enlargement as camera for wide area from the plural monitoring camera. The camera for wide area is set to a first imaging magnification to obtain a wide area image of the monitor area and the camera for enlargement is set to a second imaging magnification larger than the first imaging magnification to obtain an enlarged image of the detected object to be monitored.

Monitoring camera system, monitoring camera control device and monitoring program recorded in recording medium

An object detection apparatus is provided that detects a target object in an image more securely and at higher speed than the conventional one. The human body detection apparatus for detecting a target object in an image includes a template memory portion for memorizing a template consisting of one or more open curves indicating a part of a contour of a human model, an image data reception portion for entering an image to be detected, a head position detection portion for detecting a human body in the image by performing a matching process on the entered image using the template.

Object detection apparatus, object detection method and recording medium

An image processing apparatus first forms an edge image of an input image using, for example, SOBEL operator, and performs a smoothing process on the formed edge image by using an average filter. Thereafter, the image processing apparatus calculates a difference between the edge image and the edge smoothed image, and binarizes the edge image based on the difference.

Image processing apparatus capable of highly precise edge extraction

A counting system is for counting the number of persons passing a monitor line set in the width direction of a path, and has a laser for irradiating the monitor line with a slit ray, and an image capturing apparatus for photographing an area including the monitor line. In the image capturing apparatus, one-dimensional data indicative of a position in which the slit ray is interrupted on the monitor line is generated from an image obtained by the photographing. On the basis of the one-dimensional data, the number of passing persons is counted. By counting the number of passing persons with the one-dimensional data, a very light computation load for the counting can be achieved.

Counting system and counting method

In an image processing apparatus for performing image restoration by utilizing iterative calculations based on a captured degraded image and a degradation function obtained in or before photographing, a residual is calculated in each time of iteration in the iterative calculations and the rate of variation from the previous residual to the newly obtained residual is calculated. When the rate of variation of the residual becomes equal to or lower than a predetermined threshold value, it is judged that the iterative calculations are converged.

Image processing apparatus for performing image restoration

The transposed filter of the image degrading filter is calculated, and a residual gradient image is calculated by the convolution operation of the transposed filter and a residual image obtained from the image degrading filter, the estimated restored image and the degraded image. Then, a modified image of the estimated image is calculated from the residual gradient image, and the estimated image is modified and updated based on the modified image. Consequently, even when the degradation function is one that can be expressed as a complicated helical filter, the memory consumption is small and the calculation time is short compared to a method depending on a large matrix and an inverse matrix.

Growing community of inventors

Moriguchi, Japan

Hironori Sumitomo