Abstract: A solid-state imaging device includes a plurality of pairs of a first photoelectric conversion element and a second photoelectric conversion element which have different spectral sensitivity characteristics. A wavelength range where the first photoelectric conversion element of each pair mainly has a spectral sensitivity and a wavelength range where the second photoelectric conversion element of each pair mainly has a spectral sensitivity respectively fall within a wavelength ranges of specific colors of visible light. The plurality of pairs include a plurality of types of pairs having different specific colors of visible light. The half width in the spectral sensitivity characteristic of the first photoelectric conversion element of the each pair is wider than a half width in the spectral sensitivity characteristic of the second photoelectric conversion element of the each corresponding pair.

Abstract: A WB gain calculation unit 24 determines a light source of a Surface A on the basis of color distribution of the Surface A and a black body locus AD, determines a light source of a Surface B on the basis of color distribution of the Surface B and a black body locus BD and determines a light source of a Surface AB on the basis of color distribution of the Surface AB and a block body locus ABD and calculates the white balance gains on the basis of the determined light sources.

Abstract: An imaging apparatus includes a solid-state imaging device and a driving unit. The solid-state imaging device has a plurality of pairs of first and second photoelectric conversion elements having different spectral sensitivity characteristics. The driving unit independently controls an exposure time of a first group of the plurality of first photoelectric conversion elements and an exposure time of a second group of the plurality of second photoelectric conversion elements. A wavelength range where the first photoelectric conversion element of each pair mainly has a spectral sensitivity and a wavelength range where the second photoelectric conversion element of each pair mainly has spectral sensitivity fall within the respective wavelength ranges of specific colors of visible light.

Abstract: Provided are an inspection device and an inspection method capable of achieving improved magnetic field sensitivity by using a magnetic thin film of a small film thickness. A light-emitting unit 1 emits light of a first wavelength for acquiring magnetic field inspection information and a second wavelength for acquiring inspection object surface information. A selection unit 6 selects information from an inspection object 4 and information from a magnetophotonic crystal film 3 acquired by light irradiation performed by an irradiation unit 2. An image generation unit 9 generates image data based on the magnetic field inspection information acquired with the first wavelength and the inspection object surface information acquired with the second wavelength selected by the selection unit. Each of the generated image data is displayed on a display unit 10.

Abstract: The presence/absence of a face in an image and the type of imaging mode during imaging (such as a landscape imaging mode and a close-up imaging mode) of the image acquired by imaging and the like are detected. Images are automatically classified by groups based on the detected information (classification information). A plurality of images classified by the groups are connected to create a multi-image file (MP file), and the MP file is recorded in a recording medium. As a result, the plurality of classified images can be managed altogether (without dissipation), and the plurality of classified images can be reproduced by providing the multi-image file even if a reproducer does not have a classification function.

Abstract: A moving object detection apparatus comprising: an image acquisition device; a first moving object detection device; a difference image generation device; a second moving object detection device for detecting existence/nonexistence of the moving object based on the difference image generated by the difference image generation device; and an integration device for integrating a detection result by the first moving object detection device and a detection result by the second moving object detection device and determining that the moving object is detected in a case where the moving object is not detected by at least the first moving object detection device and the moving object is detected by the second moving object detection device.

Abstract: Two kinds of color images, obtained by a solid-state imaging device, having different color tones are combined with each other. A first captured color image due to a first pixel group (pixels in which spectral sensitivities of color filters are wide) of the solid-state imaging device is processed. A second captured color image due to a second pixel group (pixels in which spectral sensitivities of color filters are narrow) is processed. The level difference between captured image signals of the pixels of the first pixel group and captured image signals of the pixels of the second pixel group, and due to the spectral sensitivity difference between the color filters in which the spectral sensitivities are wide and narrow is obtained (steps S1 and S2). The level difference is corrected. The first captured color image and the second captured color image are combined with each other.

Abstract: Feature points are extracted from a reference frame. Matching points corresponding to the respective feature points are extracted from a tracking frame. The reference and tracking frames are consecutive in time series. An inverse vector, corresponding to motion of the whole screen, having the matching point as a starting point is obtained. An endpoint of the inverse vector is calculated as a moved point. The matching point is classified as a stationary point when a position of the moved point is within a predetermined range relative to a position of the feature point. When the position of the moved point is out of the predetermined range, whether a correlation between the moved point and the feature point is high is determined. When the correlation is high, the matching point is classified as the outlier. When the correlation is low, the matching point is classified as a moving point.

Abstract: Provided is an eddy current inspection device, an eddy current inspection probe and an eddy current inspection method that make it possible to detect defects existing in deeper parts of test objects. Three or more odd number of excitation coils are arranged at even intervals in a circumferential direction on a postulated circumference. Excitation currents applied to the excitation coils are controlled so that the phase difference between excitation currents applied to adjacent ones of the excitation coils arranged in the circumferential direction on the postulated circumference equals one cycle divided by the number of excitation coils. A magnetic field generated according to an eddy current occurring in the test object due to a magnetic field caused by the application of the excitation currents to the excitation coils is detected by use of a detector arranged on a postulated plane containing the postulated circumference but inside the postulated circumference.

Abstract: A color that readily leaves an impression is adopted as the representative color of a color image. Pixels that form a color image are distributed in L*a*b* color space and are projected onto the a*b* plane. The pixels are grouped and the color at the centroid position of each group is adopted as a representative candidate color. A first score is calculated based upon the distance from the origin, which is indicative of gray, to the representative candidate color, and a second score is calculated from the number of pixels contained in the group. A final score is calculated from the first and second scores and representative candidate colors that provide the three highest final scores are decided upon as representative colors.

Abstract: A composite image is generated by a method including: detecting at least one in a captured image; judging arrangement of the at least one subject by an arithmetic operation; selecting at least one composition-purpose image at random in accordance with the arrangement of the at least one subject; determining arrangement of the selected at least one composition-purpose image at random so that the at least one composition-purpose image does not overlap with one another and does not overlap with the detected at least one subject; and generating a composite image by combining the captured image with the at least one composition-purpose image.

Abstract: An image processing apparatus that an image input unit; a parallax acquisition unit configured to acquire a per-pixel or per-region parallax between two-viewpoint images; a main subject detection unit configured to detect a main subject on the two-viewpoint images; a parallax acquisition unit configured to acquire a parallax of the main subject; a setting unit configured to set a conversion factor of the parallax; a correction unit configured to correct the conversion factor of the parallax per pixel, per region, or per image; a multi-viewpoint image generation unit configured to convert at least one image of the two-viewpoint images in accordance with the corrected conversion factor of the parallax; an image adjustment unit configured to shift the two-viewpoint images or multi-viewpoint images to obtain a parallax appropriate for stereoscopic view; and a stereoscopically-displayed image generation unit configured to generate a stereoscopically-displayed image.

Abstract: Designated positions in a left-eye image and in a right-eye image are detected. The dominant eye is input and, if there is a position designation made by the observer, the coordinate position of the designated position is calculated in each of the left-eye and right-eye images. A template image is detected from the dominant-eye image and an image identical with the detected template image is detected from the non-dominant-eye image. Positions in respective ones of the template image and matching image become the designated positions. A distance differential between the template image detected from the dominant-eye image and the image detected from the non-dominant-eye image is parallax. The parallax of the designated positions is adjusted.

Abstract: A composite image is generated by a method including: detecting at least one in a captured image; judging arrangement of the at least one subject by an arithmetic operation; selecting at least one composition-purpose image at random in accordance with the arrangement of the at least one subject; determining arrangement of the selected at least one composition-purpose image at random so that the at least one composition-purpose image does not overlap with one another and does not overlap with the detected at least one subject; and generating a composite image by combining the captured image with the at least one composition-purpose image.

Abstract: An image processing apparatus that an image input unit; a parallax acquisition unit configured to acquire a per-pixel or per-region parallax between two-viewpoint images; a main subject detection unit configured to detect a main subject on the two-viewpoint images; a parallax acquisition unit configured to acquire a parallax of the main subject; a setting unit configured to set a conversion factor of the parallax; a correction unit configured to correct the conversion factor of the parallax per pixel, per region, or per image; a multi-viewpoint image generation unit configured to convert at least one image of the two-viewpoint images in accordance with the corrected conversion factor of the parallax; an image adjustment unit configured to shift the two-viewpoint images or multi-viewpoint images to obtain a parallax appropriate for stereoscopic view; and a stereoscopically-displayed image generation unit configured to generate a stereoscopically-displayed image.

Abstract: A composite image is generated by a method including: detecting at least one in a captured image; judging arrangement of the at least one subject by an arithmetic operation; selecting at least one composition-purpose image at random in accordance with the arrangement of the at least one subject; determining arrangement of the selected at least one composition-purpose image at random so that the at least one composition-purpose image does not overlap with one another and does not overlap with the detected at least one subject; and generating a composite image by combining the captured image with the at least one composition-purpose image.

Abstract: An eddy current testing device which confirms that a change in characteristics of a target object is detected regardless of the magnitude of the change and specifying the position of a portion from which the change is detected. The device uses an eddy current probe to inspect a bent portion of a metal body, and has an inspection controller and a display unit. The inspection controller calculates a phase angle of a signal detected by the eddy current probe and generates flaw identification image data that indicates an area (or an area of the signal detected and determined to correspond to a flaw signal, based on the phase angle of the detected signal) of a flaw signal in coordinates in which the position of the portion of the target object is plotted along a coordinate axis. The display unit displays the flaw identification image data.

Abstract: Erroneous detection of a moving body in a background region is suppressed. Specifically, the position of one input image of two input images is made to coincide with the position of the other input image so as to eliminate relative positional deviation between the two input images. The amount of residual positional deviation in the thus registered two input images is calculated. The resolution of the two input images is lowered so as to take on a resolution A if the amount of residual positional deviation is greater than a prescribed threshold value d and take on a resolution B if the amount of residual positional deviation is equal to or less than the prescribed threshold value d (resolution A<resolution B). Binary difference image data is generated using the input images of lowered resolution and a motion region is detected from a binary difference image represented by the binary difference image data.

Abstract: The presence/absence of a face in an image and the type of imaging mode during imaging (such as a landscape imaging mode and a close-up imaging mode) of the image acquired by imaging and the like are detected. Images are automatically classified by groups based on the detected information (classification information). A plurality of images classified by the groups are connected to create a multi-image file (MP file), and the MP file is recorded in a recording medium. As a result, the plurality of classified images can be managed altogether (without dissipation), and the plurality of classified images can be reproduced by providing the multi-image file even if a reproducer does not have a classification function.

Abstract: An image processing method executed in an apparatus, includes steps of: inputting a left-viewpoint image and a right-viewpoint image therebetween to the apparatus; searching for a corresponding point in the right-viewpoint image with respect to the left-viewpoint image, generating a first corresponding-point map based on a result of the searching, and generating a first intermediate-viewpoint image based on the left-viewpoint image; searching for a corresponding point in the left-viewpoint image with respect to the right-viewpoint image, generating a second corresponding-point map based on a result of the searching, and generating a second intermediate-viewpoint image based on the right-viewpoint image; extracting missing portions of pixels in the first and second intermediate-viewpoint images; determining whether a missing portion extracted is a missing portion to be interpolated or not; interpolating the missing portion to be interpolated based on a result of determination; and generating a third intermedia