Abstract: In an object tracking device that obtains image data, detects an object, and tracks a target object, a matching unit collates image data of a characteristic model that represents a tracking object, which is a tracking target, with input image data, and outputs a candidate region of the tracking object. An object region determination unit determines an object region from the candidate region that is output by the matching unit. A depth map calculation unit calculates depth information relating to the input image. An object depth setting unit obtains the object region that has been determined by the object region determination unit in the past and the depth information that has been calculated by the depth map calculation unit, and sets a predetermined depth range where the object can exist.

Abstract: In a focus adjusting apparatus, a tracking unit detects from an image signal an area of a subject to be tracked, a position of an imaging plane of the area and reliability of the position, and a prediction unit predicts a position of an imaging plane when the image signal was obtained based on a history of detected positions of imaging planes. A setting unit sets a tolerance based on the reliability, and a determination unit determines the area of the detected subject as a focus adjustment area if a difference between the predicted and detected positions is within the tolerance. The prediction unit predicts a position of an imaging plane of the subject in the focus adjustment area at a future time point.

Abstract: One aspect of the present invention provides an image processing apparatus capable of tracking an image region to be tracked with high accuracy. A distance distribution generation unit generates distance information indicating a distance of each of regions in first image data. An object likelihood distribution generation unit generates a likelihood distribution of an object in the first image data based on information indicating a position specified as the object in the first image data, the distance information, and reliability of the distance information. A feature amount extraction unit calculates a feature amount to be used to detect the object from second image data based on the likelihood distribution.

Abstract: An image processing apparatus capable of accurate area tracking, and a method for controlling the same are provided. An image area for extracting a feature amount is identified based on a specified position. A plurality of images are searched for an image that corresponds to the image area using the feature amount extracted from the identified image area. The image area is identified using distance information if distance information that satisfies a reliability condition has been obtained regarding an area that includes the specified position, and without using the distance information if the distance information that satisfies the reliability condition has not been obtained.

Abstract: An image quality adjustment apparatus includes an adjustment unit and an order controller. The adjustment unit sequentially receives instructions of adjustment details for adjustment processes of multiple of types for adjusting image quality, and sequentially performs the adjustment processes on an image. The order controller presents, in accordance with priority ranks that have been associated in advance with characteristics of an image to be processed, operation screens for receiving instruction operations for the adjustment details.

Abstract: An image forming device includes an image forming unit that forms an image on a recording medium in response to an input of image data representing the image, a reading unit that reads a first image formed by the image forming device, and a second image formed by another image forming device to respectively generate a first read image and a second read image, the first image and the second image being formed based on identical image data, a discrimination unit that discriminates the first read image and the second read image from each other, and a color correction unit that performs, on the image formed by the image forming unit, a color correction that cancels out a difference in color between the first read image and the second read image.

Abstract: A subject tracking apparatus is configured to track a subject in continuously obtained photographic images. In the photographic image, a position of the subject in the photographic image and reliability that represents subject probability are obtained. The position of the subject to a target position in the photographic image is moved a tracking state of the subject is changed on a basis of the reliability.

Abstract: An image recognition device capable of accurately recognizing an object by efficiently selecting a specific object area as a recognition target, from detected object areas. The image recognition device recognizes a specific object area from images sequentially input in time series. An object detection section detects object areas from each image. An appropriateness determination section determines whether or not each detected object area is appropriate as a recognition target. A recognition target-selecting section selects an object area for use as the recognition target, from the detected object areas, based on a result of determination by the appropriateness determination section. An image recognition device recognizes whether or not the object area selected by the recognition target-selecting section is the specific object area.

Abstract: An image recognition apparatus is arranged to extract feature information of an object area detected from among images, compare with feature information of previously-registered objects, thereby calculating a degree of similarity, when the calculated degree of similarity is equal to or greater than a predetermined threshold value, determine the object included in the object area as a registered object and set a recognition confirmed state that is successively maintained between images, and when a reliability of the object tracking is low and the degree of similarity is smaller than the predetermined threshold value, set the recognition confirmed state which is successively maintained into a recognition unconfirmed state where the object included in the object area is not determined as a registered object.

Abstract: The subject tracking apparatus comprises: a first registering unit configured to register a partial area as a template indicative of a subject in one image of supplied images; a first matching unit configured to estimate a subject area by collating a partial area in newly supplied images with the template registered by the first registering unit; a second registering unit configured to register a histogram generated based on a pixel value of a partial area indicative of the subject in one image of supplied images; a second matching unit configured to estimate a subject area by collating a histogram of a partial area in newly supplied images with the histogram registered by the second registering unit; and a tracking area determination unit configured to determine a tracking area based on estimation results by the first matching unit and the second matching unit.

Abstract: In an object tracking device that obtains image data, detects an object, and tracks a target object, a matching unit collates image data of a characteristic model that represents a tracking object, which is a tracking target, with input image data, and outputs a candidate region of the tracking object. An object region determination unit determines an object region from the candidate region that is output by the matching unit. A depth map calculation unit calculates depth information relating to the input image. An object depth setting unit obtains the object region that has been determined by the object region determination unit in the past and the depth information that has been calculated by the depth map calculation unit, and sets a predetermined depth range where the object can exist.

Abstract: An image capture apparatus functions as an object tracking apparatus for tracking an object included in provided images, registers a partial image indicating an object as a template, and performs template matching for estimating a region by using the template and histogram matching for registering a histogram of a partial image indicating the object and estimating a region by using the histogram. In a case where a distance between the estimation region based on the histogram matching and the estimation region based on the template matching is within a predetermined range, the estimation region based on the template matching is employed as an object region, and in a case where the distance between the estimation region based on the histogram matching and the estimation region based on the template matching is not within the predetermined range, the estimation region based on the histogram matching is employed as the object region.

Abstract: An object area tracking apparatus has: a face detection unit for detecting a face area on the basis of a feature amount of a face from a supplied image; a person's body detection unit for detecting an area of a person's body on the basis of a feature amount of the person's body; and a main object determination unit for obtaining a priority for each of the objects by using detection results by the face detection unit and the person's body detection unit and determining a main object of a high priority, wherein for the object detected only by the person's body detection unit, the priority is changed in accordance with a past detection result of the object in the face detection unit.

Abstract: Face regions are detected from a captured image, and a weight of each detected face region is computed based on a size and/or a position of the detected face region. Then a previous priority ranking weight is computed based on a priority ranking determined in previous processing. A priority of the face region is computed from the weight and the previous priority ranking weight. For example, if the continuous processing number exceeds the threshold the priority ranking weight is reduced. After the processing is completed for all face regions, a priority ranking of each face region is determined based on the priority computed for each face region.

Abstract: An image recognition apparatus is arranged to extract feature information of an object area detected from among images, compare with feature information of previously-registered objects, thereby calculating a degree of similarity, when the calculated degree of similarity is equal to or greater than a predetermined threshold value, determine the object included in the object area as a registered object and set a recognition confirmed state that is successively maintained between images, and when a reliability of the object tracking is low and the degree of similarity is smaller than the predetermined threshold value, set the recognition confirmed state which is successively maintained into a recognition unconfirmed state where the object included in the object area is not determined as a registered object.

Abstract: An object area tracking apparatus has: a face detection unit for detecting a face area on the basis of a feature amount of a face from a supplied image; a person's body detection unit for detecting an area of a person's body on the basis of a feature amount of the person's body; and a main object determination unit for obtaining a priority for each of the objects by using detection results by the face detection unit and the person's body detection unit and determining a main object of a high priority, wherein for the object detected only by the person's body detection unit, the priority is changed in accordance with a past detection result of the object in the face detection unit.

Abstract: Face regions are detected from a captured image, and a weight of each detected face region is computed based on a size and/or a position of the detected face region. Then a previous priority ranking weight is computed based on a priority ranking determined in previous processing. A priority of the face region is computed from the weight and the previous priority ranking weight. For example, if the continuous processing number exceeds the threshold the priority ranking weight is reduced. After the processing is completed for all face regions, a priority ranking of each face region is determined based on the priority computed for each face region.

Abstract: Face regions are detected from a captured image, and a weight of each detected face region is computed based on a size and/or a position of the detected face region. Then a previous priority ranking weight is computed based on a priority ranking determined in previous processing. A priority of the face region is computed from the weight and the previous priority ranking weight. For example, if the continuous processing number exceeds the threshold the priority ranking weight is reduced. After the processing is completed for all face regions, a priority ranking of each face region is determined based on the priority computed for each face region.

Abstract: An image recognition device capable of accurately recognizing an object by efficiently selecting a specific object area as a recognition target, from detected object areas. The image recognition device recognizes a specific object area from images sequentially input in time series. An object detection section detects object areas from each image. An appropriateness determination section determines whether or not each detected object area is appropriate as a recognition target. A recognition target-selecting section selects an object area for use as the recognition target, from the detected object areas, based on a result of determination by the appropriateness determination section. An image recognition device recognizes whether or not the object area selected by the recognition target-selecting section is the specific object area.

Abstract: A subject area detection apparatus which is capable of improving the rate of detection for a subject area in an image and detecting the subject area with ease. A first subject area which is a specific area is detected from a subject image in an image having at least one subject image. An area including at least a part of the first subject area is detected as a second subject area from the subject image in the image. An estimated area is obtained by estimating, in the second subject area, an area corresponding to the first subject area. The first subject area and the estimated area are compared with each other to obtain a correlation determination result. A subject area detection result is output according to the correlation determination result.