Abstract: Monitoring target person monitoring device, method, and system according to the present invention sense a predetermined event regarding a monitoring target person to notify the event; acquire an image including at least a video; determine, based on the acquired image, whether or not multiple persons are on the image; and start, in a case where it is determined that the multiple persons are on the image, storing the acquired video to store the video in a video storage.

Abstract: The image processing system includes a human detection unit for detecting a human region representing a person from an image, a part detection unit for detecting a part region representing a certain part of the person from the image or the human region, and a determination unit for calculating an evaluation value representing a degree by which the person is taking a predetermined action, based on image information in the human region and image information in the part region, applying the evaluation value to a determination formula for determining an action of the person, and determining the predetermined action according to a result of application. The determination unit changes the determination formula for determining the predetermined action according to a position of the human region in the image or a position of the part region in the image.

Abstract: A face authentication system includes: a data processing section for performing a predetermined data processing operation; a first data input section for inputting three-dimensional data on a face area of a subject to the data processing section; and a second data input section for inputting two-dimensional image data on the face area of the subject to the data processing section, the two-dimensional image data corresponding to the three-dimensional data to be inputted to the data processing section, wherein the data processing section includes: a quality rating section for rating the quality of the three-dimensional data based on the two-dimensional image data, and generating quality data, and an authentication processing section for executing a registration process or a verification process of authentication data based on the three-dimensional data, if the quality data satisfies a predetermined requirement.

Abstract: An apparatus for detecting a person or the like from an image extracts candidates, which are considered to human body regions, from the image. Region likelihood probabilities of the respective candidates are obtained, and a human body likelihood of a collection of the candidates is probabilistically calculated using the region likelihood probabilities and correlations among the candidates. It is thereby possible to detect the person robustly against occlusion or change in photographic environment without increasing calculation cost.

Abstract: A face authentication system includes: a data processing section for performing a predetermined data processing operation; a first data input section for inputting three-dimensional data on a face area of a subject to the data processing section; and a second data input section for inputting two-dimensional image data on the face area of the subject to the data processing section, the two-dimensional image data corresponding to the three-dimensional data to be inputted to the data processing section, wherein the data processing section includes: a quality rating section for rating the quality of the three-dimensional data based on the two-dimensional image data, and generating quality data, and an authentication processing section for executing a registration process or a verification process of authentication data based on the three-dimensional data, if the quality data satisfies a predetermined requirement.

Abstract: An object detection apparatus is provided that detects a target object correctly without decreasing a process speed so much even if there is a change in a shooting environment. The human body detection apparatus for detecting a human body in an image includes an image input portion for entering an image taken by a camera, a plurality of plane generation portions for detecting features of the image by using different methods, a section for obtaining conditions about shooting by the camera, a section for calculating reliability values of features detected by the plane generation portions in the case of the obtained condition, and a detection process portion for detecting a human body in the shot image in accordance with features detected from the shot image detected by one or more of the plane generation portions and reliability values of the features.

Abstract: 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.

Abstract: An image processing device for joining a plurality of images while preventing deterioration of a quality of joined images, and particularly for producing an entire image of a subject by joining a plurality of divided images produced from divided portions of the subject having partially overlapping portions, including a setting portion for setting a plurality of sets each including corresponding points in the two divided images having overlap regions overlapped together, a transforming portion for performing geometric transformation of one or the other of the two divided images based on the plurality of corresponding point sets, and a joining portion for joining the two divided images based on the plurality of corresponding point sets after the geometric transformation, wherein higher importance is placed on a specific portion of the overlapping regions compared with the other in the geometric transformation.

Abstract: A distortion correction device performs pre-processing on an input image data of a double page spread document or the like and produces an edge image. Upper and lower ends of the document are detected from the produced edge image. From the detected upper and lower ends of the document, a relative position of an imaging unit with respect to the document is detected. An approximate distance from the imaging unit to the document, the relative position of the imaging unit and information about the ends of the document are used to calculate respective heights of points on the document. The input image data then undergoes geometric conversion correction by means of information about the heights. In this way, the distortion correction device can be provided capable of easily and accurately correcting distortion of the image due to nonuniform height of the surface of the object, i.e., document, regardless of imaging conditions.

Abstract: An image processing apparatus comprising image capturing means for capturing a plurality of images, adjacent ones of which partially overlap with each other, overlap region determining means for determining an overlap region, first region determining means for determining a first region within the overlap region determined, first pixel value determining means for determining a pixel value in the first region based on a pixel value of one of the two adjacent images, second pixel value determining means for determining a pixel value in a second region within the overlap region based on respective pixel values of the two adjacent images, and image joining means for joining the two adjacent images with each other by utilizing the determined pixel value in the first region and the determined pixel value in the second region as pixel values in respective regions of the overlap region.

Abstract: For finding an inclination of an image rapidly and correctly, an image processing device includes a CCD picking up an image, an edge detection unit detecting an edge of an image relating to the picked-up image, a rotation unit rotating the detected edge in an image plane, an operation unit deriving a characteristic amount of rotated edge and an inclination detection unit detecting an inclination of the picked-up image based on the derived characteristic amount.

Abstract: The image processing apparatus includes a CCD that receives an image, a fold detecting unit that detects edges using a lightness component of the received image and selects a specific one of the detected edges, and a fold erasing unit that corrects the lightness component of the selected specific edge. Thus, an image processing apparatus allowing removal of a noise appearing on the lightness component is provided.

Abstract: An image processing method includes the steps of obtaining an image in which a first area including a first image and a second area including a second image are discriminated with each other, and a control step for performing a blur control to a second image using a filter to obtain a blur-controlled second image while accompanying processing for reducing an influence of the first area at a boundary between the first area and the second area and therearound when the blur control is performed to the second image.

Abstract: An object detection apparatus is provided that detects a target object correctly without decreasing a process speed so much even if there is a change in a shooting environment. The human body detection apparatus for detecting a human body in an image includes an image input portion for entering an image taken by a camera, a plurality of plane generation portions for detecting features of the image by using different methods, a section for obtaining conditions about shooting by the camera, a section for calculating reliability values of features detected by the plane generation portions in the case of the obtained condition, and a detection process portion for detecting a human body in the shot image in accordance with features detected from the shot image detected by one or more of the plane generation portions and reliability values of the features.

Abstract: 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.

Abstract: An image processing device includes: a CCD to input a plurality of images obtained from a separate-pickup of an object performed such that parts of the images are overlapped; and an image combining section to combine two images adjacent to each other in one of a first manner and a second manner; wherein in the first manner, two images adjacent to each other in a first direction are combined based on overlapping regions of two images adjacent to each other in the first direction, and in the second manner, two images adjacent to each other in a second direction are combined based on overlapping regions of two images adjacent to each other in the second direction; and the image processing device further includes a CPU to control the image combining section according to a predetermined condition such that the image combining section performs the combining in one of the first manner and the second manner.

Abstract: In an intruding object detection device which performs a background difference process, the following processings are carried out in order to increase the precision in detecting an object. A region where there is high possibility that an intruding object exists is detected from a reference image and an image (in this case, an image to be corrected) captured by a camera in a background difference process. Errors such as a deviation in the image-capturing position between the reference image and the captured image are corrected by affine transformation. The amount of deviation is calculated while excluding the region where there is high possibility that an intruding object exists. This enables appropriate deviation correction, thereby enhancing high precision in detecting an object.

Abstract: An apparatus for detecting a person or the like from an image extracts candidates, which are considered to human body regions, from the image. Region likelihood probabilities of the respective candidates are obtained, and a human body likelihood of a collection of the candidates is probabilistically calculated using the region likelihood probabilities and correlations among the candidates. It is thereby possible to detect the person robustly against occlusion or change in photographic environment without increasing calculation cost.

Abstract: An image processing device includes a first discriminator for discriminating halftone regions of an input image; and a device for smoothing only the discriminated halftone regions. An image processing device also includes a discriminator for discriminating the characteristics of a region for image compression of an image included in a corrected image; and a first device for modifying a degree of discrimination of a local area included in the region for image compression in accordance with the discriminated characteristics; and a second device for modifying a method of compression of the region for image compression in accordance with the discriminated characteristics.

Abstract: An electronic camera is provided with an image sensor, an image processor, and a controller for the image sensor and the image processor so that: the image sensor executes a first mode of sensing a whole image in a specified region of the object, and a second mode of sensing partial images in a plurality of divided parts of the specified region, and the image processor combines the plurality of sensed partial images into a single image corresponding to the whole image in the specified region, each sensed partial image being transformed with reference to the sensed whole image.