Abstract: A monitoring system includes at least one camera (1) for generating a sequence of video images, and a control circuit (2). The control circuit (2) is provided for forming a difference image of two block-structured, consecutive video images and for generating status bits allocated to a stationary or a moving object for each block of a difference image. The control circuit (2) is further used for characterizing, by means of an area mark, contiguous status bits constituting an area and allocated to a moving object for determining an area parameter from the status bits of an area and for triggering an alarm when a predetermined condition of one or more area parameters is reached.

Abstract: A circuit for recognizing a human face in a sequence of video is provided. The circuit includes a block structuring circuit in which the video is structured block-wise and wherein two consecutive frames are subtracted from one another. The circuit also includes a unit for block-wise post-processing the difference frames. The post-processing unit includes a comparator for determining the areas of the difference picture which may correspond to a moving object, a mapping unit for forming a status picture by mapping each block of the difference picture which corresponds to a moving object onto a binary "one", all other blocks being mapped to a binary "zero".The status picture is transformed by a transforming unit into a unidimensional feature function. The unidimensional feature function is then fed to a convolution unit to determine whether the feature function could include a "head-shoulder" scene.

Abstract: A correlation operational apparatus for multi-dimensional images includes: means for reducing at least two-dimensional components of the multi-dimensional images composed of at least two components by at least one component by projecting a vector representing the image onto a space perpendicular to a vector representing a large variance of distribution of the multi-dimensional image in a vector space whose orthogonal axes represent the individual dimensional component values of the multi-dimensional image; and means for performing a correlation operation for each component on the multi-dimensional images whose dimensional components have been reduced by at least one. In consequence, images having a relatively high number of high spatial frequency components and whose components have been reduced by at least one are extracted, and a correlation operation is performed thereon.

Abstract: The method of detecting corresponding regions between primary color picture images comprises respective steps of converting three primary color signals of color picture images to at least two signals based on these three primary color signals, calculating the variance of the mutual concentration distribution in the operation object regions of the three primary color picture images from these converted signals and detecting the set of regions in which the calculated variance value is minimum.

Abstract: A method of processing images taken by an endoscope having insertion section, bending section and operation section including a step of entering a first image of an object taken by the endoscope having the bending section situated in a first position, a step of entering a second image taken by the endoscope having the bending section situated in a second position, after moving the bending section into the second position by operating an operation member, the first image being partially overlapped with the second image, a step of detecting a positional relation between the first and second positions by detecting the movement of the bending section, a step of deriving a distance between corresponding points on the first and second images, a step of deriving three dimensional information of the object, and a step of displaying a three dimensional image of the object in accordance with the three dimensional information.