Abstract: Classification of a potential target is accomplished by receiving image information, detecting a potential target within the image information and determining a plurality of features forming a feature set associated with the potential target. The location of the potential target is compared with a detection database to determine if it is close to an element in the detection database. If not, a single-pass classifier receives a potential target's feature set, classifies the potential target, and transmits the location, feature set and classification to the detection database. If it is close, a fused multi-pass feature determiner determines fused multi-pass features of the potential target and a multi-pass classifier receives the potential target's feature set and fused multi-pass features, classifies the potential target, and transmits its location, feature set, fused multi-pass features and classification to the detection database.

Abstract: Techniques for identifying and enhancing colors in a digital image associated with one or more target color shades. In an embodiment, the target color shades may include a shade of blue associated with the sky, a shade of green associated with outdoor foliage, or the color red. In an embodiment, the blue chroma (Cb) and red chroma (Cr) coordinates of a pixel are evaluated to determine whether to apply an enhancement factor. The enhancement factor may incorporate an exposure index (EI) auxiliary enhancement factor, a color temperature (D) auxiliary enhancement factor, and a luminance (Y) of each pixel. Further aspects for implementing the techniques in software and hardware are disclosed.

Abstract: A method for aligning point clouds is disclosed. The method includes inputting point cloud data of a reference object and a to-be-measured object, and confirming an original position of point cloud of the to-be-measured object. The method further includes determining a virtual position of point cloud of the to-be-measured object at each iteration, determining a minimum distance between point cloud of the reference object and point cloud of the to-be-measured object, and determining coordinates of each point in point cloud of the to-be-measured object corresponding to the minimum distance. The method further includes moving each point in point cloud of the to-be-measured object.

Abstract: A method for generating a positron emission tomography (PET) attenuation correction map from magnetic resonance (MR) images includes segmenting a 3-dimensional (3D) magnetic resonance (MR) whole-body image of a patient into low-signal regions, fat regions, and soft tissue regions; classifying the low-signal regions as either lungs, bones, or air by identifying lungs, identifying an abdominal station, and identifying a lower body station; and generating an attenuation map from the segmentation result by replacing the segmentation labels with corresponding representative attenuation coefficients.

Abstract: A technique is provided for computing an atomic number of materials forming an object imaged by a radiological modality. The method includes accessing a first monochromatic image and a second monochromatic image of the object, the first monochromatic image acquired at a first energy level and the second monochromatic image acquired at a second energy level. A ratio of the mass attenuation coefficients between the first monochromatic image and the second monochromatic image may be obtained. The atomic number for a material of the object may be computed based upon the ratio of mass attenuation coefficients.

Abstract: An image processing device has: an image acquisition section that acquires a plurality of photographic images obtained by photographing the same subject at different photography times; and a part identification section that identifies a portion where a predetermined part among parts that form the subject in the photographic images appears. The device further has: a first processing section that applies first matching processing, which matches two images by transforming one or both of the two images, to two of the plurality of photographic images; and a second processing section that applies second matching processing, which matches two images by transforming one or both of the two images and whose application range of the amount of transformation required for matching is different from that of the first matching processing, to the portion identified by the part identification section in the two of the plurality of photographic images.

Abstract: An image generation apparatus generates a new video sequence with a high S/N ratio and suppressed motion blurs, from an original video sequence and a still image which are generated by capturing the same dark, moving object.

Abstract: A cycle steal mechanism for enabling a host processor to initiate and control the cycle stealing of data to or from a storage unit located in an I/O controller which is connected to the I/O channel bus of the host processor. The I/O controller also includes a microprocessor and a direct memory access (DMA) unit either or both of which can be used to control the transfer of data between the I/O controller storage unit and the host processor. Typically, the DMA unit is used for cycle stealing data between the controller storage unit and the host processor. The cycle steal mechanism of the present invention enables the host processor to also initiate and control the cycle stealing of data to or from the controller storage unit without interrupting the program running in the controller microprocessor and without interrupting the cycle stealing operations of the DMA unit.