Abstract: A method for tracking objects entering and exiting an environment comprises providing stereo vision cameras positioned proximate to the paths to enter the environment, such that the stereo vision fields associated with the cameras substantially cover the paths. Using image data obtained by the stereo vision cameras, entry signatures are determined for objects entering the environment and exit signatures are obtained for objects exiting the environment. At any time after the determination of at least one exit signature, exit signatures are matched with the set of available entry signatures. Matching may be performed according to a variety of techniques.

Abstract: A system for localizing objects has two or more measurement units. Calibration methods provide transformations to a common reference frame by locating calibration features in local reference frames of the measurement units. A calibration target spanning measurement fields of the measurement units may be used. The measurement units may have overlapping measurement fields. The measurement units may comprise stereo vision units or other units capable of locating objects or object features.

Abstract: A system for measuring dimensions of cuboid objects may be used to measure dimensions of boxes travelling on a conveyor. The system has a pair of digital cameras which obtain images of one or more lines projected by a projector on upper surfaces of the objects. The height of the upper surfaces can be determined by triangulation. Lengths and widths of the upper surfaces are measured by locating endpoints of the projected lines and fitting rectangles to the located endpoints. The system does not require rigid mounting of the projector relative to the cameras. A third camera may be used to determine speed and direction of the conveyor motion.

Abstract: A method for tracking objects entering and exiting an environment comprises providing stereo vision cameras positioned proximate to the paths to enter the environment, such that the stereo vision fields associated with the cameras substantially cover the paths. Using image data obtained by the stereo vision cameras, entry signatures are determined for objects entering the environment and exit signatures are obtained for objects exiting the environment. At any time after the determination of at least one exit signature, exit signatures are matched with the set of available entry signatures. Matching may be performed according to a variety of techniques.

Abstract: A system for localizing objects has two or more measurement units. Calibration methods provide transformations to a common reference frame by locating calibration features in local reference frames of the measurement units. A calibration target spanning measurement fields of the measurement units may be used. The measurement units may have overlapping measurement fields. The measurement units may comprise stereo vision units or other units capable of locating objects or object features.

Abstract: A stereo camera system suitable for use in high performance computer vision systems has two or more electronic cameras. Each camera has an assembly comprising a light sensing array and a lens assembly which are each directly mounted to a rigid mounting member. The rigid mounting member may be, for example, a thick metal plate penetrated by a plurality of apertures which provide light paths between the lens assemblies and the light sensing arrays. Stereo camera systems according to the invention are more rugged than current stereo camera systems. Calibration is maintained for longer periods of time. Relative motion between the two or more cameras is virtually eliminated. Relative motion between the optical components of individual cameras is virtually eliminated.