A distance measuring apparatus which can accurately measure the distance to an object and keep track of the object. In the apparatus, a pair of image-pickup optical systems pick up an object. A window including the object is set in an image picked up by one of the image-pickup optical systems. The distance to the object is measured by calculating correlation between the image in the window and an image picked up by the other image pickup optical system. Tracking an object is also possible by taking a correlation of images in the window formed in a time sequence. Adequate exposure is determined from an image signal in the window so that the image pickup optical systems are controlled to set the adequate exposure to the image in the window. Contrast of an image in the window becomes distinct, thus enabling accurate distance calculations and tracking of an object.
A method and apparatus for detecting objects from a sequence of images of a scene containing an object by using two distinct methods for object detection. One is suited for well-lit scenes (e.g., daytime), while the other is suitable for poorly-lit scenes (e.g., nighttime) where the objects have lights mounted on them. Further, the method and apparatus are adaptive to the image statistics of the objects being detected, and can be programmed to filter out "weak" detections that lie below a certain percentile in the observed statistical distribution of image measures. The specific percentile is determined based on whether the scene has been determined to be well- or poorly-lit and whether it contains shadows or not.
An optical apparatus for performing multi-point distance measurement by measuring the distance to an object using a triangulation method and light received from the object, includes first and second light-receiving systems each constituted by a multi-eye lens in which a plurality of elementary lenses are disposed adjacent to each other in a base line direction, and a photoelectric conversion element. The first and second light-receiving systems respectively form images of regions of the object, which correspond to the elementary lenses constituting the multi-eye lenses and are separated from each other, on the photoelectric conversion elements.
An ITV camera picks up an image of an image pickup region including a moving object and a fixed object which serves as a reference to measure a distance. Images are fetched through the ITV camera at a first aperture which discriminates between a background and the fixed object and a second aperture which discriminates between the background and the moving object. The images thus fetched are added together and edges (outlines) of the fixed object and the moving object from the added image. A distance between the fixed object and the moving object is calculated from the extracted edges. The optimal aperture, enlargement ratio and focal lengths are automatically determined in a short period of time in accordance with the image pickup region and an optimal image is provided.
A method and a device for maneuvering a motor vehicle out of a parking space, in the case of which neither the size nor the clearance of obstacles is known. By means of sensors mounted on the motor vehicle, the clearance between the motor vehicle and surrounding obstacles and the position and steering angle of the motor vehicle are determined, and continuously updated. On the basis of this data, an up-to-date, local map of the surroundings is placed in a computer module. A polyline is used to connect the detected obstacles to form a barrier. From this data, a model of the surroundings is produced, with which a driving strategy for maneuvering the motor vehicle out of the parking space is determined. Driving-direction pointers indicate the appropriate driving direction to the driver of the motor vehicle.
A 3D vision system utilizes a CCD camera to first generate a two dimensional image of the object. A computer analyzes the two dimensional image to determine areas of interest, such as high contrast areas which indicate sharp discontinuities. A laser range scanner, controlled by the computer, then scans the object at a variable resolution, increasing its resolution in the areas of interest to generate a computer model.
