Video distance measurement based on focus

Abstract

In monovision field- distance determination with a single camera or video- the methods for determining distances are varied and we can divide them as follows: a) based on focusfixing the lens in the position corresponding to the maximum sharpness of the acquired image (DFF- depth-from focusing); b) based on the blur / defocus of the lenses - (DFD - depth-from defocusing); c) combined methods. The DFF method scans the “scene” by taking sequences of images with various lens settings and tries to find the best focus. This paper presents a method of measuring distances based on focus. It is used a web video camera. To achieve the autofocus, the camera lens moves to a position where the clearest image is obtained. The maximum clarity is measured from the histograms of the images on which a filter with the role of highlighting the edges was initially applied. Knowing the focal length of the lens, the distances can be found from the lens equation (1). Theoretically, for each position of the lens, in which the maximum clarity is reached, it corresponds a measured distance between the object and the camera. Because the camera used has only 16 positions of the lens, each position corresponds to a range of distances, this range being of the order of centimeters. Measurement accuracy can be improved by applying a linear and parabolic interpolation process. In this way the errors reach the order of millimeters. The distance measuring device in the present paper consists of a fixed video camera, a PC with the role of information processing and an object moving in front of the camera on its optical axis. Both the camera and the object are arranged on a graduated ruler support on which the camera is calibrated.