Basic Theories and Methods of Target's Height and Distance Measurement Based on Monocular Vision

Achieving target information measurement without increasing the amount of data processing is an important while difficult issue in computer vision. In this paper, we present a target’s height and distance measurement theory of monocular vision. Firstly, based on the geometric model of camera imaging as well as the principle of converting digital signals to analog signals, we derive a model to reveal the relationships between target distance, target height, camera height, image resolution, image target size and camera parameters. Then, we theoretically analyze and prove the following three common but critical statements in computer vision: 1) under the condition that the target height is unknown, a single image is not sufficient to calculate the target distance; 2) under the conditions that the target height is unknown and "self-invariant", the target distance cannot be calculated even there are multiple views available; 3) under the condition of "self-variant", the target distance can be calculated using two different views. The correctness of the proposed theory and corresponding methods have been verified via experiments