Abstract

Context. Computational method for determining projective transformation constants of the current frame into basic coordinate system is developed. Choosing the best projective transformation allows finding the best settings for adding frames. It is the only way to prepare the license plate image to automatic distinction or distinction with the help of an expert at night. Objective. This computational method enables the transformation of the current frame license plate image into the image of the base frame license plate that is free of various types of distortions. The article provides direct assessment of the section parameters of digital image using the method of moments, as a simple, reliable and commonly used in practice. To set the projective mapping inverse mapping is used so that each reference point of the license plate on resulting photos is associated with coordinates on the license plate plane. The corners of the license plate are used as reference points. The considered metod allows adding images of license plates. The proposed method provides for reliable initial approximation of a nonlinear transformation and searching constants corresponding to it. The transition to non-linear transformation is required, e.g. when there are curvatures of license plates or other text data carriers.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call