Method and system for object tracking in multiple non-linear distortion lenses

An object tracking system, comprising:

• a database, storing an object identification model, an object comparison model and a coordinate mapping model;
• and
• a processor electrically connected to the database, being configured to execute the following steps:
• (a) reading the database to obtain the object identification model, the object comparison model and the coordinate mapping model;
• and
• (b) inputting a first image captured by a first image capturing apparatus and a second image captured by a second image capturing apparatus into the object identification model to identify a plurality of pieces of first image object information in the first image and a plurality of pieces of second image object information in the second image respectively;
• (c) inputting the plurality of pieces of first image object information and the plurality of pieces of second image object information into the object comparison model to find a plurality of groups of corresponding object information through comparison;
• (d) dividing the plurality of groups of corresponding object information into at least one group of reference corresponding object information and a group of to-be-verified corresponding object information, wherein the group of to-be-verified corresponding object information comprises a piece of first to-be-verified image object information and a piece of second to-be-verified image object information, inputting the at least one group of reference corresponding object information and the first to-be-verified image object information into the coordinate mapping model to obtain a piece of first output object information, determining that the first output object information is consistent with the second to-be-verified image object information, and repeating the step (d) until all the plurality of groups of corresponding object information have been verified;
• (e) inputting a first new image captured by the first image capturing apparatus into the object identification model to identify at least one piece of first new image object information in the first new image;
• and
• (f) inputting the plurality of groups of corresponding object information and the at least one piece of first new image object information into the coordinate mapping model to obtain at least one piece of first output new object information in the second image capturing apparatus that corresponds to the at least one piece of first new image object information.