Multifunctional inspection positioning method based on depth visual identification

The invention relates to the technical field of positioning inspection, in particular to a multifunctional inspection positioning method based on depth visual identification. The method mainly aims at solving the problems that an existing inspection robot is insufficient in error accuracy in positioning of collected images, consequently, a large amount of manual input work is caused, and the inspection positioning accuracy needs to be improved. According to the technical scheme, the multifunctional inspection positioning method comprises an image acquisition system, a coordinate establishment system, an identification judgment system, an inspection comparison system, a positioning system, a judgment system, a deep learning system and an early warning system. Compared with the prior art, the method has the advantages that equipment inspection comparison is more visual, double-path shooting comparison is achieved, the comparison precision is improved, meanwhile, the deep learning mining comparison function is a