The invention belongs to the field of machine vision, and particularly provides a three-dimensional feature extraction method and device based on machine vision. The invention aims to solve the problems of complex and time-consuming three-dimensional model reconstruction process, difficulty in popularization and the like in the prior art. In order to achieve the purpose, the three-dimensional feature extraction method based on machine vision comprises the steps of obtaining multi-angle images of preset to-be-detected feature points containing a target object; extracting position information ofthe feature points to be detected in each image; acquiring spatial position information of the to-be-tested feature point according to the position information of the to-be-tested feature point in each image; And calculating first distance information and/or second distance information corresponding to a certain to-be-measured feature point based on the spatial position information and a preset three-dimensional feature category. And obtaining different angle images containing the to-be-measured feature points through machine vision, and further obtaining spatial position information of the to-be-measured feature points so as to calculate and obtain distance information of the target object.