The invention relates to a machine vision-based automatic verification system for multipath hand-held digital multimeters. According to the technical scheme of the invention, a hand-held digital multimeter is automatically controlled to select different function range gears; according to to-de-verified items, controlling a multifunctional calibration source to output standard signals (alternating-current voltage, direct-current voltage, alternating current, direct current, resistance value and the like) to a multipath to-de-verified digital multimeter, and conveying the measurement data of the panel of the to-de-verified digital multimeter to a computer based on the image collecting, processing and recognition function at the same time. In this way, verification data can be automatically recorded, and then a verification report and a verification certificate are generated. After the verification process, a database is established and then verification records and verification data are stored over the cloud through a cloud server. Meanwhile, a verification result is generated in the form of a two-dimensional code, so that the verification result can be conveniently recorded and searched. Meanwhile, the data sharing is facilitated. Compared with the manual verification mode of a conventional digital multimeter, the working efficiency is greatly improved. Moreover, errors caused by human factors are reduced and the comprehensive quality management is conveniently realized.