Deep network and beacon assistance-based quick identification system and method

Abstract

The invention provides a deep network and beacon assistance-based quick identification system and method, and belongs to the field of artificial intelligence. The system comprises an image device, anintelligent device, a wireless receiving device, a deep server and an identification execution device; the image device transmits an image to the deep server; the intelligent device generates a beaconaccording to a rule and transmits the beacon to the wireless receiving device; the wireless receiving device transmits the beacon to the deep server; the deep server maintains an effective beacon set, processes the image through a deep identification system to obtain a deep network identification target, retrieves the identification target in the effective beacon set, obtains an effective judgment result, and sends the effective judgment result to the intelligent device and the identification execution device; the intelligent device displays the effective judgment result through a mobile APP;and the identification execution device executes subsequent businesses of the effective judgment result by executing the APP. The system and the method aim to solve the problems of wrong identification and low security caused by the fact that an existing deep network algorithm cannot identify untrained image targets by 100% due to technical defects.

Description

technical field [0001] The invention belongs to the field of artificial intelligence, and in particular relates to a fast identification system and method based on deep network and beacon assistance. Background technique [0002] The deep learning algorithm model based on neural network has become a research hotspot in the field of artificial intelligence, and the application of image recognition technology is becoming more and more extensive. When traditional image recognition technology faces a large number of categories and a complex environment, the recognition effect is often unsatisfactory. Compared with the traditional image recognition method, the deep learning system does not need complex image preprocessing, and the network can learn the generalized features by itself without artificially designing features, and its recognition effect is far greater than the recognition rate of the traditional network. However, the general deep learning algorithm will lose the str...

AI Technical Summary

Problems solved by technology

[0003] The present invention provides a fast recognition system and method based on deep networks and beacons to solve the problem that existing deep network algorithms cannot 100% identify untrained image targets due to technical reasons, which may easily lead to possible misidentifications, and based on The identification device of the deep network has a single security defect due to access authentication

Images