Two-dimension code data processing method, device and mobile terminal

Abstract

Embodiments of the present invention provide a two-dimensional code data processing method, device, and mobile terminal, wherein the two-dimensional code data processing method includes: acquiring an image frame to be processed from a video stream, and processing the image frame to be processed It is a plurality of sub-image frames with different resolutions; using the first lightweight neural network model to determine whether the plurality of sub-image frames contain two-dimensional codes, and if it is determined that two-dimensional codes are included, then output a plurality of corresponding two-dimensional codes Code candidate frame; according to the output multiple two-dimensional code candidate frames, use the second lightweight neural network model to determine the optimal two-dimensional code candidate frame, and identify and calibrate the two-dimensional code in the optimal two-dimensional code candidate frame ; According to the calibration result, use a tracking algorithm to track the two-dimensional code on other image frames after the image frame to be processed in the video stream. Through the embodiments of the present invention, the identification and tracking efficiency of the two-dimensional code is effectively improved.

Description

technical field [0001] The embodiments of the present invention relate to the field of computer technology, and in particular to a two-dimensional code data processing method, device and mobile terminal. Background technique [0002] AR (Augmented Reality, Augmented Reality) technology is a new technology that "seamlessly" integrates real world information and virtual world information. Visual information, sound information, etc.), simulated and superimposed on the real information, the real environment and virtual objects are superimposed on the same picture or space in real time and exist at the same time. [0003] With the popularity of QR codes, a method of realizing AR based on QR code recognition has emerged. This method first selects the feature points of the image; then, extracts key points and generates descriptors; then, performs feature matching; then performs 3D space positioning through the 3D model, uses the feature points obtained by feature matching, and use...

AI Technical Summary

Problems solved by technology

The implementation of the image recognition and tracking process of the above method is relatively complicated, and has high requirements for hardware, so it is difficult to apply to mobile terminals to meet the real-time needs of mobile terminals when implementing AR services.

Images