Interface identification method, device, electronic device and storage medium

Abstract

Embodiments of the present invention provide an interface recognition method, device, electronic equipment, and storage medium. The method includes: inputting the interface to be processed into a pre-trained network model to obtain interface features; using the interface features and the pre-recorded The graphic element template feature calculation response graph, the graphic element template feature has a corresponding interface type; get the matching score of the graphic element template feature according to the response graph; according to the matching score, from the graphic element template feature Determine the feature of the target graphic element template; identify the interface type of the interface as the interface type corresponding to the feature of the target graphic element template. The graphic element template feature of the embodiment of the present invention can define an interface type, so by matching the interface feature of the interface with the graphic element template feature, the target graphic element template feature can be determined, and the interface type of the interface can be determined, which improves the interface recognition accuracy.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of interface processing, and in particular, to an interface identification method, an interface identification device, an electronic device, and a storage medium. Background technique [0002] For the developed interface (User Interface, user interface), it is necessary to write and debug test scripts to perform automated testing on the interface. Among them, in the process of testing the interface, it is necessary to identify the interface, such as identifying whether the interface is a certain type of interface that has been tested in a test process, which helps to accurately describe and record the test in the interface test. process. [0003] Specifically, there are many technical solutions for current interface recognition, such as: 1) training an image classifier based on traditional machine learning methods such as support vector machines; 2) training a neural network-based ...

AI Technical Summary

Problems solved by technology

Specifically, 1) and 2) can be carried out only based on images, but the premise is that a large amount of human-labeled data is required for training, and 3) needs to provide a UI tree acquisition interface to perform normally. Although a large amount of human-labeled data is not required, the operability is poor.

It can be seen that the current existing interface recognition technology solutions either require a lot of manpower to mark the training data when identifying the interface, or the accuracy of interface recognition is not high, or the operability in actual game testing is not high.

Images