应用程序GUI组件自动化识别方法研究综述  

Overview of Research on Automated Recognition of Application GUI Components

作　　者：张中洋 ZHANG Zhongyang(College of Computer Science,Sichuan University,Chengdu 610065)

机构地区：[1]四川大学计算机学院,成都610065

出　　处：《现代计算机》2021年第17期124-127,共4页Modern Computer

摘　　要：在面向现代软件GUI(用户图形界面)开发中,开发人员需要首先将图片形式的原型设计稿中的界面组件进行识别,而这个原型迭代的过程费时费力,更由于设计师与开发者因背景领域知识而广泛存在的理解鸿沟,导致开发人员错误地理解了界面组件的设计意图,从而造成开发结果与预期不符,影响了开发效率。目前越来越多的研究人员和从业人员开始研究如何自动化地识别GUI中的组件,减轻设计师与开发者的工作负担,并将识别结果运用到GUI代码自动生成或GUI自动化测试等多种不同类型的软件工程任务中。对当前主流的GUI组件的自动化识别研究方法进行分类并综述,分析当前研究面临主要问题和挑战。In the development of modern software GUI( Graphical User Interface),the developer needs to first identify the interface components in the prototype design draft in the form of a picture,and this prototype iteration process is time-consuming and labor-intensive. The wideranging understanding gap in knowledge has caused developers to misunderstand the design intent of interface components,causing development results to be inconsistent with expectations and affecting development efficiency. At present,more and more researchers and practitioners have begun to study how to automatically identify components in GUI,reduce the workload of designers and developers,and apply the recognition results to GUI code automatic generation or GUI automatic testing. Types of software engineering tasks. Classify and summarize the current mainstream GUI component automatic identification research methods,and analyze the main problems and challenges that current research faces.

关 键 词：应用程序 用户图形界面 界面组件 自动识别 

分 类 号：TP311.56[自动化与计算机技术—计算机软件与理论]