Automated Generation of Automated Visual Inspection (AVI) Routines for Surface Mounted Assembly Systems

自动生成表面贴装装配系统的自动视觉检测 (AVI) 例程

• 批准号: 0300361
• 负责人: Jesus Villalobos
• 金额: --
• 依托单位: Arizona State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2003
• 资助国家: 美国
• 起止时间: 2003-08-01 至 2006-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0300361&HistoricalAwards=false
• 关键词: Automated; Generation; Visual; Inspection; AVI

This research proposes the development of a methodology that will result in the automated generation of inspection algorithms of new surface mounted Technology (SMT) components. The envisioned methodology will make possible the development of automated visual inspection (AVI) systems that develop and optimize their own inspection algorithms for new components. Under this notion the resulting development framework will minimize the intervention of human developers to obtain inspection algorithms that meet minimal performance measures in terms of component discrimination and sensitivity to changes in environmental variables. This methodology will be based on a vector discrimination approach for the identification of defective components whose basic characteristics have already been developed. This vector approach, in conjunction with a hierarchical classification, will be used to obtain the envisioned methodology. This methodology does not seek the replacement of the human developer; the proposed framework will free him/her from tedious tasks, so that s/he can focus on more complex tasks such as the development of morphology-based features that may be needed for more sophisticated inspection tasks. If successful, this research will significantly shorten the time required to develop inspection algorithms for SMT components. This reduction in development time will result in added manufacturing flexibility that will translate into more rapid introduction of new products into existing manufacturing lines and shorter design lead-times. This flexibility will contribute to give a competitive advantage to the electronics industry, which has a clear strategic importance for the USA. It is also expected that as a direct result of the funded activities the PI expand the concept of multi-disciplinary pyramidal research teams. This concept has given excellent results in terms of recruiting undergraduate students to continue on to graduate school and also in terms of the participation of minority students in research activities.

本研究提出了一种方法的开发，该方法将自动生成新表面安装技术（SMT）元件的检测算法。设想的方法将使自动视觉检测（AVI）系统的开发成为可能，该系统为新组件开发和优化自己的检测算法。根据这一概念，由此产生的开发框架将最大限度地减少人类开发人员的干预，以获得在组件歧视和对环境变量变化的敏感性方面满足最低性能指标的检测算法。该方法将基于矢量鉴别方法，用于识别基本特性已经确定的缺陷组件。将采用这种矢量方法，再加上分级分类，以获得设想的方法。这种方法并不寻求取代人类开发人员;所提出的框架将使他/她从繁琐的任务中解放出来，以便他/她可以专注于更复杂的任务，例如开发可能需要更复杂的检查任务的基于形态学的功能。如果成功，这项研究将大大缩短为SMT组件开发检测算法所需的时间。这种开发时间的减少将导致增加制造灵活性，这将转化为更快地将新产品引入现有生产线和更短的设计交付时间。这种灵活性将有助于为电子行业提供竞争优势，这对美国具有明显的战略重要性。预计作为资助活动的直接结果，PI扩大了多学科金字塔研究团队的概念。这一概念在招收本科生继续攻读研究生以及少数民族学生参与研究活动方面取得了很好的成果。