An Intelligent Machine Vision System for Automated Inspections Tasks

用于自动检测任务的智能机器视觉系统

• 批准号: 8714809
• 负责人: Linda Shapiro
• 金额: $ 28.03万
• 依托单位: University of Washington
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1988
• 资助国家: 美国
• 起止时间: 1988-09-01 至 1991-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=8714809&HistoricalAwards=false
• 关键词: Intelligent; Machine; Vision; System; Automated

The principal investigator will develop an automated inspection system that bridges the gap from computer aided design (CAD) models to machine vision algorithms. The system will start with a CAD model of a part augmented by information on the material and texture of the part, constraints on the camera and lighting configurations, and a set of inspection specifications. It will produce a vision model and a vision procedure that can be used to automatically inspect the part. The main thrust of this research will be to develop an off-line system that can (1) predict features that will appear in images of the part under various conditions, (2) suggest alternate lighting and sensor arrangements for certain objects, and (3) automatically generate the vision procedures for inspecting the part.

首席研究员将开发一种自动检测系统，以弥补计算机辅助设计 (CAD) 模型与机器视觉算法之间的差距。 该系统将从零件的 CAD 模型开始，并添加有关零件的材料和纹理的信息、相机和照明配置的约束以及一组检查规范。 它将生成一个视觉模型和一个视觉程序，可用于自动检查零件。 这项研究的主要目标是开发一个离线系统，该系统可以（1）预测在各种条件下零件图像中出现的特征，（2）建议某些物体的替代照明和传感器布置，以及（3）自动生成用于检查零件的视觉程序。