Research on Real-time Visual Information Processing

实时视觉信息处理研究

• 批准号: 04044197
• 负责人: YACHIDA Masahiko
• 金额: --
• 依托单位: Osaka University
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Overseas Scientific Survey.
• 财政年份: 1992
• 资助国家: 日本
• 起止时间: 1992 至 无数据
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-04044197/
• 关键词: Mobile Robot; Image Understanding; Visual Navigation; Real Time; Omnidirectional Image Sensor; Stereo Camera

In this research, we built the method for navigating the robot and understanding interesting objects in an environment by using the omnidirectional image sensor (COPIS) and binocular vision.The robot can always estimate its own location and motion precisely because COPIS observes a 360 degree view around the robot. Based on the assumption of constant linear motions of the robot and objects, the objects moving along collision paths are detected by monitoring azimuth changes.The mobile robot has binocular vision and obtains a sequence of stereo images from the environment. We extend the principle of trinocular vision to establish correspondences between a sequence of binocular images and build the 3-D wireframe model of the environment. From this wireframe model, we recognize surfaces and spatial relationships between them and understand spatial configuration of the environment.We are planning to combine both omnidirectional image sensor COPIS and binocular vision.

在本研究中，我们利用全方位影像感测器（COPIS）与双目视觉，建立机器人导航与了解环境中感兴趣物体的方法，因为COPIS可观察机器人周围的360度视野，所以机器人可以一直精确地估计自己的位置与运动。基于机器人与物体作匀速直线运动的假设，通过监测物体方位角的变化来检测物体沿着碰撞路径的运动，使移动的机器人具有双目视觉，并从环境中获取一系列立体图像。我们扩展了三目视觉的原理，建立了一系列双目图像之间的对应关系，并建立了三维线框模型的环境。从这个线框模型，我们认识到表面和它们之间的空间关系，并了解空间配置的环境。我们计划联合收割机结合全向图像传感器COPIS和双目视觉。

项目成果

A. Utsumi: "Acquiring Surface Information from Wire-ffame Model" MIRU'92 (In Japanese). Vol.II. II423-II430 (1992)

A. Utsumi：“从线模型获取表面信息”MIRU92（日语）。

A.Utsumi: "Recognizing Surface and Spatial Structure of Environment from Wire-frame Model Obtained by Motion Stereo" Scandinavian Conference on Image Analysis(1993). (1993)

A.Utsumi：“从运动立体获得的线框模型中识别环境的表面和空间结构”斯堪的纳维亚图像分析会议（1993）。

Y.Yagi: "Map Based Navigation of the Mobile Robot Using Omnidirectional Image Sensor COPIS" Proc.IEEE Int. Conf.Robotics and Automation. Vol.1. 47-52 (1992)

Y.Yagi：“使用全向图像传感器 COPIS 的移动机器人的基于地图的导航”Proc.IEEE Int。

Y. Yagi: "Obstacle Avoidance for Mobile Robot Integrating Omnidirectional Image Sensor COPIS and Ultrasonic Sensor" Proc. 2nd Int Conf.Automation Robotics and Computer Vision. Vol.3. 11.7.1-11.7.5 (1992)

Y. Yagi：“集成全向图像传感器COPIS和超声波传感器的移动机器人避障”Proc。

A.Utsumi: "Estimating Surface and Spatial Structure from Wire-frame Model Using Geometrical & Heuristical Relation" Proc.IAPR Workshop on Machine Vision and Applications. 25-28 (1992)

A.Utsumi：“使用几何模型从线框模型估计表面和空间结构