Masato Iehara

Human behavior recognition using unconscious cameras and a visible robot in a network robot system

We developed a network robot system integrating various types of robots via ubiquitous networks that introduce an interactive robot into areas in which people are located. In this paper, we present a human behavior recognition method necessary for providing guidance in a public space, which uses a tangible network robot system composed of a mobile robot and vision sensors embedded in an environment. We define some basic features that a person typically exhibits when in need of guidance. Various human behaviors were successfully interpreted by first recognizing the basic features and then forming their different combinations.

Non-uniform azimuth image shift caused by the pitching motion of ships observed in Radarsat images

Non-uniform azimuth image shift of a rigid body is observed, for the first time, in the Radarsat images of moving ships in pitching motion. The pitching velocity is estimated from the SAR image of a known ship, and compared with that computed using a numerical model with the meteorological and ships data; further comparison is made with the wave orbital velocity. The results are shown to be all in excellent agreement. Comparisons are also made between the pitching velocities of unknown ships with the wave orbital velocity, and a reasonable agreement is obtained.