detecting objects of  interest  that  may  be  moving  at  • Systems which detect objects of interest only while times  and  stationary  at other times。 they are moving。 Once objects of interest stop, they

This  research  goes  beyond  merely  detecting  the become invisible to the motion detection   scheme。 presence  of an object。 We also connect the  detection • Systems  which  function  properly  only  when  the module  to other important  sensory components  of  a camera  is  either   stationary   or  moving,  but  not vision-based robot control scheme。   Particularly both。   To   the   contrary,   our   system    generally significant is the ability to find landmarks on   objects operates  under  static  camera  conditions,  but also of interest  and to know about the projected  shape  of allows the freedom  of visually servoing an  eye-in- objects。  In addition,  tracking  techniques  are used to hand  system based  upon  target location。

monitor objects without human intervention。  Our  • Systems  which  cope  with  a  single  moving  target, solution    to    the    tracking    problem    follows  the even   though   several   application   areas  involve controlled  active vision  framework,'  which  avoids a images with several targets。

heavy  reliance  on  a priori  information  through  the     • Systems  which  assume  that  a  moving  object  is a use  of  optical  flow。 Optical  flow is induced  by any rigid  body。  Further  assumptions  may  include    a combination  of camera  or  object motion。 specific pattern of motion for the object of  interest。

The  primary   contribution   of  this  research   is  a

complete software and hardware implementation of our detection framework。 In the process of con- structing this system, we have selected and modified computer vision techniques that are  appropriate  to the visual detection problem。 Many of the techniques used by our framework (e。g。 frame-differencing) have also appeared in similar forms in existing research, which is a demonstration of their usefulness。 Our solution to the detection problem is innovative in the way in which it has uniquely combined these techniques into a general framework that can be directly applied to real-world situations。 We have made modifications to these techniques where necessary, and we have also incorporated our own ideas where the existing literature was lacking (e。g。 dynamic segmentation domains)。 Finally, we have customized the theory to the specihc industrial needs of robotics。 This has demonstrated that our frame- work can provide precisely the type of information required to manage effectively a situation requiring visual detection。 Results from experimentation with our visual servoing systems show the potential of our approach  under  general conditions。