We have proposed a new approach for the detection of events in image sequences. Our method relies on a number of logical sensors that can be defined over specific regions of interest in the viewed scene. These sensors measure time varying image properties that can be attributed to primitive events of interest. Thus, the logical sensors can be viewed as a means to transform image data to a set of symbols that can assist event detection and activities interpretation. On top of these elementary sensors, temporal and logical a ggregation mechanisms are used to define hierarchies of progressively more complex sensors, able to detect events having more complex semantics. Finally, scenario verification mechanisms are employed to achieve process monitoring, by checking whether events occur according to a predetermined order.
The framework proposed in this paper is particularly suited to the application area of monitoring of automated processes. In most such processes, things occur in a strict, predetermined way. For example, in an assembly automation process, mechanical parts move on a conveyor belt and are being manipulated by actuators in a process that, typically, presents no considerable deviations. The fact that these processes have considerable structure, permit us to turn difficult detection problems into much simpler verification problems. More precisely, instead of trying to detect what is going on in the viewed scene, the VBLS approach can be used to verify that things proceed as expected. This results in several advantages:
The proposed framework has been tested and validated in an application involving monitoring of automated processes. The obtained results demonstrate that the proposed approach, despite its simplicity, provides a promising framework for vision based event detection in the context of such applications.
The prototype software developed for testing the VBLS approach, while in operation.
Thomas Sarmis, Antonis Argyros
The electronic versions of the above publications can be downloaded from my publications page.