An Artificial Perceptual Model Approach - Artificial Human Sensors: Science and Applications - page 115

a cognitive communication with the sensors, in order to achieve an optimal goal
and to receive the best available information.
The citation below, given in Bajcsy (1988), depicts an illustrative description
of the intention connected to active perceptual systems. However, in this citation
with reference to the visual sense in general;
we do not just see, we look
makes the conclusion that a corresponding passive perception sensor system then
can be illustrated as;
we do just see
In Bothe (1999), an illustrative example of the principles of an active system
shows the benefits of an active perceptual system by fusing auditory and visual
information. Two perceptual sensor types, a microphone set and a camera are used
to visually localise and follow a sound source, carried around a room by a person.
The auditory-vision sensor fusion approach illustrates the theory of merging two
human related sensing abilities in a corresponding fused picture. Other examples
of active perception-based models have been presented in a variety of applications,
e.g., fire detection, food and water quality assessment, Biel (2002a) and Iliev (2006),
with successful results.
During the last decade, a number of perception related artificial models have
been proposed at various levels of fusion as described earlier in Section 5.1.1. The
characteristic properties and the primary goal is clearly to provide a human in-
volved solution when fusing artificial sensor information. By those means a sys-
tem designer has also to consider, not only the human participation but also to
measure human-like perceptual properties.
The generic perceptual model proposed below has been developed and tested
in different situations. The perceptual model is split in different sub-processes into
separate units, each of them corresponding to a specific task, e.g., sensing, process-
ing and decision-making. The model has been successfully tested in a variety of
applications, for example, in a perceptual sensing device, comprising five separate
artificial sensors jointly integrated in a perceptual head device, Wide (2000).
The artificial based perceptual conception is often compared with the func-
tionality of the human brain and has in this topic only been intended to be used as
illustrative views of related functions. The biological outstanding ability, with its
highly specified performance will not be comparable to the artificial analogue sen-
sor systems. Therefore some sub-processes and functions are biologically inspired,
however, not restricted to these expected boundaries. Maybe the future will pro-
vide us with tighter biological-artificial solutions that will complement each other
in a blended integration of performance.
A comprehensive view describing the proposed generic sensor-based percep-
tion model with human related abilities, as shown in Fig. 5.8, is also given in Wide
(2007), concerning the background and intentions, in Loutfi (2005), the given con-
cept of the design.