Information Technology Reference
In-Depth Information
Fig. 13.6
Sample frames of video snapshots misclassified by our method.
Blue
-bordered frames are
of
heavy
traffic videos misclassified as
medium
traffic.
Red
-bordered frames are of
medium
traffic
snapshots misclassified as
heavy
traffic.
Green
-bordered frames are of
medium
traffic snapshots
misclassified as
light
traffic.
Yellow
-bordered frames are of
light
traffic videos misclassified as
medium
traffic
In Fig.
13.6
, sample frames of misclassified video snapshots are presented. Our
method was unable to properly discriminate 13 out of 254 video snapshots. We
observe that construction of motion descriptors on the video snaphot level instead
of the frame level is one promising possibility to further proceed in order to reduce
confusion between different traffic situations.
13.4 Scenario Simulation
In this section, we present a scenario where we explain the advantages of using
different information sources such as surveillance video analysis in the IMA system.
As in many other huge cities, living in Berlin, Germany can be sometimes exhausting.
Every “Berliner” spends around 70 min on average in the traffic per day [
11
]. This
means that an efficient and intelligent routing solution is expected to have a big
impact in our daily lives. It also explains why we need a system like IMA.
Search WWH ::
Custom Search