Intelligence Mobile Robotics Home Work 02

Intelligence Mobile Robotics
Home Work 02

1 Task One
Figure 1: An example scanned view of a laser
Task one is related line the extraction techniques which you have studied in class. Your task is
to scan your room with the provided laser and try to extract the lines and segment them properly.
There are two HOKUYO laser sensors at your service. You can get them in the laboratory (room
466a) and bring them back during working hours. There is a data acquisition software available
for this sensor (check the manufacturer’s website) or else you may use any software for processing,
but provide them with the links and instructions on running your code. In your report, include the
room description, used implementations of algorithms. For the line extraction, you must develop
your own implementation of split-and-merge and line-regression. You may use any existing implementation for Hough transform. In the end, compare the results of the line extraction in terms of
algorithm speed (absolute time). (Points: 2.5)
1
2 Task Two
Id Name
01 Sabirova Adelya
02 Ahmed Nawaz
03 Andrey Stepanov
04 Arslan Siddique
05 Aydar Ahmetzyanov
06 Dmitriy Desyatkin
07 Lyailya Aminova
08 Maksim Rassabin
09 Oleg Balakhnov
10 Sami Sellami
11 Valeriya Skvortsova
12 Victor Massague Respall
13 V´ıctor Fernando Perez Garc ´ ´ıa
In this task, you are going to touch upon a step into a tracking problem especially into data association. Each of you is provided with an image sequence of a scenario. After browsing your image
set, try to seek an object which is moving through a few images (at least 5 images) continuously.
Locate a bounding box for a chosen object which can be done either manually or automatically.
Assuming object is not going away from the initial bounding box, try to find a your own feature
descriptor such as SIFT for the bounding box and calculate same feature descriptor for successive
images as well. Then, apply a similarity measurement technique to see the similarity. Explain characteristics of your feature descriptor while mentioning why you choose those characteristics result
in the report. (Points: 5.5, if you use SIFT without your own feature descriptor, you would get
up to 3.5 at max.)
3 Submit
What should you turn in? Please, upload the single zip file which includes your source code (task
01 and task 02), dataset you used task 01 and report for both tasks.