51. Clustering object on surveillance radar display by combining ADBSCAN algorithm and marked boundary object
- Author
-
Tajul Miftahushudur, Arie Setiawan, Vicky Zilvan, and Arief Nur Rahman
- Subjects
DBSCAN ,Computer science ,business.industry ,010401 analytical chemistry ,020206 networking & telecommunications ,02 engineering and technology ,Object (computer science) ,01 natural sciences ,0104 chemical sciences ,law.invention ,law ,0202 electrical engineering, electronic engineering, information engineering ,Canopy clustering algorithm ,Computer vision ,Noise (video) ,Artificial intelligence ,Radar ,Radar display ,business ,Cluster analysis ,Algorithm ,Secondary surveillance radar - Abstract
The received information on radar processor that collected from radar transmitter sometimes are not good enough to be displayed on the radar display. This problem makes user difficult to distinguish noise and object. One technique that can be used to solve this problem is by grouping the objects based its density. One of clustering method used to handle object clustering automatically is DBSCAN algorithm. Unfortunately, this algorithm has a high complexity that will interfere the radar performance. To overcome both of the problems, need a quick method that able to classify the object and the noise. This paper proposed a new clustering algorithm by combining ADBSCAN algorithm using Marked Boundary Object (MBO) as the policy connectivity. The developed algorithm has successfully implemented on PPET-LIPI radar display. This algorithm is shown promising running time performance than the original DBSCAN algorithm (increase of approximation 71%).
- Published
- 2016
- Full Text
- View/download PDF