Contour Search

Finding the contour of an object, known as contour search, has a large variety of applications. Beside the conventional applications in image processing, there are other applications in robotics. Imagine a scenario of identifying the contour of wildfire or oil spill on the ocean. The major challenges concern searching for an object and then identifying the corresponding contour. The existing algorithms typically rely on image processing technologies, which typically require knowledge about the entire search space. In our research, we work on a new algorithm for contour search which is based on Particle Swarm Optimization (PSO). We unify both of the search processes for the object and the contour identification in one algorithm. We consider the contour search of an object with an unknown shape which is located in an unknown position in a two-dimensional search space. Similar to optimization scenarios, the goal is to find and identify the contour (border) of any encountered part of the object. This research is being published in the following paper: 

Here are two videos on a sample test problem with both convex and concave contours: 

 

 

 

Last Modification: 17.01.2024 - Contact Person: Webmaster