Line-Circle: A Geometric Filter for Single Camera Edge-Based Object Detection

Abstract

This paper presents a state-of-the-art approach in object detection for being applied in future SLAM problems. Although, many SLAM methods are proposed to create suitable autonomy for mobile robots namely ground vehicles, they still face overconfidence and large computations during entrance to immense spaces with many landmarks. In particular, they suffer from impractical applications via sole reliance on the limited sensors like camera. Proposed method claims that unmanned ground vehicles without having huge amount of database for object definition and highly advance prediction parameters can deal with incoming objects during straight motion of camera in real-time. Line-Circle (LC) filter tries to apply detection, tracking and learning to each defined experts to obtain more information for judging scene without over-calculation. In this filter, circle expert let us summarize edges in groups. The Interactive feedback learning between each expert creates minimal error that fights against overwhelming landmark signs in crowded scenes without mapping. Our experts basically are dependent on trust factors' covariance with geometric definitions to ignore, emerge and compare detected landmarks. The experiment for validating the model is taken place utilizing a camera beside an IMU sensor for location estimation.

Line–Circle–Square (LCS): A multilayered geometric filter for edge-based detection

“…The kinematics of a moving robot is sketched as Figure 2, where a holonomic robot with velocity of travels distance in the direction of axis. In this work, we consider a motion field model rather than a linear rotational models that was used in our previous study [36]. The rotation is getting updated by the motion field kinematics where the translation is decoupled from the rotation [6].…”

“…The rotation is getting updated by the motion field kinematics where the translation is decoupled from the rotation [6]. The location of landmarks/layers ( − 1) = [ , ] from − 1 step including the origin points (rebel landmarks/layers have their own origin [36]) are updated by…”

“…Also, newly appeared unmatched edges with are added as a single landmark in . Grouping temporary parameter detects candidate edges in by their distance ∥ − ∥ from distance radius [36]. Next, the ignorance parameter removes the edges via including the expressed regional constraints for different geometric flag as…”

“…The LCS filter is a multilayered geometric filter for edge-based detection. In our previous work, we did a preliminary study on the Line-Circle filter [36], the predecessor of the LCS filter. The algorithm of the LC filter was run offline with limited performance evaluation in the scene.…”

“…The algorithm of the LC filter was run offline with limited performance evaluation in the scene. Compared with the LC filter [36], we have optimized computation flow, restructured the algorithmic conditions and improved the code for real-time applications in all of the experts. The LCS filter is advanced with more inter-active information transfer between each expert.…”