An extended-HCT semantic description for visual place recognition

Abstract

We describe a new semantic descriptor for robots to recognize visual places. The descriptor integrates image features and color information via the hull census transform (HCT) and image histogram indexing. Our approach extracts the semantic description based on the convex hull points and statistical calculation. The color histograms are then formed by four indices and added to the descriptor. The semantic codebook consists of several places with many image descriptors. Finally, a one-versus-one (OVO) multi-class support vector machine (SVM) is used to model the places. The proposed technique is achieved by using a high-level cue integration scheme based on the learning information over the color and feature space to optimally combine the weighted cues. It is suitable for visual place recognition, particularly for the images captured by an omnidirectional camera. The experimental results show that the codebook with less vectors is as robust as most popular codebooks under varying environments. The performance is evaluated and compared with several state-of-the-art descriptors.

Bubble space and place representation in topological maps

“…The experiments are similar to those that were done in recent and related work (Ullah et al, 2008; Wang and Lin, 2011). The robot first learns each place at each laboratory in one illumination condition via constructing the corresponding bubble descriptors at a set of base points using one route sequence.…”

“…The first set of experiments is done with COLD data set (Pronobis and Caputo, 2009) that has been used extensively to evaluate previous approaches (Ullah et al, 2008; Wang and Lin, 2011). The COLD data set consists of visual data from three different laboratories with different illumination conditions (the cloudy, sunny and night conditions).…”

“…Egocentric representation have also been used via projection onto spheres which are then compactly represented via spherical harmonics (Makadia and Daniilidis, 2006; Friedrich et al, 2008; Rossi et al, 2008). Semantic descriptors that are based on the convex hull of image features encode spatial relations between neighbouring features (Wang and Lin, 2011). A different approach, called a bubble memory, simultaneously encodes visual feature values and their local S 2 -metric relation from robot’s viewpoint (Soyer et al, 2000, 2006).…”

“…The images are acquired either via a two-degree-of-freedom camera head or an omnidirectional camera mounted on a mobile robot. This problem is of great interest to the machine vision and robotics communities (Wang and Lin, 2011). The proposed approaches either represent each place directly based on visual data or first find a set of image features which are then used to generate the final representation.…”

HOPIS: Hybrid Omnidirectional and Perspective Imaging System for Mobile Robots

“…The proposed technique provides the robot vision system with the capability of omnidirectional surveillance and 3D reconstruction. It can be used for mobile robot applications, such as obstacle detection, with the derived 3D information and vision-guided navigation using the omnidirectional images [ 35 ]. The imaging formation of the hybrid camera system is formulated using a unifying projection model.…”