AKAZE-Based Visual Odometry From Floor Images Supported by Acceleration Models

Abstract

To realize the self-localization of autonomous robots, methods for the 2D motion measurement of robots are required. In this research, a self-localization system using a CCD camera is proposed. In the proposed system, the self-localization is estimated by movement tracking using some keypoints detected from the floor images captured by the CCD camera. For the illumination of floor image, two LED illuminate are used. These lighting systems are installed in such a way that lit from both sides of the floor and parallel to the floor so that some minimal bumps or original veins are captured from the floor. An accelerated KAZE feature is applied in the proposed system for keypoint detection and for computing resulting in the generation of the descriptor. In the estimation, the location information from the previous step and the estimated position one frame ahead are used. Using the estimated result and computed descriptor, the proposed system matches the nearest keypoint between the estimated keypoint detected from the previous floor image and the keypoint detected from the next floor image at the beginning. The Hamming distance is employed in the proposed system to evaluate the matching. If the Hamming distance is longer than a threshold, the proposed system tries to match from the 2nd nearest keypoint. Based on an experiment in which the measurement distance and computation time are investigated, the effectiveness of the proposed method is confirmed. INDEX TERMS Visual odometry, vision measurement, image motion analysis, odometry measurement, AKAZE features, image matching.

Review of Bioinspired Vision-Tactile Fusion Perception (VTFP): From Humans to Humanoids

“…1) Visual Sensors: In robot vision, image sensors such as the charge-coupled devices (CCDs) [63] and the complementary metal-oxide silicon (CMOS) [64], [65] that produce video frames have been applied to visual tracking, localization, and navigation [66]- [73] of robots. They have the advantages of low cost and high resolution of small pixels.…”

The Use of a 3D Image Comparison Program for Dental Identification

“…In this paper, the authors report a new identification method using an image comparison program (AKAZE) [26][27][28][29] and binarized images created from cross-sectional images of left unilateral STL data of the upper and lower jaws, obtained by using an intraoral scanner.…”

Ground texture based localization

“…Ground texture based localization can be performed with appearance-based approaches [Aqel et al, 2016, Zaman, 2007, Kelly et al, 2007, Nagai and Watanabe, 2015], e. g. using normalized cross-correlation to find reoccurring image patches, and with feature-based approaches that find feature correspondences [Swank, 2012, Nakashima et al, 2019, Kozak and Alban, 2016, Zhang et al, 2019, Chen et al, 2018. Furthermore, localization methods can be divided into approaches for map-based absolute localization with or without available prior pose estimate, and approaches for incremental relative localization to estimate the pose of the current camera image in respect to the previous one [Desouza and Kak, 2002].…”

“…Feature-based approaches, on the other hand, track local visual features in consecutively recorded images. For example, Nakashima et al [2019] proposed a solution based on AKAZE [Alcantarilla et al, 2013] features. They propose to search for correspondences of a given feature from a previously recorded image by predicting its position in the current image, which avoids to match the feature with all features of the current image, improving robustness and efficiency of the method.…”

“…State-of-the-art methods to ground texture based localization employ featurebased localization [Swank, 2012, Nakashima et al, 2019, Chen et al, 2018, Kozak and Alban, 2016] that relies on the extraction of similar features from varying views of the same location. While several feature extraction methods were evaluated in these works, this survey is an extension.…”