Robust Robot Pose Estimation for Challenging Scenes With an RGB-D Camera

Huang, Yu; Fu, Qiang; Yang, Zhengeng; Tan, Lei; Sun, Wei; Sun, Mingui

doi:10.1109/jsen.2018.2884321

Cited by 32 publications

(13 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results of image convolution as shown in Figure 7 are adopted as the horizontal and vertical Harr wavelet response of each pixel point. The size of the Harr wavelet is 4S, as shown in Equation (12). The scale value S where the feature point is located is calculated according to the size of the current template.…”

Section: Determination Of Foreground Feature Pointsmentioning

confidence: 99%

“…It is invariant to geometric characteristics and noise, and it is also stable to viewpoint changes [11]. However, its feature vector is 128-dimensional, so it has a large amount of calculation and a slower calculation speed [12,13], which is not suitable for navigation and other situations with high real-time requirements. Researchers have proposed many improvement methods on its basis.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Research on Feature Extraction Method of Indoor Visual Positioning Image Based on Area Division of Foreground and Background

Zheng

Qin

Han

et al. 2021

IJGI

View full text Add to dashboard Cite

In the process of indoor visual positioning and navigation, difficult points often exist in corridors, stairwells, and other scenes that contain large areas of white walls, strong consistent background, and sparse feature points. Aiming at the problem of positioning and navigation in the real physical world where the walls with sparse feature points are difficult to be filled with pictures, this paper designs a feature extraction method, ARAC (Adaptive Region Adjustment based on Consistency) using Free and Open-Source Software and tools. It divides the image into foreground and background and extracts their features respectively, to achieve not only retain positioning information but also focus more energy on the foreground area which is favourable for navigation. In the test phase, under the combined conditions of illumination, scale and affine changes, the feature matching maps by the feature extraction algorithm proposed in this paper are compared with those by SIFT and SURF. Experiments show that the number of correctly matched feature pairs obtained by ARAC is better than SIFT and SURF, and whose time of feature extraction and matching is comparable to SURF, which verifies the accuracy and efficiency of the ARAC feature extraction method.

show abstract

Section: Determination Of Foreground Feature Pointsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Research on Feature Extraction Method of Indoor Visual Positioning Image Based on Area Division of Foreground and Background

Zheng

Qin

Han

et al. 2021

IJGI

View full text Add to dashboard Cite

show abstract

“…When considering some factors within feature extraction algorithms (e.g., SIFT, SURF) such as velocity, stability, rotation invariance and so on [48], the ORB algorithm [15] is selected to extract and match point features by using the Opencv library. By combing the RANSAC [43] algorithm with a fundamental matrix constraint, we reject the mismatches and obtain an initial set of matching points.…”

Section: B Extraction and Matching Of Point And Line Featuresmentioning

confidence: 99%

PL-GM:RGB-D SLAM With a Novel 2D and 3D Geometric Constraint Model of Point and Line Features

Zhang

2021

IEEE Access

View full text Add to dashboard Cite

In the study of RGB-D SLAM (Simultaneous Localization and Mapping), two types of primary visual features, point and line features, have been widely utilized to calculate the camera pose. As an RGB-D camera can capture RGB and depth information simultaneously, most RGB-D SLAM methods only utilize the 2D information within the point and line features. To obtain a higher accuracy camera pose and utilize the 2D and 3D information within points and lines better, a novel geometric constraint model of points and lines (PL-GM) using an RGB-D camera is proposed in this paper. Our contributions are threefold. Firstly, the 3D points and lines generated by an RGB-D camera combining with 2D point and line features are utilized to establish the PL-GM, which is different from most models of point-line SLAM (PL-SLAM). Secondly, in addition to the 2D re-projection error of point and line features, the constraint errors of 3D points and lines are constructed and minimized likewise, and then a unified optimization model based on PL-GM is extended to the bundle adjustment model (BA). Finally, extensive experiments have been performed on two public benchmark RGB-D datasets and a real scenario sequence. These experimental results demonstrate that our method achieves a comparable or better performance than the state-of-the-art SLAM methods based on point and line features, and point features.

show abstract

“…In recent years, environmental perception technology for robotic system has attracted a lot of attention from researchers, and most of the studies on it are focused on the space above ground. With the development of computer science, there are more and more environmental perception sensors developed for robotics and artificial intelligence School of Instrumentation Science and Engineering, Harbin Institute of Technology, Heilongjiang, China fields, including radio frequency identification (RFID) with its sensors networks, 1,2 lidar sensors, [3][4][5][6][7] red-green-bluedepth (RGB-D) camera, 8,9 millimeter radar, 3,10 and so on. In addition, the related perception technology above can be classified into two main categories: the one is vision based and the other one is sensor fusion based.…”

Section: Introductionmentioning

confidence: 99%

“…11 Recently, color-depth cameras have been widely used in robotic system for the applications of environmental perception, such as pose estimation, 9 detection and tracking of humans, 12 autonomous navigation, 13 and simultaneous localization and mapping (SLAM). 8 Compared with the other cameras, a single color-depth camera is able to produce a transformation matrix with an absolute scale because of its additional depth information, 8,14,15 which helps robots achieve higher sensing accuracy with lower system cost. However, the fact that cameras are highly affected by lighting conditions makes those optical sensors unsuitable to be used in dark environment.…”

Section: Introductionmentioning

confidence: 99%

Ground-penetrating radar-based underground environmental perception radar for robotic system

Shen

2020

International Journal of Advanced Robotic Systems

View full text Add to dashboard Cite

In recent years, the environmental perception technology for robotic system has attracted a lot of attention from researchers, but only a little of studies on environmental perception technology are focused on the space underground. Meanwhile, in the field of mobile robotic systems, with the development of research on underground emergency hedging and buried targets’ high-resolution fault imaging, more and more attention has also been paid to underground environmental detection and perception. This article proposes a ground-penetrating radar-based underground environmental perception radar (UEPR) for mobile robotic system indoors. The underground environmental perception radar can achieve noncontact and real-time perception, which helps people detect buried targets and get the image of targets more conveniently and precisely. Major contributions of this work are threefold. Firstly, a stepped frequency continuous wave modulation and demodulation scheme is proposed; secondly, a switch device for a six-channel antenna array is designed and contributed; thirdly, based on a linear antenna array and a signal processing platform, the underground environmental perception radar is supposed to achieve three-dimensional imaging in underground space indoors with its low power consumption. For the experiment of three-dimensional imaging on the copper box and underground environment indoors, the process of imaging is successful, although the size of them is a little bigger than the real size. In addition, the comparison experiment shows that the resolution of underground environmental perception radar system is similar with that of sound wave methods, and the working range of underground environmental perception radar system is deeper than the others. It can be concluded that the underground environmental perception radar can detect the copper box underground and perceive something special within 1.5 m depth.

show abstract

Robust Robot Pose Estimation for Challenging Scenes With an RGB-D Camera

Cited by 32 publications

References 31 publications

Research on Feature Extraction Method of Indoor Visual Positioning Image Based on Area Division of Foreground and Background

Research on Feature Extraction Method of Indoor Visual Positioning Image Based on Area Division of Foreground and Background

PL-GM:RGB-D SLAM With a Novel 2D and 3D Geometric Constraint Model of Point and Line Features

Ground-penetrating radar-based underground environmental perception radar for robotic system

Contact Info

Product

Resources

About