A Novel Monocular Visual Odometer Method Based on Kinect and Improved SURF Algorithm

2014 ◽  
Vol 556-562 ◽  
pp. 4081-4084
Author(s):  
Li Jun Zhang ◽  
Fei Chen
Keyword(s):  
Dynamic Environment ◽  
Mean Value ◽  
Color Images ◽  
Least Square ◽  
Depth Image ◽  
Mean Value Theorem ◽  
Feature Points ◽  
Kinect Sensor ◽  
Depth Images ◽  
Visual Odometer

The paper proposes a novel monocular visual odometer method based on Kinect sensor made by Microsoft and the improved SURF algorithm. Firstly the Kinect sensor capture color images and depth images of the surrounding environment, then we use the improved SURF algorithm to extract feature points of the color images and match for them. At last, map what we get with the depth image and estimate the path information of the robot by doing 3D reconstruction and using the the least square mean value theorem. Experimental results show that by using this new method, the average matching accuracy reaches 92.6%. And even in a dynamic environment, it shows good robustness, so it comes down to the conclusion that the combination of the Kinect sensor and the improved SURF algorithm applied to visual odometer is a simple and effective method.

scholarly journals HRDepthNet: Depth Image-Based Marker-Less Tracking of Body Joints

Sensors ◽  
2021 ◽  
Vol 21 (4) ◽  
pp. 1356
Author(s):  
Linda Christin Büker ◽  
Finnja Zuber ◽  
Andreas Hein ◽  
Sebastian Fudickar
Keyword(s):  
Color Images ◽  
Depth Image ◽  
Accuracy Evaluation ◽  
Timed Up And Go ◽  
Position Errors ◽  
Depth Images ◽  
Upper And Lower Extremities ◽  
Rgb Images ◽  
Human Joints ◽  
Body Joints

With approaches for the detection of joint positions in color images such as HRNet and OpenPose being available, consideration of corresponding approaches for depth images is limited even though depth images have several advantages over color images like robustness to light variation or color- and texture invariance. Correspondingly, we introduce High- Resolution Depth Net (HRDepthNet)—a machine learning driven approach to detect human joints (body, head, and upper and lower extremities) in purely depth images. HRDepthNet retrains the original HRNet for depth images. Therefore, a dataset is created holding depth (and RGB) images recorded with subjects conducting the timed up and go test—an established geriatric assessment. The images were manually annotated RGB images. The training and evaluation were conducted with this dataset. For accuracy evaluation, detection of body joints was evaluated via COCO’s evaluation metrics and indicated that the resulting depth image-based model achieved better results than the HRNet trained and applied on corresponding RGB images. An additional evaluation of the position errors showed a median deviation of 1.619 cm (x-axis), 2.342 cm (y-axis) and 2.4 cm (z-axis).

scholarly journals ANINet: a deep neural network for skull ancestry estimation

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Lin Pengyue ◽  
Xia Siyuan ◽  
Jiang Yi ◽  
Yang Wen ◽  
Liu Xiaoning ◽  
...  
Keyword(s):  
Higher Learning ◽  
Depth Image ◽  
Estimation Methods ◽  
Data Sets ◽  
Feature Points ◽  
Depth Images ◽  
Wide Range ◽  
Ancestry Estimation ◽  
Object Use ◽  
Small Parameters

Abstract Background Ancestry estimation of skulls is under a wide range of applications in forensic science, anthropology, and facial reconstruction. This study aims to avoid defects in traditional skull ancestry estimation methods, such as time-consuming and labor-intensive manual calibration of feature points, and subjective results. Results This paper uses the skull depth image as input, based on AlexNet, introduces the Wide module and SE-block to improve the network, designs and proposes ANINet, and realizes the ancestry classification. Such a unified model architecture of ANINet overcomes the subjectivity of manually calibrating feature points, of which the accuracy and efficiency are improved. We use depth projection to obtain the local depth image and the global depth image of the skull, take the skull depth image as the object, use global, local, and local + global methods respectively to experiment on the 95 cases of Han skull and 110 cases of Uyghur skull data sets, and perform cross-validation. The experimental results show that the accuracies of the three methods for skull ancestry estimation reached 98.21%, 98.04% and 99.03%, respectively. Compared with the classic networks AlexNet, Vgg-16, GoogLenet, ResNet-50, DenseNet-121, and SqueezeNet, the network proposed in this paper has the advantages of high accuracy and small parameters; compared with state-of-the-art methods, the method in this paper has a higher learning rate and better ability to estimate. Conclusions In summary, skull depth images have an excellent performance in estimation, and ANINet is an effective approach for skull ancestry estimation.

Online identification method for a system with unknown varying time-delay and parameter

2021 ◽  
pp. 014233122110224
Author(s):  
Yifei Hu ◽  
Jinbo Wu
Keyword(s):  
Time Delay ◽  
System Parameter ◽  
Estimation Error ◽  
Lyapunov Stability Theory ◽  
Mean Value ◽  
Least Square ◽  
Mean Value Theorem ◽  
Unknown Parameters ◽  
Online Identification ◽  
Identification Method

An online identification method that can simultaneously estimate the unknown system parameter and the unknown time-delay is proposed. Firstly, with the help of Lagrange mean value theorem, the system with time-delay can be transformed into two terms that can be identified by modified least-square algorithm and one term that represents an approximate error. Then, a modified least-square algorithm is introduced to estimate all the unknown parameters in case of external disturbances. Additionally, an restrain term are added in the covariance matrix to enhance the robustness to deal with the approximate error which is related to the estimated error of system parameter and time-delay. Also, the boundedness of the estimation error is guaranteed via Lyapunov stability theory. Finally, the effectivity of the proposed method is verified by simulations results.

scholarly journals Promising Depth Map Prediction Method from a Single Image Based on Conditional Generative Adversarial Network

2021 ◽  
Author(s):  
Saddam Abdulwahab ◽  
Hatem A. Rashwan ◽  
Armin Masoumian ◽  
Najwa Sharaf ◽  
Domenec Puig
Keyword(s):  
Pose Estimation ◽  
Prediction Method ◽  
Depth Estimation ◽  
Color Images ◽  
Depth Image ◽  
3D Pose Estimation ◽  
3D Cad ◽  
Depth Images ◽  
Proposed Model ◽  
Rgb Image

Pose estimation is typically performed through 3D images. In contrast, estimating the pose from a single RGB image is still a difficult task. RGB images do not only represent objects’ shape, but also represent the intensity that is relative to the viewpoint, texture, and lighting condition. While the 3D pose estimation from depth images is considered a promising approach since the depth image only represents objects’ shape. Thus, it is necessary to know what is the appropriate method that can be used for predicting the depth image from a 2D RGB image and then to use for getting the 3D pose estimation. In this paper, we propose a promising approach based on a deep learning model for depth estimation in order to improve the 3D pose estimation. The proposed model consists of two successive networks. The first network is an autoencoder network that maps from the RGB domain to the depth domain. The second network is a discriminator network that compares a real depth image to a generated depth image to support the first network to generate an accurate depth image. In this work, we do not use real depth images corresponding to the input color images. Our contribution is to use 3D CAD models corresponding to objects appearing in color images to render depth images from different viewpoints. These rendered images are then used as ground truth and to guide the autoencoder network to learn the mapping from the image domain to the depth domain. The proposed model outperforms state-of-the-art models on the publicly PASCAL 3D+ dataset.

Depth Image Vibration Filtering and Shadow Detection Based on Fusion and Fractional Differential

2020 ◽  
pp. 2150002
Author(s):  
Ting Cao ◽  
Pengjia Tu ◽  
Weixing Wang
Keyword(s):  
Image Fusion ◽  
State Of The Art ◽  
Shadow Detection ◽  
Depth Image ◽  
Fusion Method ◽  
Kinect Sensor ◽  
Depth Images ◽  
Fractional Differential ◽  
Image Fusion Method

The depth image generated by Kinect sensor always contains vibration and shadow noises which limit the related usage. In this research, a method based on image fusion and fractional differential is proposed for the vibration filtering and shadow detection. First, an image fusion method based on pixel level is put forward to filter the vibration noises. This method can achieve the best quality of every pixel according to the depth images sequence. Second, an improved operator based on fractional differential is studied to extract the shadow noises, which can enhance the boundaries of shadow regions significantly to accomplish the shadow detection effectively. Finally, a comparison is made with other traditional and state-of-the-art methods and our experimental results indicate that the proposed method can filter out the vibration and shadow noises effectively based on the [Formula: see text]-measure system.

Dynamic Obstacle Avoidance Algorithm for the Mobile Robot Based on Depth Image of Kinect Sensor

2014 ◽  
Vol 541-542 ◽  
pp. 1072-1078
Author(s):  
Yi Zhang ◽  
Xue Rong Tong ◽  
Yuan Luo
Keyword(s):  
Mobile Robot ◽  
Obstacle Avoidance ◽  
Dynamic Environment ◽  
Path Selection ◽  
Depth Image ◽  
Depth Information ◽  
Kinect Sensor ◽  
Dynamic Obstacle Avoidance ◽  
Kalman Filter Algorithm ◽  
Dynamic Obstacle

In order to solve the problem of the dynamic obstacle avoidance of the mobile robot in indoor environment, a new approach based on depth information is presented in this paper. The depth information of surrounding environment was collected and used to set the robots obstacle avoidance warning area by a Kinect sensor. When the moving obstacle accessed into the warning area, the robots obstacle avoidance direction was determined preliminary by the obstacles position, and then an improved Kalman filter algorithm was used to optimize the avoidance path. Experiments show that this approach can overcome the potential problem of path selection, and realize the mobile robot obstacle avoidance behavior in the dynamic environment.

scholarly journals On Riemann—Liouville and Caputo Fractional Forward Difference Monotonicity Analysis

Mathematics ◽  
2021 ◽  
Vol 9 (11) ◽  
pp. 1303
Author(s):  
Pshtiwan Othman Mohammed ◽  
Thabet Abdeljawad ◽  
Faraidun Kadir Hamasalh
Keyword(s):  
Time Scale ◽  
Initial Value Problems ◽  
Mean Value ◽  
Mean Value Theorem ◽  
Discrete Fractional Calculus ◽  
Initial Value ◽  
Forward Difference ◽  
The Mean ◽  
Monotonicity Analysis ◽  
Monotonicity Results

Monotonicity analysis of delta fractional sums and differences of order υ∈(0,1] on the time scale hZ are presented in this study. For this analysis, two models of discrete fractional calculus, Riemann–Liouville and Caputo, are considered. There is a relationship between the delta Riemann–Liouville fractional h-difference and delta Caputo fractional h-differences, which we find in this study. Therefore, after we solve one, we can apply the same method to the other one due to their correlation. We show that y(z) is υ-increasing on Ma+υh,h, where the delta Riemann–Liouville fractional h-difference of order υ of a function y(z) starting at a+υh is greater or equal to zero, and then, we can show that y(z) is υ-increasing on Ma+υh,h, where the delta Caputo fractional h-difference of order υ of a function y(z) starting at a+υh is greater or equal to −1Γ(1−υ)(z−(a+υh))h(−υ)y(a+υh) for each z∈Ma+h,h. Conversely, if y(a+υh) is greater or equal to zero and y(z) is increasing on Ma+υh,h, we show that the delta Riemann–Liouville fractional h-difference of order υ of a function y(z) starting at a+υh is greater or equal to zero, and consequently, we can show that the delta Caputo fractional h-difference of order υ of a function y(z) starting at a+υh is greater or equal to −1Γ(1−υ)(z−(a+υh))h(−υ)y(a+υh) on Ma,h. Furthermore, we consider some related results for strictly increasing, decreasing, and strictly decreasing cases. Finally, the fractional forward difference initial value problems and their solutions are investigated to test the mean value theorem on the time scale hZ utilizing the monotonicity results.

scholarly journals RobotP: A Benchmark Dataset for 6D Object Pose Estimation

Sensors ◽  
2021 ◽  
Vol 21 (4) ◽  
pp. 1299
Author(s):  
Honglin Yuan ◽  
Tim Hoogenkamp ◽  
Remco C. Veltkamp
Keyword(s):  
Pose Estimation ◽  
Ground Truth ◽  
3D Models ◽  
Depth Image ◽  
Great Success ◽  
Estimation Algorithms ◽  
Depth Images ◽  
Object Pose Estimation ◽  
Image Pairs ◽  
Bounding Boxes

Deep learning has achieved great success on robotic vision tasks. However, when compared with other vision-based tasks, it is difficult to collect a representative and sufficiently large training set for six-dimensional (6D) object pose estimation, due to the inherent difficulty of data collection. In this paper, we propose the RobotP dataset consisting of commonly used objects for benchmarking in 6D object pose estimation. To create the dataset, we apply a 3D reconstruction pipeline to produce high-quality depth images, ground truth poses, and 3D models for well-selected objects. Subsequently, based on the generated data, we produce object segmentation masks and two-dimensional (2D) bounding boxes automatically. To further enrich the data, we synthesize a large number of photo-realistic color-and-depth image pairs with ground truth 6D poses. Our dataset is freely distributed to research groups by the Shape Retrieval Challenge benchmark on 6D pose estimation. Based on our benchmark, different learning-based approaches are trained and tested by the unified dataset. The evaluation results indicate that there is considerable room for improvement in 6D object pose estimation, particularly for objects with dark colors, and photo-realistic images are helpful in increasing the performance of pose estimation algorithms.

scholarly journals On a hybrid version of the Vinogradov mean value theorem

2021 ◽  
Vol 163 (1) ◽  
pp. 1-17
Author(s):  
C. Chen ◽  
I. E. Shparlinski
Keyword(s):  
Mean Value ◽  
Mean Value Theorem

scholarly journals Arithmetic of higher-dimensional orbifolds and a mixed Waring problem

2021 ◽  
Author(s):  
Tim Browning ◽  
Shuntaro Yamagishi
Keyword(s):  
Circle Method ◽  
Rational Points ◽  
Mean Value ◽  
Mean Value Theorem ◽  
Waring’S Problem ◽  
Waring Problem ◽  
Waring's Problem ◽  
Main Conjecture ◽  
Higher Dimensional ◽  
Thin Set

AbstractWe study the density of rational points on a higher-dimensional orbifold $$(\mathbb {P}^{n-1},\Delta )$$ ( P n - 1 , Δ ) when $$\Delta $$ Δ is a $$\mathbb {Q}$$ Q -divisor involving hyperplanes. This allows us to address a question of Tanimoto about whether the set of rational points on such an orbifold constitutes a thin set. Our approach relies on the Hardy–Littlewood circle method to first study an asymptotic version of Waring’s problem for mixed powers. In doing so we make crucial use of the recent resolution of the main conjecture in Vinogradov’s mean value theorem, due to Bourgain–Demeter–Guth and Wooley.

Sign in / Sign up

Export Citation Format

Share Document