No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Evaluating 3D Depth Sensors: A Study and an Open Source Data Set of Logistic Scenarios*
ResearchGate has not been able to resolve any citations for this publication.
The relationship between the disparity and depth information of corresponding pixels is inversely proportional. Thus, in order to accurately estimate depth from stereo vision, it is important to obtain accurate disparity maps, which encode the difference between horizontal coordinates of corresponding image points. Stereo vision can be classified as either passive or active. Active stereo vision generates pattern texture, which passive stereo vision does not have, on the image to fill the textureless regions. In passive stereo vision, many surveys have discovered that disparity accuracy is heavily reliant on attributes, such as radiometric variation and color variation, and have found the best-performing conditions. However, in active stereo matching, the accuracy of the disparity map is influenced not only by those affecting the passive stereo technique, but also by the attributes of the generated pattern textures. Therefore, in this paper, we analyze and evaluate the relationship between the performance of the active stereo technique and the attributes of pattern texture. When evaluating, experiments are conducted under various settings, such as changing the pattern intensity, pattern contrast, number of pattern dots, and global gain, that may affect the overall performance of the active stereo matching technique. Through this evaluation, our discovery can act as a noteworthy reference for constructing an active stereo system.
The accurate and reliable fruit detection in orchards is one of the most crucial tasks for supporting higher level agriculture tasks such as yield mapping and robotic harvesting. However, detecting and counting small fruit is a very challenging task under variable lighting conditions, low-resolutions and heavy occlusion by neighboring fruits or foliage. To robustly detect small fruits, an improved method is proposed based on multiple scale faster region-based convolutional neural networks (MS-FRCNN) approach using the color and depth images acquired with an RGB-D camera. The architecture of MS-FRCNN is improved to detect lower-level features by incorporating feature maps from shallower convolution feature maps for regions of interest (ROI) pooling. The detection framework consists of three phases. Firstly, multiple scale feature extractors are used to extract low and high features from RGB and depth images respectively. Then, RGB-detector and depth-detector are trained separately using MS-FRCNN. Finally, late fusion methods are explored for combining the RGB and depth detector. The detection framework was demonstrated and evaluated on two datasets that include passion fruit images under variable illumination conditions and occlusion. Compared with the faster R-CNN detector of RGB-D images, the recall, the precision and F1-score of MS-FRCNN method increased from 0.922 to 0.962, 0.850 to 0.931 and 0.885 to 0.946, respectively. Furthermore, the MS-FRCNN method effectively improves small passion fruit detection by achieving 0.909 of the F1 score. It is concluded that the detector based on MS-FRCNN can be applied practically in the actual orchard environment.
Robotic weeding enables weed control near or within crop rows automatically, precisely and effectively. A computer‐vision system was developed for detecting crop plants at different growth stages for robotic weed control. Fusion of color images and depth images was investigated as a means of enhancing the detection accuracy of crop plants under conditions of high weed population. In‐field images of broccoli and lettuce were acquired 3–27 days after transplanting with a Kinect v2 sensor. The image processing pipeline included data preprocessing, vegetation pixel segmentation, plant extraction, feature extraction, feature‐based localization refinement, and crop plant classification. For the detection of broccoli and lettuce, the color‐depth fusion algorithm produced high true‐positive detection rates (91.7% and 90.8%, respectively) and low average false discovery rates (1.1% and 4.0%, respectively). Mean absolute localization errors of the crop plant stems were 26.8 and 7.4 mm for broccoli and lettuce, respectively. The fusion of color and depth was proved beneficial to the segmentation of crop plants from background, which improved the average segmentation success rates from 87.2% (depth‐based) and 76.4% (color‐based) to 96.6% for broccoli, and from 74.2% (depth‐based) and 81.2% (color‐based) to 92.4% for lettuce, respectively. The fusion‐based algorithm had reduced performance in detecting crop plants at early growth stages.
This chapter discusses the use of computer models for such diverse applications as safety assessments for geologic isolation of radioactive waste and for nuclear power plants; loss cost projections for hurricanes; reliability analyses for manufacturing equipment; transmission of HIV; and subsurface storm flow modelling. Such models are usually characterized by a large number of input variables (perhaps as many as a few hundred), and usually, only a handful of these inputs are important for a given response. In addition, the model response is frequently multivariate and time dependent. Latin hypercube sampling (LHS) uses a stratified sampling scheme to improve on the coverage of the k-dimensional input space for such computer models. This means that a single sample will provide useful information when some input variable(s) dominate certain responses (or certain time intervals), while other input variables dominate other responses (or time intervals). By sampling over the entire range, each variable has the opportunity to show up as important, if it indeed is important. If an input variable is not important, then the method of sampling is of little or no concern. The values of the stratified sampling scheme can be paired to ensure a desired correlation structure among the k input variables. LHS is more efficient than simple random sampling in a large range of conditions.Keywords:Latin hypercube sampling;uncertainty analysis;sensitivity analysis;rank correlation;hurricane loss projection;uncertainty importance
The problem of geometric alignment of two roughly preregistered, partially overlapping, rigid, noisy 3D point sets is considered. A new natural and simple, robustified extension of the popular Iterative Closest Point (ICP) algorithm (Besl and McKay, 1992) is presented, called the Trimmed ICP (TrICP). The new algorithm is based on the consistent use of the least trimmed squares (LTS) approach in all phases of the operation. Convergence is proved and an efficient implementation is discussed. TrICP is fast, applicable to overlaps under 50%, robust to erroneous measurements and shape defects, and has easy-to-set parameters. ICP is a special case of TrICP when the overlap parameter is 100%. Results of testing the new algorithm are shown.
This paper presents a semantic mapping system for mobile robots in intralogistics environments. The sensor data captured with a solid-state RGB-D LiDAR camera serves as input data for state-of-the-art deep learning based object recognition systems. The presented approach uses two deep learning algorithms, one for the 2D and one for the 3D space. The premier system YOLACT performs an instance segmentation on the RGB images. In an additional step, point clouds from the RGB-D sensor are converted into so called bird’s eye view images. The key idea is to feed these bird’s eye view images in combination with the prior knowledge generated by YOLACT into the latter system Complex-YOLO (CY) to obtain the spatial extension of the detected object. Moreover, a conventional Minimum Bounding Rectangle (MBR) approach is presented. Both approaches are evaluated by specific scenarios and selected metrics in combination with a highly precise motion capturing system, which provides ground truth data.
The reliable fusion of depth maps from multiple viewpoints has become an important problem in many 3D reconstruction pipelines. In this work, we investigate its impact on robotic bin-picking tasks such as 6D object pose estimation. The performance of object pose estimation relies heavily on the quality of depth data. However, due to the prevalence of shiny surfaces and cluttered scenes, industrial grade depth cameras often fail to sense depth or generate unreliable measurements from a single viewpoint. To this end, we propose a novel probabilistic framework for scene reconstruction in robotic bin-picking. Based on active stereo camera data, we first explicitly estimate the uncertainty of depth measurements for mitigating the adverse effects of both noise and outliers. The uncertainty estimates are then incorporated into a probabilistic model for incrementally updating the scene. To extensively evaluate the traditional fusion approach alongside our own approach, we will release a novel representative dataset with multiple views for each bin and curated parts. Over the entire dataset, we demonstrate that our framework outperforms a traditional fusion approach by a 12.8% reduction in reconstruction error, and 6.1% improvement in detection rate. The dataset will be available at
https://www.trailab.utias.utoronto.ca/robi
.
In computational science and computer graphics, there is a strong requirement to represent and visualize information in the real domain, and many visualization data structures and algorithms have been proposed to achieve this aim. Unfortunately, the dataflow model that is often selected to address this issue in visualization systems is not flexible enough to visualize newly invented data structures and algorithms because this scheme can accept only specific data structures. To address this problem, we propose a new visualization tool, RViz, which is independent of the input information data structures. Since there is no requirement for additional efforts to manage the flow networks and the interface to abstracted information is simple in RViz, any scientific information visualization algorithms are easier to implement than the dataflow model. In this paper, we provide case studies in which we have successfully implemented new data structures and related algorithms using RViz, including geometry synthesis, distance field representation, and implicit surface reconstruction. Through these cases, we show how RViz helps users visualize and understand any hidden insights in input information.
Dinar, Boualem Merabet, and Samir Ghouali. “NTP Server Clock Adjustment with Chrony
- Amina Elbatoul
A time-of-flight depth sensor-system description, issues and solutions
- S Burak
- Hakan Gokturk
- Cyrus Bamji
Mastering ROS for Robotics Programming: Design, build, and simulate complex robots using the Robot Operating System
- Lentin Joseph
- Jonathan Cacace