- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
New approach to enhancing the performance of cloud-based vision system of mobile robots
摘要： Mobile robots require real-time performance, high computation power, and a shared computing environment. Although cloud computing offers computation power, it may adversely affect real-time performance owing to network lag. The main objective of this study is to allow a mobile robot vision system to reliably achieve real-time constraints using cloud computing. A human cloud mobile robot architecture is proposed as well as a data flow mechanism organized on both the mobile robot and the cloud server sides. Two algorithms are proposed: (i) A real-time image clustering algorithm, applied on the mobile robot side, and (ii) A modified growing neural gas algorithm, applied on the cloud server side. The experimental results demonstrate that there is a 25% to 45% enhancement in the total response time, depending on the communication bandwidth and image resolution. Moreover, better performance in terms of data size, path planning time, and accuracy is demonstrated over other state-of-the-art techniques.
关键词： Computation offloading,Computer vision,3D point cloud,Mobile robot,Stereo vision,Real-time networking,Cloud computing,Cloud robotics
更新于2025-09-23 15:23:52
-
Factors influencing the choice of low-vision devices for visual rehabilitation in Stargardt disease
摘要： Background: Stargardt disease is the most common cause of juvenile macular degeneration leading to early central visual loss. Dispensing of correct low-vision devices based on the residual visual function and specific visual requirements of a patient can result in a positive outcome. It is important to know the factors involved in the selection of these devices. This study was undertaken to assess these factors. Methods: Patients with Stargardt disease referred to a low-vision clinic underwent evaluation of visual status, disease stage, visual requirements and lifestyle. They were evaluated for suitability for successful use of various low-vision devices. Their education level and occupation were noted. They were counselled regarding the proper use of the devices and lifestyle modifications. For patients with extensive use of computers, modifications related to contrast, font size and audio software were explained. Results: Among the 97 patients in the study (age range 7–66 years, mean 23.7 ± 13.1), there were 49 (50.5 per cent) students, 36 (37.1 per cent) employed, eight (8.2 per cent) unemployed, and four (4.2 per cent) homemakers. Except for seven patients (7.2 per cent), all were literate, with education ranging from primary school to college graduation. The presenting visual acuity for distance was 0.10–1.47 (0.79 ± 0.28), and for near was N4–N40 (N9.95 ± 6.65). This acuity showed positive correlation with age (p < 0.0001, R2 = 0.16) and with magnification required (p < 0.0001, R2 = 0.26). Patients above 40 years preferred higher adds and half-eye glasses, whereas younger patients preferred dome magnifiers. The occupation also influenced the choice. Disease stage was seen to affect the choice with advanced stages requiring higher magnifiers (p = 0.03, R2 = 0.11). Duration of disease and the magnification of low-vision devices showed a positive correlation (p = 0.03, R2 = 0.049). Conclusions: Age, presenting visual acuity, disease stage, duration, education and occupation influenced the choice of low-vision devices in patients with Stargardt disease. However, they appear to be interdependent and a larger, controlled study is required to provide information on the individual effect.
关键词： occupation,visual rehabilitation,heredomacular degeneration,Stargardt disease,low vision rehabilitation,low vision devices
更新于2025-09-23 15:23:52
-
Mapping Forest Structure Using UAS inside Flight Capabilities
摘要： We evaluated two unmanned aerial systems (UASs), namely the DJI Phantom 4 Pro and DJI Mavic Pro, for 3D forest structure mapping of the forest stand interior with the use of close-range photogrammetry techniques. Assisted flights were performed within two research plots established in mature pure Norway spruce (Picea abies (L.) H. Karst.) and European beech (Fagus sylvatica L.) forest stands. Geotagged images were used to produce georeferenced 3D point clouds representing tree stem surfaces. With a flight height of 8 m above the ground, the stems were precisely modeled up to a height of 10 m, which represents a considerably larger portion of the stem when compared with terrestrial close-range photogrammetry. Accuracy of the point clouds was evaluated by comparing field-measured tree diameters at breast height (DBH) with diameter estimates derived from the point cloud using four different fitting methods, including the bounding circle, convex hull, least squares circle, and least squares ellipse methods. The accuracy of DBH estimation varied with the UAS model and the diameter fitting method utilized. With the Phantom 4 Pro and the least squares ellipse method to estimate diameter, the mean error of diameter estimates was ?1.17 cm (?3.14%) and 0.27 cm (0.69%) for spruce and beech stands, respectively.
关键词： point cloud,diameter at breast height (DBH),photogrammetry,obstacle sensing,forestry,unmanned aerial system (UAS),vision positioning system
更新于2025-09-23 15:23:52
-
Methodology and implementation of a vision-oriented open CNC system for profile grinding
摘要： CNC systems with the vision function have become very valuable for intelligence machine tools because machine vision is a fast-growing intelligent feature for machines. A novel vision-oriented open CNC system was developed in this study and used in profile grinding machines for the precise machining of parts with contour surfaces, such as complex molds and cutting tools. The system is an innovation to the conventional profile grinding method and enabled the profile error to be visually detected and compensated during the machining process. In this study, a novel design methodology for a machine-vision-oriented CNC system was proposed. An Ethernet-based hardware architecture was constructed for the vision-oriented CNC system. The software characteristics of the developed CNC system were analyzed, including a new type of multi-thread software architecture, a seamless handover approach for multi-thread accessing of the memory space, the integration of the human-machine interface with image processing, and virtual-axis-based online error compensation. The running efficiency test of the software development platforms, time-consumption analysis of the measurement and control in the vision-oriented CNC system, and verification of the effectiveness of the developed vision-oriented CNC system were performed. The results indicate that the proposed vision-oriented open CNC system can effectively fuse image processing with motion control, meet the efficiency requirement, and improve the machining precision of profile grinding. The results are also valuable for developing other machine-vision-based intelligent machine tools.
关键词： Contour error,Profile grinding,Image processing,On-machine measurement,Open CNC,Machine vision
更新于2025-09-23 15:23:52
-
Experimental verification of turbidity tolerance of stereo-vision-based 3D pose estimation system
摘要： This paper presents the verification of the turbidity tolerance of a stereo-vision-based 3D pose estimation system for underwater docking applications. To the best of the authors’ knowledge, no studies have yet been conducted on 3D pose (position and orientation) estimation against turbidity for underwater vehicles. Therefore, the effect of turbidity on the 3D pose estimation performance of underwater vehicles and a method of operating under turbid conditions were studied in this work. A 3D pose estimation method using the real-time multi-step genetic algorithm (RM-GA) proposed by the authors in the previous works shows robust pose estimation performance against changing environmental conditions. This paper discusses how and why the RM-GA is well suited to effective 3D pose estimation, even when turbid conditions disturb visual servoing. The experimental results confirm the performance of the proposed 3D pose estimation system under different levels of turbidity. To demonstrate the practical usefulness of the RM-GA, docking experiments were conducted in a turbid pool and a real sea environment to verify the performance and tolerance of the proposed system under turbid conditions. The experimental results verify the robustness of the system against turbidity, presenting a possible solution to a major problem in the field of robotics.
关键词： Robustness against turbidity,Real-time multi-step genetic algorithm,Sea docking,3D pose estimation,Stereo-vision,Visual servoing
更新于2025-09-23 15:23:52
-
Intelligent Indoor Mobile Robot Navigation Using Stereo Vision
摘要： Majority of the existing robot navigation systems, which facilitate the use of laser range finders, sonar sensors or artificial landmarks, has the ability to locate itself in an unknown environment and then build a map of the corresponding environment. Stereo vision,while still being a rapidly developing technique in the field of autonomous mobile robots, are currently less preferable due to its high implementation cost. This paper aims at describing an experimental approach for the building of a stereo vision system that helps the robots to avoid obstacles and navigate through indoor environments and at the same time remaining very much cost effective. This paper discusses the fusion techniques of stereo vision and ultrasound sensors which helps in the successful navigation through different types of complex environments. The data from the sensor enables the robot to create the two dimensional topological map of unknown environments and stereo vision systems models the three dimension model of the same environment.
关键词： Point clouds,Triangulation,SLAM,Arduino,Stereo vision system
更新于2025-09-23 15:23:52
-
DeepPrimitive: Image decomposition by layered primitive detection
摘要： The perception of the visual world through basic building blocks, such as cubes, spheres, and cones, gives human beings a parsimonious understanding of the visual world. Thus, efforts to find primitive-based geometric interpretations of visual data date back to 1970s studies of visual media. However, due to the difficulty of primitive fitting in the pre-deep learning age, this research approach faded from the main stage, and the vision community turned primarily to semantic image understanding. In this paper, we revisit the classical problem of building geometric interpretations of images, using supervised deep learning tools. We build a framework to detect primitives from images in a layered manner by modifying the YOLO network; an RNN with a novel loss function is then used to equip this network with the capability to predict primitives with a variable number of parameters. We compare our pipeline to traditional and other baseline learning methods, demonstrating that our layered detection model has higher accuracy and performs better reconstruction.
关键词： biologically inspired vision,primitive detection,deep learning,layered image decomposition
更新于2025-09-23 15:23:52
-
Ground based hyperspectral imaging for extensive mango yield estimation
摘要： Fruit yield estimation in orchard blocks is an important objective in the context of precision agriculture, as it makes it easier for the farmer to plan ahead and efficiently use resources. Nevertheless, its implementation is labour-intensive and involves the manual counting of the fruit present in the trees. While colour (RGB) has been widely shown to be successful and arguably sufficient for yield estimation in orchards, hyperspectral imaging (HSI) shows promise for more nuanced tasks such as disease detection, cultivar classification and fruit maturity estimation. Therefore, it is important to ask how appropriate is HSI for the task of yield estimation, with a view to performing all of these tasks with just one sensor. This paper presents a novel mango yield estimation pipeline using ground based line-scan HSI acquired from an unmanned ground vehicle. Hyperspectral images were collected on a commercial mango orchard block in December 2017 and pre-processed for illumination compensation. After tree delimitation and mango pixel identification, an optimisation process was carried out to obtain the best models for fruit counting, using mango counts obtained by manually counting the fruit on-tree, and using state-of-the-art RGB techniques for yield estimation. Models were validated and tested on hundreds of trees, and subsequently mapped. In testing, determination coefficients reached values of up to 0.75 against field counts (predicting 18 trees) and 0.83 against RGB mango counts (predicting 216 trees). These results suggest that line-scan HSI can be used to accurately estimate yield in orchards, especially in scenarios in which this technology is already chosen for the determination of other traits.
关键词： Field robotics,Computer vision,Lidar,Hyperspectral,Fruit counting
更新于2025-09-23 15:23:52
-
The first deaf-blind patient in Russia with Argus II retinal prosthesis system. What he sees and why
摘要： A 59-year-old man with retinitis pigmentosa and hearing loss (clinical Usher syndrome) underwent surgery to implant the Argus II retinal prosthesis system in the right eye. This is the first case of retinal implant surgery in the Russian Federation. In this study we propose a new method for evaluating the functional results after implantation of the Argus II, the statistical analysis was performed to find correlations between the sizes of phosphenes that the patient drew and then were digitalized, and the height of the subimplant interface, retinal thickness, impedance, and perceptual threshold, which showed interesting mixed results. Correlation analysis showed the weak dependency between the size of a phosphene and the perceptual threshold and we did not find significant correlations between a phosphene and the height of the interface, impedance and retinal thickness.
关键词： Argus II,Usher syndrome,retinitis pigmentosa,deaf-blind,blind,artificial vision,retinal prosthesis system,retinal implant,deafblind
更新于2025-09-23 15:23:52
-
[IEEE 2018 15th International Conference on Control, Automation, Robotics and Vision (ICARCV) - Singapore, Singapore (2018.11.18-2018.11.21)] 2018 15th International Conference on Control, Automation, Robotics and Vision (ICARCV) - Quadrotors' Low-cost Vision-based Autonomous Landing Architecture on a Moving Platform
摘要： In this paper, a low-cost vision-based autonomous landing architecture for an unmanned aerial vehicle (UAV) on a moving platform is presented. First, a novel landing pad was designed for a monocular camera to robustly detect the pad in both high and low altitudes. In order to solve mirror effect and occasional misidentification, a 3D points cluster algorithm for relative position estimation is presented. Second, the dynamics of the quadrotor is simplified for this landing task and a PD controller is designed based on the estimated relative position. Finally, the low-cost system architecture of the quadrotor and the experiment results are both presented to show the effectiveness of the proposed method.
关键词： landing pad design,PD controller,quadrotor,moving platform,low-cost,vision-based,3D points cluster algorithm,autonomous landing
更新于2025-09-23 15:23:52