Search results for "Computer vision"
showing 10 items of 2353 documents
Developing a framework for using structure-from-motion techniques for road distress applications
2020
On Urban road networks, road agencies need to quickly identify road pavement distresses in order to identify appropriate maintenance and rehabilitation strategies. This is integral as agencies are plagued with financial and time constraint issues. There have been several attempts over the last few years to identify new solutions and techniques to solve these issues. Several of these have shown merit and accuracy in identifying distresses. However, the techniques in many instances are not correlated to available distress identification standards. One of the considered techniques is the use of Structure-from-Motion, which tries to recreate 3D distress models for identification and analysis. T…
Online Intercultural Exchanges Through Digital Storytelling
2017
This article focuses on the affordances of a digital storytelling project in developing students' language, digital and other skills: learning and innovation, creativity, critical thinking, problem solving, team working, and life and career skills. The project was undertaken by university English for Specific Purposes students and was conducted within an Online Intercultural Exchange between the Cyprus University of Technology and the University of Valencia. Its design was based on a Project-Based Learning (PBL) methodology. It incorporated active learning and multimodal resources and capabilities. The need for transforming language teaching pedagogies was borne in mind, as it is necessary …
Real-time Multispectral Image Processing and Registration on 3D Point Cloud for Vineyard Analysis
2021
International audience; Nowadays, precision agriculture and precision viticulture are under strong development. In order to accomplish effective actions, robots require robust perception of the culture and the surrounding environment. Computer vision systems have to identify plant parts (branches, stems, leaves, flowers, fruits, vegetables, etc.) and their respective health status. Moreover, they must merge various plant information, to measure agronomic indices, to classify them and finally to extract data to enable the agriculturist or expert to make a relevant decision. We propose a real-time method to acquire, process and register multispectral images fused to 3D. The sensors system, co…
2D size, position and shape definition of defects by B-scan image analysis
2015
The non-destructive evaluation of defects by automatic procedures is of great importance for structural components. Thanks to the developments of the non-contact ultrasonic techniques, the automation of the inspections is gaining a progressively important role. In this work, an automatic inspection technique for the evaluation of defects by the analysis of B-scan images obtained by a laser ultrasonic system is presented. The data are extracted directly from a B-scan map obtained for a panel with internal defects, and are used to build an image of the cross section of the panel. The proposed automatic procedure allows the definition of size, position and shape of defects in panels of known t…
High Quality Reconstruction of Dynamic Objects using 2D-3D Camera Fusion
2017
International audience; In this paper, we propose a complete pipeline for high quality reconstruction of dynamic objects using 2D-3D camera setup attached to a moving vehicle. Starting from the segmented motion trajectories of individual objects, we compute their precise motion parameters, register multiple sparse point clouds to increase the density, and develop a smooth and textured surface from the dense (but scattered) point cloud. The success of our method relies on the proposed optimization framework for accurate motion estimation between two sparse point clouds. Our formulation for fusing it closest-point and it consensus based motion estimations, respectively in the absence and pres…
A coarse to fine 3D acquisition system
2015
International audience; The 3D chain (acquisition-processing-compression) is , most of the time , sequenced into several steps. Such approaches result into an one-dense acquisition of 3D points. In large scope of applications , the first processing step consists in simplifying the data. In this paper , we propose a coarse to fine acquisition system which permits to obtain simplified data directly from the acquisition. By calculating some complementary information from 2D images , such as 3D normals , multiple homogeneous regions will be segmented and affected to a given primitive class. Contrary to other studies , the whole process is not based on a mesh. The obtained model is simplified di…
Multisensory texture exploration at the tip of the pen
2016
A tool for the multisensory stylus-based exploration of virtual textures was used to investigate how different feedback modalities (static or dynamically deformed images, vibration, sound) affect exploratory gestures. To this end, we ran an experiment where participants had to steer a path with the stylus through a curved corridor on the surface of a graphic tablet/display, and we measured steering time, dispersion of trajectories, and applied force. Despite the variety of subjective impressions elicited by the different feedback conditions, we found that only nonvisual feedback induced significant variations in trajectories and an increase in movement time. In a post-experiment, using a pa…
Toward a real-time tracking of dense point-sampled geometry
2012
4 pages; International audience; In this paper, we address the problem of tracking temporal deformations between two arbitrary densely sampled point-based surfaces. We propose an intuitive and efficient resolution to the point matching problem within two frames of a sequence. The proposed method utilizes two distinct space partition trees, one for each point cloud, which both are defined on a unique discrete space. Our method takes advantage of multi-resolution concerns, voxel adjacency relations, and a specific distance function. Experimental results obtained from both simulated and real reconstructed data sets demonstrate that the proposed method can handle efficiently the tracking proces…
SAMSLAM: Simulated Annealing Monocular SLAM
2013
This paper proposes a novel monocular SLAM approach. For a triplet of successive keyframes, the approach inteleaves the registration of the three 3D maps associated to each image pair in the triplet and the refinement of the corresponding poses, by progressively limiting the allowable reprojection error according to a simulated annealing scheme. This approach computes only local overlapping maps of almost constant size, thus avoiding problems of 3D map growth. It does not require global optimization, loop closure and back-correction of the poses.
PERFORMANCE EVALUATIONS OF 3D WEB-SERVICES AND OPEN-SOURCE SOFTWARE FOR DIGITAL MODELING OF ARCHAEOLOGICAL FINDS
2016
The Structure From Motion (SFM) algorithms derived from Computer Vision community make image-based techniques a competitive and user friendly approach to obtain 3D information of real objects. These algorithms have been implemented in many 3D web-services and open-source software to develop free instruments for the reconstruction of very detailed 3D digital models. Their use is becoming very common above all in archaeological studies but their potentiality for metric application has not been tested at all. The aim of the work is to investigate the quality and the accuracy of 3D models obtained through 3D web-services and SFM open-source software. Different tests have been performed using so…