Robust and adaptive merge of multiple range images with photometric attribute

Polishing operations often require constant normal contact forces, making robotic systems difficult to apply in polishing operations with offline programming methods that the geometric information of workpieces is lacking. This paper proposes a new method to achieve normal contact control between a polishing tool and workpieces without priori geometric information and designs a robot end-effector applicable to this method. The method employs three laser distance sensors and a linear encoder to form a measurement system. The theoretical analysis and simulation verification of the proposed method shows that the proposed method can obtain a higher normal tracking accuracy with an average error of $$0.37^\circ$$ . An experimental platform built based on the proposed method also achieves good normal tracking in the actual polishing process. Both simulation and experimental results show that the proposed method provides a viable robotic polishing method when workpieces have no precise geometric models.

Presents a method for the reconstruction of real-world objects from multiple range images. One major contribution of our approach is the simultaneous registration of all range images acquired from different scanner views. Thus, registration errors are not accumulated, and it is even possible to reconstruct large objects from an arbitrary number of small range images. The registration process is based on a least-squares approach where a distance metric between the overlapping range images is minimized. A resolution hierarchy accelerates the registration substantially. After registration, a volumetric model of the object is carved out. This step is based on the idea that no part of the object can lie between the measured surface and the camera of the scanner. With the marching cube algorithm, a polygonal representation is generated. The accuracy of this polygonal mesh is improved by moving the vertices of the mesh on to the surface implicitly defined by the registered range images.

One major challenge of 3DTV is content acquisition. Here, we present a method to acquire a realistic, visually convincing 3D model of indoor environments based on a mobile platform that is equipped with a laser range scanner and a panoramic camera. The data of the 2D laser scans are used to solve the simultaneous localization and mapping problem and to extract walls. Textures for walls and floor are built from the images of a calibrated panoramic camera. Multiresolution blending is used to hide seams in the generated textures. The scene is further enriched by 3D-geometry calculated from a graph cut stereo technique. We present experimental results from a moderately large real environment. 1

This study identifies a way to accurately estimate extrinsic calibration parameters between stereo vision camera and 2D laser range finder (LRF) based on 3D reconstruction of monochromatic calibration board and geometric co-planarity constraints between the views from these two sensors. It supports automatic extraction of plane-line correspondences between camera and LRF using monochromatic board, which is further improved by selecting optimal threshold values for laser scan dissection to extract line features from LRF data. Calibration parameters are then obtained by solving co-planarity constraints between the estimated plane and line. Furthermore, the obtained parameters are refined by minimising reprojection error and error from the co-planarity constraints. Moreover, calibration accuracy is achieved because of extraction of reliable plane-line correspondence using monochromatic board which reduces the impact of range-reflectivity-bias observed in LRF data on checkerboard . As the proposed method supports to automatically extract feature correspondences, it provides a major reduction in time required from an operator in comparison to manual methods. The performance is validated by extensive experimentation and simulation, and estimated parameters from the proposed method demonstrate better accuracy than conventional methods.

In augmented virtuality, estimating object surface reflectance properties is important when rendering objects under arbitrary illumination conditions. However, faithfully estimating surface reflectance properties is difficult for objects having interreflections. The present paper describes a new method for densely estimating the non-uniform surface reflectance properties of real objects constructed of convex and concave surfaces having diffuse and specular interreflections. The registered range and surface color texture images were obtained using a laser rangefinder. In the proposed method, the light positions are first determined in order to take color images, which are then used to discriminate diffuse and specular reflection components of surface reflection. Surface reflectance parameters are then estimated based on an inverse global illumination rendering. Experiments were conducted to reveal the usefulness of the proposed method.

In this paper, we propose a three-dimensional (3-D) measurement method of objects in liquid with a laser range finder. When applying vision sensors to measuring objects in liquid, we meet the problem of an image distortion. It is caused by the refraction of the light on the boundary between the air and the liquid, and the distorted image brings errors in a triangulation for the range measurement. Our proposed method can measure the accurate 3-D coordinates of objects surfaces in liquid taken for calculating the refraction effect. The effectiveness of the proposed technique is shown through experiments. The accuracy of the 3-D measurement is 0.7 mm for objects located about 250 mm from the laser range finder when considering the refraction of the light, although that is 2.9 mm without the consideration of it.

These days, robust autonomous navigation incorporates the information on the digital map including variety of road features such as position of lane markings, road boundaries and etc. In this study, a method of extracting road features are developed using two dimensional laser range finder (LRF). In particular, range measurements are employed to identify the road surface using split and merge method. In addition, the reflectivity measurements are used to identify lane markings on the road surface. Finally a digital map of road surface is constructed which is useful for robust autonomous navigation. The proposed method is verified using experimental LRF data measured on the road environments.

This paper proposes a new method to estimate the 3D motion of a vehicle based on car-like structured motion model using an omnidirectional camera and a laser rangefinder. In recent years, motion estimation using vision sensor has improved by assuming planar motion in most conventional research to reduce requirement parameters and computational cost. However, for real applications in environment of outdoor terrain, the motion does not satisfy this condition. In contrast, our proposed method uses one corresponding image point and motion orientation to estimate the vehicle motion in 3D. In order to reduce requirement parameters for speedup computational systems, the vehicle moves under car-like structured motion model assumption. The system consists of a camera and a laser rangefinder mounted on the vehicle. The laser rangefinder is used to estimate motion orientation and absolute translation of the vehicle. An omnidirectional image-based one-point correspondence is used for combining with motion orientation and absolute translation to estimate rotation components of yaw, pitch angles and three translation components of T <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</sub> , T <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">y</sub> , and T <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">z</sub> . Real experiments in sloping terrain demonstrate the accuracy of vehicle localization estimation using the proposed method. The error at the end of travel position of our method, one-point RANSAC are 1.1%, 5.1%, respectively.

This paper presents a method to track and follow target person. A laser range finder (LRF) is used to measure highly accurate distance of objects with the range of 180 degrees. First of all, the erroneous data are excluded due to the error of LRF. Then all the raw sensor data are separated into many groups when the difference of the measuring distances of two adjacent laser points are beyond a limited value. For each group, the width is calculated, and it is considered as human legs if the defined conditions are satisfied. Finally, a real-time human following experiment with a SICK LRF and PIONEER mobile robot is done to validate our proposed method.

Recently, various deep learning-based high dynamic range (HDR) imaging methods have been proposed to overcome the performance limitation of simply merging or fusion of multiple low dynamic range (LDR) images. However, early deep learning-based methods did not fully exploit the inter-image relationship between different exposures. T o consider inter-image features without the dataset dependency, this paper presents a novel HDR method using three-dimensional residual dense network (3D RDN). We first reconstruct a set of differently exposed LDR images in video format. Next, we take the reconstructed video as an input to the 3D RDN, which generates an HDR image as output. Experimental results show that the proposed method provides higher score compared with state-of-the-arts HDR methods. To the best of authors knowledge, the proposed work first adopted the 3D RDN in the field of HDR imaging using more than three LDR images without any additional pre-or post-processing.

Successful operation of a mobile robot in the navigation area highly depends on the environmental information gathered by its exteroceptive sensors such as laser rangefinder, ultrasonic sensors, CCD cameras, etc. Typically, geometric information derives from the surface of objects in the environment, such as planes, edges, vertices, etc. In general, only partial and imprecise information can be obtained from real sensors, thus integration of multiple observations of the same geometric entity is required to compute an estimation of its location with respect to the vehicle. Partiality refers to the degrees of freedom associated to the representation of different geometric entities and how they determine the location of other entities related to them. Imprecision refers to the accuracy in the estimation of the location of geometric entities. Adequately representing such environmental information is a crucial aspect in mobile robotics. Dealing with uncertain geometric information has given rise to a variety of models which can be classified into either set-based models or probabilistic-models. On the one hand, set-based models describe the imprecision in the localization of a geometric entity by a bounded region, which correspond to the set of feasible locations for the geometric entity. Fusion of geometric information is performed using algebraic operations on the regions limited by the error bounds. On the other hand, probabilistic models represent the uncertain location of a geometric element using a probability distribution, usually Gaussian. Here, fusion is carried out using estimation methods.

In our previous report on an integrated localization using laser range scanners (LRS) and odometry, a variable covariance depending on the number of data, predicted incident angle, and predicted distance of LRS is introduced to EKF to deal with their effects on the localization. In the present paper, this method is applied to a front steering vehicle navigated by the LRS localization and odometry data. Its experimental results show that the proposed method localize the vehicle position and the outlier of LRS data is properly vanished.

We describe a method for aligning multiple range images given by a range finder. We used range images of the insides and outsides of buildings, which often contain many planes. Our method for registering range images is to refine the iterative closest point (ICP) algorithm using not only 3-D information but normal vectors of planes in the scene since structures of buildings often consist of planes. Analysis showed that multiple range images were aligned more accurately and faster using our method compared with the standard ICP algorithm.

A novel brain shift tracking protocol is introduced in this paper which utilizes laser range scan (LRS) data and 2D deformable image registration. This protocol builds on previous efforts to incorporate intra-operative LRS data into a model-updated image guided surgery paradigm for brain shift compensation. The shift tracking method employs the use of a LRS system capable of capturing textures of the intra-operative scene during range data acquisition. Textures from serial range images are then registered using a 2D deformable registration approach that uses local support radial basis functions and mutual information. Given the deformation field provided by the registration, 3D points in serial LRS datasets can then be tracked. Results from this paper indicate that the error associated with tracking brain movement is 1.1mm on average given brain shifts of approximately 20.5mm. Equally important, a strategy is presented to rapidly acquire intra-operative measurements of shift which are compatible with model-based strategies for brain deformation compensation.

High dynamic range (HDR) images are highly desirable and widely applied in our daily life. However, most conventional digital cameras are only able to capture low dynamic range (LDR) images. To produce HDR images, a popular and effective method is to merge multiple LDR images with different exposure settings. To achieve high quality HDR images, determining the exposure settings when capturing the multiple LDR images is of great importance. Though many researchers have paid attention to the exposure determination issue for HDR imaging, most of them have focused on DSLR cameras and surveillance video cameras. Meanwhile, mobile devices nowadays have increasing use but few studies in literature focus on them. Since cameras of mobile devices have some features different from DSLR cameras and surveillance video cameras, many existing exposure determination methods may not be directly applied on them. To fill this research gap, this paper proposes a dynamic exposure determination method for HDR photography on mobile devices, in particular on iOS mobile devices. In the proposed method, three LDR images are captured on iOS devices, which are then merged to generate an HDR result image that can cover more details of the target scene. Considering the time and memory limits of mobile devices, the proposed method analyzes the histogram of the target scene thumbnail to determine the three exposure settings with suitable exposure gaps. Based on our proposed method, we have developed an iOS HDR Camera App. The experimental results demonstrate that the HDR result images of this App are capable to contain more details in both bright area and dark area than those of Apple's default HDR Camera App on iPhone 5s for real-world target scenes.