Xia Renbo

• Professor
• Chinese Academy of Sciences

To improve the performance and reduce the weight of new-generation transparent windshields in aircraft, composite structures are commonly used. The thickness of each layer in a composite windshield has a significant impact on its optical properties and load-bearing capacity. However, existing thickness measurement methods for transparent windshield...

Calibration is vital for the binocular visual system. The high precision measurement volume of the traditional binocular visual system is very small, therefore, it is difficult to measure large objects accurately. Accurate calibration for the binocular visual system is also extremely difficult in a large-scale field-of-view (FOV). Several attempts...

Bend tubes have been widely used in aerospace equipment due to their excellent mechanical properties and chemical properties. However, elliptic deformation will occur in the process of bending, which will reduce the pressure bearing capacity and service life of the bend tube. In order to ensure the production quality of the bend tube, a method of d...

As 3D acquisition technology develops and 3D sensors become increasingly affordable, large quantities of 3D point cloud data are emerging. How to effectively learn and extract the geometric features from these point clouds has become an urgent problem to be solved. The point cloud geometric information is hidden in disordered, unstructured points,...

High-precision measurement of sizes and key parameters of industrial parts is crucial to ensure manufacturing accuracy and assembly reliability. Low coherence measurement methods, offer advantages including high-precision, relatively large measuring range, and nondestructive nature. These measurement methods have shown great potential for applicati...

Tubes are widely used in the aerospace field for such things as engines, fuel pipelines, cooling systems, etc. To improve manufacturing efficiency and accuracy, researchers develop multi-camera vision methods for measuring tube geometry parameters. However, these methods cannot measure the exact position of the endpoints of the tube because the pro...

Inspired by the idea of the coordinate transformation of the robot arm, the tube structure is analogized to the robot arm structure, and the conversion method of the tube bend processing parameters LRA to the three-dimensional coordinate points is obtained. The method first regards the tube as a “mechanical arm”; then, using the similarity of the L...

3D scanning is a crucial step to ensure the machining quality of the workpiece and is an essential part of intelligent manufacturing. However, existing scanning systems usually have only one profiler, which must be combined with a dynamic tracking system to achieve a complete scan of a workpiece. This scanning method has low efficiency and complica...

Aiming at the problem of fuzzy edge segmentation and incomplete division in DeepLabV3+ segmentation results of parts, we propose an improved DeepLabV3+ semantic segmentation algorithm. Firstly, We replace Xception in the original network with lightweight MobileNet_V2 on the backbone network, improving the lightness of the network model. Then, chann...

Multi-camera systems have important applications in industrial online measurement, attracting wide interest due to their encouraging performance. However, how to develop a convenient and high-accuracy multi-camera calibration model is a big challenge. Traditional calibration methods based on 2D object require the preparation of high-precision and l...

For high-value aircraft, in order to ensure their stealth, pneumatic and safety characteristics, surface geometric defects such as unacceptable rivet height and seam width must be accurately detected during the manufacturing process. These defects need to be controlled within a very small scope of error. Traditional defect detection methods are dif...

Circular holes on industrial parts are a very important feature, and are often used as assembly references for other parts, so accurate measurement is required. Circular hole measurement based on computer vision is a promising method and very challenging task because it is tricky to find the edge of a hole accurately in an image due to the diversit...

Traditional point cloud registration methods require large overlap between scans, which imposes strict constraints on data acquisition. To facilitate registration, users have to carefully position scanners to ensure sufficient overlap. In this work, we propose to use high-level structural information (i.e., plane/line features and their interrelati...

Using 5-axis CNC machine tools for high speed machining, once a mutation occurs somewhere in the processing interference and mutations of surface corner, the mutation of tool axis vector may occur rapidly, while the use of traditional methods including retracting and then engaging again, will inevitably lead to reduce the product quality and proces...

Multi-camera vision systems are often needed to achieve large-scale and high-precision measurement because these systems have larger fields of view (FOV) than a single camera. Multiple cameras may have no or narrow overlapping FOVs in many applications, which pose a huge challenge to global calibration. This paper presents a global calibration meth...

The measurement accuracy of a phase-shifting measurement system is adversely affected by phase errors. This paper presents a theoretical analysis of phase errors caused by nonuniform surface reflectivity, such as varying reflectivity and a sharp change in reflectivity. Based on the analysis, a method to adaptively adjust the maximum input gray leve...

Many ellipse detection methods have been proposed for detecting ellipses in images. However, they are unsuitable for industrial images due to low signal-to-noise ratios (SNR). This paper presents an ellipse detection method combining the advantages of Hough transform(HT) based methods and the advantages of edge following methods, which is capable o...

A new extrinsic calibration method of a display-camera system that the display is not in the direct view of the camera is proposed. An annular mirror is used so that the camera can capture the virtual image of the display. The position of the mirror can be obtained after outer circle is detected. The position parameters are determined uniquely by p...

Complex pockets with one or more islands have been widely used in industrial and manufacturing production. In this paper, a new double spiral tool-path generation and linking method are proposed for complex pockets with islands which can be used for high-speed machining (HSM) is used. Taking into account the path interval, step length and other pro...

Complex pockets have been widely used in industry for the production. At present, the basic forms of tool-path which most commercial CAM software systems used are still limited to direction-parallel and contour-parallel offset. If the pocket contains multiple islands, the tool-path will include many cutter lifting and retraction motions. In this pa...

This paper focuses on developing an algorithm that can generate toolpaths in NURBS form for smooth, high speed and accurate machining. The initial toolpaths are obtained by tool radius compensation method which is based on the workpiece boundary offsetting. According to different lengths and the continuous short block (CSB) criterion, these offset...

Our work focuses on the extrinsic calibration of a display-camera system where the camera has no direct view of the display. An annular planar mirror is used to reflect the display so that the camera can capture the reflection images. The position of the mirror can be obtained after outer circle is detected. The position parameters are determined u...

Aiming at the extrinsic camera calibration with respect to a reference object without a direct view, an annular mirror based automatic calibration method is proposed. The camera captures the images of the reference object through the reflection of a known-radius annular mirror. The equations of the inner and outer circles of the mirror in the image...

To solve the problems in computer vision based measurement for smooth and continuous specular surface, an imaging simulation measurement method is proposed. The geometrical optics principle for specular surface is studied. A setup which consists of a camera and a LCD display is used to measure the specular objects. A parametric surface is used to s...

In this paper, a new method is presented to match a pair of visible and infrared images of same scene based on hybrid visual features including line segments and interest points. First, improved Harris corner extraction method and line segment detector method is used to extract feature points and segments. Then, a novel descriptor integrating the i...

Airport runway recognition technology would play an important role in developing intelligent weapon systems in the future. In this letter, a method of automatically finding runways in forward looking infrared (FLIR) images is proposed based on the knowledge of vision. First, the line segments in the images are extracted by a fast line segment detec...

Aiming at the calibration problem of the position parameter of camera and screen in vision-based specular surface measurement, we propose a fast and simple calibration method based on mirror. A LCD screen is used to project a new circle-based pattern, then a camera captures the image of the pattern in the mirror, and the camera calibration is perfo...

This paper presents a new feature matching algorithm for nonrigid multimodal image registration. The proposed algorithm first constructs phase congruency representations (PCR) of images to be registered. Then scale invariant feature transform (SIFT) method is applied to capture significant feature points from PCR. Subsequently, the putative matchin...

Docking ring is a typical component on space vehicle, which can provide a single circle feature. But monocular vision pose estimation based on single circle with two solutions, which can not be applied to practical engineering application. In order to exclude the interference of false solution, a new method of removing the false solution using Eucl...

To estimate UAV (unmanned aerial vehicle) pose relative to airfield runway, a posterior parameter estimation method using the runway image and onboard pose sensors is presented. Firstly, pose parameters of UAV are measured by onboard sensors such as altimeter, GPS (global positioning system) and INS (inertial navigation system). Supposing the prior...

Monocular pose determination from point and line mixed features is a typical problem in computer vision. Given one point feature and two coplanar line features on an object and their correspondences on the image plane of a calibrated camera, the pose parameters between the camera and the object can be calculated. The problem is studied in two cases...

An algorithm of effective feature line extraction for triangle mesh model is proposed. Feature line extraction of triangle mesh model is not only the key technology of revere engineering, but also an important research project of computer vision and machine intelligence. In the paper, firstly the principal curvatures and principal directions of ver...

For a wrap-around surface, many issues, such as difficulty in parameterisation of data points, incapability of automatic modelling, and high-accuracy surface reconstruction, need to be resolved. Therefore, an automatic accurate non-uniform rational B-spline (NURBS) surface reconstruction method is proposed in the present paper. First, after a metho...

This paper presents a simple and computationally efficient saliency extraction method for detecting dim small target from single frame in heterogeneous background. The proposed method is based on background subtraction (BS), which identifies targets from the portion of a image that differs significantly from a background model. A set of horizon-dir...

Pose measurement is an important problem in computer vision. This paper presents a method for finding the closed form solutions to camera pose measurement problem by using 2D-3D correspondences of three Z shaped lines. This problem is a special case of Perspective-three-Line (P3L) problem. By introducing two undetermined parameters, we translate th...

Surface mesh denoising is an important preprocess for many digital geometrical applications. Unlike most previous smoothing algorithms, which are vertex- or normal-based, we present a smoothing approach based on gradient field. The proposed algorithm first applies an extended non-local means image filter to diffuse mesh gradient field and then reco...

In this paper, a robust non-rigid multimodal image registration method is proposed using the local phase coherence representation (LPCR) as well as the residual complexity (RC) similarity measure. By integrating the LPCR and RC measure, we can find and track the non-rigid transformation of each pixel under diffeomorphism between the source and targ...

The STL model has become a standard in rapid prototyping (RP) technology, but slice errors are high because of tessellation. To improve the precision of slice contours, a novel slicing method is presented to compute accurate contour curves based on an STL model. First, the normal vector of a triangle facet to be sliced is calculated. An interpolati...

Camera calibration is a critical process to acquire 3D metric information from 2D images, and is a time consuming step because the first step of finding correspondences is often performed manually. In this paper a new calibration method is presented, which allows a camera to be calibrated fully automatically. In this method, five big circles are ad...

Camera calibration is one of critical steps in computer vision, also an exhaustive process because substantial human computer interactions are frequently required to deal with the matching problem. In this paper, an automatic matching method of markers for camera calibration is presented based on the local architecture characteristics of a new plan...

Part-building orientation (PBO) and scanning direction of path planning (SDPP) are two tasks of process plan in rapid prototyping technology (RPT). Through investigating the geometric issues of STL model and process planning of RPT. This paper establishes optimizing model based on the considerations of staircase effect, support area and production...

A method of scanning path plan of cylinder cross-section is proposed for layered manufacturing of special material for inner surface of revolving workpieces. At first, the fundamental theory of the manufacturing technology and the method of cylinder surface slicing based on triangular mesh are introduced in brief. Through analyzing the type of scan...

In this paper, we propose a new robust algorithm for the detection of the maximum inscribed circles (MIC) in images. We first use a vector distance transformation (VDT) strategy to create a distance field. Then, we globally search the maximal value in distance field to extract the medial axes. Finally, we give a procedure to determine the center an...

In level set method, initialization mode not only influences evidently the implementation efficiency but also relates directly to the final results. The paper presents an new initialization scheme for improving the segmentation performance of Chan-Vese model. The proposed initialization scheme consists of two stages. The first stage computes rough...

This paper presents an algorithm of the point constraint for B-spline surfaces. Based on static energy model, the shape of the surface can be sculpted in an intuitive manner by changing the force distribution vector and control point constraint, thus this method provides the user with a familiar interface for modifying the shape. Furthermore, in or...

Rapid prototyping is a technology of layered manufacturing; therefore, the scanning path of certain layer is quite important. An algorithm of sub-regional scanning path generation is presented in the paper. The algorithm divides complex sectional contours into some simple non-hole sub-region, the scanning path in every sub-region is continuous and...

This paper introduces some key technologies in laser path generation including contour scanning path and internal scanning path for stereolithography. To generate the contour scanning path the cross-sectional contours must be offset a distance inward solid area in the course of machining parts because of the size of laser light spot. A simple calcu...

In this paper, we proposed an improved version of the marching cube algorithm which gives a topologically correct triangular approximation of the isosurface for any cube configuration. First, a classification and characterization of critical points on the isosurface of trilinear functions is studied in detail. Then, unlike the past work on marching...

The initialization is an important step in the level set method. However, it is a computational time-consuming step. In order to speed up the level set evolution procedures, we first introduce a new initialization algorithm based on the vector distance transform, which propagates a vector with a position of the nearest object pixel instead of the s...