The optical triangulation probe (OTP), which consists of a light spot projector and a camera, has found widespread applications for three-dimensional (3D) measurement and quality control of products in the industrial manufacturing. The OTP calibration is an extremely important issue, since the performances such as high accuracy and repeatability are crucially depended on the calibration results. This paper presents a flexible approach for modeling and calibration of the OTP, which only requires planar patterns observed from a few different orientations and light spots projected on the planes as well. For the calibration procedure, the structure parameters of the OTP are calculated, such as the camera extrinsic and intrinsic parameters which include the coefficients of the lens distortion, and the directional equation for the light axis of the projector. For the measuring procedure, the formulations of 3D computation are concisely described using the calibration results. Experimental tests of the real system confirm the suitable accuracy and repeatability. Furthermore, the technique proposed here is easily generalized for the OTP integration in robot arms or Coordinate Measuring Machines (CMMs).
KEYWORDS: Calibration, Cameras, Imaging systems, 3D acquisition, Optimization (mathematics), Distortion, Visual process modeling, 3D modeling, 3D metrology, Optoelectronics
In the field of vision measuring, camera calibration is a necessary precondition of the three-dimensional measurements. For the sake of obtaining high-precision external parameters, the inadequacies which cycling conditions restrict external parameters calculation in traditional calibration method was founded and elaborated in detail. The re-projection error was re-selected as cycling conditions and a new amended method was designed to solve the above shortcomings. Then Binocular vision system calibration experiments was carried , it has been found that standard deviation and the maximum error of three-dimensional reconstruction target after the amendments for external parameters is less than the calibration errors before the amendments.
A photoelectric autocollimator, which consists of an optical autocollimator and an area CCD and whose standard
deviation is less than 0.5" calibrated by a dual-frequency laser interferometer HP5528A, is proposed in the paper. The
positional precision of a NC motorized stage is detected automatically and quickly by applying the photoelectric
autocollimator calibrated and an optical polyhedron to finish the error compensation of the stage. According to GB/T
17421.2-2000, when the polyhedron and the stage both revolve with the same axis, the positional error of the stage is
measured by the photoelectric autocollimator and then sent back to the controlling system of the stage manually or
automatically. Experimental results show that an angle can be measured by the photoelectric autocollimator and the
calculated position accuracy agrees with that which is measured by a Triangle autocollimator 500-57 at the same time.
KEYWORDS: Clouds, Information technology, 3D metrology, Information science, Image registration, Reverse engineering, Computing systems, Reverse modeling, Computer graphics, 3D scanning
3D point clouds registration is a crucial problem in reverse engineering. In order to register point clouds without manual
information on objects, a novel method is proposed based on geometric properties of point clouds. The method consists
four parts: selecting public areas, computing normal vectors and curvatures, finding corresponding points and further
precise registration. Public areas of point clouds are selected through a scan order method. Normal vector and curvature
of each point are calculated through surface fitting. Curvatures of points are taken as the registration relationship and all
the pair-wise points with the same or similar curvature are extracted. The property of distance invariance in rigid body
transformation is used to match the pair-wise points. Matching points are picked out and mismatched ones are excluded
by comparing the curvatures in the k-neighborhood. After that the algorithm of quaternion is used to compute transform
matrix and ICP algorithm is introduced to improve registration precision. Experimental results show that the proposed
method is robust when registering point clouds of different scans.
KEYWORDS: Computer aided design, Solid modeling, Distributed interactive simulations, Information technology, Information science, Data modeling, Control systems, Inspection, Feature extraction, Visual process modeling
Based on the CAD model of the free-form surface workpiece, an automatic method to obtain the position and orientation
of the workpiece without contact is proposed. The method consists of two steps. The first step is rough localization
which gets the approximate correspondence between workpiece and CAD model based on surface features. The second
step is fine localization. It is achieved by searching the nearest point on the CAD surface to the light dot by gridsubdivide
surface of CAD model. ICP (iterative closest point) algorithm that obtains precise correspondences between
the workpiece and the CAD model is used for this step. Experiments were carried out and the results show that the
method is efficient and robust.
KEYWORDS: Field programmable gate arrays, Computing systems, Image processing, Digital signal processing, Edge detection, Digital filtering, 3D image processing, Signal processing, Image filtering, Imaging systems
Image processing is necessary for three-dimensional information recovering of stereovision measurement system and it
is always bottleneck for real-time applications. In order to accelerate system computational power, the design of SOPC
system which can fulfills image processing tasks parallel is discussed. As a part of high-speed stereovision measurement
system, the application specific SOPC is designed as an embedded PCI board card of hosts PC. This paper focuses on
three aspects. Firstly, Principles of SOPC system designing and SOPC features selecting are analyzed with measuring
requirements under consideration. Then the realization of SOPC system is described in detail. The embedded processor,
special IPs (Intelligent Properties), several custom logic modules are included in a single FPGA. All units are seamlessly
integrated into the overall system using the system builder interface. The parallel processing is illustrated by examples.
In the end, simulation and debugging results of SOPC system are introduced. Elements that influence running time are
analyzed and final results are given. Experiment and test results show that all the functions needed were realized with
much higher efficiency and processing speed in our SOPC system than conventional software.
This paper presents an approach to measuring the coaxiality of a shafting and bearing system for large and medium-sized
ship using CMOS (Complementary Metal Oxide Semiconductor) image sensor. The measurement system consists of a
laser diode with single mode fibre pigtail, which generates a reference datum-line, and position sensor of centre of
shafting or bearings that is composed of a CMOS image sensor and a rotating inductance displacement transducer. The
experimental results show that the approach meets the precision requirement of the coaxiality measurement. It provides
flexibility and costs less. The method can also be applied to the positioning and accurate coaxiality measurement of
similar bearing holes of large-scale axles.
KEYWORDS: Cameras, Image registration, 3D modeling, Imaging systems, Sensors, Calibration, 3D metrology, Data modeling, Visual process modeling, 3D acquisition
The checkerboard target is precisely processed and fixed on the measurement system. The feature points' coordinates on
the target in the measurement coordinate system are invariable with the motion of the measurement sensors, and obtained
by using the intermediary coordinate transformation method. The feature points are captured by an accurately calibrated
camera. The three collinear and equidistant points are used to build the simplest perspective-three-point model, and the
coordinates can been calculated by the model. Not less than three non-collinear points, which can be obtained by two
groups of models, are used to coarsely estimate the exterior parameters from the target coordinate system to the camera
coordinate system. To improve the exterior parameters' precision, an optimization scheme is used to minimize the cost
function of re-projection error. Every feature point' coordinates in the camera coordinate system can be accurately
calculated. Finally, the registration matrix is obtained. Experimental results show that the new method is simple, flexible
and effective.
Four issues of the method of photoelectric autocollimation for two-dimensional small angle measurement such as
adjustment of calibration system, elimination of the effect on the measurement result induced by the installing position of
the CCD camera relative to the autocollimator, processing methods of experimental data and repeatability of the system
are studied in the paper. The calibration system which is composed of a precision motorized stage, a dual-frequency laser
interferometer and a photoelectric autocollimator, is regulated to accomplish two separate calibrations around Z-axis and
X-axis in the measuring coordinates system which is established by considering horizontal, vertical line of the crosshair
and optical axis as X-axis, Z-axis and Y-axis respectively. Experimental results show that the formula method which can
reduce the random error and remove the system error effectively with the average error of zero and the standard
deviation 0.59". Compared with traditional optical autocollimator, the improvement of the photoelectric autocollimation
method can be used to achieve rapid and automatic positional detection and error compensation of CNC rotary shaft.
A method of distortion correction based on parallel lines is presented, not only for radial distortion coefficient but also for tangential distortion coefficients. Firstly, an approximate line passing through the center is approached by anamorphic curves. The coefficient of distortion is combined with the vision process to correct an anamorphic image for the final calibration of the liner model. Secondly, the point with only tangential distortion is gained by the actual point minus radial distortion. The distances from the point with tangential distortion and the ideal point to the principal point are equivalent. The Least-squares method is used to fit lines in theory, and the coefficients of tangential distortion are gotten.
The coded target design is geometrically constructed only of circular elements, i.e. circular marked point, circular reference point and circular coding points. Marked point is surrounded by a reference point and some coding points with bit position at equally spaced angular interval. The circular radius of the reference point is bigger than that of coding points. Marked point represents the point location itself. Reference point provides a start bit for coding points. Coding points provide robust identification of the target anti-clockwise, starting from reference point. The design method provides a sufficient number of identification points by introducing a reference point. Finally, the application of the proposed coded targets to 3D data registration is described. Experimental results show that the developed coded targets are independent to location, rotation and change of scale, and the marked points are easily and accurately detected.
In digital photogrammetry and computer vision, for high accuracy, several images obtained from the different positions are required. The targets in the images are recognized and matched by the computer automatically. This paper present a new method based on the bundle adjustment for the targets matching. It transforms the problem solving space from the 2D images into the 3D world space. With a few targets matched manually, the projection lights can be reconstructed with the interior and exterior parameters provided by the bundle adjustment. More targets are matched based on the correlations of the reconstructed lights. With the new matched targets, the bundle adjustment runs again and the parameters are revised. Repeat these steps until all targets are matched. Experimental results are shown to testify its feasibility and validity. In the experiments, the matching rate of the new method can reach 100%, while the missing rate is very low.
In order to get good accuracy, designer used to consider how to place cameras. Usually, cameras placement design is a
multidimensional optimal problem, so people used genetic algorithms to solve it. But genetic algorithms could result in
premature or convergent problem. Sometime we get local minimum and observe vibrating phenomenon. Those will get
inaccurate design. So we try to solve the problem using the changing environment genetic algorithms. The work
proposes giving those species groups difference environment during difference stage to improve the property. Computer
simulation result shows the acceleration in convergent speed and ability of selecting good individual. This work would
be used in other application.
This work proposes an algorithm to measure the smoothness of three-dimensional object. Firstly use computer vision
technology to reconstruct the three-dimensional surface. Then construct high order vanishing moment wavelet to analyze
the data. In order to detect the discontinuous derivative points, this paper proposes an algorithm based on wavelet
transform. Wavelet is used to descript surface smoothness and detect abnormal data. When using different vanishing
moment wavelets to deal with data, we would get different value at those points with different smoothness value. That
surface with same smoothness would become a plane. Meanwhile, in order to improve the algorithm validity, this paper
proposes using Daubechies wavelet. This algorithm resolved our project problem. It can be used in other projects that
require the surface with good smoothness.
In order to accurately position the nodes of a SAR(synthetic aperture radar), a special assemble workbench is designed
and a laser tracker named LTS-1100 from the API is used to measure it. The two stations layout of laser tracker along the
lengthways center line of the workbench are adopted. The six common points used when the laser tracker moving from
one station to the other are selected based on the stability experiment under the working environment. The laser tracker
measures the positions of all the points real time and directs their adjustments. The distance error of the points from the
ideal on the workbench is less than ±0.105mm. The close range photogrammetric method is then used to measure the
nodes position of the SAR antenna assembled on the workbench. The result indicates the coincidence with that obtained
by the laser tracker.
Analyzes the interferogram of a system for measuring the dispersion of CFBGs (chirped fiber Bragg grating) through
Fourier spectrometric method. Theoretical representation of the interferogram is provided. A numerical model for
simulating the interferogram is established under the common cognition that the recorded interfogram during the
scanning of OPD (optical path difference) along the light beam axis is the sum of separate symmetrical interferogram of
all quasi-monochromatic light components according to power. This analysis method will facilitate the analysis of
interferogram of several interferometric systems in which the light transmits via a dispersive medium.
The measurement system currently used in Phase Measuring Profilometry(PMP) consists of a Digital Light
Projector(DLP), a CCD camera, and a computer system. However, the inherent gamma nonlinearity of the DLP and
CCD camera can affect the output with a nonsinusoidal fringe image. In the same tine the systematic noise is an
important error source. Some conventional filtering algorithms may either make the fringe more blurring or be
inefficient for fringe images. Eventually, the obtained fringe image is non-sinusoidal and with systematic random
noise inevitably. Aimed at these problems, a pre-processing method for the fringe image is presented in this paper.
Firstly, an anti-deforming light model is designed and then projected by the DLP. Through the gamma nonlinear
response of the whole system, the waveform is corrected. Secondly, an improved orientation filter-based method is
designed to overcome the systematic random noise and can achieve better effect than other algorithm does.
Experiment, according to aforementioned two steps, is carried out and preferable fringe images can be gained. The
fringe waveform is more close to the ideal sinusoidal wave. Also the systematic noise is reduced effectively while
the fringe image is still clear. In the paper, two steps of this method are detailed and some experimental results are
also reported.
Many vision tasks such as 3D measurement, scene reconstruction, object recognition, etc., rely on feature
correspondence among images. This paper presents a novel binocular method taking full account of the causes of
matching ambiguities. On the basis of rectification which increases the distribution similarity of the points from the two
images, the dynamic programming technique is used to search for an optimal path with the matching fashions of the
most appropriate correspondences along each scan line as its nodes, which makes a global minimization of the
summation of all the intra- and inter-scan line disparity gradients. This method was applied to the 3D surface
measurement for an inflatable parabolic reflector with satisfactory results despite the significantly different viewpoints
and occlusion that indicates the robustness, efficiency and application feasibility in real-time measurement field of this
method.
A recognizing and sub-pixel locating algorithm of a crosshair image and a calibrating method of the setup are studied. A
sub-pixel locating accuracy and 2" within the full range are achieved. The crosshair captured with a CCD camera was
processed and calculated by a computer. The gray values of each cross section of each line were fitted to Gaussian curve
with the least squares method, the central points of the curves were fitted to a line with the same method, the
intersections of two pair of lines were obtained and 2D displacement in pixels was figured out. With the precision stage
and HP5528A dual-frequency laser interferometer, the setup within the full range was calibrated with piecewise linear
interpolation method. Eye fatigue and errors by eye sighting and reading are eliminated and measuring efficiency is
raised greatly. It is suitable for NC machine stage precision measurement and stage error automatic compensation.
Remote medical monitoring network for long-term monitoring of physiological variables would be helpful for recovery of patients as people are monitored at more comfortable conditions. Furthermore, long-term monitoring would be beneficial to investigate slowly developing deterioration in wellness status of a subject and provide medical treatment as soon as possible. The home monitor runs on an embedded microcomputer Rabbit3000 and interfaces with different medical monitoring module through serial ports. The network based on asymmetric digital subscriber line (ADSL) or local area network (LAN) is established and a client - server model, each embedded home medical monitor is client and the monitoring center is the server, is applied to the system design. The client is able to provide its information to the server when client's request of connection to the server is permitted. The monitoring center focuses on the management of the communications, the acquisition of medical data, and the visualization and analysis of the data, etc. Diagnosing model of sleep apnea syndrome is built basing on ECG, heart rate, respiration wave, blood pressure, oxygen saturation, air temperature of mouth cavity or nasal cavity, so sleep status can be analyzed by physiological data acquired as people in sleep. Remote medical monitoring network based on embedded micro Internetworking technology have advantages of lower price, convenience and feasibility, which have been tested by the prototype.
Reports a laser diode alignment system for monitoring dam, building, bridge, etc. It comprises a light source and a
receiving unit. The former is mainly composed of a laser diode for emitting light in red spectrum band at wavelength of
680nm, a single mode fiber for spot transforming and mode selecting, and a reverse telescope for collimating the
emission light beam. The whole light source is mounted on a 4-dimension adjustable table fixed on a tripod. The latter is
an embedded computer based on PC104 architecture. It contains a mainboard, an image grabber, a LCD display and a
CCD. Central position of the light spot is calculated according to barycenter method. It is proved by experiments in lab
that within a span of 8m the direction stability of the alignment system is up to 1X10-6 magnitude order.
The measurement of helical surfaces is used not only to evaluate the precision of the surfaces, but also to analyze the sources of its deviation from the specification. In this article, many aspects of helical surfaces are contained, as the mathematical principle of forming the surface, the measuring equipment, the measuring methods and our research status. First the forming principle of the helical surface is given and the classification methods by the mathematical equations are shown. Then the equipment is summarized, and the classification and the working principles of the measurement of the helical surface is given. At third we introduce our research status, and also give a measurement example of a hobbing cutter and a concise analysis. From the analysis we derive a method to improve the quality of the helical surface. Finally, we discuss the future development of the measurement of the helical surface.
In order to enhance the precision of videogrammetry for the profile of large-scale antenna, multi-station network is
necessary, which will also affect the reliability, efficiency and stability of the results. In this paper, the impacts of normal
and convergent configurations on the precision of space targets are discussed. Constraints, including image scale, depth
of field, field of view, and distribution of image points, which affect the network configurations, must be taken into
account. Specific network configurations for the large-scale gossamer inflatable antenna are designed and experimental
data results verify the conclusions.
A point matching method in photogrammetry for microwave antenna with non-metric cameras is presented. The method with coded points and epipolar constrain is described. To solve the point matching in large measurement range, the idea of the dividing area is presented. The circular coded points as identifiers are used to identify different subareas, and subarea matching is preformed. Then, the measured points are matched with epipolar constraint in each subarea of left and right images. In order to reduce the influence of lens distortion to epipolar constraint, a control points grid as frame of reference is used to correct all image points using a distortion model. Experimental result indicates that this method is valid and reliable to reduce data processing time, and to improve image matching veracity and automatic degree. The points matching ratio is 100%. This method has provided one excellent solution for point matching problem in large object photogrammetry with non-metric cameras.
KEYWORDS: Cameras, Distortion, Calibration, 3D modeling, Photogrammetry, 3D metrology, Digital cameras, Visual process modeling, Nonlinear optics, Electronics engineering
Recent progress of images taken by non-metric digital cameras encourages amateurs to utilize it for 3D measurements, but also introduces the difficulties of calibration. The ideal camera projection model is based on the pinhole linear model which proceeds on the assumption that the object point, the perspective center of the lens and the ideal image point all lie on a straight line. However, real camera lenses introduce nonlinear distortions that affect the accuracy of the transformation unless proper corrections are applied. In this paper, we present a new camera calibration method adopting a linear projection model and a nonlinear lens distortion model. Experiments are conducted on real data to compare the proposed method with the non-linear bundle adjustment and DLT (Direct Linear Translation), which shows that the accuracy obtained with the proposed method is close to the former, and much higher than the latter.
2-D vision approach used to measure dimensions of complex cross-section of an object is presented in the paper. Considering the complexity of profile in cross section, some methods for edge detection and geometric elements recognition are studied. The means that can acquire a fine boundary image is adopted. A special approach for recognizing all geometric elements in image is presented. The recognition is processed by matching elements in image with corresponding elements in a standard graph. The standard graph is drawn according to the machine drawing. The measurement for total parameters is accomplished at the same time. The measure precision is ± 0.05mm.
A digital close range photogrammetric method for space deployable microwave antenna is presented. The planning of multi-camera stations, 3D data connection of images of different view field and the calibration of the cameras are discussed. The procedure of calibration has two steps: direct linear transformation and the bundle adjustment. The measurement error resulted from lens distortion can be corrected. The relative accuracy of 1 part in 1000 can be achieved using on site automatic calibration during measurement. The profile of the space deployable microwave antenna was obtained through the least squares fitting of the adjusted observations.
KEYWORDS: Stereo vision systems, 3D image processing, 3D metrology, 3D vision, Artificial neural networks, Distortion, Associative arrays, 3D image reconstruction, Neural networks, 3D acquisition
With a Damman grating and an artificial neural network back-propagation network (BP network), a fast and parallel 3-D non-contact measurement method using a stereo vision system is introduced. Some detail discussions on optical setup, data acquisition and some method to raise the training precision of BP network are proposed. Experiments have been completed and the results proved the feasibility of the method. Using the method, one can get the object profile information rapidly, deal with data almost parallel and need not to consider the effect of lens distortion.
A single mode fiber (SMF) pigtailed laser diode alignment system used for coaxiality measurement of large scale holes is studied. The drift of angle of laser beam is restricted obviously by mode restriction of single mode fiber. The stability of laser reference line is the order of 1.5 x 10-6. It can be used in measurements of position and form errors of large scale workpieces.
A new method of stereo vision measurement using artificial neural network is presented. The experiment results show its feasibility. It can reduce the effect of nonlinearity of optical system.
This paper introduces 3D profiles noncontact measurement method by using a kind of artificial neural network- backpropagation network. It makes a detail discussion on optical system, data acquisition, and how to train the BP network. An object is projected by a laser stripe light, and the lien image of the object is captured by a CCD camera. The input data of BP network is given from a series of image on CCD by certain image processing technology. Once a mapping relationship of coordinates between CCD camera image plane and related position in object's space is set up by sing BP network, the image plane coordinates of any objects are used as the input data of BP network it will immediately obtain the corresponding coordinates in space according to the output. Thus the profile can be easily established by these coordinates. This method reduces the requirement for system accuracy, and no need to test and a light the whole system in advance accurately.
On the basis of photogrammetry, a stereo vision system has been built for 3D profile measurement. It consists of two CCD cameras, a frame grabber and a personal computer. It will be extensively used in industrial inspection and robot vision, etc. In this paper the calibration of system, the measurement procedure of a 3D object, experimental results and the factors to affect the accuracy are discussed in detail. The stereo vision system needn't precise alignment of optical system and precise measurement of system's parameters in advance. It is suitable especially for the noncontact measurement of large scale objects with complex surface.
This paper presents a holographic diffuser for diffusion to definite direction. It can be used as objective carrier in an imaging system. The holographic diffuser has several functions of diffuser, prism and field lens. The angle of diffraction can be controlled. So the holographic diffusers have many advantages over conventional diffusers.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.