Visual measurement plays an increasingly important role in the field o f aerospace, ship and machinery manufacturing. Camera calibration of large field-of-view is a critical part of visual measurement . For the issue a large scale target is difficult to be produced, and the precision can not to be guaranteed. While a small target has the advantage of produced of high precision, but only local optimal solutions can be obtained . Therefore, studying the most suitable ratio of the target size to the camera field of view to ensure the calibration precision requirement of the wide field-of-view is required. In this paper, the cameras are calibrated by a series of different dimensions of checkerboard calibration target s and round calibration targets, respectively. The ratios of the target size to the camera field-of-view are 9%, 18%, 27%, 36%, 45%, 54%, 63%, 72%, 81% and 90%. The target is placed in different positions in the camera field to obtain the camera parameters of different positions . Then, the distribution curves of the reprojection mean error of the feature points’ restructure in different ratios are analyzed. The experimental data demonstrate that with the ratio of the target size to the camera field-of-view increas ing, the precision of calibration is accordingly improved, and the reprojection mean error changes slightly when the ratio is above 45%.
Compound eye has the merits of large FOV (Field of View), high acuity to motion and compact structure. In order to
achieve large FOV, most natural compound eyes have curved structures. However, a 3D microlens array alone cannot
work properly with a planar image sensor, as a result, very complex relay optical design is required for beam-steering
and image formation. On the other hand, artificial compound eyes with planar structure are easy to design and fabricated,
but the field of view is very small. To address this issue, an innovated design is presented in this paper. The system
comprises of a planar structured microlens array and two curved folded mirrors. A very high fill factor can be achieved
by using planar microlens array. The design was verified with Zemax simulation and preliminary experiment. The results
show that the system can achieve large FOV imaging without significant lens distortion and ghost image, demonstrating
the feasibility and flexibility of the proposed method.
The traditional vibration measurement method usually uses contact sensors, which induce unwanted mass and have limits for moving parts. Currently, using stereo vision for vibration measurement is developing fast because of noncontact and full field. However, for stereo vision, the high-accuracy reconstruction for the vibration motion is a significant challenge and the factors that affect the construction accuracy have not been thoroughly studied. The accuracy analysis of sinusoidal motion reconstruction is important because it can provide guidance for the reconstruction of other vibration motions. The high accuracy reconstruction for sinusoidal motion and the factors that affect the accuracy of the reconstruction are presented. First, the error model of reconstruction using stereo vision considering the delay time, frequency, amplitude, and disparity is established. The accuracy of sinusoidal motion reconstruction in the whole period is analyzed theoretically and experimentally considering subpixel interpolation. Peak identification is essential for the sinusoidal motion reconstruction and has the highest resolution requirement for the reconstruction resolution. This requirement is systematically investigated for the sinusoidal motions with different frequencies and amplitudes. The relationships between the reconstruction resolution and parameters of cameras are analyzed.
KEYWORDS: Cameras, Imaging systems, 3D modeling, Calibration, Image processing, Data transmission, Data compression, Precision measurement, MATLAB, 3D vision
As a method of non-contact, binocular vision technique is important for vibration test. When measuring the vibrating objects, the frame rate of cameras must match the vibration frequency of the object. If the objects vibrate in high-frequency, it will lead massive image data. However, the speed of data transmission is limited by massive data. Thus, in order to deal with these data, data compression is inevitable before data transmission. Binarization is a simple and fast process to minimize the size of the image data. But the pixel locations of the marked points may change after binarization, which will inevitably affect the reconstruction of the marked points and further influence the precision of the measurement data. In this paper, the parameters which influence the position of various types of marked points are studied after binarization for dynamic test. A vibration table is employed to provide a standard moving motion. The frequency and amplitude is given by the controller of the vibration table. Three types of marked points, dot, circle and cross, are studied in this research. The obtained images in the dynamic test are minimized after binarization, and the pixel locations of the marked points are obtained. The change of pixel location is studied for the different types of the marked points with the comparison experiments and the precision of the reconstruction is investigated later.
Multi-cameras calibration using spheres is more convenient than using planar target because it has an obvious advantage in imaging in different angles. The internal and external parameters of multi-cameras can be obtained through once calibrat ion. In this paper, a novel mult i-cameras calibration method is proposed based on multiple spheres. A calibration target with fixed multiple balls is applied in this method and the geometric propert ies of the sphere projection model will be analyzed. During the experiment, the spherical target is placed in the public field of mult i-cameras system. Then the corresponding data can be stored when the cameras are triggered by signal generator. The contours of the balls are detected by Hough transform and the center coordinates are determined with sub-pixel accuracy. Then the center coordinates are used as input information for calibrat ion and the internal as well as external parameters can be calculated by Zhang's theory. When mult iple cameras are calibrated simultaneously from different angles using multiple spheres, the center coordinates of each sphere can be determined accurately even the target images taken out of focus. So this method can improve the calibration precision. Meanwhile, Zhang's plane template method is added to the contrast calibrat ion experiment. And the error sources of the experiment are analyzed. The results indicate that the method proposed in this paper is suitable for mult i-cameras calibrat ion.
Calibration is the basis of three-dimensional (3D) reconstruction for machine vision technology. Nowadays, the most widely used calibration method among computer vision is the technique for binocular stereo measurement. However, binocular stereo vision has limited view field which is difficult to measure large-scale mechanical components synchronously. Thus, enlarging the view field is urgent in need for the large scale measurement. With the application of multi-camera system, the calibration for cameras with different focal lengths is required. In this paper, a method aiming at calibration problems for multi-camera system of different focal lengths is proposed. An imaging model for multi-camera system with various focal lengths is analyzed. The Harris corner detector is applied to determine the relationship between signal camera and checkerboard. Finally, the external parameters of different cameras can be obtained by the link with the checkerboard. The calibration results indicate that the calculation method used in this work can calibrate multi-camera with various focal lengths.
Energy harvesting (EH) derives from the idea of converting the ambient energy into electric energy, which can solve the problem of DC supply for some electronic equipment. PZT is a typical piezoelectric material of inorganic, which has been developed as EH devices to transfer ambient vibration energy into electric energy. However, these PZT devices require relatively violent excitation, and easy to be fatigue fracture under the resonance condition. In this paper, PVDF, which is a kind of soft piezoelectric polymer, is adopted for developing transducer. The PVDF devices are flexible and have longer life time than PZT devices under the harmonic environment. The EH researches are mainly focused on the development of energy transfer efficiency either by the mechanical structure of transducer or the improvement of circuit. However, the practicality and stability of the EH devices are important in the practical engineering applications. In this paper, a charge amplifier is introduced in the circuit in order to guarantee the stability of the battery charging under small ambient vibration conditions. The model of the mechanical structure of PVDF and the electric performance of circuit are developed. The experimental results and simulation show that the stability of battery charging is improved and the working time of DC equipment is prolonged.
Considering the relatively low price, convenience and potential high accuracy of Digital Fringe Projection techniques (DFP), it has begun widely to use in static or even in dynamic scene for object three-dimensional reconstruction. A conventional DFP system utilizes projector to project a sinusoidal fringe patterns onto object surface, which is generated by computer. By further fringe analysis, it could retrieve the phase for reconstruction. However, this technique is not trouble free for various factors such as camera noise, nonlinear intensity response between camera and projector, especially Gamma effect, which seriously distort the sinusoidal wave. Since 1-bit binary structure patterns would not be affected by Gamma effect, S Lei proposed squared binary defocusing techniques to overcome the limitation of DFP techniques above. However, by defocusing the projector, squared binary defocusing technique has low SNR property, whose precision will be affected by various error factors easily. This paper proposes two types of error elimination algorithms of squared binary defocusing techniques. Principle of each technique will be presented, and the experimental results show the differences between the two algorithms.
KEYWORDS: Tolerancing, Manufacturing, Modal analysis, 3D modeling, Dynamical systems, Monte Carlo methods, Structural dynamics, Statistical analysis, Chaos, Finite element methods
Uncertainty is the degree of suspicious for correctness or accuracy of measurement results. Parts of the manufacturing error directly affect the structural dynamic characteristics. The propagation of uncertainty in the static measurement from characteristics to the measurement results has been investigated for decades. However, the propagation of manufacture uncertainty to the results of dynamic measurement has been rarely investigated, which is an important issue for the design and tests of structures and equipment. In this paper, the propagation of manufacturing uncertainty to the dynamic measurement is investigated on the basis of modal analysis method. The manufacturing uncertainty here is represented by the tolerance of form and position, which is the difference between actual shape or position and the digital 3D model of geometry shape and position. Tolerance of form and position is an< important indicator to evaluate object geometry size and the relative position in the mechanical design. The uncertainty of the tolerance of form and position is assumed to satisfy Gaussian distribution. The relationship between modal parameters and tolerance is predicted by analyzing transition rules of modal parameters. A series of beams with tolerance of form and position analyzed experiment to verify transitive relation between varied modal parameters in modal analysis and tolerance. The investigation found that the distribution of the uncertainty of the parameters of dynamic measurement, such as the natural frequencies and modal shapes, are not inherently satisfied with Gaussian distribution, although the uncertainty of characteristics are Gaussian distributed, which confirms the difference of uncertainty propagation in the static measurement and dynamic measurement.
That sensors are used to collect experimental data is still prevalent nowadays. A common way to analyze the mode of vibration of a structure, lots of sensors may required to be installed on the structure in order to ensure completeness of data. However, if these data can be obtained based on machine vision, the effects caused by sensors on the beam would be removed. A new kind of non-contact method will be used to measure the mode of vibration of simply supported beam. In this paper, the basic theory of the simply supported beam and machine vision will be introduced. It is different from traditional way that is based on a large number of sensors to collect experimental data. Two cameras record the vibration process of simply supported beam while the beam is vibrating caused by an exciter. After those images that have recorded the vibration process of simply supported beam are processed, calibration and registration included, those data collected by sensors also are reconstructed by traditional modal test method for comparison. Through comparing the machine vision method and sensor based method, errors caused by the process of reconstruction might be analyzed. The first order modal vibration modes by using two ways and combining with finite element method to produce can also be analyzed what their differences are.
The dynamic testing of structures and components is an important area of research. Extensive researches on the methods of using sensors for vibration parameters have been studied for years. With the rapid development of industrial high-speed camera and computer hardware, the method of using stereo vision for dynamic testing has been the focus of the research since the advantages of non-contact, full-field, high resolution and high accuracy. But in the country there is not much research about the dynamic testing based on stereo vision, and yet few people publish articles about the three-dimensional (3D) reconstruction of feature points in the case of dynamic. It is essential to the following analysis whether it can obtain accurate movement of target objects. In this paper, an object with sinusoidal motion is detected by stereo vision and the accuracy with different feature detection algorithms is investigated. Three different marks including dot, square and circle are stuck on the object and the object is doing sinusoidal motion by vibration table. Then use feature detection algorithm speed-up robust feature (SURF) to detect point, detect square corners by Harris and position the center by Hough transform. After obtaining the pixel coordinate values of the feature point, the stereo calibration parameters are used to achieve three-dimensional reconstruction through triangulation principle. The trajectories of the specific direction according to the vibration frequency and the frequency camera acquisition are obtained. At last, the reconstruction accuracy of different feature detection algorithms is compared.
Vibration is a source to induce uncertainty for the measurement. The traditional passive vibration control method has low efficiency and limited working conditions. The active vibration control method is not practical for its power demanding, complexity and instability. In this paper, a novel semi-active vibration control technology based on magnetorheological
(MR) fluid is presented with dual variable stiffness and damping capability. Because of the rheological behavior depending on the magnetic field intensity, MR fluid is used in many damping semi-active vibration control systems. The paper proposed a structure to allow the both overall damping and stiffness variable. The equivalent damping and stiffness of the structure are analyzed and the influences of the parameters on the stiffness and damping changing are further discussed.
Dynamic measurement normally is point test using sensors such as accelerators. With the development of binocular stereo vision, the full field dynamic measurement becomes possible. However the dynamic measurement based on binocular stereo vision encounters several challenges. One of the major challenges is that the error of dynamical testing using binocular stereo vision is unknown. In this paper, two models of dynamic testing based on binocular stereo vision are established. The proposed models mainly analyze non-synchronized errors in dynamical measurement and can be used to minimize errors for improving the measurement accuracy.
Specialized video measuring system (SVMS) is a kind of instruments diagnosing only one special type of part with high performance. In the design of a SVMS, it is necessary to consider the special requirements during the inspection of the parts. We take the video-based inspecting instrument for watch escapement (VIIWE) as an example of SVMS. A method based on least interferences is proposed to determine an optimized movement layout. The quality of image taken by an imaging lens with a limited depth of field (100 μm) may be degraded by two parameters, i.e., the perpendicularity of optical axis to workbench and the parallelism of probe movement plane to workbench. To optimize the two parameters, we calibrate the perpendicularity with a definition function, measure the parallelism with a dial indicator, and develop a mechanism of adjustment. The scheme about movement layout and adjustment has been successfully applied in the VIIWE.
Escape wheel as a typical precision micro-machinery part is one of the most precision parts in one mechanical watch. A
new inspecting instrument based on machine vision technology used to achieve semi-automatic inspection of watch
escape wheel is introduced in this paper. This instrument makes use of high resolution CCD sensor and independent designed
lens as the imaging system. It can not only achieve to image an area with 7mm diameter once, but also has the resolving
power in micrometer and cooperates with two-dimensional moving station to achieve a continuous and automatic
measurement of the work pieces placed in array type. In which, the following aspects are highlighted: measuring princeple
and process, the basic components of array type measuring workbench, positioning process and verticality, parallelism
and other precision adjusting mechanism. Cooperating with novelty escape wheel preparation tool this instrument
forms an array type semi-automatic measuring mode. At present, the instrument has been successfully running in the
industry field.
The significant disadvantage of traditional sub-pixel edge location algorithm is difficult to achieve high accurate edge
location of micro-parts, because reflective characteristic of edge is widely divergent for material, chamfer, and any other
factors. This paper introduces a sub-pixel location algorithm for micro-parts edge which is based on matching template.
First, it studies the edge imaging characteristic for a specific practical engineering part, and constructs one dimension
matching operator. Second, we abstract the pixel edge points on the image of practical engineering parts, and get the
normal direction information of the whole points. Finally, according to the correlation operation between matching
operator and a series of pixel points by the normal direction of the edge points, found the best position for matching. This
position is the location point of the edge point under sub-pixel accuracy. Compared with traditional sub-pixel algorithm,
this algorithm requires different matching operators for different measured objects, not a universal operator to all types of
the edges. Therefore, the effect which brings by the parts' edge characteristic and the imaging characteristic can be
reduced. This algorithm has prominence significance for using sub-pixel technology to measure practical parts.
Experiment result shows that, for the practical engineering part which has clear measurement aim and conformable
character of the edge, this method will achieve high accurate edge location, and it is suitable for wholesale measurement
of edge for complex micro-parts which require high accurate.
Optical array technology is an attractive method of achieving high power output using diode bars, which is different from using traditional diode stacks. Reflectors are the main optical elements in optical array diode lasers, and their influences of alignment errors on the output beam are analyzed using the matrix method. Analytical expressions of the beam size and deviation influenced by the misalignment of the reflectors are deduced, and the misalignment tolerances are also given. Based on the results, it is helpful to adjust the reflectors to achieve an output beam of good beam quality from the optical array diode lasers.
KEYWORDS: Dispersion, Microwave radiation, Telecommunications, Signal processing, Digital signal processing, Signal detection, Sensors, Transmitters, Fiber optic communications, Optical amplifiers
A feasible and cost-effective online chromatic dispersion (CD) monitoring scheme for high speed optical communication system is demonstrated in this paper. Based on the formerly verified theoretical model, the output electrical power of a specific frequency band is tested to reflect the residual chromatic dispersion online. Thus the microwave devices are selected to form the electrical power detecting circuit, which consists of a PIN photodiode, a microwave power filter and a power detector in tandem. In the experiments, frequency band center were chosen at 5 GHz for the 10 Gbit/s system and 12 GHz for the 40 Gbit/s system. Then the output voltage of the detector was processed and converted to digital signal and the signal was processed to obtain the amount of the chromatic dispersion of the tested systems. For the 40 Gbit/s system, the maximum detectable chromatic dispersion was around 130-ps/nm and a resolution of 5-ps/nm/db was achieved at the chosen frequency band centered at 12 GHz. Performance of the microwave devices and comparison of the experimental results at different frequency bands are also discussed, which verified that the chosen center frequency was suitable for detection of CD in optical communication system beyong 10 Gbit/s speed.
A multichannel bidirectional dynamic data transmission system (DDTS) through a rotary interface with one fiber is designed based on a fabricated single pass fiber optic rotary joint. The feasibility of transmission system is tested at both 1310- and 1550-nm wavelength bands. The performance of this DDTS was measured using optical spectrum analyzer and lightwave multimeter. The insertion losses of DDTS were 1.55 and 1.20 dB at 1310- and 1550-nm wavelength bands, respectively. The total bandwidth of the DDTS is more than 170 nm.
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.