Based on the polarization splitting principle of the only-phase liquid crystal device, a common path phase-shifting interferometry is built to measure the whole and local phase modulation characteristics of a liquid-crystal spatial light modulator (LC-SLM) with high spatial resolution and high gray level. In this technique, the SLM is not only a device to be measured but also an optical component that is used to splitting the incident linearly polarized light. A specially designed control grayscale image is loaded into the SLM, and then several frames of phase-shifted interferograms are recorded. The wavefront phase can be calculated by using the phase-shifting algorithm. Through changing the loaded images, the phase modulation curve is obtained. The experimental steps and results are presented in the paper, and the uniformity of local phase modulation is also evaluated. The proposed method has stable measuring results and has the advantage of simple principle and convenient operation.
The line-structured light vision sensor is widely used in 3D reconstruction and scanning, quality inspection and other fields. In this technique, the segmentation of light stripe image is an important preprocessing steps, which separates light stripe and extract ROI from an original image. The segmentation of light stripe is the basis of centerline extraction and also affects the efficiency of centerline extraction. In our work, the 3D reconstruction of wood surface needs to be realized through using line-structured light sensor, and then the 3D defects need to be detected. The measured wood has a curved, rough and scattering surface. The acquired light stripe is relative complex. Several algorithms of image segmentation are studied, and the segmentation performance and operation efficiency are compared and analyzed, so as to help us find the optimum technique.
In the real-time structured light 3D measurement, there are issues such as low throughput and slow measurement speed in the extraction of light stripe centerline by computer software and hardware processors like ARM. This paper designs a real-time system for extracting the light stripe centerline based on the ZYNQ-7000. Firstly, the thresholding segmentation of light stripe image is implemented by using the Otsu’s method. Secondly, the Gaussian filter is used to remove the effect of noise. Finally, the centerline of light stripe is extracted by using the grayscale gravity method. The image processing algorithm is converted from C++ to Verilog language and packaged into an IP core using the Xilinx HLS tool. This paper successfully processes the input video signal with the spatial resolution of 1920×1080 pixels and displays the extracted centerline of light stripe on the HDMI display at the frame rate of 30 fps. The test results show that the system can meet the real-time requirement for light stripe centerline extraction in the engineering application.
KEYWORDS: Point clouds, Tunable filters, 3D vision, 3D modeling, 3D image processing, Optical filters, 3D acquisition, 3D scanning, Binocular vision, Structured light
The paper focuses on the emerging three-dimensional (3D) vision detection technology using in pavement crack detection, and takes 3D data acquisition, data preprocessing, crack feature extraction and identification as the main lines of idea, then provides a systematic review of 3D vision detection technology development for pavement cracks. Firstly, three commonly used 3D vision imaging techniques for pavement crack detection are analyzed, including measurement principle, measurement steps and technical characteristics. Secondly, 3D point cloud filtering methods are classified and compared. It lays a foundation for subsequent crack feature extraction and crack identification. Finally, the pavement crack detection method based on point cloud data is systematically sorted out. This paper provides reference for relevant researchers and application personnel to comprehensively and systematically understand the 3D vision detection technology of pavement cracks.
The ice cream stick is a kind of wood product with a plane surface. For the qualified stick it has the nearly flat surface. However, due to slender and thin shape the bend defect is prone to occur during machining of ice cream stick. At present, manual technique is often used to identify the bend defect in the industrial fields, which holds many problems such as low efficiency, low automation and unreliable detection result. In the paper, the PCA-based bend feature selection method is proposed which is applied in defect detection of ice cream stick. At first, the 3D data of object surface is obtained based on the three-dimensional (3D) measurement principle of line structured light. Then four kinds of bend features are designed according to the shape characteristics of the light stripe, which include variation coefficient, correlation coefficient, determination coefficient and straightness metric. At last, the bend feature selection is operated based on the PCA-based method. The research results provide valuable reference for the engineering application in the intelligent defect detection.
Starlight navigation is a navigation technology with strong anti-interference ability, which can output aircraft attitude, azimuth and other information. The star sensors not only have the highest measurement accuracy, but also have no time drift error, so that they become the most widely used attitude sensors in the aerospace field. The star sensor shall be calibrated before being put into use to ensure the accuracy of the star sensor. When the star simulator is used to calibrate the magnitude of the star sensor, the mismatch between the spectrum of the star simulator and the simulated star spectrum will produce a large magnitude simulation error. Therefore, a spectral tunable light source based on pulse width modulation is studied and designed in this paper, which can simulate the stellar radiation at different color temperatures, in order to solve the high-precision magnitude calibration of star sensor under different color temperatures. The spectrum tunable light source based on PWM control can realize 3000K ~ 20000k spectrum modulation within 360 ~ 900nm, the output optical power is stable, the error of spectrum simulation is better than ±0.1, and the uniformity of light source is better than 97%. It can be used for high-precision radiation calibration of star sensor with different color temperature. The problem of spectrum mismatch between star sensor and star simulator is solved, and the error of magnitude simulation is reduced.
In order to measure the local phase modulation characteristic of a phase-only reflective Liquid crystal spatial light modulator (LCSLM) and accurately evaluate the nonuniformity of phase modulation, a Twyman-Green interferometer is built to simultaneously measure the local phase modulation of multi-region in LCSLM. A series of carrier fringe patterns can be captured, which is modulated by the local control pattern loaded to the LCSLM. The phase distribution from all fringe patterns are extracted respectively with the Fourier transform method. By calculating the variance values of all phase values in each pixel position, a phase variance matrix is obtained. Then the regions of interests (ROIs) from the fringe pattern are extracted with the Otsu segmentation algorithm and the connected region labeling algorithm. At last, the phase modulation of local regions is obtained synchronously with the phase-shift algorithm based on Fourier transform method. The proposed method can greatly improve the measuring efficiency and reduce the environmental impact to great degree.
A method based on the local gradient direction information is proposed to locate the center of circular fringe. The new method is developed which based on the idea that the normal directions of any point on the circular fringe are always pointing towards the center. Besides, the local gradient direction of fringe is related to the normal line of fringe. We deduce theoretically the principle of local gradient direction estimation. Then a new circular fringe center location algorithm is developed with the help of digital image processing technology and statistical ideas. The simulation results show that new method can locate accurately the center of different types of circular fringes with less or without filtering. It also holds good robustness and can meet the requirements of actual engineering application.
In the 3D phase measurement with large view field, when the number of fringes is not too many, the period broadening problem of projection fringes will seriously affect the accuracy of measurement. In this paper, an accurate and convenient 3D shape measurement method based on phase shifting fringe projection is prop osed. Firstly, in the fringe projection measurement system based on the triangulation principle, the fringe position coordinates are taken as the input and output parameters of the system, and the linear mathematical model of fringe period correction is de rived. Secondly, the model parameters are obtained by simple calibration process. Through using the idea of reverse fringe projection, the new fringe to be projected is calculated from the correction model, and then periodic four-step phase shifting projection fringes can be produced. Finally, a four-step phase shifting method is used to restore the 3D shape of object. The experiments of fringe period correction and 3D profile measurement show that the proposed method can easily generate the phase-shifting projection fringes with equal period distribution and hence improve the measurement accuracy of phase-shifting method.
In the accuracy measurement of phase from interferometers with adjustable fringe contrast, it needs to estimate the contrast of experimental patterns so as to obtain the interference patterns with the maximum contrast. We develop the Fourier-polar transform and combine the directional projection to estimate the global contrast of carrier fringe pattern. The technique is especially used for low-quality fringe pattern such as low contrast and low signal to noise ratio (SNR) that often appear in the interferometric experiment. An illustrative experiment based on the radial shearing interferometer is given. Results generated from this technique are compared with the derived values from theoretical model, and exemplary agreement between both is demonstrated.
A global direction metric method based on Fourier-polar transform is proposed in this paper, which calculates the global fringe direction according to the directional distribution of intensity from the power spectrum of fringe pattern. By introducing polar coordinate transform, the rotation of power spectrum is transformed into translation component, which can make the calculation process simple and fast. Then the original image is projected along the global fringe direction and the mean value of pixel gray is calculated. Also, the fringe pitch can be calculated from the projection curve close to cosine distribution. This detection method of optical fringe parameters uses overall information of the image, and holds good adaptability and robustness to noise and degraded image. Moreover, without any pre-processing operations such as smoothing filter and threshold segmentation is required in this method. It can directly detect two parameters of global fringe direction and fringe spacing, which is of great significance for quantitative analysis of fringe image.
Optical fringe is one of important output information from the optical systems. Some important optical or system parameters can be obtained by analyzing the fringe information from optical system such as interferometer system or diffraction setup. The straight fringe is a kind of optical fringes frequently appearing in Young’s double slit interference and single-slit diffraction and other optical structures. For the information extraction of straight fringes, it is often necessary to calculate the fringe spacing parameters. Popular straight fringes analysis methods include the fringe center method and the Fourier transform method. In addition, some image processing methods realized line detection can also be used to analyze this straight fringes image, which include Hough transform and Radon transform. In this paper, four algorithms for fringe analysis are discussed and compared, which focus on method’s principle, algorithm’s simulation and performance when they be applied to detect the fringes spacing. At the same time, the anti-noise performance of two image processing algorithms including Hough transform and Radon transform are analyzed.
In visual detection fields based on line-structured light, the analysis of optical stripe image is a key problem. For the cross-line target image, through measuring the angle between two linear optical stripes the target position or some system’s parameters can be obtained. The traditional technique usually needs many preprocessing steps including image filtering, threshold segmentation, thinning processing and so on. For the images with low signal noise ratio or non-uniform intensity distribution, their application performance will be challenged. Based on the characteristic of translation invariance and rotation synchronization of two-dimensional Fourier transform, the paper combines Fourier transform with polar transform to form new Fourier-polar transform algorithm. It implements the angle measurement in the frequency-domain replaced in the spatial domain. At the same time, to improve the convenient of compute, the polar transform is adopted to calculate the distribution direction of amplitude spectrum energy. The proposed Fourier-polar transform algorithm uses the overall information of the image, and the calculating process is simple and no requirement of image preprocessing. Therefore, it can be applied to measure the angle of cross-line target image in low quality image such as low signal-to-noise ratio or with noise.
A surface roughness measurement method is introduced in the paper, which is based on laser triangulation and digital image processing technique. In the measuring system, we use the line-structured light as light source, microscope lens and high-accuracy CCD sensor as displacement sensor as well. In addition, the working angle corresponding to the optimal sensitivity is considered in the optical structure design to improve the measuring accuracy. Through necessary image processing operation for the light strip image, such as center-line extraction with the barycenter algorithm, Gaussian filtering, the value of roughness is calculated. A standard planing surface is measured experimentally with the proposed method and the stylus method (Mitutoyo SJ-410) respectively. The profilograms of surface appearance are greatly similar in the shape and the amplitude to two methods. Also, the roughness statistics values are close. The results indicate that the laser triangulation with the line-structured light can be applied to measure the surface roughness with the advantages of rapid measurement and visualized display of surface roughness profile.
Pyramid wavefront sensor (PWFS) without modulation is prevailing over one with modulation. So far how to describe
measured signals of non-modulation PWFS needs deeply research. In this paper, the theory of the non-modulation PWFS
is briefly presented according to wave optics. This paper analyses the existing four approaches in theory. By numerical
simulation this paper further verifies the performance of four approaches under the experiment condition. The result
shows that the approach with total intensity of pixels conjugate to the same spot in the pupil as signal denominator is the
best choice for the non-modulation PWFS in closed-loop correction.
The paper firstly discusses the derivation of interference fringes and FFT method that is a conventional method for measuring fringes with equal spacing. And then, a new method for measuring the interference line fringes is presented. On the basis of Hough transform and digital image processing technology, this method accomplishes fringes image preprocessing (i.e., fringes skeletonizing) and fringes spacing measurement. It has been shown, with both theoretical analysis and experimental measurements, that this method is a universal method for fringes measurement and has more remarkable advantage than the FFT method because of the application of the former not being limited by some regulations such as equidistant fringes and sampling length, and so on. In addition, it also holds a good noise rejection capability and great reliability.
To overcome the difficulty in the non-contact angle measurement especially for large-scaled workpieces, the paper presents a new angle measurement method based on Hough transform. The realization process of the method consists of workpiece image acquisition & preprocessing and angle measurement. With the preprocessing one can obtain an ideal binarized edge image of the workpiece image. The angle measurement is then carried out by employing the classical Hough transform. Except the discussion of measurement principle, both practical examples and simulated experiments are introduced to demonstrate the performance of the method proposed in the paper. The experiment results indicate that this method can be applied to measure the workpiece angle under many occasions and can achieve the aim of the non-contact angle measurement. At the same time, it does still possess great advantages in terms of reliability and noise rejection capability.
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.