The inspection of particles plays an important role in quality control of smooth surfaces, such as optics and silicon wafers. The difficulty of inspection lies not only in the high-resolution detectivity for small-size particles, but also in the need to distinguish among different types of particles due to their different damages and re-work processes. This paper proposes a detection and discrimination method of the particles on (surface particles) and below smooth surfaces (subsurface particles) by laser scattering with polarization measurement. The incident laser beam passes through a polarization state generator (PSG) to illuminate the smooth surface. A large proportion of the particle-induced scattered light is detected after collected by an ellipsoidal mirror. Meanwhile, in the blind area of the ellipsoidal mirror, a small amount of scattered light enters the polarization measurement channel, where a polarization state analyzer (PSA) is used to finally obtain a two-dimensional polarization signal for discriminating the two types of particles. Then, by a nonlinear global search combined with the theoretical models for polarized light scattering from particles, the optimal polarization measurement state of the PSG and the PSA is obtained to maximize the separability between the surface particles and subsurface particles on the acquired polarization signals. This method takes full advantage of the polarization of scattered light, and with one single scan the inspection for the whole surface can be accomplished. Finally, experimental results demonstrate the effectiveness and accuracy of the method.
Aimed at the problem of strong background interference introduced in digital image processing from complex surfaces under industrial defect detection, a method for complex surface defect detection based on human visual characteristics and feature extracting is proposed. Inspired by the visual attention mechanism, defect areas can be identified from the background noise conveniently by human eyes. We introduce the improved grayscale adjustment and frequency-tuned saliency algorithm combined with the salient region mask obtained by dilation and differential operation to eliminate the background noise and extract defect areas. Meanwhile the directional feature matching and merging algorithm is applied to enhance directional features and retain details of defects. Testing images are captured by our established detecting system. Experimental results show that our method can retain defect information completely and achieve considerable extracting efficiency and detecting accuracy.
In inertial confinement fusion system, the intermittent scratches on the polished surface of single-sided polished and bottom surface frosted optical components are complex, and it’s of great difficulty to extract them completely. In order to solve this problem, established in the light-field surface detection system, this paper brings forward a novel intermittent scratch detection method based on adaptive sector scanning algorithm (ASC) cascading mean variance threshold algorithm (MVTH). In the preprocessing step, dividing the original image into subimages with a number of integer multiple of cpu cores so as to fully compress image processing time utilizing parallel processing, using mean filter to balance background and then obtaining binary subimages utilizing morphology and threshold operations, finally, utilizing Two-pass algorithm to label the connected domains of binary subimages. In the detection step, considering the complexity of the pattern of intermittent scratches, ASC is first used for routine intermittent scratches stitching and then supplemented by MVTH. In the verification step, in order to prove that the detected intermittent scratches satisfy the criteria for scratches in human eyes, the method of support vector machine (SVM) pattern recognition is utilized to compare the detected results with the continuous scratch samples detected by human eyes. This algorithm has high degree of parallelism, high speed and strong robustness. The experimental results illustrate that the complete extraction rate of intermittent scratches is 93.59% , the average processing time of single image is merely 0.029 second and the accuracy rate of detection is up to 98.72% by SVM verification.
This paper introduces a spherical optical surface defects evaluation system (SSDES) based on the dark-field microscopic scattering imaging (DFMSI) method. The specially designed annular illuminant with variable aperture angles ensures the condition of DFMSI for spherical optical components with variable surface shapes and radii of curvature. On account of the small imaging field of view (FOV) of the SSDES relative to the large spherical optical component under test, the scanning path for subaperture images is planned along longitudes and latitudes of the spherical surface to detect the whole surface. Besides, for avoiding the misplaced subaperture images stitching due to the decenter error, a centering system is utilized to perform the alignment of the optical axis of the spherical optics in relation to the reference axis before capturing subaperture images. Then we propose a defect evaluation method, primarily involving the threedimensional (3D) image reconstruction and global coordinate transformation, the projective stitching of 3D subaperture images, and the quantitative evaluation of defects, to process the captured spherical subaperture images. Experiments results are shown in good accordance with the OLYMPUS microscope for the relative error within 5%, and validate the SSDES to the micrometer resolution.
Surface defect detection is one of the essential steps of quality control. However, it is hard to illuminate defects on curved surface, as the surface properties and geometrical shapes lead to uneven background light distribution on the captured image. In this paper, all the devices used by machine vision including the sample are regarded as part of the whole illuminating scene. Models for each component of the illuminating scene are established separately and Monte Carlo ray tracing is used for generating the image. At last, a specific illuminating scene is designed for illuminating a part of curved cylinder surface. the entire surface is lightened and the non-uniform image caused by background light is greatly improved. The paper analyses how the components of illuminating scene, such as light source placement, surface properties, influence the grayscale distribution on image background and provides method to design illuminating scene for real-time accurate curved surface defect detection.
The inspection of surface defects is one of significant sections of optical surface quality evaluation. Based on microscopic scattering dark-field imaging, sub-aperture scanning and stitching, the Surface Defects Evaluating System (SDES) can acquire full-aperture image of defects on optical elements surface and then extract geometric size and position information of defects with image processing such as feature recognization. However, optical distortion existing in the SDES badly affects the inspection precision of surface defects. In this paper, a distortion correction algorithm based on standard lattice pattern is proposed. Feature extraction, polynomial fitting and bilinear interpolation techniques in combination with adjacent sub-aperture stitching are employed to correct the optical distortion of the SDES automatically in high accuracy. Subsequently, in order to digitally evaluate surface defects with American standard by using American military standards MIL-PRF-13830B to judge the surface defects information obtained from the SDES, an American standard-based digital evaluation algorithm is proposed, which mainly includes a judgment method of surface defects concentration. The judgment method establishes weight region for each defect and adopts the method of overlap of weight region to calculate defects concentration. This algorithm takes full advantage of convenience of matrix operations and has merits of low complexity and fast in running, which makes itself suitable very well for highefficiency inspection of surface defects. Finally, various experiments are conducted and the correctness of these algorithms are verified. At present, these algorithms have been used in SDES.
The principle of microscopic scattering dark-field imaging is adopted in surface defects evaluation system (SDES) for large fine optics. However, since defects are of micron or submicron scale, scattering imaging cannot be described simply by geometrical imaging. In this paper, the simulation model of the electromagnetic field in defect scattering imaging is established on the basis of Finite-Difference Time-Domain (FDTD) method to study the scattering imaging properties of rectangular and triangular defects with different sizes by simulation. The criterion board with scribed lines and dots on it is used to carry out experiments scattering imaging and obtain grayscale value distributions of scattering dark-field images of scribed lines. The experiment results are in good agreement with the simulation results. Based on the above analysis, defect width extraction width is preliminary discussed. Findings in this paper could provide theoretical references for defect calibration in optical fabrication and inspection.
For the Spherical Surface Defects Evaluation System (SSDES), lens centering is essential to obtain the precise scanning trace and defect features without mismatch. Based on a combination of auto-collimating microscopy and Computer-Aided Alignment (CAA), an auto-centering system that can measure the deviation of large spherical center with respect to a reference rotation axis rapidly and accurately is established in this paper. The auto-centering system allows the closedloop feedback control of spherical center according to the different images of the crosshair reticle on CCD. Image entropy algorithm is employed to evaluate image clarity determined by the auto-focus experiment of 50μm step-length. Subsequently, an improved algorithm that can search the crosshair center automatically is proposed to make the trajectory of crosshair images and the position of rotation axis more reliable based on original circle fitting algorithm by the least square method (LSM). The comparison results indicates to show the high accuracy and efficiency of the proposed fitting method with LSM.
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.