No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Abstract
This paper studied the influence law of image quality caused by different illumination methods and lighting intensity, proved the internal relationship between image quality and measuring accuracy. In order to control the lighting source automatically and ensure the measuring accuracy, the paper modeled the image quality assessment system under different illumination conditions and proposed a method to evaluate the image quality. That is, under the back light illumination, using the image contrast as the criterion while under the front light illumination, using the low frequency ratio as the criterion to evaluate the image quality. Through adjusting the contrast and low frequency ratio, the image quality can be controlled and the measuring accuracy can be satisfied.
... With the development of image sensing technology and image process algorithm, measurement based on image process has been more and more widely used. Computer vision, phototypography, stripe projection measurement are different application area based on image process to measure objet dimension size or displacement [2][3][4]. Digital image correlation method is the technology by using computer to process digital image and with the help of digital image correlation processing to get the characteristics as we need. ...
... The most common forms of definition of correction functions include standardize correlation function, standardized covariance correlation function and least square distance correlation function. In this paper, we choose least square distance correlation function shown as formula (2). ...
Strain measurement is an important means to study structural strength. Strain gauge measurement method is widely used in different industry area. The strain gauge should be stuck on the surface of the measured object to detect displacement or strain. As the measurement mode is contact measurement, strain gauge can have a certain impact on the measured object itself. So research of strain detection is increasingly moving from contact method to non-contact method. In this paper, we design a digital image correlation non-contact method used for strain detection. The basic principle of image correlation method is introduced. And the image process algorithm is detailed described. A set of laser lighting hardware is especially used for object illumination to get a random distribution light spot instead of sticking target bar on object. The measurement results show that this method can detect strain with good effect and can be applied to non-contact strain detection test.
... As a result, illumination quality can directly impact image analysis results [11]. There are multiple methods for illumination system design; as an example of an optical path-based analysis, Lu [12] determined three color ring LED light source ray angles and light source geometric parameters, which were used for detecting circuit board weld defects; as a method based on optimization techniques, Sunil [13] studied the energy function of an optimum light source position, and calculated the minimum energy function and estimated light source position by a simulated annealing algorithm; for design based on ray directions, Nicolas [14] designed a light source with the same shape as a shower head, which enhances defect contrast and detects parts surface defects with random direction scratches; as an example of design based on dynamic illumination, Ng [15] designed a moving ring light source and judged surface defects by bright ring changes on a slippery surface for the purpose of defect detection of bearing surfaces, and in an example of designs based on imaging quality analysis, Wu [16] associated image quality with measurement precision, and adjusted light illuminance by analyzing image quality to get highest measurement precision. In a word, one needs to study the light source optimization choice and design method combination for different detection tasks and work environments. ...
Because of the perceived lack of systematic analysis in illumination system design processes and a lack of criteria for design methods in vision detection a method for the design of a task-oriented illumination system is proposed. After detecting the micro-defects of a gyroscope pivot bearing with a high curvature glabrous surface and analyzing the characteristics of the surface detection and reflection model, a complex illumination system with coaxial and ring lights is proposed. The illumination system is then optimized based on the analysis of illuminance uniformity of target regions by simulation and grey scale uniformity and articulation that are calculated from grey imagery. Currently, in order to apply the Pulse Coupled Neural Network (PCNN) method, structural parameters must be tested and adjusted repeatedly. Therefore, this paper proposes the use of a particle swarm optimization (PSO) algorithm, in which the maximum between cluster variance rules is used as fitness function with a linearily reduced inertia factor. This algorithm is used to adaptively set PCNN connection coefficients and dynamic threshold, which avoids algorithmic precocity and local oscillations. The proposed method is used for pivot bearing defect image processing. The segmentation results of the maximum entropy and minimum error method and the one described in this paper are compared using buffer region matching, and the experimental results show that the method of this paper is effective.
There are some weaknesses for traditional microscope measurement, including small scope field, complex operation and subjective errors. To avoid these issues, a measurement system combining video and microscope is made. With the magnification of microscope and video, the accuracy of grating micrometer and the real-time notification of chamfer location, this system achieves high precise measurement. By displaying with two monitors, the computer can collect and display other measurement data while the chamfer is measured, which makes full use of the display resolution and enhances the efficiency of instruments. The method is valid for measuring size of special work-pieces.
In order to always obtain the optimal images of parts in the assembly process, the problem of automatic optimization of microscopic vision illumination intensity was investigated. To realize the illumination of parts edges and surface detail, the illumination automatic control system based on coaxial and ring light source was developed. The illumination optimize performance of four types of image quality evaluation function was compared and analyzed, the image quality evaluation methods based on Roberts gradient sum function was proposed and applied in developed assembly systems. The results show that the process of illumination optimization is fast and accurate, the contrast of the target edge is significantly enhanced and the parts recognition error is eliminated after illumination optimization. Moreover, the precision automatic assembly requirements of miniature parts can be met.
Positioning of forward symmetrically installed square light-emitting diode (LED) arrays are studied to achieve uniform illumination with which high quality images can be gathered for machine vision measurement. The optimization method is used to solve this problem because the optical axes of the arrays are not perpendicular to the irradiated plane, which can't be solved by analytic method. The variance is chosen to establish the mathematical model of goal function on the base of the structure characteristics of LEDs. The simulated annealing (SA) algorithm is chosen for solving the non-convex objective function. An experiment system is developed for illumination photometry. The sizes of area with 95% of uniformity of the analytic method are a little larger than those of the optimization based method. The maximum difference is within 4%. The results show that the optimization objective is reasonable and the method is feasible.
Specialized lighting is required in the majority of machine vision systems for quality inspection of various products. Machine vision systems require a crisp, clear image of the inspected object to which their complex algorithms and routines are applied. The angle of lighting in an application is important as light reacts differently with different materials. The lighting should help to create as much uniform contrast between the display background and the displayed characters while minimizing or eliminating any reflections. Backlighting is another type of specialized lighting used to illuminate a part's overall outline to minimize visibility of clear or translucent housings. Square Continuous Diffuse Illuminator is a diffused type of lighting in which the light source is sent through a diffuser, which softens the impact of the light and then a beam-splitter spreads the light uniformly throughout the field.
Non-uniform illumination is one of the key points that affects the result of face recognition, and noise is the common issue in the image processing problems. A face recognition under non-uniform illumination and noise was addressed and the scheme that combines Gabor feature, Enhanced FLD Model (EFM) and Maximum A Posteriori (MAP) was proposed. The experimental result shows that this method is superior to other traditional method and robust under non-uniform illumination and noise.
The selection of a suitable illumination subsystem is seldom practicable in the automated visual inspection of highly reflective surfaces. The paper presents an algorithmical approach in the form of a framework for enhancing images of objects with such surfaces. This framework is based on the application of so-called Intelligent Localized Fusion Operators (ILFOs), whose formalization is herein undertaken for the first time. Furthermore the guidelines for its implementation are given and different aspects of the resulting pre-processing system are systematically analyzed. The framework successfully performs in the automated visual inspection of different objects presenting highly reflective surfaces, namely headlamp reflectors, plastic bundled packages, and electric bulbs.
An optimization approach to automatic sensor and light source positioning for a machine vision task, where geometric measurement and/or object verification is important, is discussed. The goal of the vision task is assumed to be specified in terms of measurements related to edges. The optimal sensor and light source positions are defined in such a way that when the sensor and light source are placed in the optimal positions, we can obtain a picture which produces the minimum variance for the required measurement. Experiments show that the uncertainty in the edge point position is inversely proportional to the contrast across the edge. Using a variant of the Torrance-Sparrow model that takes into account the polarization of the light, the contrasts across edges are computed and used to estimate the variance of the required 2D measurement. An optimization procedure employing mathematical programming techniques uses this information to determine the best positions for the light source and sensor in order to perform the required measurement. A series of experiments was conducted to demonstrate the feasibility of our optimization approach. The optimal positions computed by the program were found to be the best ones in the real experiments. Furthermore, the correlation coefficient between the expected variance and the variance computed from the real pictures was 0.768.
The image reference approach is very popular in industrial inspection due to its generality for different inspection tasks. Unfortunately, this approach is sensitive to illumination variations. A novel illumination compensation algorithm is proposed in this paper for correcting smooth intensity variations due to illumination changes. By using the proposed algorithm as a preprocessing step in the image reference based inspection or localization, we can make the image inspection or localization algorithm robust against spatially smooth illumination changes. This technique is very useful to achieve a reliable automated visual inspection system under different illumination conditions. The proposed illumination compensation algorithm is based on the assumption that the underlining image reflectance function is approximately piecewise constant and the image irradiance function is spatially smooth. Reliable gradient constraints on the smooth irradiance function are computed and selected from the image brightness function by using a local uniformity test. Two surface fitting algorithms are presented to recover the smooth image irradiance function from the selected reliable gradient constraints. One is a polynomial surface fitting algorithm and the other is a spline surface fitting algorithm. The spline surface fitting formulation leads to solving a large linear system, which is accomplished by an efficient preconditioned conjugate gradient algorithm. Once the image irradiance function is estimated, the spatial intensity inhomogeneities can be easily compensated. Some experimental results are shown to demonstrate the usefulness of the proposed algorithm.