Mean Intensity Gradient Research Articles

Naturalization and evaluation of synthetic random stripes pattern for digital image correlation

Abstract The morphology of speckle patterns commonly used in the Digital Image Correlation (DIC) community can be categorized into artificially generated free shapes and computer-synthesized circular dots. Due to the non-repeatability of manually sprayed patterns, researchers tend to favour code-controlled dot patterns, which exist in two forms: “white dots on a black background” and “black dots on a white background”. However, from the perspective of biological evolution, these two dot patterns can be hybridized “in silico” using the Turing model to create an intermediate form—the “striped-like” pattern. This hybridization significantly improves species identification and environmental adaptability. Although there have been reports on producing stripe patterns based on the kernel-based Turing model, there is no specific speckle pattern generation specification or comprehensive evaluation research for DIC applications. This paper naturalizes a novel striped speckle pattern and a corresponding generation approach. The pattern quality was assessed and compared with circular dots and hand-sprayed speckles. The stripes pattern outperformed the other two forms regarding Mean Intensity Gradient (MIG), q-factor, systematic bias, and random error. Sub-pixel displacement simulation and actual translation test results demonstrate that the proposed striped pattern offers higher precision and robustness in displacement and static strain measurements. Therefore, this striped pattern provides a preferred alternative for DIC technicians.

Towards an image quality criterion to optimize Digital image correlation. Use of an analytical model to optimize acquisition conditions

Digital image correlation (DIC) is expanding in many fields. In particular, it requires that the acquisition conditions and speckle be adapted to each application. A new methodology allowing to define the quality of the images in a fast and precise way is proposed. An analytical model based on mean intensity gradient of speckle pattern criterion and the standard deviation of displacement on images has been proposed. The model can be used on the whole image or by creating subsets in the image. Standard deviation maps of displacement are obtained. The methodology makes it possible to save a considerable amount of time compared to the full DIC calculation to optimize the image acquisition conditions, the speckle and the DIC parameters.

A New Pattern Quality Assessment Criterion and Defocusing Degree Determination of Laser Speckle Correlation Method

The laser speckle correlation method has found widespread application for obtaining information from vibrating objects. However, the resolution and accuracy of the laser speckle correlation method as they relate to the defocusing degree have not been analyzed sufficiently. Furthermore, the possible methods for speckle pattern quality assessment and enhancement have not been studied. In this study, the resolution and accuracy of the laser speckle correlation method are analyzed, and it is found that they are affected by the defocusing degree and speckle pattern quality, respectively. A new speckle pattern quality criterion combining the mean intensity gradient and frequency spectrum was proposed, called CMZ. The quality of the speckle pattern is higher when the CMZ is closer to zero. The proposed criterion was verified by simulated speckle patterns and real speckle patterns with different speckle sizes, densities, and gray contrasts. In the experimental setup stage, a suitable defocusing degree can be selected based on the resolution requirement and optimal speckle size, and other experimental parameters can be determined according to the CMZ criterion. Rotation and vibration experiments verified the effectiveness of the laser speckle correlation method and confirmed the reliability of the experiment preparation based on proposed CMZ criterion.

Speckle pattern creation methods for two‐dimensional digital image correlation strain measurements applied to mechanical tensile tests up to 700°C

AbstractThe purpose of this study is to develop novel speckle pattern techniques for digital image correlation (DIC) kinematic measurements of mechanical tests at high temperatures, typically from 400 to 700°C. In this context, the speckle pattern should not only meet morphological criteria (size, density, distance) in order to improve spatial resolution, but it should also present a high contrast and resist high temperature and strain levels. To find a speckle pattern matching these specifications, a comparison was performed on six types of speckle made using different techniques. First, a computer‐generated speckle pattern that meets DIC criteria was numerically designed to produce six types of speckle pattern. Next, the speckle patterns produced using these six techniques were compared in terms of speckle morphology, image quality and adherence to titanium alloy TA6V material at high temperatures. From 25 to 600°C, the speckle pattern made by the technique combining anodisation and laser engraving named M5 technique gave the best contrast (highest value of mean intensity gradient [MIG] and Shannon entropy value) and the adherence of 200% of strain measurements to the TA6V material. At 700°C, speckle image quality is considerably reduced due to oxidation of the titanium alloy, and this may not be suitable for DIC measurements. Only the speckles produced by painting in which the paint plays a protective role provide with a better speckle contrast compared with other techniques. However, these speckle patterns enable only a strain measurement of 22% by the DIC method. This article concludes with guidelines for producing a speckle pattern suitable for high‐temperature mechanical tests.

Fractal Pattern for Multiscale Digital Image Correlation

Digital Image Correlation (DIC) is based on the matching, between reference and deformed state images, of features contained in patterns that are deposited on test sample surfaces. These features are often suitable for a single scale, and there is a current lack of multiscale patterns capable of providing reliable displacement measurements over a wide range of scales. Here, we aim to demonstrate that a pattern based on a fractal (self-affine) surface would make a suitable pattern for multiscale DIC. A method to numerically generate patterns directly from a desired auto-correlation function is introduced. It is then enhanced by a Mean Intensity Gradient (MIG) improvement process based on grey level redistribution. Numerical experiments at multiple scales are performed for two different imposed displacement fields and results for one of the patterns generated are compared with those obtained for a random pattern and a Perlin noise one. The proposed pattern is shown to lead to DIC errors comparable to those found with the two others for the first scales, but has much greater robustness. More importantly, the pattern generated here exhibits stable errors and robustness with respect to the scale whereas these two outputs become significantly degraded for the other two patterns as the scale increases. As a result, scale invariance properties of the pattern based on fractal surfaces correspond to scale invariance in DIC errors as well. This is of great interest regarding the use of such patterns in multiscale DIC.

Assessment of speckle pattern quality in digital image correlation from the perspective of mean bias error

Digital image correlation (DIC) technology is severely affected by the speckle pattern quality. Most of the existing assessment methods always use only one parameter to evaluate both mean bias error and standard deviation error, such as the mean intensity gradient (MIG). However, the principles of these two error models are quite different. The mean bias error is closely related to the first-order and second-order gradients of speckle pattern intensity. A parameter named Ef, based on the mean bias error, is proposed to evaluate the speckle pattern quality in this work. Numerical translations are applied to eight different speckle patterns to justify its correctness. The results indicate that the existing MIG is efficient to assess speckle pattern quality by evaluating the standard deviation error, while the parameter Ef is efficient to assess the speckle pattern quality from the perspective of mean bias error.

Quality assessment of laser speckle patterns for digital image correlation by a Multi-Factor Fusion Index

Digital image correlation based on laser speckle has its unique advantages in ultra-high temperature deformation measurement. The traditional quality assessment criteria of artificial speckle which consider only the contrast of the speckle pattern or the morphology of the speckle particles are not suitable for laser speckle. Here, a new index, called Multi-Factor Fusion Index (MFFI), which takes the inhomogeneity of gray distribution, the mean square deviation of gray and the standard deviation of speckle particles size into consideration, is proposed to evaluate the quality of laser speckle patterns. Multi-Factor Fusion Index was compared with three commonly used criteria, the mean intensity gradient, the mean intensity of the second derivative, and the pattern quality metric combining the sum of square of subset intensity gradients (SSSIG) and the secondary auto-correlation peak height. The results showed that the proposed index had better capability of assessing the quality of laser speckle patterns. And the influence by external factors such as apertures, laser powers and temperatures and material factors such as roughness and light reflectance were analyzed by MFFI. Experiments on artificial speckle showed that MFFI was also suitable for artificial speckle patterns in most cases.

Study of the quality of wood texture patterns in digital image correlation

Wood texture (WT) is sometimes used as speckle patterns in digital image correlation (DIC) method. However, the various WT due to orthotropic behavior of wood cause it vary difficult to assess the quality of WT and the accuracy of DIC analysis. In this paper, the mean directional intensity derivative (MID) was used to qualify the quality of the WT patterns for digital image correlation. To verify the effectiveness and correctness of this parameter, numerical displacement experiments in x and y directions were performed. The results show that the measurement errors of WT patterns in x and y directions are related to the mean directional intensity derivatives. The WT pattern with a lower MID in x or y direction has a larger measurement error when the mean intensity gradient are nearly equal to each other, and a higher values of the mean directional intensity derivative in one direction leads to the standard deviation error decrease. Furthermore, to produce equivalent measurement errors in the two directions, an elliptic (or rectangular) subset whose ratio of the radius between x and y directions should be approximately equal to the inverse ratio between MIDx and MIDy may be more appropriate for WT patterns.

Optimum Paint Sequence for Speckle Patterns in Digital Image Correlation

This note answers whether it is better to use black-on-white or white-on-black painted speckle patterns, and provides recommendations for optimum painted patterns for digital image correlation (DIC). Although DIC algorithms have no preference between tracking the patterns of either bright features or dark features on contrasting backgrounds, we show that paint sequence can be important due to fundamental differences in paint pigments. If the sample and experiment conditions call for two painting steps (basecoat and speckles), then applying a basecoat of white paint followed by black speckles is better than the converse. Black speckles are preferred because they increase contrast and provide a higher mean intensity gradient, lower correlation confidence interval, and lower displacement error. The primary reason for the increased contrast is the greater hiding power of black paint versus white paint. A secondary effect that also reduces contrast in white speckles is undertone, or a slight blue hue shift from Rayleigh scattering.

Development of microscale pattern for digital image correlation up to 1400 °C

Speckle patterns to be used for digital image correlation (DIC) at the micrometer level up to 1400°C were fabricated by several methods. The quality of the patterns before and after heating was evaluated in terms of the mean intensity gradient (MIG) and the speckle size distribution. The displacement accuracy in simulative translation of images showed that the MIG alone was not enough to evaluate the pattern properties; a large MIG, an even speckle size distribution, and a wide speckle size range pattern were required for a good DIC. The reaction between the patterning material and substrate, the cracking of speckles, and the plastic flow of patterning material may cause changes in the pattern morphology at high temperature. Two patterning methods, spraying a mixture of ceramics powder and binder by a fine-nozzle air brush and abrading a polished surface, yielded a small pattern with high MIG values and even size distributions that was stable at 1400°C. The potential of the fabricated patterns was shown by measuring the coefficient of thermal expansion of polycrystalline Al2O3 from 800°C to 1400°C.

Application of the mean intensity of the second derivative in evaluating the speckle patterns in digital image correlation

In this work, the mean intensity of the second derivative of speckle pattern was used to quantify the quality of the speckle patterns for digital image correlation. Numerical experimental studies were performed to justify the correctness and effectiveness of this new global parameter. The results indicate that the measured displacement error was related to the mean intensity of the second derivative of the speckle patterns when they had equal mean intensity gradients. The results also indicate that the measurement accuracy was affected by both the mean intensity gradient and the mean intensity of the second derivative. Therefore, high quality speckle patterns should have a large mean intensity gradient and a small mean intensity of the second derivative.

Fabrication technique of micro/nano-scale speckle patterns with focused ion beam

The fabrication technique of micro/nano-scale speckle patterns with focused ion beam (FIB) system is studied for digital image correlation (DIC) measurement under a scanning electron microscope (SEM). The speckle patterns are fabricated by directly etching the counterpart of the specimen to the black part of a template. Mean intensity gradient is used to evaluate the quality of these SEM images of speckle patterns fabricated based on different templates to select an optimum template. The pattern size depending on the displacement measurement sensitivity is adjusted by altering the magnification of FIB according to the relation curve of the etching size versus magnification. The influencing factors including etching time and ion beam current are discussed. Rigid body translation tests and rotation tests are carried out under SEM to verify the reliability of the fabricated speckle patterns. The calculated values are in good agreement with the imposed ones.

Fabrication of micro-scale speckle pattern and its applications for deformation measurement

A novel and effective method of fabricating a micro-scale speckle pattern applied to metallic and polymeric materials is described. It can be used for measuring deformation under an optical microscope or other high magnification imaging systems. In this paper, the mean intensity gradient (MIG) is used to evaluate the quality of speckle pattern under different fabrication conditions, such as proportion of compound, spinning rates and centrifugal solidifying time, and the optimal fabrication parameters are proposed. Finally, deformation of a PET thin film and interface of an optical fiber device are measured with fabricated micro-scale speckle pattern. Further, it is proved that the speckle pattern produced by the novel method can be used with digital image correlation to measure deformation at micro-length scale.

Recent Progress in Digital Image Correlation

In this paper, we report the following important progress recently made in the basic theory and implementation of digital image correlation (DIC) for deformation and shape measurement. First, we answer a basic but confusing question to the users of DIC: what is a good speckle pattern for DIC? We present a simple local parameter, called the sum of squared subset intensity gradient, and an easy-to-compute yet effective global parameter, called mean intensity gradient, for quality assessment of the local speckle pattern within each subset and entire speckle pattern, respectively. Second, we provide an overview of various correlation criteria used in DIC for evaluating the similarity of the reference and deformed subsets, and demonstrate the equivalence of three robust and mostly widely used correlation criteria, i.e., a zero-mean normalized cross-correlation (ZNCC) criterion, a zero-mean normalized sum of squared difference (ZNSSD) criterion and a parametric zero-mean normalized sum of squared difference (PZNSSD) criterion with two additional unknown parameters, which elegantly unifies these correlation criteria for pattern matching. Finally, to overcome the limitation of the existing DIC techniques, we introduce a robust and generally applicable reliability-guided DIC technique, in which the calculation path is guided by the ZNCC coefficients of computed points, to determine the genuine full-field deformation or shape of objects containing geometrical discontinuities and discontinuous deformation.

Mean intensity gradient: An effective global parameter for quality assessment of the speckle patterns used in digital image correlation

Digital image correlation (DIC) is an image-based optical metrology for full-field deformation measurement. In DIC technique, the test object surface must be covered with a random speckle pattern, which deforms together with the object surface as a carrier of deformation information. In practice, the speckle patterns may show distinctly different intensity distribution characteristics and have an important influence on DIC measurements. How to assess the overall quality of different speckle patterns with a simple yet effective parameter is an interesting but confusing problem, and is also helpful to the optimal use of the technique. In this paper, a novel, simple, easy-to-calculate yet effective global parameter, called mean intensity gradient, is proposed for quality assessment of the speckle patterns used in DIC. To verify the correctness and effectiveness of the new concept, five different speckle patterns are numerically translated, and the displacements measured with DIC are compared with the exact ones. The errors are evaluated in terms of mean bias error and standard deviation error. It is shown that both mean bias error and standard deviation of the measured displacement are closely related to the mean intensity gradient of the speckle pattern used, and a so-called good speckle pattern should be of large mean intensity gradient.