A Comparison of 3 Digital Image Correlation Techniques on Necessarily Suboptimal Random Patterns Recorded By X-Ray

Abstract

Dynamic x-rays have been used to follow the deformation ahead of a steel ball fired into a mock-up of a generic cylindrical rocket motor. The impact was arranged to intersect a sparse lead powder layer within the mock explosive that created a random speckle pattern on x-ray film. Three different digital image correlation programs are compared to examine any sensitivity to the sub-optimal speckle pattern produced by the lead powder. An identical output data reduction method was used in all cases to aid comparison. All three correlation methods were able analyze the deformation, but all had intricacies that would require more detailed optimization of the data reduction in order to fully exploit the technique. Quantitative analysis showed that the three methods agreed closely in estimations of rigid body displacements between a pair of representative x-ray images. It was discovered that the deformation caused by the ball impact was highly localized and the useful data available about the deformation pattern was sparse. This limits the applicability of this technique to this specific application. Extensive cracking was not observed that would have aided the development of computer-based models for prediction of such impact events. The x-ray technique was however excellent for determining the ball position as a function of time after impact.

Theoretical evaluation of sensitivity and thermal stability for high explosives based on quantum chemistry methods: A brief review

“…Such a study on explosives could provide considerable insight into understanding the factors that affect their behavior. [3][4][5][6][7][8][9] In the past few decades, many efforts have been made to improve the survivability and safety of munitions, during which a new class of insensitive munitions (IM) explosives has been developed. However, these IM formulations differ somewhat from each other in a variety of aspects especially in sensitivity and stability.…”

Visual Methods to Assess Strain Fields in Armour Materials Subjected to Dynamic Deformation—A Review

“…Advances in X-ray technology have allowed for 3D DIC to be extended to tracking the motion within objects, by seeding them with particles of significantly higher density. One example of this is a study conducted by Rae et al [20], where the authors examined the displacement ahead of a spherical projectile being fired at a mock (non-reactive) motor casing. The mock motor the authors used was made of 6061 aluminium tubing, a Sylgard 184 insulation layer and a mock explosive (PBS 9501) on the inside.…”

“…Due to a lack of easy optimization of the speckle size and contrast, the authors decided to compare three different correlation techniques on the X-ray images. The three correlation methods were: (i) the Vic-2D (Correlated Solutions, Inc., Irmo, SC, USA) commercial software package, (ii) making use of a fast Fourier transform to change the data to the frequency domain, with sophisticated sub-pixel image shifting to optimize the overlap between sub-images and (iii) also making use of a fast Fourier transform to perform the correlation, but utilizing a simple bi-cubic fit to find the maximum correlation peak, to find the average displacement between sub-images [20]. Rae et al found that all three correlation methods produced some valid data from the sub-optimal speckle patterns around the projectile in the case of the solid sample, but method (ii) produced poor results around the impact region for the case with the 'burn' hole [20].…”

Flow Measurements in a Blood-Perfused Collagen Vessel Using X-Ray Micro-Particle Image Velocimetry

“…A significant limitation of phase-contrast x-ray imaging is that it provides extremely low signal to noise ratio images that are generally inadequate for PIV cross-correlation. Several groups have developed complex techniques to improve the quality of x-ray PIV, including optimization of experimental conditions [39] , [40] , alternative image processing algorithms [41] , and application of various correlation algorithms [38] , [42] . However, the techniques previously published are application-specific.…”