Image analysis framework with focus evaluation for in situ characterisation of particle size and shape attributes (original) (raw)
Related papers
A simple and fast matlab-based particle size distribution analysis tool
International Journal of Computational Methods and Experimental Measurements, 2021
Particle size distribution is one of the most important physical properties of a particulate sample. Traditional particle-sizing methods to estimate a geometrical particle size distribution employ a sieve analysis (or gradation test), which entails filtering the particles through a series of sieves and measuring the weight remaining on each sieve to estimate the number-weighted particle size distribution. However, these two quantities have the same value only if particles are perfectly spherical and round. On the other hand, a particle sizer such as the Malvern particle size analyzer, which uses laser diagnostics to measure the particle sizes, can be a hefty investment. Alternatively, imaging techniques can be applied to estimate the size of these particles by scaling a reference dimension to the pixel size, which in turn is used to estimate the size of the visible particles. The focus of this work is to present a simple methodology using a DSLR camera and an illuminated LED panel to generate enough contrast. Using the camera and lens properties, the scale, or size, of any image can be obtained based on the mounting distance of the camera with respect to the target. An analysis tool was developed in MATLAB where the images are processed automatically based on the prescribed camera and lens properties embedded within the same image file and requiring the user to only input the mounting distance of the camera. So far, results show a positive agreement when comparing to measurements using ImageJ imaging tools and a sieve analysis. Future tests will analyze different particle sizes and types, as well as using a Malvern particle size analyzer to corroborate the results.
2021
The two objectives of this paper were to demonstrate use the of the discrete element method for generating synthetic images of spherical particle configurations, and to compare the performance of 9 classic feature extraction methods for predicting the particle size distributions (PSD) from these images. The discrete element code YADE was used to generate synthetic images of granular materials to build the dataset. Nine feature extraction methods were compared: Haralick features, Histograms of Oriented Gradients, Entropy, Local Binary Patterns, Local Configuration Pattern, Complete Local Binary Patterns, the Fast Fourier transform, Gabor filters, and Discrete Haar Wavelets. The feature extraction methods were used to generate the inputs of neural networks to predict the PSD. The results show that feature extraction methods can predict the percentage passing with a root-mean-square error (RMSE) on the percentage passing as low as 1.7%. CLBP showed the best result for all particle size...
Chemical Engineering Science, 2016
Efficient processing of particulate products across various manufacturing steps requires that particles possess desired attributes such as size and shape. Controlling the particle production process to obtain required attributes will be greatly facilitated using robust algorithms providing the size and shape information of the particles from in situ measurements. However, obtaining particle size and shape information in situ during manufacturing has been a big challenge. This is because the problem of estimating particle size and shape (aspect ratio) from signals provided by in-line measuring tools is often ill posed, and therefore it calls for appropriate constraints to be imposed on the problem. One way to constrain uncertainty in estimation of particle size and shape from in-line measurements is to combine data from different measurements such as chord length distribution (CLD) and imaging. This paper presents two different methods for combining imaging and CLD data obtained with in-line tools in order to get reliable estimates of particle size distribution and aspect ratio, where the imaging data is used to constrain the search space for an aspect ratio from the CLD data.
Sensitivity of particle size and shape parameters with respect to digitization
The growing success of image analysis based instruments for particle characterization demonstrates the importance of size and shape analysis in operations involving particulate materials. ISO norms for particle sizing using image analysis are being elaborated to clarify nomenclature and measurement principles. But despite this, there is still a lack of understanding of how the digital representation of a particle affects different size and shape parameters. It is the purpose of this paper to explore the magnitude of estimation errors of a series of size and shape parameters from different digital image representations of a single particle. These images are simulated from grey level images of black particles presenting a Gaussian transition towards their white background. Particles themselves are generated from analytical functions sampled by digital grids with variable densities, positions and orientations. Results of inscribed disk, elongation, circularity, roughness, roundness, etc. are plotted as a function of grid density (magnification) with error bars corresponding to the scattering of results for variable thresholds, grid translations and rotations As a conclusion, confidence intervals are given for parameters as a function of magnification and the most sensitive and robust methods of shape analysis are put forward.
AAPS PharmSciTech, 2006
The purpose of this paper is to describe results from the use of a set of Excel macros written to facilitate the comparison of image analysis (IA) and laser diffraction (LD) particle size analysis (psa) data. Measurements were made on particle systems of differing morphological characteristics including differing average aspect ratios, particle size distribution widths and modalities. The IA and LD psa data were plotted on the same graph treating both the weighting and the size unit of the LD psa data as unknowns. Congruency of the IA and LD plots was considered to indicate successful experimental determination of the weighting and size unit. The weighting of the resulting LD psa data (so-called volume-weighted) is shown to be better correlated with IA area-weighted data. The size unit of LD psa data is shown to be a function of particle shape. In the case of high aspect ratio particles characterized by approximately rectangular faces the LD psa data is shown to be a function of multiple particle dimensions being related to IA size descriptors through a simple variation of the law of mixtures. The results demonstrate that successful correlations between IA and LD psa data can be realized in the case of non-spherical particle systems even in the case of high aspect ratio particles; however, the inappropriateness of the application of the Equivalent Spherical Volume Diameter and the Random Particle Orientation assumptions to the interpretation of the LD psa results must first be acknowledged. Correlation permits cross validation of IA and LD psa results increasing confidence in the accuracy of the data from each orthogonal technique.
Rapid Particle Size Measurement Using 3D Surface Imaging
AAPS PharmSciTech, 2011
The present study introduces a new three-dimensional (3D) surface image analysis technique in which white light illumination from different incident angles is used to create 3D surfaces with a photometric approach. The three-dimensional features of the surface images created are then used in the characterization of particle size distributions of granules. This surface image analysis method is compared to sieve analysis and a particle sizing method based on spatial filtering technique with nearly 30 granule batches. The aim is also to evaluate the technique in flowability screening of granular materials. Overall, the new 3D imaging approach allows a rapid analysis of large amounts of sample and gives valuable visual information on the granule surfaces in terms of surface roughness and particle shape.
Quantification of particle size and concentration using in-line techniques and multivariate analysis
Powder Technology, 2020
We study means of extracting quantitative information about particle attributes using state-ofart in-line and off-line particle measurements and analysis techniques. The approach comprises a combination of image analysis, laser diffraction, inversion of chord length distribution, and multivariate analysis. Polystyrene particle suspensions are used as the model system to provide a wide range of particle loadings (up to 10 wt%), sizes (<90 to 800 µm) and shapes. We identify key challenges and limitation of the in-line imaging and chord length measurements; particularly, an upper limit of particle number density of 10,000 g-1 is observed, as well as the impact of internal reflections from large and transparent particles. The latter phenomena deteriorate the accuracy of the chord length distribution and the subsequent particle size estimation using inversion algorithms. The study demonstrates the use of multivariate analysis 2 for quantifying particle size and concentration, which yields relative errors of 6 and 11 %, respectively.
Improving particle size measurement using multi-flash imaging
Minerals Engineering, 2009
An accurate calculation of the size distribution of coarse particles is crucial in the mining and quarrying industry. Machine vision has the capability to overcome many inherent limitations of traditional sieving methods and is an active research area. However, inaccurate image segmentation of particles through software based algorithms is a significant source of error. In this paper, a hardware based approach to improving image segmentation is demonstrated using multi-flash Imaging (MFI), where multiple images captured with different illumination allows depth edges around a particle to be captured through shadow information. The MFI method is compared with conventional segmentation methods such as watershed and Canny edge detection. In order to provide more accurate evaluation of performance wooden spheres of known diameter were evaluated. Imaging the size distribution of pebbles provided a practical scenario in the evaluation of MFI. The results revealed that MFI produced more accurate size estimations than conventional segmentation techniques for both the wooden spheres and pebbles, demonstrating the potential for future use in the mining industry.
Artificial vision system for particle size characterization from bulk materials
Chemical Engineering Science, 2017
This study shows how to develop a fast, reliable, and non-invasive artificial vision system to quantitatively estimate the particle size distribution of granular products. The system, based on multivariate and multiresolution texture analysis, uses digital images of the bulk material to extract quantitative information on the particle size ranges appearing in each image and on their weight proportion independently of the shape of the particle distribution (mono-or multi-modal). The method is applied to a wet-granulated product (namely, microcrystalline cellulose), and it is shown that the size distributions can be estimated accurately. The system performance is discussed in the light of an application in the automated monitoring of particle size distribution in industrial processes.
Application of Imaging Techniques to Geometry Analysis of Aggregate Particles
Journal of Computing in Civil Engineering, 2004
This paper presents image analysis techniques by which to characterize the texture, angularity, and form of aggregate particles used in highway construction and geotechnical applications. For texture analysis, wavelet decomposition in gray scale images of particles is performed. The results demonstrate that multiscale wavelet representation is a powerful tool by which to capture the texture and to differentiate ''true'' texture from ''false'' texture caused by variations of natural color on a particle surface. Angularity and form analyses of particles are done using binary images. A gradient-based method is employed to describe angularity. This method is shown to differentiate between particles with different angularity characteristics. Form analysis of the particles includes computing the shape factor and sphericity index, which are based on measurements of the shortest, intermediate, and longest axis of the particle. Particle thickness is measured using the feature of an autofocus microscope. The width and length are calculated by an eigenvalue decomposition method of two-dimensional particle projections. Details of an interactive software developed to compute the different aggregate shape factors are discussed. The results indicate that these calculated values of the particle dimensions match very closely the values measured manually using a digital caliper.