Image analysis of particles is one of the key methods for particle characterization. This article discusses the major steps during the image analysis of particles.
Image Credit: Gorodenkoff/Shutterstock.com
What is Image Analysis of Particles?
Image analysis of particles is performed to determine the physical parameters, such as size and shape, of a particle effectively from images. The image analysis method can be classified into static image analysis method and dynamic image analysis method based on the particle movement state during the measurement.
Automated imaging methods offer several advantages for determining the particle size distribution of material over alternative methods such as static light scattering. For instance, every particle can be photographed and analyzed individually using imaging methods, which enables statistical calculations on both particle size and particle shape.
The analysis of both particle shape and size leads to the detection of fine and oversized particles, more meaningful shape and size parameters of every particle, flexible changeovers between volume/area/number distribution types based on the particular task, virtual assessment of the dispersing sample state, and higher differentiation of materials, which are beneficial for determining the particle size/shape distribution.
Major Steps in Image Analysis of Particles
The major steps involved during the image analysis of particles include image acquisition, image processing and particle detection, and particle size and shape calculation.
Image Acquisition
Image acquisition is the first step in image analysis in which the particles are imaged using different imaging devices, such as high-speed digital cameras or compound microscopes equipped with digital cameras, at a sufficiently high resolution to resolve the particle features accurately.
Adequate resolution is ensured using proper magnification with a suitable numerical aperture. In both static and dynamic image analysis, the process involves capturing a two-dimensional (2D) image of three-dimensional (3D) particles. In dynamic image analysis, the particle presentation to the detection optics depends on several parameters, including the viscosity and flow rate of diluents and sample holder thickness.
In static image analysis, the particle presents themselves with their largest cross-sectional area that is perpendicular to the detection system and light. Thus, the detection system analyzed image is primarily the largest cross-sectional projection. Refractive index matching must be avoided by improving the contrast during image acquisition.
Image Processing and Particles Detection
The captured image is digitized into pixels, with each pixel containing two types of information, including the intensity and location. The image can be in black and white, also known as grayscale, or in color, depending on the style and quality of the camera.
Color image is typically converted into a grayscale image for image analysis of particles. In most systems, grayscale is defined by the white light intensity, with a zero intensity value indicating no white light, while the maximum 255 intensity value indicates white light of the highest intensity.
The edge of the particle is defined by thresholding, a process that converts a grayscale image to a binary image. In this technique, the defined threshold must be surpassed to obtain the edge of the article.
For instance, if the intensity value of an image background is measured as 150 and the threshold is selected to be 100, then the pixel intensity value above 100 anywhere within the field of view is considered the interior or edge of a particle.
However, thresholding can lead to biased data when the selected threshold value is too low or too high. In a binary image, the particle edge must be smooth and dark without a lighter halo around it and no erosion must occur in the edge.
An extremely low threshold can lead to erosion in the actual particle edge, which can affect several shape measurements, while a high threshold leads to the formation of a halo around the particle data.
The issue can be resolved by selecting a constant threshold for all fields of view. Specifically, a consistent grayscale or intensity value must be maintained in the background by measuring and adjusting the incident light intensity automatically to a constant value before every analysis and sample preparation.
Particle Size and Shape Calculation
Particle size and shape measurements can be performed after the particle edges have been clearly defined. The shape and size of the particles can be measured using different parameters, such as area, perimeter, Feret’s diameter, aspect ratio, convexity, circularity, circular equivalent diameter, and spherical equivalent volume.
Although both equivalent area diameter and Feret’s diameter can be used to measure the particle size distribution, Feret's diameter displays a greater accuracy in measuring the particle size distribution compared to the equivalent area diameter. The shape and size parameters can be analyzed to obtain information about the particle behavior and characteristics.
More from AZoOptics: Surface Topography Imaging with AFM
References and Further Reading
Shanthi, C., Porpatham, R. K., Pappa, N. (2014). Image Analysis for Particle Size Distribution. International Journal of Engineering and Technology, 6, 1340-1345. https://www.researchgate.net/publication/287319918_Image_Analysis_for_Particle_Size_Distribution
Image Analysis of Particles [Online] (Accessed on 18 September 2023)
Olson, E. (2011). Particle Shape Factors and Their Use in Image Analysis – Part 1: Theory [Online] (Accessed on 18 September 2023)
Image Analysis [Online] Available at https://www.bettersizeinstruments.com/products/by-technology/image-analysis/ (Accessed on 18 September 2023)
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.