SEM/X-ray imaging of cement-based materials

SEM/X-ray imaging of cement-based materials Next: Stereological Parameters Up: Sample Preparation and Imaging Techniques Previous: Image Acquisition

2.3 Image Processing

To process the input SEM/X-ray images and determine the distribution of phases, a decision tree is traversed for each pixel location in the images. Example decision trees for a typical cement powder and a fly ash containing substantial chloride are shown in Figures 1 and 2, respectively. In these figures, X^* represents a critical threshold greylevel value, pixels having a greylevel greater than the value of X^* being considered to contain the element of interest and those with a greylevel below X^* being classified as not containing the element. To determine the values of X^* for each element, the corresponding greylevel histogram [7] for each X-ray image is viewed. The greylevel histogram is a plot of the fraction of pixels containing each greylevel value (e.g., 0 to 255). For example, Figure 3 shows the greylevel histogram for the calcium signal image for a portland cement. The local minimum value between the two peaks would serve as a good estimate of Ca^* for this particular image. For some of the elements, the greylevel histogram contains only one peak and the analyst must estimate the appropriate value of X^* from the shoulder present on the right side of the peak. In all cases, the raw X-ray image can also be thresholded at different greylevel values, to produce a binary image that should represent the locations of each element in the 2-D image. If this value is set too low, many small isolated regions of element X (basically random noise) will appear. If this value is set too high, very little of element X will be present. Many modern image processing systems allow this threshold to be changed dynamically using a mouse or dial so that the analyst may quickly determine the "best" threshold value for a particular element.

Figure 1: Segmentation algorithm for separating portland cement into its components. C₃S denotes tricalcium silicate, C₂S denotes dicalcium silicate, C₃A denotes tricalcium aluminate, C₄AF denotes tetracalcium aluminoferrite, and CaO corresponds to free lime.

Figure 2: Segmentation algorithm for separating a municipal waste fly ash into its components. AS denotes an aluminosilicate phase and CAS₂ denotes calcium alumino-disilicate.

Figure 3: Greylevel histogram for calcium X-ray image for an ordinary portland cement.

After the segmentation tree is traversed, the segmented image produced will still contain a substantial amount of random noise. To improve the image quality, three "filters" are applied in succession to the processed image. First, all isolated one pixel "solid" pixels are converted to porosity. Second, all isolated one pixel "pores" (totally surrounded by solids) are converted to the majority surrounding solid phase. Finally, a median filter is applied to replace each solid pixel by the majority solid phase present in the surrounding neighborhood, typically 5 pixels x 5 pixels. This three-fold process removes the remaining noise present in the segmented image, producing an image ready for quantitative stereological analysis.

Next: Stereological Parameters Up: Sample Preparation and Imaging Techniques Previous: Image Acquisition