Image Segmentation and Grain Size Measurements of Palm Kernel Shell Powder

Image segmentation is one of the essential tools to determine the foreground from background and at the same time enhance visual perception for better understanding through image manipulation. It is very useful for pattern recognition and image processing. This enables users to determine the high quality and high resolutions of the final result of the analysis. In this present study, PKS-Powder has been characterized with the use of Scanning Electron Microscopy (SEM), and Energy Dispersive X-ray (EDX) analyses. Digital Vibratory milling machine was employed for the mechanical milling at the time interval of 0, 10, 15 and 20 min. ImageJ software was employed for image processing by sectioning an image into various regions using thresholding segmentation method. It was revealed that at 0 min (i.e. 300 µm sieved), it has the highest mean area value of 127.169 µm² and area standard deviation of 4,091.487 µm² with the least value of a number of particle size distribution of 458 µm. In contrast, 20 min milled has the lowest values for mean area and area standard deviation of 52.913 µm² and 795.413 µm² respectively with the highest number of particle size distribution of 1,315 µm. It was observed that milling time increases the number of particle sizes distributions and reduces the area of particle size. EDX analysis revealed that Ca, Al, Si, Fe, C, K, and O are the main elemental constituents of PKS-powder.