Robust image segmentation technique for rock fragmentation analysis

Fragmentation analysis of blast or crushed rock material is a time-consuming and costly process. During the last two decades, the mining industry has been investigating image-based analysis systems as an alternative to generate fragmentation results. Many image-based techniques and commercial products have emerged during the last few years, and the mining industry has recently begun using these tools for real-time estimation of size distributions. Some existing systems require manual image editing to add or delete edges (or net) before executing the fragmentation analysis routine. This manual procedure may take hours of time for each image analyzed. Reliable and robust image segmentation should be able to handle a wide range of rock textures and sizes under a variety of lighting conditions. This paper describes a novel software application, which has the capability to autonomously capture images and analyze them to generate the particle size distribution. The system can also process a batch of images captured during a fixed duration of time and produce the overall particle size distribution. The new method has different layers of segmentation modules, which allows the system to operate under a wide range of rock textures and lighting conditions. A new grey-level slicing technique is developed which can perform under a range of illuminating conditions. The Canny-based edge detection technique is used to segment rocks appearing in dark regions.