Multithreaded two-pass connected components labelling and particle analysis in ImageJ

Sequential region labelling, also known as connected components is a standard image segmentation problem that joins contiguous foreground pixels into blobs. Despite its long development history and widespread use across diverse domains such bone biology, materials science geology, labelling can still form bottleneck in processing pipelines. Here, I describe multithreaded implementation of classical two-pass sequential introduce an efficient collision resolution step, ‘ bucket fountain’ . Code was validated on test images against commercial software (Avizo). It performance tested from 2 MB (161 particles) to 6.5 GB (437 508 determine whether theoretical linear scaling ( O (n)) had been achieved, 1–40 CPU threads measure speed improvements due multithreading. The new achieves b = 0.905–1.052, time ∝ ; R 0.985–0.996), which improves with increasing thread number up 8–16 threads, suggesting it memory bandwidth limited. This reduces the required hours few tens seconds for several GB, limited only by hardware scale. available open source free charge BoneJ.