A New Connected-Component Labeling Algorithm Article Swipe

View

Related Concepts

Pixel Line (geometry) Scan line Artificial intelligence Computer science Equivalence (formal languages) Algorithm Connected-component labeling Component (thermodynamics) Computer vision Connected component Image (mathematics) Pattern recognition (psychology) Mathematics Image segmentation Discrete mathematics Physics Geometry Grayscale Scale-space segmentation Thermodynamics

Yuyan Chao , Lifeng He , Kenji Suzuki , Qian Yu , Wei Tang ·

YOU? · · 2016 · Open Access · · DOI: https://doi.org/10.21307/ijanmc-2016-008 · OA: W2802429252

This paper proposes a new first-scan method for two-scan labeling algorithms. In the first scan, our proposed method first scans image lines three by three with a leaving line, and for foreground pixels among each three lines, assigns them provisional labels, and finds and resolves label equivalences among them. Then, it processes the leaving lines from top to bottom one by one, and for each line, assigns foreground pixels on the line provisional labels, finding and resolving label equivalences between the foreground pixels and those on the lines immediately above and below the current line. Experimental results demonstrated that our method is more efficient than conventional label-equivalence-based labeling algorithms.