YACCLAB - Yet Another Connected Components Labeling Benchmark

Abstract

The problem of labeling the connected components (CCL) of a binary image is well-defined and several proposals have been presented in the past. Since an exact solution to the problem exists and should be mandatory provided as output, algorithms mainly differ on their execution speed. In this paper, we propose and describe YACCLAB, Yet Another Connected Components Labeling Benchmark. Together with a rich and varied dataset, YACCLAB contains an open source platform to test new proposals and to compare them with publicly available competitors. Textual and graphical outputs are automatically generated for three kinds of test, which analyze the methods from different perspectives. The fairness of the comparisons is guaranteed by running on the same system and over the same datasets. Examples of usage and the corresponding comparisons among state-of-the-art techniques are reported to confirm the potentiality of the benchmark

Connected Components Labeling on DRAGs

“…In order to evaluate the benefit of the proposed strategy we tested the DRAG algorithm using YACCLAB, an open source C++ benchmarking framework for CCL algorithms. The original version of the benchmark has been presented in [21], and it allows to test state-of-the-art algorithms on a wide range of datasets covering most of the fields in which CCL could be exploited. The fairness of the comparison is guaranteed by compiling the algorithms with the same optimizations and by running them on the same data and over the same machine.…”

Spaghetti Labeling: Directed Acyclic Graphs for Block-Based Connected Components Labeling

“…In this section the benefits of the proposed strategy are evaluated using YACCLAB [26], [32], [33], a widely used [34], [35] open source C++ benchmarking framework for CCL algorithms. YACCLAB allows researchers to test state-of-theart algorithms on real and synthetic generated datasets.…”

Mono-vision based moving object detection in complex traffic scenes

“…Then, considering a third view, P also can be represented by the projective structure between the second view and the third one: 23 is the projective depth to a new reference plane connecting the second view and the third view. There exists a relationship that links the two projective structures from a static point in quadratic form.…”

“…14) is applied for combining information from the different constraints to segment the moving pixels. After removing the moving pixels outside the driving space area, on-road moving pixels are then clustered using connected components labeler [23].…”