Skeletonizing Digital Images with Cellular Automata - Cellular Automata in Image Processing and Geometry - page 55

Fig. 3.6 Flowchart of the implementation on CUDA TM of the CA algorithm

Fig. 3.7 The binarized STOP signal, its skeletonization, its inverted binarization and its
skeletonization

composed by 12 Stream Processors with a total of 96 cores to 1340 MHz. It has 1
GB DDR3 main memory in a 128 bits bus to 700 MHz. So, the transfer rate obtained
is by 54.4 Gbps. The used Constant Memory is 64 KB and the Shared Memory is 16
KB. Figure 3.6 shows a flowchart of the implementation on CUDA of the algorithm.
Next, we show the results of some experiments of skeletonizing with our parallel
CA implementation of the Guo and Hall algorithm. As a first example 6 ,Fig.3.7,
shows the skeletonization of the STOP traffic signal. This example illustrates an
old problem in black and white images. In such images, the meaning is usually
associated to black pixels, whereas white pixels constitute the background. In this

6

These examples are borrowed from [35].