Title :
Fast and exact boundary extraction of high resolution PCB images
Author :
Hu, Tao ; Guo, Baoping
Author_Institution :
Inst. of Optoelectron. Sci. & Eng., Huazhong Univ. of Sci. & Technol., Wuhan
Abstract :
An improved algorithm was proposed and applied to the fast and exact boundary extraction of high resolution PCB images. Firstly, two problems in the original algorithm were mentioned and solved respectively by introducing splitting points and merging points and by designing an extended data structure. Secondly, the MMX/SSE instructions were easily introduced to optimize the run-length encoding procedure of the improved algorithm, and a parameter called run average compression ratio was introduced and used to qualitatively analyze the range of application of the improved algorithm and the optimization efficiency to be achieved. Experimental results show that the optimization can make it run more than two times faster in the application of the boundary extraction for high resolution PCB images.
Keywords :
data structures; edge detection; electronic engineering computing; image resolution; printed circuits; runlength codes; MMX-SSE instructions; boundary extraction; extended data structure; high resolution PCB images; merging points; run average compression ratio; run-length encoding; splitting points; Algorithm design and analysis; Computer aided manufacturing; Data mining; Image coding; Image processing; Image resolution; Machine vision; Merging; Pixel; Testing; MMX/SSE; PCB; boundary extraction; merging point; run average compression ratio; splitting point;
Conference_Titel :
Industrial Electronics and Applications, 2009. ICIEA 2009. 4th IEEE Conference on
Conference_Location :
Xi´an
Print_ISBN :
978-1-4244-2799-4
Electronic_ISBN :
978-1-4244-2800-7
DOI :
10.1109/ICIEA.2009.5138628
