Title :
Object recognition for industrial image
Author :
Licev, Lacezar ; Tomecek, Jan ; Babiuch, Marek
Author_Institution :
Dept. of Comput. Sci., VSB Tech. Univ. of Ostrava, Ostrava, Czech Republic
Abstract :
The contribution is dealing with the design and implementation of software modules that will assist the user to the advanced recognition of objects of interest, such as the profile of mining and other industrial objects. Modules allow initial adjustment with a broad range of different image filters, which will allow for more precise work of the second module for automatic segmentation, which ensures active contour method with using the Gradient Vector Flow and Fast Marching Level Sets. Both modules are implemented in Java with the JAI framework and are included in the software FOTOMNG system.
Keywords :
edge detection; filtering theory; gradient methods; image segmentation; object recognition; FOTOMNG system; JAI framework; Java; active contour method; automatic segmentation; fast marching level sets; gradient vector flow; image filters; industrial image; object recognition; software modules; Active contours; Detectors; Image edge detection; Image segmentation; Level set; Noise; Smoothing methods; FOTOMNG; Java; active contours; digital image processing; edge detection; fast marching; gradient vector flow; image filters; noise reduction;
Conference_Titel :
Control Conference (ICCC), 2014 15th International Carpathian
Conference_Location :
Velke Karlovice
Print_ISBN :
978-1-4799-3527-7
DOI :
10.1109/CarpathianCC.2014.6843620
