Title :
Neural network based computer leather matching system
Author :
Min, Cai ; Liqun, Han
Author_Institution :
Dept. of Autom. Eng., Beijing Inst. of Light Ind., China
Abstract :
The structure and design of a neural network-based computerized leather matching system are introduced. Measurements for stable conditions of taking pictures, drawing and selecting of color and texture characteristics, and two self-organizing map (SOM)-based leather clustering methods are discussed. More than 84% of the online leather clustering results using the system are consistent with the results of manual matching
Keywords :
automatic optical inspection; colour photography; image colour analysis; image matching; materials; online operation; production engineering computing; self-organising feature maps; texture; color characteristics selection; drawing; neural network-based computerized leather matching system; online leather clustering; photography; self-organizing feature map; stable conditions; texture characteristics selection; Belts; Cameras; Charge coupled devices; Charge-coupled image sensors; Clothing; Computer networks; Decoding; Neural networks; Prototypes; Signal processing;
Conference_Titel :
Intelligent Processing Systems, 1997. ICIPS '97. 1997 IEEE International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-4253-4
DOI :
10.1109/ICIPS.1997.672804
