Affiliation:
1. Huzhou Teachers College
Abstract
Because the fusion rules which are based on coefficient of a single point and coefficient of region has one-sidedness and are easy to make blending images lack fidelity, a new fusion rule is put forward. When this fusion rule fuses the low frequency sub-band resolved by wavelet method, it adequately consider the local areas regional energy and the regional energys similarity of the image which is need to fuse, and it use threshold to choose appropriate fusion rule. Experimental result indicated that this way can preferably keep detail features of the image, and it has some engineering application value.
Publisher
Trans Tech Publications, Ltd.
Reference7 articles.
1. Carisetti C A, Fong T Y, Fromm C. I Learn self-learning defect classifier [J]. Iron and Steel Engineer, 1998, 75(8): 50-53.
2. Bilstein W, Enderle W, Moreas G, et al. Two systems for on-line oil film and surface roughness measurement for strip steel production [J]. Revue de Metallurgies, 2007, 104(7/8): 347-353.
3. Wei Y, Yan Y, Li B , et al. Development of a Surface Defect Inspection System for Cold Rolled Strip Based on Bright-dark Field Mode [C]/ 2nd International Conference on Advances in Product Development and Reliability, 2010, July, Shenyang, China, 2010: 762-765.
4. G Piella. A region-based multiresolution image fusion algorithm [C]/ in Proceedings of the 5th International Conference on Information Fusion, Annapolis, USA, July 2002, 1557-1564.
5. Jin L, Leshi Zh, Ke-xin X. Multimodal Face Recognition Based on Images Fusion on Feature and Decision Levels [J]. Nanotechnology and Precision Engineering, 2009, 7(1): 65-70.