A Multi-focus Image Fusion Classifier

作者：Siddiqui Abdul Basit; Rashid Muhammad^*; Jaffar M Arfan; Hussain Ayyaz; Mirza Anwar M

来源：Information-An International Interdisciplinary Journal, 2012, 15(4): 1757-1764.

• Multi-focus image fusion
• Support vector machine
• Image processing

Many image processing tasks such as image segmentation, restoration, enhancement, edge detection etc. require an image in which all the objects are developed. The focal length of optical lenses is limited which results in an image containing partially developed objects. This paper deals with obtaining an image where all the objects are in focus: A SVM based classifier is used for multi-focus image fusion. The classifier is trained using feature vectors of blocks of ten different pairs of partially developed images. Different measures are used to characterize the detail present in image blocks. The results of proposed technique are compared using different quantitative measures with some of the existing techniques which show its performance.

• 出版日期2012-4