Region-of-interest extraction in low depth of field images using ensemble clustering and difference of Gaussian approaches
作者：
Highlights：
• An automatic segmentation approach is proposed for segmenting low DOF images.
• It is time-efficient as it relies on texture details rather than color information.
• A reliable block-based ensemble clustering approach at two levels is developed.
• A threshold is optimized in a difference of Gaussian image.
• We achieve an average F-measure of 91.3% to extract the ROI in an image.
摘要
•An automatic segmentation approach is proposed for segmenting low DOF images.•It is time-efficient as it relies on texture details rather than color information.•A reliable block-based ensemble clustering approach at two levels is developed.•A threshold is optimized in a difference of Gaussian image.•We achieve an average F-measure of 91.3% to extract the ROI in an image.
论文关键词：Low depth-of-field,Difference of Gaussian method,Ensemble clustering,Expectation-maximization algorithm,Region-of-interest extraction
论文评审过程：Received 8 October 2012, Revised 9 January 2013, Accepted 6 March 2013, Available online 3 April 2013.
论文官网地址：https://doi.org/10.1016/j.patcog.2013.03.006
