A new technique for image enhancement using digital fractional-order Savitzky–Golay differentiator

详细信息    查看全文

• 作者：Suraj Suman ; Rajib Kumar Jha
• 关键词：Riemann–Liouville fractional derivative ; Savitzky–Golay differentiator ; Mask ; Image enhancement
• 刊名：Multidimensional Systems and Signal Processing
• 出版年：2017
• 出版时间：April 2017
• 年：2017
• 卷：28
• 期：2
• 页码：709-733
• 全文大小：
• 刊物类别：Engineering
• 刊物主题：Circuits and Systems; Electrical Engineering; Signal,Image and Speech Processing; Artificial Intelligence (incl. Robotics);
• 出版者：Springer US
• ISSN：1573-0824
• 卷排序：28

文摘

In this manuscript, digital fractional-order Savitzky–Golay differentiator (DFOSGD) is explored in the image enhancement application. In DFOSGD, moving window is responsible for estimating the fractional-derivative of desired fractional-order of a given signal. Here, this moving window is employed as a mask and masking is done with given input image in eight different directions to make it rotation independent. This mask is utilized to characterize the fractional derivative for an image as well as to enhance dark and low contrast images. The response of the mask against fractional-order on different profile of images is correlated with the fundamental properties of given input image to select the optimal value of fractional-order to get better enhanced image. Visual as well as quantitative assessment of the simulation results are performed using different no-reference quantitative metrics. Comparative analysis reveals that the proposed mask produces the better or comparable quality of enhanced image with less computational complexity than several well-known and state of the art techniques.