As time passes, there are more and more stains in the ancient works for being soiled and destroyed, which make it difficult to process ancient works. A projection operator is proposed and applied in distinguishing the stains from the binary images of ancient works. Firstly, during the image traversal of the ancient works, the projection operator is designed to deal with the vectors constructed on the base of the local contents of binary image of ancient work, which contributes much to the detection of stain complying with the regulated size; Secondly, the projection operator is designed to deal with the vectors constructed on the base of the whole contents of binary image of ancient works, which can detect the lines and borders of the image; On the basis of the above, the information of the lines and borders will be wipe off from the detected stains, which will avoid incorrect operation to the lines and borders. Experiments validate the projection operator proposed in the paper can not only quickly detect the stains with high accuracy rate but also can avoid the incorrect operation to lines and borders of images, which will benefit much to the incorrect processing of ancient works.
[1]
WU Xiong-ying.
Component labeling and noise eliminating of binary image
,
2006
.
[2]
Qiang Chen,et al.
A moment-based nonlocal-means algorithm for image denoising
,
2009,
Inf. Process. Lett..
[3]
李绍滋,et al.
Adaptive fractal-wavelet image denoising base on multivariate statistical model
,
2014
.
[4]
Joachim Weickert,et al.
Rotationally invariant similarity measures for nonlocal image denoising
,
2011,
J. Vis. Commun. Image Represent..
[5]
Zhou Ji-liu.
Research on image denoising based on fractional order integral
,
2011
.
[6]
Yang Hui,et al.
Infrared Image Denoising Based on Improved Threshold and Inter-scale Correlations of Wavelet Transform
,
2011
.