提出一种反锐化掩模(USM)锐化检验方法。采用局部二值模式方法(LBP)检测图像边缘特征,利用支持向量机进行分类,然后检测图像是否被锐化。针对不同的锐化强度,对比不同LBP检测模式,筛选出最优检测方式。实验结果表明,LBP方法能达到较好的USM锐化检测效果,其中旋转不变模式的检测效果最好,在弱锐化情况下检出率仍能达到90%,优于现有文献中各类方法的检测性能。

Herein, a method for unsharp masking (USM) sharpening detection is proposed. First, the local binary pattern (LBP) method is used to detect the edge features in an image. Then, a support vector machine is used for classification to detect whether the image is sharpened. Subsequently, the different LBP detection modes are compared in terms of the resulting sharpening intensity to select the optimal detection method. The experimental results show that the LBP method can achieve a relatively good USM sharpening detection effect. The rotation-invariant mode provides the best detection performance, providing a detection rate of up to 90% under the condition of weak sharpening, which is better than those achieved by the existing methods.