αβ■(ABO)图像去噪模型边缘检测算子的改进

Improvement of Edge Detection Operator for αβ■(ABO) Image Denoising Model

为了能够更好地检测图像的边缘和纹理,并有效地去除噪声,在αβ■(ABO)图像去噪模型的基础上,提出一种综合利用梯度模和水平集曲率模作为图像边缘检测算子的去噪模型。该模型保持了曲线边缘、角点等图像特征,同时,和传统的PM模型、TV模型、Y-K模型以及αβ■(ABO)模型相比,使用该模型去噪后,图像的峰值信噪比(PSNR)有了明显提高,而均方根误差(RMSE)则明显降低。最后,仿真实验证明了该方法的有效性。

In order to detect the edge and texture of the image effectively,and to remove the noise efficiently,an image denoising model is proposed,which is based on αβω（ ABO） model. In the new model,gradient modules and level set curvature modules are used as the image edge detection operator. The edge texture details,such as the curve edges,corners and so on,can be retained by the new model. At the same time,the PSNR have been enhanced by comparing with traditional PM model,TV model,Y-K model and αβω（ ABO） model. Meanwhile,the RMSE have been significantly reduced. Finally,the validity of the proposed model has been proved through the experiments.