基于图像边缘检测的变系数误差扩散方法

误差扩散算法是一种重要的图像半色调化技术,被广泛应用于各种二值化的输出设备当中,但传统的误差扩散算法生成的半色调图像易出现"蠕虫"现象和边缘模糊。针对此问题,本文提出一种结合图像边缘检测的变系数误差扩散方法,该方法以Ostromoukhov提出的变系数误差扩散算法为基础,利用其优化后的误差扩散系数,可以较好地克服"蠕虫"现象。在此基础上,采用Sobel算子检测图像的边缘并沿边缘方向修正误差扩散滤波系数,以较好地保持原图像的边缘特性。实验结果表明,利用该方法产生的半色调图像不仅具有蓝噪声特性,而且边缘结构清晰。

A New Variable-Coefficient Error Diffusion Method Based on Image Edge Detection

The error diffusion algorithm is widely used in many binary output devices,however,the worm artifacts and edge blur often appear in the halftone image generated by the traditional error diffusion algorithm.In order to overcome the above drawbacks of the algorithm,this paper proposes a new variable-coefficient error diffusion method based on image edge detection.The proposed method is based on the well-known Ostromoukhov algorithm which can overcome the worm artifacts by the optimized filter coefficients.In addition,the image edge is detected by the Sobel operator and the error diffusion filter coefficients are adjusted according to the edge direction for keeping the image edge.The experimental results show that the halftone images generated by the proposed method have an ideal blue noise spectrum,as well as the clear edge structures.