中文摘要：

工业CT成像过程中,金属伪影会降低成像质量,影响图像结构信息的判断和提取。针对射束硬化和光子散射这两个引起金属伪影的主要因素进行研究,提出了一种基于能谱滤波和图像残差重投影的金属伪影校正方法。首先根据透射衰减信息对初始射线能谱进行能谱估计,然后根据能量相关的物质衰减系数变化率曲线定量计算滤光片的厚度,并通过比较不同材料滤光片的有效射线透过率选择滤光片材料,最后根据原始重建图像目标区域进行散射估计,通过图像残差重投影去除散射噪声,实现CT图像的金属伪影校正。实验结果表明,文章方法能够有效抑制射束硬化和光子散射导致的金属伪影影响,提高图像对比度和改善成像质量。

英文摘要：

The metal artifact occurred in industrial Computed Tomography （CT） reduces image quality, influencing the estimation and extraction of image structure information. Focusing on beam hard- ening and photon scattering, which are two major causes, a metal artifact correction method through spectra filtering and residual image reprojection is proposed. First, it estimates the original spectra based on the information of transmission attenuation. Then, it quantitatively calculates the filter thickness from the energy-relevant attenuation coefficient curve, and selects the optimal filter materi- al after comparison of the effective ray transmissions with different filter materials. Finally, it esti- mates the scattering according to the original reconstruction image target region, and removes the scattering noise through image residual reprojection, realizing the CT image metal artifact correction. Experimental results show that the method can effectively suppress the beam hardening and photon scattering of metal artifacts, as well as improve the image contrast and quality.