图像融合技术在Computed Radiography影像中的应用研究

Image Fusion in Computed Radiography

将图像融合技术应用于CR(ComputedRadiography)技术中 ,有效改善影像的信噪比 (signal-to -noiseratio)。运用双IP(ImagePlate)技术采集人体体模影像后 ,在变换域进行图像配准。建立融合图像的数学模型 ,基于此模型对配准完成的源图像实施不同权重的加权计算得到融合图像。结果 融合图像信噪比源图像可提高 30 %左右。结论 将图像融合技术应用于CR技术可有效改善影像质量 ,有助于提高诊断的准确性。

A new technique for image fusion was developed to improve and optimize the image signal-to-noise ratio(SNR) in dual-plate computed radiography(CR). A front and a back phantom image were acquired using two CR image plates. The two images were then superimposed to form a fused image. In a mathematical model,the front and back images were weighted in proportion to their SNR squared to optimize the SNR of the fused image. Imaging experiments were performed to validate the model. The results showed that the SNR of the fused image was increased by approximately 30%. Conclusions: Application of image fusion in CR can improve image quality and diagnostic accuracyl.