Dual-energy CT super-resolution image reconstruction method and system based on sparse representation

Abstract

The invention belongs to the technical field of image super-resolution reconstruction, and particularly relates to a dual-energy CT (Computed Tomography) super-resolution image reconstruction method and a dual-energy CT super-resolution image reconstruction system based on sparse representation, which are characterized in that paired high-resolution and low-resolution dual-energy CT images are obtained through de-downsampling operation of an original truth value image fuzzy filter, and paired matched image segmentation blocks are extracted from CT images in a training set; in order to obtain similarity of sparse representation, two pairs of dictionaries are jointly trained for the high-energy-spectrum CT image and the low-energy-spectrum CT image respectively, and each pair of dictionaries comprises dictionaries of low-resolution image blocks and dictionaries of high-resolution image blocks; the sparse representation coefficients extracted from the low-resolution input block are then multiplied by the corresponding high-resolution dictionary for reconstructing the high-resolution output. According to the method, better image quality and clearer visual effect can be obtained by reconstructing the high-resolution image in combination with a proper amount of iterative operation, the resolution of the dual-energy CT image can be effectively improved according to the image quality evaluation index and the peak signal-to-noise ratio, and the method is convenient for actual scene application.

Description

technical field [0001] The invention belongs to the technical field of image super-resolution reconstruction, in particular to a dual-energy CT super-resolution image reconstruction method and system based on sparse representation. Background technique [0002] Computed Tomography (CT) technology can non-destructively and non-contactly obtain the internal structure information of the imaging object, and has played an important role in the fields of medical diagnosis, industrial inspection and safety inspection. CT technology is also one of the key technologies of computer-aided diagnosis based on medical imaging. Efficiently obtaining high-quality medical CT images is very important for the clinical application of computer-aided diagnosis technology. Computed tomography technology obtains material and structural information by collecting projection information of X-rays penetrating objects, but due to the similar attenuation coefficients between materials for X-rays, single...

AI Technical Summary

Problems solved by technology

The basic constraint of SR reconstruction is that after applying the same generative model to the restored image, the observed low-resolution image should have been regenerated. However, super-resolution image reconstruction itself is a pathological problem. A kno

Images