Global matching optimized ultrasound image three-dimension reconstruction method

The invention discloses a global matching optimized ultrasound image three-dimension reconstruction method. The method comprises the following steps of reading two-dimensional ultrasound images acquired by an ultrasonic probe in real time, and acquiring two parameters of sizes and intervals of the two-dimensional ultrasound images; tracking and positioning the space position of each two-dimensional ultrasound image through a space positioning system, and acquiring the relative position relationship of each two-dimensional ultrasound angiography image and ultrasonic three-dimensional data; traversing each pixel in the two-dimensional ultrasound images, and assigning pixel gray values to voxels in the ultrasonic three-dimensional data according to preset thresholds; traversing the ultrasonic three-dimensional data, acquiring voxel vacant areas, and detecting the boundaries of the voxel vacant areas; calculating repaired weight coefficient of each voxel on the boundaries of the vacant areas; traversing the whole three-dimensional data, finding a module which is the best match for a module with the biggest repaired weight coefficient, repairing the module with the biggest weight coefficient, and finishing the reconstruction of the ultrasonic three-dimensional data. The method can be used for quickly and accurately reconstructing the ultrasonic three-dimensional data, and provides theoretical guidance for clinical focus diagnosis.