| Ultrasound imaging which is portability, noninvasive, security, low cost, and real-time,and it has wide application prospect in early diagnosis of breast cancer. However, compared to X-ray, CT and MRI, ultrasound imaging suffers from its poor resolution, and exhibits coarse speckle sand many types of artifacts. In order to achieve an even better resolution and further reduce angular-dependent artifacts, The research of this paper is carried on around Ultrasound Compounding.First, the traditional ultrasound imaging and the basic working principle of k-Wave ultrasound simulation platform is analysed and studied. A model contains fat, soft tissue, and tumour, reflecting the characteristics of sound speed and density of tissues is established. In order to obtain good signal-to-noise ratio and provide a stable premise for later work, wavelet threshold shrinking method is adopted for denoising of the received RF signal. Several threshold functions are analysed and compared, and the simulation results indicate that denosing with half soft threshold function is the best.Then, on the basis of the RMS compound method, this paper investigates a new type of compounding we call multiplicative compounding, in which compound images are produced by a summation of multiplied pairs of component images acquired at different angles. Point spread functions created from a set of simulated B-scan PSF by different compounding methods are compared and analysed, the results show that the lateral resolution can be improved in the compounding PSF and further improved in each of the MC images with the sharpest occurring at a angle of 90°. By comparing the compounding images of modeles demonstrates that compounding reduce artifacts and suppress the speckle, and achieve a better edge definition. What’s more, the CSR and SR are improved in MC image with a larger angular.At last, designed quality matrices have the potential to improve image quality parameters of contrastand target edge definition over conventional beamforming computational expense. Standard spatial compounding gives a large improvement in image quality, which can be modestly improved by designed matrix weighting. |
