A hybrid method based on anisotropic diffusion (AD) is proposed for ultrasound speckle suppression and edge enhancement. This method is designed to utilize the different denoising properties of three techniques: median filtering, improved AD filtering and isotropic diffusion filtering. The gradient matrix is analyzed, and thresholds are chosen by experiments. The hybrid method is made by combining the three filtering methods for three different grayscale gradient ranges respectively. The filtering program is realized by iteration, with iteration time determined by iteration stopping criterion (ISC) accurately. In the experiment hundreds of images are processed by hybrid method, in contrast with median filtering and speckle reducing anisotropic diffusion (SRAD) filtering. The experimental results shows hybrid method can greatly improve processing speed, while successfully suppress speckle and enhance edge, therefore suitable for high-speed noise elimination of 3D images with huge data.

Edge Enhancement and Filtering of Medical Ultrasonic Images Using a Hybrid Method

Local-structure-adapted Image Diffusion

According to the characteristics of the liver image and the shortcomings of the traditional region growing algorithm, a modified region growing algorithm is proposed based on the adaptive anisotropic filtering. The first, a anisotropic filtering algorithm is proposed based on adaptive anisotropic filtering and we make images noise reduction through it. Second, we studied a method which can select seeds automatically, then the parameters which are in the region growing algorithm will be obtained through Otsu. These innovations and improvements enable segmentation algorithm automatic, fast, accurate. Experiments show that, the algorithm that this article proposed the region growing based on adaptive image segmentation of the liver, the result is better than the traditional region growing algorithm.

A Region Growing Algorithm Based on Anisotropic Filtering for Image Segmentation of the Liver

For the characteristics of liver images and the shortcomings of the traditional region growing algorithm, a liver segmentation method based on RBF-CI (RBF-Confidence Interval) is proposed. On the basis of the application of window adjusting technology and the anisotropic diffusion method, the RBF neural network learning algorithm is introduced to calculate the coefficient of the confidence interval in order to reduce users' interaction amount and realize adaptive liver image segmentation. Simulation results show that, the proposed RBF-CI region growth segmentation algorithm can achieve effective segmentation results of liver images. For the ten sets of the liver images, while our proposed segmentation algorithm result achieves an average accuracy rate of 90.86%, which indicates the segmentation is accurate.

http://www.joics.com/publishedpapers/2015_12_5_1703_1711.pdf

Liver Image Segmentation Algorithm Based on RBF Confidence Interval

Local image information is crucial for accurate segmentation of images with intensity inhomogeneity which usually occurs in medical images. However, image information in local region is not incorporated in popular region-based active contour models, such as piecewise constant models and piecewise smooth models. In this paper, a method which is able to use local information is proposed. The main point is the introduction of the local fitting information expressed by a kernel function which is crucial for segmentation. Our method is based on piecewise constant Chan-Vese model, and compare with different methods for several synthetic images and medical images.

A Variation Level Set Formulation Local Based C-V Model for Medical Images Segmentation

Utilizing the echoic intension and distribution of different organizations and local details,ultrasonic image catches the important medical pathological changes.However ultrasonic image may be contaminated by the speckle noise in its forming process,which degrades image quality specially concealing some details,and works disadvantages to image segmentation,character extraction and image recognition,disease diagnosis and quantitative analysis.Using local coordinate transform,the first and second order normal derivatives of edge and local detail and the hyperbolic tangent function,also combining the anisotropic diffusion equation,we have put forth an ultrasonic image denoising and edge enhancement scheme,which can preserve edges,local details and ultrasonic echoic bright strips on denoising.This has been indicated theoretically and experimentally.

FU Shu-jun

Papers

Anisotropic Diffusion Equation Based Ultrasonic Image Denoising and Edge Enhancement