4 results about "Image texture" patented technology

The invention provides a lung 4D-CT image super-resolution processing method based on variational optical flow estimation. A variational optical flow model used for solving an optical flow field among different lung 4D-CT phase images is constructed; the model is solved through utilizing a fast alternating direction multiplier method to acquire the optical flow field among the different phase images; based on the optical flow field, a non-local iteration back projection super-resolution reconstruction algorithm is utilized to realize reconstruction of the high resolution lung images. The method is advantaged in that on the condition that the image texture structure is enhanced, image contours can be better preserved; the main factor which influences the super-resolution reconstruction algorithm effect is image registration precision, the combined local and global CLG variational optical flow estimation model is utilized to acquire the more precise and dense optical flow field, and the fast alternating direction multiplier method is further employed to realize fast solution of the optical flow model; secondly, the non-local iteration back projection algorithm based on an edge is employed, so edges and fine grains of the super-resolution images are effectively enhanced.

Embodiments relate to biasing an image noise filter to reduce edge and texture blurring of image data. Pixel values used to determine photometric coefficients for a bilateral filter are modified by offset values. The offset value for a pixel value is determined by applying a high pass filter to the pixel (referred to as the center pixel) and neighboring pixels of the center pixel. By adding the offset value to the center pixel value, the pixel value difference between the neighboring pixels and the center pixel becomes smaller for pixels on the same side of an edge as the center pixel. Thus, pixels on the same side of the edge get more weight in the bilateral noise filter. Conversely, pixels on the opposite side of the edge as the center pixel get less weight in the bilateral filter. As a result, the biased bilateral filter reduces blurring of edges and increases preservation of texture in the image data.

The invention provides a prostate activity appearance model establishing method based on a TURS image, and belongs to the technical field of image processing. The method comprises the following steps: firstly, acquiring prostate 3D MR and TRUS image data, marking shape feature points in the image data, and carrying out shape coordinate alignment by adopting a Protein analysis method; secondly, a principal component method is adopted to carry out dimension reduction on data, features are analyzed, and texture information is obtained through triangulation; and finally, performing feature analysis on the image texture by using a principal component analysis method, establishing a shape texture combination model, and performing parameter solution and optimization on the combination model. According to the method, the shape and appearance change characteristics of the prostate glands are considered, the front AAM model is established on the basis of the TRUS training set images of the prostate glands and serves as the basis of image segmentation positioning, and real-time accurate segmentation of the 3D MR and TRUS images can be achieved.

The embodiment of the invention provides a distortion effect generation method, apparatus and device, and a storage medium. The method comprises the steps: obtaining an image texture; carrying out distortion processing on texture coordinates corresponding to the image texture to obtain first distortion texture coordinates, wherein distortion processing sequentially comprises water ripple distortion, annular ripple distortion and spiral distortion; obtaining a second distorted texture coordinate input by the user, and superposing the first distorted texture coordinate and the second distorted texture coordinate to obtain a final sampling coordinate; sampling the image texture through the final sampling coordinates, and outputting a sampling result to a temporary texture cache; performing radial distortion processing on the sampling result in the temporary texture cache to obtain a mixed and distorted image texture, and displaying the mixed and distorted image texture in a screen. According to the method disclosed in the invention, a plurality of distortion algorithms can be freely combined, and more abundant and vivid distortion effects such as image rotation and fluctuation can berealized.