Medical image texture analysis system

Abstract

The present invention relates to a medical image texture analysis system. By utilization of MATLAB programming, Fourier transform and autocorrelation function texture analysis methods are integrated in a GUI interface, a generated GUI module is compiled into an EXE file independent of the MATLAB software, and image analysis is achieved by directly operating the EXE file. The image texture analysis system comprises an image input module, an image adjusting module, an image processing module, a status module and an image output module. According to the medical image texture analysis system, multiphoton microscopic images of skin injury can be qualitatively analyzed and processed, and images and curve graphs which are visual and convenient for judgment can be obtained.

Description

technical field [0001] The invention relates to the field of medical image texture analysis, in particular to a medical image texture analysis system. Background technique [0002] Image texture is an important research content in the field of pattern recognition, and has a very wide application background in the field of biomedical research. A core issue of texture analysis is texture description, that is, texture feature extraction. Studies have shown that the three most important latitudes for natural texture discrimination are periodicity, directionality, and randomness. At present, there are many texture feature extraction methods in the field of image texture analysis. , where the Fourier transform is an important signal processing texture analysis method, and the Fourier transform spectrum contains rich image information, which can roughly describe the texture mode. Extracting the features of projected energy distribution through the texture image Fourier spectrogra...

AI Technical Summary

Problems solved by technology

On the one hand, MATLAB software has a low processing rate for texture images, and the user interface is relatively simple. Users need to use manual methods and can only search and call a single function to realize texture recognition. Multiple texture analysis methods have not yet been integrated into the same human-machine A complete module with a friendly interface, which cannot directly call existing modules to analyze image textures in multiple dimensions; on the other hand, image texture analysis is usually implemented by another general software ImageJ, which has rich functions and a simple interface. However, the research and application of this project showed functional redundancy, did not fully consider the functional interface deployment and ease of use of the software, and there is no unified standardized analysis process so far, which is very unsuitable for clinical diagnosis. convenient

In addition, the image generated by the second harmonic contains a heavy calculation process. In order to facilitate real-time and effective medical evaluation and treatment in the clinic, it is urgent to improve the functional integrity of the image texture analysis system.

Images