A feature selection method for breast x-ray images based on bfba and elm

Abstract

The present invention is a mammary X-ray image feature selection method based on BFBA and ELM, which relates to the technical field of image processing. The problem of falling into a local optimal solution; the technical solution adopted is as follows: the first step is to collect the data set MIAS used; the second step is to set the BFBA parameters; the third step is to initialize the bat population; the fourth step is to The position encoding of bats generates corresponding feature subsets; the fifth step is to update the search pulse frequency, speed and position of each bat; the sixth and seventh steps are to generate uniformly distributed random numbers rand; the eighth step is to adapt to all bats Degree values ​​are sorted to find the current optimal solution and the optimal value; Step 9, determine whether the optimal solution has changed; Step 10, determine whether stagnant_count is equal to stagnant_max; Step 11, repeat Step 4 to The tenth step; the twelfth step, output the global optimal value and optimal solution.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a mammogram X-ray image feature selection method based on Bird Flock Bat Algorithm (BFBA) and Extreme Learning Machine (Extreme Learning Machine, ELM). Background technique [0002] Breast disease is one of the common diseases in women. At the same time, the multiple and harmful effects of breast cancer seriously affect women's health and even life. Therefore, early diagnosis of breast disease is directly related to women's personal health. Especially for breast cancer, people still can't fully determine its pathogenesis. The current clinical diagnosis methods of breast cancer mainly include touch diagnosis, histological diagnosis, cytological diagnosis and imaging diagnosis. Imaging is widely used for its convenience, scientificity and relatively high operability in diagnosis. Mammography is the most common technique for early diagnosis of breast cancer. For t...

AI Technical Summary

Problems solved by technology

[0004] The present invention overcomes the deficiencies in the prior art, and the technical problem to be solved is to provide a feature selection method for mammary gland X-ray images, which avoids the "exponential explosion" problem encountered by the dynamic programming method, and the analytical method or bat algorithm is easy to fall into local optimal solution problem

Images