Gingivitis Detection by Fractional Fourier Entropy and Standard Genetic Algorithm

Abstract

Early detection of gingivitis is crucial for oral health. Dental diagnosis and treatment require very high standards of care and a great deal of experience. To reduce the diagnostic difficulties of dentists, this work proposes a classification method for gingivitis based on fractional Fourier entropy and standard genetic algorithm. The fractional Fourier transform was used to extract the eigenvector fractional Fourier entropy, and the eigencoefficient was substituted into the classifier of standard genetic algorithm for the classification of healthy gingival images and pathological gingival images. Experimental results show that this approach has better performance than existing methods and is effective in image classification of gingivitis.