Fractal and Recognition Method of Animal Disease Image Based on Generalized Structure Element

  • Dongfang Lv
Keywords: Generalized Structural Elements, Animal Germs, Image Segmentation, Digital Image Processing

Abstract

For the analysis and research of animal pathogens, manual classification and statistics are mainly
used at present. The labor intensity is large, the work efficiency is low, and it is greatly influenced by human
beings. It is not objective and does not meet the requirements of current standardization. Based on the above
background, the purpose of this paper is to study the fractal and recognition methods of animal pathogen images
based on generalized structural elements. In this paper, the idea of generalized structural element is introduced
into the detection and identification of animal pathogenic images, and three generalized structural elements are
constructed to perform extended morphological filtering on the three animal pathogenic images. According to
the constructed generalized structure element, the red circular pattern, blue circular pattern and black circular
pattern are detected respectively. This paper first introduces several calculation methods of fractal dimension,
and finds a scheme to calculate the fractal dimension of animal pathogen images. Then grayscale, binarization
and histogram equalization are carried out on the animal germ images, and the animal germ images are
enhanced in order to highlight their texture features. Finally, the fractal theory is used to perform segmentation
processing on the processed image to extract the features of animal pathogenic images. The experimental results
show that the data obtained by using the pixel method for rough statistics has a larger error than the actual data,
and there is a difference of more than 120 germs; while the method using the generalized structural element
method is very close to the actual results.

Published
2020-03-01