Recognition Of Cancer Cells Based On Gray Level Co-occurrence Matrix

With the development of robotics, atomic force microscopy is widely used in biomedical research because of high resolution imaging, especially the study of cells at the nanoscale. Gray level co-occurrence matrix feature parameters are the main characteristic parameters to identify normal cells and cancer cells, and they are the main research parts of the paper.Based on the analysis and research of image segmentation algorithm, image feature extraction method and cell identification technology at home and abroad, the paper presents a cancer cell recognition method based on gray level co-occurrence matrix. Firstly, the paper uses the medium filter to achieve cell image processing. Secondly, morphological operations, global threshold segmentation, Otsu algorithm and watershed algorithm are used to achieve cell image segmentation. By comparing the analysis, the optimum segmentation can be obtained when the global threshold is 0.3. Then cell height and cell surface roughness are extracted, and cell texture is extracted by gray level co-occurrence matrix. Comparative analysis the biological characteristics between cancer cells and normal cells is regarded as the basis for cancer cell discrimination. Finally, the curve fitting and bar diagram in MATLAB are used to show the species of testing cell.It is proved that cancer cell texture feature extraction by gray level co-occurrence matrix and cells identification method through the MATLAB can accurately identify cancer cells.