Huang, D., 2018. Intelligent recognition method of micro image feature recognition in large data environment. In: Liu, Z.L. and Mi, C. (eds.), Advances in Sustainable Port and Ocean Engineering. Journal of Coastal Research, Special Issue No. 83, pp. 697–705. Coconut Creek (Florida), ISSN 0749-0208.

The intelligent recognition of features of micro image can effectively improve the recognition rate of the essential features of micro images, which has great significance to the study of micro images in various fields. For the recognition of large data features, we need to fuse the features of the micro images with the same scale and different directions, and then preprocess them completing the intelligent recognition of the micro image features. The traditional method filters the noise of the micro image to get the smooth image, and obtain the features of illumination invariant data of the micro image, but we ignore to preprocess the image, which leads to low recognition accuracy. This paper proposes the intelligent identification method of the features of micro images based on the theory of clonal selection and K nearest neighbor discrimination theory in large environment. Firstly, this method performed Gabor transform on the micro image, and fused features of micro images with the same scale in different directions according to the two fusion rules, and obtained the fusion feature of micro image on the five scales, and used the clonal selection theory to preprocess the fusion feature of micro image, forming antibody library of micro images, and combined with K nearest neighbor discriminant theory to complete the intelligent recognition of feature of micro image to be identified. Experimental results show that the proposed method can effectively improve the average recognition rate of micro image features, and has good robustness.

This content is only available as a PDF.
You do not currently have access to this content.