Holographic diffraction label image recognition algorithm based on double tensors

Abstract

The invention discloses a holographic diffraction label image recognition algorithm based on double tensors, which comprises the following steps: S11, acquiring an original image of a holographic diffraction label, the original image comprising an original image of a training sample and an original image of a test sample; S12, preprocessing the obtained original image of the test sample to generate an HSV tensor and extract an HOG tensor; preprocessing the obtained original image of the training sample to generate an HSV tensor and extract an HOG tensor; s13, combining the obtained HSV tensorand HOG tensor of the test sample into double tensors; combining the obtained HSV tensor and HOG tensor of the training sample into double tensors; measuring the similarity between different decomposition matrixes of the training sample and the test sample through canonical correlation analysis; and S14, classifying the similarities between the different decomposition matrixes by using a nearest neighbor algorithm. According to the method, the similarity vectors of different decomposition matrixes are projected to the PCA subspace for KNN classification, so that the recognition performance between different samples is effectively improved.

Description

technical field [0001] The invention relates to the technical field of computer image recognition, in particular to a double tensor-based holographic diffraction label image recognition algorithm. Background technique [0002] With the development of multimedia and printing technology, digital images have been widely used in science, industry, commerce and other fields. In the medical field, digital image storage technologies (such as CT (Computed Tomography) and MRI (Magnetic Resonance Imaging)) are used to image images on solid film, and automatic diagnosis is realized through content-based image information retrieval [1]; law enforcement agencies Use digital images to record the facial features and fingerprint information of victims and suspects, as well as record crime scenes, and conduct image similarity analysis in the database to quickly find criminals; the meteorological department obtains satellite remote sensing images to analyze the spectral characteristics of ima...

AI Technical Summary

Problems solved by technology

Although the above-mentioned patent can identify the authenticity of commodities or bills based on barcodes, it uses traditional image algorithms

Due to the physical characteristics of holographic diffraction tags that change with light, traditional image recognition algorithms cannot obtain stable recognition results

Images