Vessel image compression and dimension reduction method of venipuncture blood sampling robot

Abstract

The invention relates to a vessel image compression and dimension reduction method for a venipuncture blood sampling robot, and the method specifically comprises the following steps: S1, obtaining anRGB image of a hand vein and a preset proportion, converting the RGB image into a gray scale image, and completing the gray scale image into a square matrix; S2, standardizing a square matrix, settingeach column as a feature, and calculating a covariance matrix of each feature; S3, calculating the eigenvalues and eigenvectors of the covariance matrix through eigenvalue decomposition, sorting theeigenvalues, accumulating the eigenvalues from front to back according to a sorting result, calculating an eigenproportion, comparing the eigenproportion with a preset proportion, and recording the number of the eigenvalues when the eigenproportion exceeds the preset proportion as a target number; and S4, forming a transformation matrix by taking the eigenvectors of the eigenvalues of the front target number in the sequencing result as column vectors, and combining the transformation matrix with the features and the square matrix to form a target dimension reduction matrix. Compared with the prior art, the method has the advantages of improving the recognition speed of the hand vein image, improving the stability of blood vessel image recognition and the like.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a blood vessel image compression and dimensionality reduction method for a venipuncture blood collection robot. Background technique [0002] In the context of the rapid development of artificial intelligence image recognition technology, vein recognition is in a unique and important position. From vein identity authentication to fully automatic venipuncture, more efficient, safe and high-end technologies have promoted the development of vein image recognition technology. [0003] In the process of automatic blood collection by venipuncture and blood collection robots, convolutional neural networks are usually used for image recognition of hand dorsal vein images. In the process, it is necessary to analyze each frame of image collected by a near-infrared (NIR) camera, perform semantic segmentation on the image, and mark the position of the blood vessel. Therefore, in t...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a blood vessel image compression and dimensionality reduction method for a venipuncture and blood collection robot in order to overcome the defect in the above-mentioned prior art that the image recognition speed does not match the image acquisition speed due to the lack of dimensionality reduction of pictures.

Images