Gene image processing estimation method, system, medium and equipment based on deep learning

Abstract

The invention discloses a gene image processing method, system, medium and equipment based on deep learning, and the method comprises the steps: obtaining an original image, and building and marking the original image as a sample set; dividing the sample set into a training sample and a test sample, obtaining and extracting a pre-selected model, and obtaining an image prediction model according tothe transfer learning training sample; calculating the test sample according to the image model to obtain a prediction calculation result; and obtaining an actual monitoring result, comparing the actual monitoring result with the prediction calculation result to obtain index calculation information, and calculating performance index data according to the index calculation information. The technical problems that in the prior art, too many noise pictures exist, the model performance is poor, and the picture marking effect is poor are solved.

Description

technical field [0001] The present invention relates to an image processing method, in particular to a genetic image processing estimation classification method, system, medium and equipment based on deep learning. Background technique [0002] In the process of gene expression analysis, the spatiotemporal expression forms of genes in different developmental processes are extremely important for understanding the function of genes and the development mechanism of embryos. Drosophila is easy to raise, has strong fecundity, small number of chromosomes, and mutation traits. Many and other characteristics have become classic materials for genetics research. Similar to target detection, each label corresponds to a certain picture or the local position of the picture, but this local correspondence is not explicitly marked in the database. This increases the difficulty of implementing automatic annotation algorithms for Drosophila biological images. The automatic annotation of Dr...

AI Technical Summary

Problems solved by technology

This assumption will affect the predictive performance of the model and introduce many noisy pictures), and there is no deep model for automatic labeling of Drosophila images at the gene level.

[0003] To sum up, the existing technology has technical problems such as too many noisy pictures, poor model performance and poor picture labeling effect

Images