Annotation data generation method and device and computer readable storage medium

Abstract

The invention provides an annotation data generation method and device and a computer readable storage medium. The method comprises following steps of S100: acquiring a data universal set and annotated annotation data sets included in the data universal set; S200: analyzing data characteristics of the annotation data sets and manufacturing pseudo-data sets satisfying the data characteristics according to the data characteristics; S300: based on a GAN neural network, expanding the pseudo-data sets so as to form an expanded data sets; S400: identifying whether data in the expanded data sets needs to be annotated, and screening annotated data so as to form a training data set; S500: carrying out neural network trainings on the training data set so as to form a training model; and S600: basedon the training model, cleaning data in the data universal set except for the data in the annotation data sets, annotating data satisfying the training model and arranging the data in the annotation data sets. Thus, based on little data, a training data set which highly matches the sample data and is highly random can be rapidly and highly efficiently generated, and data quantity of the annotationdata is improved.

Description

technical field [0001] The present invention relates to the field of data models, in particular to a method, device and computer-readable storage medium for generating labeled data. Background technique [0002] With the rapid development of applications on smart terminals, and artificial intelligence technology based on applications, it has entered people's lives more and more widely. Whether it is daily use, games, work, etc., it will need to learn based on original sample data to understand the usage habits in this field, so as to make intelligent judgments. [0003] For the learning of original sample data, deep neural network technology can be used. The deep neural network technology has developed rapidly in recent years, and has achieved far-beyond-expected accuracy in the field of image recognition, and has achieved gratifying applications in many fields. However, in practical engineering applications, many special image recognition requirements lack training data s...

AI Technical Summary

Problems solved by technology

[0004] 1. The length and width of the original image data corresponding to some models are small, and the amount of data that can be expanded by random cropping is limited

[0005] 2. When the original sample data is small, the data obtained by these methods is easy to overfit the model because the characteristics are not scattered enough

[0006] 3. Some models are sensitive to data stretching, and the recognition rate decreases significantly after stretching;

[0007] 4. Manually collecting and labeling data will consume a lot of manpower and energy

Images