Label automatic definition method based on image content

Abstract

The invention discloses a label automatic definition method based on image content, and belongs to the field of image processing and natural language processing, and particularly relates to an automatic image labeling method based on image mutual information analysis. For problems in the prior art that accurate labeling cannot be realized, and the model is influenced by wrong annotation information, the technical scheme is that the method comprises the steps of firstly generating a pre-training model, then crawling pictures in the Internet, putting the crawled pictures into the generated pre-training model to output corresponding labels, assessing the labels output by the pre-training model through a space vector, and finally updating the model and a vector space according to obtained newdata. According to the method, the model can be automatically updated according to the network data, the annotation is more accurate through mutual learning of the annotation information between the images, and the model can be prevented from being influenced by the wrong annotation information through a special label evaluation module.

Description

technical field [0001] The method belongs to the fields of image processing and natural language processing, and in particular relates to an automatic label definition method based on image content. Background technique [0002] Image annotation is usually a technology with practical value in the field of image processing. It is widely used in image retrieval and recommendation systems. Its practical scenarios include image search by image, precise positioning of advertisements, image pornography and piracy detection, etc. Traditional methods usually extract the information of the image itself for image annotation. Due to the explosive growth of images on the Internet, it is impossible to annotate new images and new information. [0003] At present, the automatic image labeling technology adopts the method of probability and statistics in the early stage. For example, in the document "Mori Y, Takahashi H, Oka R. Image-to-word transformation based on dividing and vector quant...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problems in the prior art that accurate labeling is impossible and incorrect labeling information affects the model, the present invention provides an automatic label definition method based on image content, the purpose of which is to automatically update the model according to network data, and to Labeling information among them, learning from each other, making their labeling more accurate, and the unique label evaluation module can prevent wrong labeling information from affecting the model

Images