Image fine-grained classification method and device, storage medium and equipment

Abstract

The embodiment of the invention discloses an image fine-grained classification method and device, a storage medium and equipment. The method comprises the steps of acquiring at least two sample images; performing feature extraction on the at least two sample images to obtain feature information corresponding to the at least two sample images; determining first label values corresponding to the atleast two sample images; constructing a preset measurement model according to the obtained feature information and the first label value, wherein the preset measurement model represents a measurementmodel for optimizing the distance between at least two pieces of feature information; combining the preset measurement model with a pre-trained preset classification model to obtain a target classification model, wherein the target classification model is used for realizing fine-grained classification of the to-be-processed image.

Description

technical field [0001] The present application relates to the technical field of image classification, and in particular to an image fine-grained classification method, device, storage medium and equipment. Background technique [0002] The difference between fine-grained image classification and general image classification tasks is that the granularity of the category to which the image belongs is finer, and the difference between different fine-grained object categories is only reflected in the subtleties. Due to the influence of conditions such as environment, location, background and shape, even objects of the same category may have large intra-class visual differences. For example, dogs of different breeds belong to the broad category of dogs, so the differences between categories are small; however, due to the diversity of conditions such as background and appearance, there are still large differences between categories. [0003] The current solution is to improve on...

AI Technical Summary

Problems solved by technology

For example, different breeds of dogs all belong to the category of dogs, so the differences between the categories are small; however, due to the diversity of conditions such as background and appearance, there are still large differences between the categories

[0003] The current solution is to improve on the basis of the general classification algorithm, but the general classification algorithm does not have a high degree of distinction between classes for fine-grained classification, resulting in a relatively short center distance between inter-class features, and insufficient clustering of intra-class features. There is overlap in the feature distribution between multiple categories, which is easy to cause misclassification between categories; in addition, the calculation in the current scheme is more complicated, and it will bring more time delay

Images