Scene classification method, device and electronic equipment

Abstract

Embodiments of the present disclosure provide a scene classification method, device, and electronic equipment, which belong to the field of data processing technology. The method includes: setting type values ​​for multiple acquired images containing different scenes; The classification of each image in is predicted, and the classification prediction value of each image is obtained; based on the type value and classification prediction value of each image, the minimum objective function is constructed, and the minimum objective function is used to determine the performance of different classification scenarios. Weight value; Based on the calculation function containing the weight value of different classification scenes, classify and predict the newly input image. Through the solution disclosed in the present disclosure, the accuracy of picture scene classification is improved.

Description

technical field [0001] The present disclosure relates to the technical field of data processing, and in particular to a scene classification method, device and electronic equipment. Background technique [0002] Scene recognition or scene classification, as a basic task in the field of understanding, has been applied to many fields. The classification of scenes is to simulate the visual perception of scenes according to the way of human thinking, and to distinguish various target scenes. Human visual perception scenes and machine vision are similar in semantics. The semantic information in the scene not only contains the prior information of biological vision, but also explains the environmental context. This semantic understanding provides a basis for future scene understanding and perception. theoretical basis. [0003] Scene classification corresponds to the classification problem in pattern recognition. It is a typical supervised learning method. Supervised learning is...

AI Technical Summary

Problems solved by technology

[0004] In the prior art, when performing scene classification on images, the accuracy of prediction is usually not high due to the difference in the number of samples of different categories in the training samples, which affects the accuracy of scene classification calculations.

Images