An image classification training method with incremental learning in big data scenarios

Abstract

The invention discloses an image classification training method capable of incremental learning in a big data scene. The method includes: step 1, using original image data to train an initial image classifier; step 2, if a new category image appears during the application process, Incremental training is then performed on the initial model to obtain an updated image classifier; step 3, using the trained image classifier to identify the image to be classified to obtain the test image category. The method of the invention uses a convolutional neural network to extract features from an image, can avoid manually defining features, and has high recognition; uses the neural network to identify images, occupies less memory, and has a fast calculation speed. The image classification training method of the present invention can carry out incremental learning for new categories of images without saving the original training data, which can save a lot of training time and storage space, and thus is especially suitable for big data image classification scenarios.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to an image classification training method capable of incremental learning in a big data scene. Background technique [0002] In recent years, the field of computer vision is in a period of rapid development. As an important technology in this field, image classification is designed to allow computers to replace humans to process a large amount of physical information, thereby automatically identifying the subject category in the image. [0003] At present, there are related patents in the field of image classification, such as the automatic image classification method proposed by the patent CN103577475A, which includes the following steps: receiving the image to be classified; reading the feature categories in the feature library; Classify the feature data of the picture; match the extracted feature data with the preset feature data corresponding to the feature category, and merge th...

AI Technical Summary

Problems solved by technology

If the original image data is required for each new category image, it will require a lot of training time and storage, and it cannot be implemented on a machine with lower performance.

Images