Method for improving image classification accuracy by using partition decision mechanism

Abstract

The invention discloses a method for improving image classification accuracy by using a partition decision mechanism, and the method comprises the steps: enabling a model to recognize different regions of an image based on an ensemble learning thought by using the partition decision mechanism, summarizing a plurality of recognition results, and deducing the category of the whole image. The invention provides a model improvement method capable of stably and reliably improving the image classification accuracy, the model training process is simple, the accuracy of the convolutional neural network model during image classification is improved, and meanwhile, excessive extra operation overhead is not brought to training.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence computer vision image recognition, and more particularly to a method for improving the accuracy of image classification by using a partition decision mechanism. Background technique [0002] Image recognition refers to the use of computers and other equipment to process and analyze images, extract image features, and complete tasks such as classification, target detection, and matching. Image recognition is an important research direction in the field of computer vision. With the development of artificial intelligence technology in recent years, more and more methods and application results have emerged. Image classification is an important sub-task in the field of image recognition. Many computer vision tasks are based on image classification. For example, a core problem of target detection is how to correctly identify the category of sub-images in the detection frame. At present...

AI Technical Summary

Problems solved by technology

In different data sets and application scenarios, it is difficult to guarantee the improvement effect of the improved method on the accuracy of the model, and it may even be lower than the accuracy of the original model;

[0004] 2) Poor portability

Most improvement methods are mutually exclusive and cannot be used at the same time, resulting in the validity of a study often being based on the negation of other studies;

[0005] 3) Poor interpretability

Many improvement methods essentially rely on the stacking of computing power and data scale, the improvement effect is difficult to be reasonably explained, and may lead to an increase in the cost of system operation

Images