Fine-grained image recognition method, convolutional neural network and training method thereof

Abstract

The invention relates to a fine-grained image recognition method, a convolutional neural network and a training method thereof, computer equipment and a computer readable storage medium. The method comprises the steps of obtaining an original image; extracting global features of the original image by adopting a convolutional network; determining a plurality of candidate regions in the original image, and determining a feature value of each candidate region according to the global feature; sorting the plurality of feature values to obtain a first sorting result, and determining at least one candidate region corresponding to the maximum N feature values in the first sorting result; extracting features of the at least one candidate region to obtain at least one local region feature, each candidate region corresponding to one local region feature; cascading the global feature and the at least one local region feature to obtain a cascading feature; classifying the original images accordingto the cascade features to obtain the classification result, so that the problems of complexity and high labeling cost of a fine-grained image recognition method are solved, and the fine-grained imagerecognition method is simplified.

Description

technical field [0001] The present application relates to the fields of computer vision technology and deep learning technology, and in particular to a fine-grained image recognition method, a convolutional neural network, a training method for a convolutional neural network, computer equipment, and a computer-readable storage medium. Background technique [0002] The goal of fine-grained image recognition is to classify object subcategories at a fine-grained level. Since the differences between different subcategories are very subtle, fine-grained image recognition is very challenging. Compared with traditional image classification, the differences and difficulties of fine-grained image recognition are: [0003] (1) The difference between classes is very subtle, such as different subclasses of birds or different subclasses of cars, the differences between objects of different subclasses are mainly reflected in local details; (2) For fine-grained images, intra-class differen...

AI Technical Summary

Problems solved by technology

[0008] Embodiments of the present application provide a fine-grained image recognition method, a convolutional neural network, a training method for a convolutional neural network, a compute...

Images