Weak supervision fine-grained image classification method of multi-branch neural network model

Abstract

The invention discloses a weak supervision fine-grained image classification method of a multi-branch neural network model. The weak supervision fine-grained image classification method is characterized by the steps of: firstly, randomly dividing a fine-grained image data set into a training set and a test set in proportion; secondly, positioning a local region with potential semantic informationby using a local region positioning network; thirdly, respectively inputting an original image and the positioned local region into a deformable convolution residual network and a rotation invariant coding direction response network to form a feature network of three branches, respectively training the three branches, and respectively carrying out backward propagation learning on the three branches based on cross entropy loss; and finally, combining intra-branch loss and inter-branch loss to optimize the whole network, and performing classification prediction on the test set. According to theweak supervision fine-grained image classification method, the negative influence of various changes such as posture, visual angle and background interference on a classification result is reduced, and a better effect is achieved on a fine-grained image classification task.

Description

technical field [0001] The invention belongs to the field of fine-grained image classification, in particular to a weakly supervised fine-grained image classification method of a multi-branch neural network model. Background technique [0002] Fine-grained image classification is an important branch of computer vision, and it is of great significance in both military and civilian fields. Its goal is to classify images belonging to the same basic category (such as birds, dogs, airplanes, etc.) Subclassing. Compared with traditional image classification tasks, the differences between fine-grained image classes are more subtle, and often only small local differences can be used to distinguish different classes. At the same time, there are many changes in the object's posture, viewing angle, occlusion, and background interference, resulting in huge differences within the class. These factors bring great difficulties to fine-grained image classification. [0003] Most of the ea...

AI Technical Summary

Problems solved by technology

The disadvantage is that it focuses on a multi-scale local area, and there is a large redundancy between features

[0005] To sum up, the above methods still have some limitations, the main difficulties are: 1) how to effectively focus on the latent semantic regions and locate discriminative foreground objects; 2) if the non-rigid structural objects have large shape changes , how to extract rich feature information; 3) how to reduce the impact on classification due to many changes in posture, perspective and background interference

Images