A self-adaptive weight deep learning target classification method based on feature fusion

Abstract

The invention provides a self-adaptive weight deep learning target classification method based on feature fusion. The method comprises performing target coarse detection; extracting image convolutionfeatures and HOG features, and performing dimension expansion processing on the HOG features; embedding the SENet into a Resnet network framework, and establishing a network framework for extracting multi-feature weights of the image; calculating adaptive weight vectors of the convolutional features and the HOG features, making a feature fusion strategy, and calculating image fusion features; andestablishing a multi-target classification framework based on the precise binary classification network set. According to the method, image convolution features and HOG features are fused, adaptive weight vectors of the image features are extracted, deep learning network configurations and parameters are designed, an accurate classification network is constructed, the network obtains more candidate frames by reducing a score threshold value, and the recall rate of target detection is increased; by designing a plurality of binary classification networks, the method has higher accuracy in termsof multi-classification problems.

Description

technical field [0001] The invention relates to a deep learning object classification method, in particular to an adaptive weight deep learning object classification method based on feature fusion, which belongs to the technical field of image recognition. Background technique [0002] Object classification technology is widely used in many fields. In recent years, the field of artificial intelligence has developed rapidly. Object classification technology has become an indispensable technical foundation in the field of artificial intelligence. Object classification can provide important information sources for video surveillance and automatic driving. For example, through object classification, it is provided whether there are pedestrians, vehicles, and buildings in the image. It can be said that accurate object classification technology is a technical bottleneck that needs to be solved in many fields. In the early days, people often used hand-designed features to extract i...

AI Technical Summary

Problems solved by technology

In the past, some scholars used to combine convolutional features with HOG features, often extracting one of the features first, and then extracting another feature on this basis, and classifying it through the support vector machine, but there are two problems in this method: first, The process of extracting one of the features will inevitably weaken the other feature; secondly, this process does not change the influence weight and loss function of each feature, and does not take into account that different features have different gains in classification accuracy

Therefore, the classification effect of the previous method is not ideal.

Images