A Multi-Layer Feature Fusion Method for Few-Shot Image Classification

Abstract

In image classification, few-shot learning deals with recognizing visual categories from a few tagged examples. The degree of expressiveness of the encoded features in this scenario is a crucial question that needs to be addressed in the models being trained. Recent approaches have achieved encouraging results in improving few-shot models in deep learning, but designing a competitive and simple architecture is challenging, especially considering its requirement in many practical applications. This work proposes an improved few-shot model based on a multi-layer feature fusion (FMLF) method. The presented approach includes extended feature extraction and fusion mechanisms in the Convolutional Neural Network (CNN) backbone, as well as an effective metric to compute the divergences in the end. In order to evaluate the proposed method, a challenging visual classification problem, maize crop insect classification with specific pests and beneficial categories, is addressed, serving both as a test of our model and as a means to propose a novel dataset. Experiments were carried out to compare the results with ResNet50, VGG16, and MobileNetv2, used as feature extraction backbones, and the FMLF method demonstrated higher accuracy with fewer parameters. The proposed FMLF method improved accuracy scores by up to 3.62% in one-shot and 2.82% in five-shot classification tasks compared to a traditional backbone, which uses only global image features.

Enhancing Few-Shot Learning in Lightweight Models via Dual-Faceted Knowledge Distillation

“…According to the process setup of the few-shot image classification task [ 33 ], the given image dataset is randomly partitioned into three subsets, i.e., the base class dataset , the validation dataset , and the novel class dataset . These three subsets, which have different class labels, are used for the training of the model, the validation of the model, and the few-shot testing, respectively.…”

Advanced Technologies and Artificial Intelligence in Agriculture

“…No research has been found where disease classification is conducted on a self-collected dataset comprising more than 70 classes. However, the research by Cui et al [156] and Gomes et al [157] employs very interesting techniques that deserve mention. The same classification approach was utilized in a joint project with the Temiryazev Academy as part of the World-class Scientific Center "Agrotechnologies of the Future" [158].…”