Deep metric learning for few-shot image classification: A Review of recent developments

Li, Xiaoxu; Yang, Xiaochen; Ma, Zhanyu; Xue, Jing-Hao

doi:10.1016/j.patcog.2023.109381

Cited by 47 publications

(19 citation statements)

References 99 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to verify the performance of the proposed method for few-shot fault diagnosis, five methods are compared: 1) Finetune-Last [32] ; 2)Finetune-Whole [32] ; 3)Siamese Net [16] ; 4) Protot Net [18] ; 5) Matching Net [14] .The Adam method is used to optimize the network parameters. For the fine-tuning based models (Finetune-Last and Finetune-Whole), the fine-tuning iteration number is set to 100, and the network model is a four-layer convolutional neural network.…”

Section: Experimental Setting and Parameter Definitionmentioning

confidence: 99%

“…In this paper, the meta-learning method is used to make the network model learn general metaknowledge from the source domain and adapt to new tasks quickly. At present, many ideas based on metalearning [6] have been proposed, mainly including metric-based algorithms [7][8] 、 algorithms based on data augmentation [9] and model-based optimization algorithms [10][11][12] .Among them, metric-based metalearning is considered to be one of the simplest and effective methods, Wang et al [13] proposed a metalearning model based on feature space metrics, which combines general supervised learning and metric meta-learning [14] (Matching Net), using the similarity information of single sample and sample group to learn the fault category of unknown samples. Zhang et al [15] proposed a few-shot model based on Siamese network [16]( Siamese network) for bearing fault diagnosis.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A few-shot fault diagnosis method for aircraft engine casing

Xie

Liu

Chen

et al. 2023

Preprint

View full text Add to dashboard Cite

In recent years, the label monitoring data collected from aeroengine accessory casing system are limited, and there are differences between the training data domain and the target data domain in practical application, which leads to the degradation of the model diagnosis performance. To solve the above problems, a few-shot fault diagnosis method based on meta-learning is proposed in this paper. Firstly, the original signal is converted into time-frequency image, and the multi-channel time-frequency image is fused as the input of the network model. Secondly, an autoencoder is trained using the source data domain, and the encoder is used as the feature extraction network. Thirdly, the extracted sample features are taken as nodes, and the distance between features is taken as edge weight, and the graphs are constructed for the known samples and unknown samples. Finally, the meta-task-based graph learning method is used to learn the distribution rules of edge weights of graph nodes, and the label information of the known samples in the task is propagated to the unknown samples. The experimental results show that the proposed method outperforms the traditional small-sample fault diagnosis methods on both public datasets and aircraft engine casing datasets. Document identification code: A Chinese graphic classification number: TP315.69

show abstract

Section: Experimental Setting and Parameter Definitionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A few-shot fault diagnosis method for aircraft engine casing

Xie

Liu

Chen

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The recognition task can be completed by comparing the distance to labelled samples. This method can effectively avoid the problem of over fitting because it does not require additional parameters for new classes [19]. Different distance metrics can be used: Manhattan distance, Euclidean distance, cosine similarity, and others.…”

Section: Metric Learningmentioning

confidence: 99%

“…However, the number of support samples is limited, so they cannot adequately represent the novel class [19].…”

Section: Metric Learningmentioning

confidence: 99%

A review of few-shot image recognition using semantic information

Guo,

Marlisah,

Ibrahim

et al. 2023

View full text Add to dashboard Cite

In recent years, the utilization of deep learning techniques has been employed in the field of image recognition with the aim of improving performance. However, deep learning demands a substantial amount of labeled data for model training, a process that is both expensive and time-consuming. In order to tackle this particular difficulty, the approach of few-shot learning (FSL) has emerged as a viable alternative. FSL, or Few-Shot Learning, is a computational approach that aims to replicate the cognitive processes observed in humans. By using a small set of examples and experiences, FSL enables the acquisition of new concepts. Research in the field of FSL has investigated many approaches to extracting the highest amount of information from limited data or making use of affordable and easily accessible sources of information. Researchers have been incorporating outside data into FSL techniques more frequently. This paper conducts an in-depth exploration of leveraging semantic information to enhance few-shot learning. By reviewing papers from the last five years in WOS, IEEE, and Science Direct (some papers in arXiv are also used), this study delves into the strategies employed to bridge the gap between visual and semantic information. The review extends to encompass zero-shot learning, which is considered a subcategory of FSL, enriching the analysis. Moreover, this paper identifies the potential of employing semantic information to enhance fine-grained few-shot (FGFS) learning. Techniques such as direct projection and the application of generative adversarial networks (GANs) emerge as promising avenues to accomplish this enhancement.

show abstract

“…Various triplet selection strategies, like semi-hard triplet mining, as used on FaceNet [29], aim to strike a balance between excessively challenging and overly simplistic examples, located within a predefined margin hyperparameter. DML has been applied to tasks such as face classification, hyperspectral image classification, and few-shot classifications of diseases [25,[29][30][31][32][33][34], where it has outperformed the baseline methods.…”

Section: Introductionmentioning

confidence: 99%

Pine Wilt Disease Segmentation with Deep Metric Learning Species Classification for Early-Stage Disease and Potential False Positive Identification

Thapa,

Khanal,

Bhattarai

et al. 2024

Electronics

View full text Add to dashboard Cite

Pine Wilt Disease poses a significant global threat to forests, necessitating swift detection methods. Conventional approaches are resource-intensive but utilizing deep learning on ortho-mapped images obtained from Unmanned Aerial Vehicles offers cost-effective and scalable solutions. This study presents a novel method for Pine Wilt Disease detection and classification using YOLOv8 for segmenting diseased areas, followed by cropping the diseased regions from the original image and applying Deep Metric Learning for classification. We trained a ResNet50 model using semi-hard triplet loss to obtain embeddings, and subsequently trained a Random Forest classifier tasked with identifying tree species and distinguishing false positives. Segmentation was favored over object detection due to its ability to provide pixel-level information, enabling the flexible extension of subsequent bounding boxes. Deep Metric Learning-based classification after segmentation was chosen for its effectiveness in handling visually similar images. The results indicate a mean Intersection over Union of 83.12% for segmentation, with classification accuracies of 98.7% and 90.7% on the validation and test sets, respectively.

show abstract

Deep metric learning for few-shot image classification: A Review of recent developments

Cited by 47 publications

References 99 publications

A few-shot fault diagnosis method for aircraft engine casing

A few-shot fault diagnosis method for aircraft engine casing

A review of few-shot image recognition using semantic information

Pine Wilt Disease Segmentation with Deep Metric Learning Species Classification for Early-Stage Disease and Potential False Positive Identification

Contact Info

Product

Resources

About