An Attention Cascade Global–Local Network for Remote Sensing Scene Classification

Shen, Junge; Yu, Tianwei; Yang, Haopeng; Wang, Ruxin; Wang, Qi

doi:10.3390/rs14092042

Cited by 26 publications

(16 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(2022) and Shen et al. (2022a, b) proposed three CNN architectures obtained by neural architecture search. Speaking of advantages, these four methods both have fewer parameters compared to the off-the-shelf CNNs developed on ImageNet-1K.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

MBC-Net: long-range enhanced feature fusion for classifying remote sensing images

Song

2023

IJICC

View full text Add to dashboard Cite

PurposeClassification of remote sensing images (RSI) is a challenging task in computer vision. Recently, researchers have proposed a variety of creative methods for automatic recognition of RSI, and feature fusion is a research hotspot for its great potential to boost performance. However, RSI has a unique imaging condition and cluttered scenes with complicated backgrounds. This larger difference from nature images has made the previous feature fusion methods present insignificant performance improvements.Design/methodology/approachThis work proposed a two-convolutional neural network (CNN) fusion method named main and branch CNN fusion network (MBC-Net) as an improved solution for classifying RSI. In detail, the MBC-Net employs an EfficientNet-B3 as its main CNN stream and an EfficientNet-B0 as a branch, named MC-B3 and BC-B0, respectively. In particular, MBC-Net includes a long-range derivation (LRD) module, which is specially designed to learn the dependence of different features. Meanwhile, MBC-Net also uses some unique ideas to tackle the problems coming from the two-CNN fusion and the inherent nature of RSI.FindingsExtensive experiments on three RSI sets prove that MBC-Net outperforms the other 38 state-of-the-art (STOA) methods published from 2020 to 2023, with a noticeable increase in overall accuracy (OA) values. MBC-Net not only presents a 0.7% increased OA value on the most confusing NWPU set but also has 62% fewer parameters compared to the leading approach that ranks first in the literature.Originality/valueMBC-Net is a more effective and efficient feature fusion approach compared to other STOA methods in the literature. Given the visualizations of grad class activation mapping (Grad-CAM), it reveals that MBC-Net can learn the long-range dependence of features that a single CNN cannot. Based on the tendency stochastic neighbor embedding (t-SNE) results, it demonstrates that the feature representation of MBC-Net is more effective than other methods. In addition, the ablation tests indicate that MBC-Net is effective and efficient for fusing features from two CNNs.

show abstract

Section: Related Workmentioning

confidence: 99%

“…E.g., Shen et al. (2022a, b) and Tang et al. (2021) proposed two different methods by fusing features from two CNNs.…”

Section: Related Workmentioning

confidence: 99%

MBC-Net: long-range enhanced feature fusion for classifying remote sensing images

Song

2023

IJICC

View full text Add to dashboard Cite

show abstract

“…For instance, Xu et al (2022aXu et al ( , 2022bXu et al ( , 2022cXu et al ( , 2022d and Li et al (2022) introduced cascaded multi-model methods by integrating a CNN with either a graph convolutional network or deep-gated recurrent units. In a similar vein, some researchers (Shen et al, 2022b;Tang et al, 2021;Wang et al, 2022b;Xu et al, 2022aXu et al, , 2022bXu et al, , 2022cXu et al, , 2022d proposed parallel multi-model methods that incorporate two CNNs. Moreover, some other researchers (Deng et al, 2022;Ma et al, 2022;Wang et al, 2023aWang et al, , 2023bYang et al, 2023;Zhang et al, 2021;Zhao et al, 2023) proposed more complex multimodel methods by combining CNNs and ViTs.…”

Section: Related Workmentioning

confidence: 99%

Efficient knowledge distillation for remote sensing image classification: a CNN-based approach

Song,

Wei,

Yong

2023

IJWIS

View full text Add to dashboard Cite

Purpose The paper aims to tackle the classification of Remote Sensing Images (RSIs), which presents a significant challenge for computer algorithms due to the inherent characteristics of clustered ground objects and noisy backgrounds. Recent research typically leverages larger volume models to achieve advanced performance. However, the operating environments of remote sensing commonly cannot provide unconstrained computational and storage resources. It requires lightweight algorithms with exceptional generalization capabilities. Design/methodology/approach This study introduces an efficient knowledge distillation (KD) method to build a lightweight yet precise convolutional neural network (CNN) classifier. This method also aims to substantially decrease the training time expenses commonly linked with traditional KD techniques. This approach entails extensive alterations to both the model training framework and the distillation process, each tailored to the unique characteristics of RSIs. In particular, this study establishes a robust ensemble teacher by independently training two CNN models using a customized, efficient training algorithm. Following this, this study modifies a KD loss function to mitigate the suppression of non-target category predictions, which are essential for capturing the inter- and intra-similarity of RSIs. Findings This study validated the student model, termed KD-enhanced network (KDE-Net), obtained through the KD process on three benchmark RSI data sets. The KDE-Net surpasses 42 other state-of-the-art methods in the literature published from 2020 to 2023. Compared to the top-ranked method’s performance on the challenging NWPU45 data set, KDE-Net demonstrated a noticeable 0.4% increase in overall accuracy with a significant 88% reduction in parameters. Meanwhile, this study’s reformed KD framework significantly enhances the knowledge transfer speed by at least three times. Originality/value This study illustrates that the logit-based KD technique can effectively develop lightweight CNN classifiers for RSI classification without substantial sacrifices in computation and storage costs. Compared to neural architecture search or other methods aiming to provide lightweight solutions, this study’s KDE-Net, based on the inherent characteristics of RSIs, is currently more efficient in constructing accurate yet lightweight classifiers for RSI classification.

show abstract

“…The method optimizes the pre-trained model by adjusting the data side and achieves an effective improvement in classification performance. Shen et al [13] proposed a method to combine dual model features by bilinear fusion, which improves the scale adaptation of the model and improves the ability of the model to resist the effect of complex background redundancy. These innovative classification efforts for complex scenes do optimize the performance of the models, but there are still some potential problems.…”

Section: Introductionmentioning

confidence: 99%

Optimizing Remote Sensing Image Scene Classification Through Brain-Inspired Feature Bias Estimation and Semantic Representation Analysis

2023

View full text Add to dashboard Cite

In the realm of remote sensing image classification and detection, deep learning has emerged as a highly effective approach, owing to the remarkable advancements in object perception models and the availability of abundant annotated data. Nevertheless, for specific remote sensing image scene classification tasks, obtaining diverse and large amounts of data remains a daunting challenge, leading to limitations in the applicability of trained models. Consequently, researchers are increasingly focusing on optimal data utilization and interpretability of learning. Drawing inspiration from brain neural perception research, researchers have proposed novel approaches for deeper interpretation and optimization of deep learning models from diverse perspectives. In this paper, we present a brain-inspired network optimization model for remote sensing image scene classification, which considers both shape and texture features and reconstructs feature scaling of data through feature bias estimation. The model achieves greater robustness through complementary training. We evaluate our optimized model on general datasets by integrating it into an existing benchmark method and compare its performance with the original approach. Our results demonstrate that the proposed model is highly effective, with dynamically reconstructed data leading to a significant enhancement of model learning.INDEX TERMS Remote sensing image, scene classification, brain-inspired learning, feature bias, data enhancement.

show abstract

An Attention Cascade Global–Local Network for Remote Sensing Scene Classification

Cited by 26 publications

References 55 publications

MBC-Net: long-range enhanced feature fusion for classifying remote sensing images

MBC-Net: long-range enhanced feature fusion for classifying remote sensing images

Efficient knowledge distillation for remote sensing image classification: a CNN-based approach

Optimizing Remote Sensing Image Scene Classification Through Brain-Inspired Feature Bias Estimation and Semantic Representation Analysis

Contact Info

Product

Resources

About