Research on Image Classification of Lightweight Convolutional Neural Network

Yuan, Zhengwu; Lu, Chenxing

doi:10.1109/icbaie52039.2021.9389911

Cited by 2 publications

(3 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared to existing research, light CNN sacrifices some model accuracy to reduce the consumption of limited memory and computational resources. Therefore, the challenge of maintaining high accuracy with fewer parameters has become an urgent issue [31]. Multiscale feature extraction provides an effective solution to retain more relevant information within a limited number of parameters.…”

Section: Introductionmentioning

confidence: 99%

MADANet: A Lightweight Hyperspectral Image Classification Network with Multiscale Feature Aggregation and a Dual Attention Mechanism

Cui,

Wen,

Song

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

Hyperspectral remote sensing images, with their continuous, narrow, and rich spectra, hold distinct significance in the precise classification of land cover. Deep convolutional neural networks (CNNs) and their variants are increasingly utilized for hyperspectral classification, but solving the conflict between the number of model parameters, performance, and accuracy has become a pressing challenge. To alleviate this problem, we propose MADANet, a lightweight hyperspectral image classification network that combines multiscale feature aggregation and a dual attention mechanism. By employing depthwise separable convolution, multiscale features can be extracted and aggregated to capture local contextual information effectively. Simultaneously, the dual attention mechanism harnesses both channel and spatial dimensions to acquire comprehensive global semantic information. Ultimately, techniques such as global average pooling (GAP) and full connection (FC) are employed to integrate local contextual information with global semantic knowledge, thereby enabling the accurate classification of hyperspectral pixels. The results from the experiments conducted on representative hyperspectral images demonstrate that MADANet not only attains the highest classification accuracy but also maintains significantly fewer parameters compared to the other methods. Experimental results show that our proposed framework significantly reduces the number of model parameters while still achieving the highest classification accuracy. As an example, the model has only 0.16 M model parameters in the Indian Pines (IP) dataset, but the overall accuracy is as high as 98.34%. Similarly, the framework achieves an overall accuracy of 99.13%, 99.17%, and 99.08% on the University of Pavia (PU), Salinas (SA), and WHU Hi LongKou (LongKou) datasets, respectively. This result exceeds the classification accuracy of existing state-of-the-art frameworks under the same conditions.

show abstract

Section: Introductionmentioning

confidence: 99%

MADANet: A Lightweight Hyperspectral Image Classification Network with Multiscale Feature Aggregation and a Dual Attention Mechanism

Cui,

Wen,

Song

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Some lightweight modeling methods based on VGG-16 also have high accuracy and compression rates, as shown in recent studies [21][22][23][24]. In [21], a model pruning method incorporating a scaling factor and mutual information value order was proposed for the lightweight research of models in target detection scenarios.…”

Section: Introductionmentioning

confidence: 99%

“…Ref. [23] designed a lightweight convolutional neural network based on the VGG-16 network structure, replacing traditional convolutional units with lightweight convolutional units, and making lightweight improvements on a large deep convolutional neural network. The results showed that the method outperformed other state-of-the-art lightweight convolutional neural networks and had an accuracy of 93.27%.…”

Section: Introductionmentioning

confidence: 99%

A Lightweight Identification Method for Complex Power Industry Tasks Based on Knowledge Distillation and Network Pruning

Wang,

Zhou,

Jiang

et al. 2023

Processes

View full text Add to dashboard Cite

Lightweight service identification models are very important for resource-constrained distribution grid systems. To address the increasingly larger deep learning models, we provide a method for the lightweight identification of complex power services based on knowledge distillation and network pruning. Specifically, a pruning method based on Taylor expansion is first used to rank the importance of the parameters of the small-scale network and delete some of the parameters, compressing the model parameters and reducing the amount of operation and complexity. Then, knowledge distillation is used to migrate the knowledge from the large-scale network ResNet50 to the small-scale network so that the small-scale network can fit the soft-label information output from the large-scale neural network through the loss function to complete the knowledge migration of the large-scale neural network. Experimental results show that this method can compress the model size of the small network and improve the recognition accuracy. Compared with the original small network, the model accuracy is improved by 2.24 percentage points to 97.24%. The number of model parameters is compressed by 81.9% and the number of floating-point operations is compressed by 92.1%, making it more suitable for deployment in resource-constrained devices.

show abstract

Research on Image Classification of Lightweight Convolutional Neural Network

Cited by 2 publications

References 6 publications

MADANet: A Lightweight Hyperspectral Image Classification Network with Multiscale Feature Aggregation and a Dual Attention Mechanism

MADANet: A Lightweight Hyperspectral Image Classification Network with Multiscale Feature Aggregation and a Dual Attention Mechanism

A Lightweight Identification Method for Complex Power Industry Tasks Based on Knowledge Distillation and Network Pruning

Contact Info

Product

Resources

About