Search

Your search keyword '"lightweight network"' showing total 2 results
Start Over Descriptor "lightweight network" Publication Type Reviews
2 results on '"lightweight network"'

1. ConvFishNet: An efficient backbone for fish classification from composited underwater images.

Author

Qu, Huishan, Wang, Gai-Ge, Li, Yun, Qi, Xin, and Zhang, Mengjie

Subjects
*CLASSIFICATION of fish, *AUTOMATIC classification, *MARICULTURE, *MARINE ecology, *CONVOLUTIONAL neural networks, *SPINE, *UNDERWATER photography
Abstract

For the purpose of monitoring fish health, managing aquaculture, and comprehending marine ecology, there is a growing interest in the automatic classification of different fish species. Recent developments in machine vision-based classification methods, known for their speed and non-destructive nature, have led to the introduction of various automatic categorization approaches. Drawing inspiration from FishNet, a new architecture model named ConvFishNet has been proposed. This model incorporates large convolutional kernels and depth-wise separable convolutions to reduce the number of parameters in the model. Additionally, the PixelShuffle has been developed to enhance the upsampling information and improve fish classification performance. While maintaining precision in fish classification, the model achieves a lightweight design with only 0.83G. Compared with the FishNet model, the new model reduces parameters by 80 %. The model demonstrates a precision of 88.44 % on the WildFish dataset subset and 99.8 % on the Fish4knowledge dataset, surpassing existing methods including FishNet. This method shows promise for fish classification in challenging underwater environments, such as marine and aquaculture settings, and further investigation is planned for the future. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

2. MMViT-Seg: A lightweight transformer and CNN fusion network for COVID-19 segmentation.

Author

Yang, Yuan, Zhang, Lin, Ren, Lei, and Wang, Xiaohan

Subjects
*IMAGE segmentation, *COMPUTER vision, *COVID-19, *CONVOLUTIONAL neural networks, *SUPERVISED learning
Abstract

• A light weight CNN+transformer segmentation network named Mini-MobileViT-Seg. • Multi Query Attention module to fuse the multi-scale features of different levels. • Compared with traditional segmentation methods, it has only 1M parameters. • This model performance reaches the level of the state-of-art segmentation model. Background and objective: COVID-19 is a serious threat to human health. Traditional convolutional neural networks (CNNs) can realize medical image segmentation, whilst transformers can be used to perform machine vision tasks, because they have a better ability to capture long-range relationships than CNNs. The combination of CNN and transformers to complete the task of semantic segmentation has attracted intense research. Currently, it is challenging to segment medical images on limited data sets like that on COVID-19. Methods: This study proposes a lightweight transformer+CNN model, in which the encoder sub-network is a two-path design that enables both the global dependence of image features and the low layer spatial details to be effectively captured. Using CNN and MobileViT to jointly extract image features reduces the amount of computation and complexity of the model as well as improves the segmentation performance. So this model is titled Mini-MobileViT-Seg (MMViT-Seg). In addition, a multi query attention (MQA) module is proposed to fuse the multi-scale features from different levels of decoder sub-network, further improving the performance of the model. MQA can simultaneously fuse multi-input, multi-scale low-level feature maps and high-level feature maps as well as conduct end-to-end supervised learning guided by ground truth. Results: The two-class infection labeling experiments were conducted based on three datasets. The final results show that the proposed model has the best performance and the minimum number of parameters among five popular semantic segmentation algorithms. In multi-class infection labeling results, the proposed model also achieved competitive performance. Conclusions: The proposed MMViT-Seg is tested on three COVID-19 segmentation datasets, with results showing that this model has better performance than other models. In addition, the proposed MQA module, which can effectively fuse multi-scale features of different levels further improves the segmentation accuracy. [ABSTRACT FROM AUTHOR]

Published
2023
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results