Education, Science, Technology, Innovation and Life
Open Access
Sign In
  1. Home
  2. Journals
  3. Computing, Performance and Communication Systems
  4. Vol 9, Issue 1, 2025

Application of Lightweight CNN in Garbage Image Classification

Download as PDF

DOI: 10.23977/cpcs.2025.090109 | Downloads: 1 | Views: 63

Author(s)

Peng Yin 1

Affiliation(s)

1 School of Electronic and Information Engineering, University of Science and Technology Liaoning, Anshan, China

Corresponding Author

Peng Yin

ABSTRACT

With urbanization accelerating, domestic waste output has surged, and garbage classification is crucial for alleviating environmental pressure and recycling resources. Traditional manual classification is inefficient, costly, and subjective, so automated garbage classification technology is necessary. To solve problems of existing CNN models like large parameter size, slow inference, and difficulty in edge - device deployment, this paper proposes a lightweight CNN based on a simplified ResNet for garbage image classification. The public TrashNet dataset with 6 common domestic waste categories is used. Data augmentation and transfer learning are employed to optimize the model's adaptation to garbage image features. Experimental results show the simplified ResNet model achieves 91.2% classification accuracy on the TrashNet dataset, with precision of 90.8%, recall of 90.5%, and an F1 - score of 90.6%. Its parameter number is only 48% of the traditional ResNet18, and the per - image inference time is shortened to 12.3ms. Compared with mainstream models, it reduces computational complexity and storage requirements while ensuring performance, making it more suitable for edge - computing devices like smart trash cans and classification robots, and providing an efficient solution for practical automated garbage classification. 

KEYWORDS

Lightweight Convolutional Neural Network; Garbage Image Classification; Transfer Learning; ResNet Simplification; TrashNet Dataset

CITE THIS PAPER

Peng Yin, Application of Lightweight CNN in Garbage Image Classification. Computing, Performance and Communication Systems (2025) Vol. 9: 65-72. DOI: http://dx.doi.org/10.23977/cpcs.2025.090109.

REFERENCES

[1] He, Kaiming, et al. "Deep residual learning for image recognition." Proceedings of the IEEE conference on computer vision and pattern recognition. 2016.
[2] Howard, Andrew G., et al. "Mobilenets: Efficient convolutional neural networks for mobile vision applications." arXiv preprint arXiv:1704.04861 (2017).
[3] Sandler, Mark, et al. "Mobilenetv2: Inverted residuals and linear bottlenecks." Proceedings of the IEEE conference on computer vision and pattern recognition. 2018. 
[4] Zhang, Xiangyu, et al. "Shufflenet: An extremely efficient convolutional neural network for mobile devices." Proceedings of the IEEE conference on computer vision and pattern recognition. 2018.
[5] Iandola, Forrest N., et al. "SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and< 0.5 MB model size." arXiv preprint arXiv:1602.07360 (2016).
[6] Pan S J, Yang Q. A survey on transfer learning[J]. IEEE Transactions on knowledge and data engineering, 2009, 22(10): 1345-1359.
[7] Deng J, Dong W, Socher R, et al. ImageNet: A large-scale hierarchical image database[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2009: 248-255.

Subscription
E-Mail Alert
Most popular papers
Call for Special Issue
Publication Ethics & OA Statement
Plagiarism Policy
Contact Us
Downloads: 3343
Visits: 207414

Sponsors, Associates, and Links

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.