Education, Science, Technology, Innovation and Life
Open Access
Sign In

Design and Implementation of a Fruit Image Recognition System Based on CNN

Download as PDF

DOI: 10.23977/cpcs.2025.090110 | Downloads: 0 | Views: 34

Author(s)

Ranning Deng 1

Affiliation(s)

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

Corresponding Author

Ranning Deng

ABSTRACT

Fruit classification and recognition have significant application value in fields such as agricultural sorting and fresh food retail. To address the poor robustness of traditional manual feature extraction methods, this paper designs and implements a fruit image recognition system based on lightweight convolutional neural networks (CNNs). The system adopts 20 common types of fruit images from the Fruit-360 dataset. After preprocessing, an improved LeNet-5 model and a simplified AlexNet model are constructed, with a comparative analysis of the performance differences between ReLU and Sigmoid activation functions. Experimental results show that the simplified AlexNet combined with the ReLU activation function achieves the optimal performance, with an average test set accuracy of 96.87%. It also features fast training convergence, making it suitable for real-time recognition requirements in low-computing-power scenarios.

KEYWORDS

Convolutional Neural Network; Fruit Image Recognition; Lightweight Model; Activation Function; Image Classification

CITE THIS PAPER

Ranning Deng, Design and Implementation of a Fruit Image Recognition System Based on CNN. Computing, Performance and Communication Systems (2025) Vol. 9: 73-77. DOI: http://dx.doi.org/10.23977/cpcs.2025.090110.

REFERENCES

[1] Szegedy, Christian, et al. "Going deeper with convolutions." Proceedings of the IEEE conference on computer vision and pattern recognition. 2015.
[2] Simonyan, Karen, and Andrew Zisserman. "Very deep convolutional networks for large-scale image recognition." arXiv preprint arXiv:1409.1556 (2014). 
[3] Pan, Sinno Jialin, and Qiang Yang. "A survey on transfer learning." IEEE Transactions on knowledge and data engineering 22.10 (2009): 1345-1359.
[4] Yosinski, Jason, et al. "How transferable are features in deep neural networks?" Advances in neural information processing systems 27 (2014).
[5] Miao Yujie, Zhu Shiping, Pu Jing, Li Junxian, Ma Lingkai, Huang Hua. Recognition of Furniture Wood Image Species Based on Convolutional Neural Networks [J]. Scientia Silvae Sinicae, 2023, 59 (8): 133-140. 
[6] Xiaoyue Qin, Ruwei Huang, and Bei Hua. "Research and implementation of yield recognition of Citrus reticulata based on target detection." Journal of physics: conference series. Vol. 1820. No. 1. IOP Publishing, 2021.
[7] Girshick, Ross, et al. "Rich feature hierarchies for accurate object detection and semantic segmentation." Proceedings of the IEEE conference on computer vision and pattern recognition. 2014.

Downloads: 3343
Visits: 207415

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.