Article Details

Title · Authors · Abstract · Keywords · DOI · References

Cherry Image Classification Detection Based on Deep Learning

LIU Qing, WU Zhongxiao, ZHANG Yaya, HE Bingwei, WEI Kaibin, ZHAO Limin, ZHAO Yuxiang

(School of Electronic Information and Electrical Engineering, Tianshui Normal University, Tianshui 741001, China)

Section:Artificial Intelligence · 2025, Vol. 9, No. 11 · Pages:64-69 · 2096-4706(2025)11-0064-06 · CLC number:TP391.4 · Document code:A · DOI:10.19850/j.cnki.2096-4706.2025.11.013

Abstract

In order to detect cherry images in multiple types of fruit images and lay the foundation for the classification detection of subsequent cherry image high-quality fruits and defective fruits, this paper proposes model analysis, parameter optimization, training and testing of eight commonly used Deep Neural Networks, and evaluates the models by using objective evaluation criteria, and then uses the optimal model to train the image classification of cherry high-quality and defective fruits. Through detection experiments, it is verified that the T2T_ViT model achieves average accuracies of 99.40% and 99.12% for high-quality and defective fruits of cherry images, respectively. There are good classification detection results of cherry images with high-quality and defective fruits.

Keywords

Deep Learning; cherry image classification detection; T2T_ViT model; objective evaluation

References

  1. [1] Xu Mingchen, Niu Yuanyuan, Yu Yongchang A Review of Research on Fruit and Vegetable Picking Robots [J]. Anhui Agricultural Science, 2014, 42 (31): 11024-11027+11057
  2. [2] NIE M Y,ZHAO Q J,BI S H,et al. Apple External Quality Analysis Based on BP Neural Network [C]//2019 1st International Conference on Industrial Artificial Intelligence (IAI).Shenyang:IEEE,2019:1-5.
  3. [3] HUANG Y-P,WANG T-H,BASANTA H. Using Fuzzy Mask R-CNN Model to Automatically Identify Tomato Ripeness [J].IEEE Access,2020,8:207672-207682.
  4. [4] GAO F F,FANG W T,SUN X M,et al. A Novel Apple Fruit Detection and Counting Methodology Based on Deep Learning and trunk tracking in Modern Orchard [J].Computers and Electronics in Agriculture,2022,197:107000.
  5. [5] LU S L,CHEN W K,ZHANG X,et al. Canopy-attention-YOLOv4-based Immature/Mature Apple Fruit Detection on Dense-foliage Tree Architectures for Early Crop Load Estimation [J/OL].Computers and Electronics in Agriculture,2022,193:106696[2025-01-06].https://doi.org/10.1016/j.compag.2022.106696.
  6. [6] KRIZHEVSKY A,SUTSKEVER I,HINTON G E. ImageNet Classification with Deep Convolutional Neural Networks [J].Communications of the ACM,2017,60(6):84-90.
  7. [7] Wang Bei Novel Pain Expression Recognition Based on Transfer Learning [D]. Nanjing: Nanjing University of Posts and Telecommunications, 2020
  8. [8] Wang Huiling, Qi Xiaolong, Wugang Mountain Research progress on object detection technology based on deep convolutional neural networks [J]. Computer Science, 2018, 45 (9): 11-19
  9. [9] Yang Hongye Multi spectral remote sensing image super-resolution reconstruction using convolutional neural networks [D]. Xuzhou: China University of Mining and Technology, 2020
  10. [10] HUANG G,LIU Z,MAATEN L V D,et al. Densely Connected Convolutional Networks [C]//2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).Honolulu:IEEE,2017:4700-4708.
  11. [11] YUAN L,CHEN Y P,WANG T,et al. Tokens-to-token ViT: Training Vision Transformers from Scratch on ImageNet [C]//2021 IEEE/CVF International Conference on Computer Vision(ICCV).Montreal:IEEE,2021:558-567.
  12. [12] TROCKMAN A,KOLTER J Z. Patches are All You Need? [J/OL].arXiv:2201.09792 [cs.CV].(2022-01-24).https://arxiv.org/abs/2201.09792.
  13. [13] GUO M-H,LU C-Z,LIU Z-N,et al. Visual Attention Network [J/OL].arXiv:2202.09741 [cs.CV].(2022-02-20).https://arxiv.org/abs/2202.09741.

Funding

Gansu Province Science and Technology Plan Projects (23JRRE0740, 25JRRE004); Qinzhou District Science and Technology Plan Project of Tianshui City (2023-SHFZG-6476); Graduate Innovation Guidance Program of Tianshui Normal University (TYCX2330, TYCX2331)

About the authors

Liu Cong (1970-), male, Han ethnicity, from Tianshui, Gansu Province, professor, PhD, research direction: intelligent image information processing.