Jujube Variety Recognition Method Based on Multi-organ Feature Fusion

XU Nan, YUAN Yingchun, LEI Hao, MENG Xi, HE Zhenxue


Aiming at the problem of jujube variety identification in natural scenes, machine vision technology with jujube fruit as the research object has become one of the mainstream methods for accurate identification of jujube varieties. However, due to the small inter-class difference and large intra-class difference of jujube varieties, it is difficult for a single organ to fully express the different characteristics of jujube varieties. A method of jujube varieties recognition based on multi-organ feature fusion was proposed. Firstly, the YOLO v3 detection algorithm was used to segment and extract the jujube fruit and leaf organs in the collected natural scene images, and a multi-sample dataset of jujube varieties based on Cartesian product was proposed to construct two organ combination pairs, and then based on the EfficientNetV2 network model, a fusion strategy that can fully learn the correlation between the characteristics of the two organs was designed to improve the model performance, and a stepwise transfer training method was introduced to improve the recognition efficiency of jujube varieties. Finally, a large number of experiments were carried out on the constructed dataset containing 20 jujube varieties, and the recognition accuracy of 97.04% was obtained, which was significantly better than that of the existing research results, and the training time and convergence speed of the proposed method were also improved. The results showed that this method can effectively integrate the characteristic information of jujube fruit and leaf organs of jujube cultivars, which can provide valuable reference for other variety identification research.


Keywords:jujube variety recognition;Cartesian product;feature fusion;transfer learning;YOLO v3

Download Full Text:



MENG X, YUAN Y C, TENG G F, et al. Deep learning for fine-grained classification of jujube fruit in the natural environment J. Journal of Food Measurement and Characterization ,2021 , 15(5) ; 4150 -4165.

ZHANG R C, YUAN Y C, MENG X, et al. A multitask model based on MobileNetV3 for fine-grained classification of jujube varieties [J] . Journal of Food Measurement and Characterization ,2023 , 17(5) : 4305 -4317.

LIU Yuanvuan. Research on apple classification approach based on feature fusion[ D ] . Fuzhou: Fujian Normal University, 2020. (in Chinese)

LEI Huan, JIAO Zeyu, MA Jingqi, et al. Fast recognition algorithm of apple varieties based on multi feature fusion and SVM [J ]. Automation & Information Engineering, 2020,41(4) ; 13 — 17. (in Chinese)

WANG Jiaqi. Research and implementation of fruit detection and tracking system based on deep learning[ D]. Beijing: Beijing University of Technology, 2021. (in Chinese)

LI Xiuhao, MA Xu, LI Zehua, et al. Identification of rice variety based on multi-feature fusion and SVM [J ]. Journal of Chinese Agricultural Mechanization, 2019, 40(7) ; 97 - 102. (in Chinese)

YOU Jiawei. Cultivar recognition methods of soybean leaf image based on deep learning [ D]. Nanjing; Nanjing University of Finance and Economics, 2021. (in Chinese)

LIU Xinvue. Muti-feature flower recognition system based on mobile terminal [ D]. Changchun: Jilin University, 2019. (in Chinese)

ZHAO Y F, GAO X, HU J F, et al. Tree species identification based on the fusion of bark and leaves [ J ]. Mathematical Biosciences and Engineering, 2020, 17(4) ; 4018 -4033.

HU II Q, DAI В S, SHEN W Z, et al. Cow identification based on fusion of deep parts features[ J ]. Biosystems Engineering, 2020, 192: 245 -256.

RZANNY M, MADER P, DEGGELMANN A, et al. Flowers, leaves or both? How to obtain suitable images for automated plant identification [ J ]. Plant Methods, 2019, 15(1): 77.

LUO Juan, CAI Cheng. Multi-cue plant species identification [J ]. Computer Engineering and Applications, 2020, 56(5) : 160 -165. (in Chinese)

LEE S II, CHAN С S, MAYO S J, et al. How deep learning extracts and learns leaf features for plant classification [ J ]. Pattern Recognition ,2017, 71; 1 -13.

MENG Xi, YUAN Yingchun, LIU Bo, et al. Classification of jujube by integrating fruit with leaf information[ J]. Journal of Hebei Agricultural University, 2022, 45(2) ; 107 - 113. (in Chinese)

REDMON J, FARHADI A. YOLO v3; an incremental improvement[ J]. arXiv Preprint arXiv; 1804. 02767 , 2018.

TAN M, LE Q. EfficientNetV2: smaller models and faster training [ С ]/ International Conference on Machine Learning. PMLR, 2021 : 10096 - 10106.

DENG J, DONG W, SOCHER R, et al. Imagenet: a large-scale hierarchical image databasef С]/У2009 IEEE Conference on Computer Vision and Pattern Recognition. IEEE, 2009; 248 -255.

IOFFE S, SZEGEDY C. Batch normalization: accelerating deep network training by reducing internal covariate shift [ С] // International Conference on Machine Learning. PMLR, 2015: 448 -456.

SIFRE L. Rigid-motion scattering for image classification[ D]. Paris; Ecole Polytechnique, 2014.

HU J, SHEN L, SUN G. Squeeze-and-excitation networks [ С ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018; 7132-7141.

HUANG G, SUN Y, LIU Z, et al. Deep networks with stochastic depth [ С ] // Computer Vision-ECCV 2016: 14th European Conference, 2016; 646 -661.

ALEX K, ILYA S, GEOFFREY E. ImageNet classification with deep convolutional neural networks [ J]. Communications of the ACM, 2017,60(6) :84 -90.

SIMONYAN K, ZISSERMA A. Very deep convolutional networks for large-scale image recognition [J] . arXiv Preprint arXiv; 1409.1556, 2014.

HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition [С ]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770 -778.

SZEGEDY C, LIU W, JIA Y, et al. Going deeper with convolutions [ С ] // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015; 1 -9.


  • There are currently no refbacks.