Abstract
Early disease detection in greenhouses is an important part of integrated disease management in modern agriculture. A real-time object detection method of melon leaf disease, Pruned-YOLO v5s+Shuffle (PYSS) is proposed in this research. First, for enhancing the feature extraction capability, the backbone of the YOLO v5s is reconstructed with ShuffleNet v2 Inverted Residual block. Then, to further downsize the model, the channel pruning method is used to prune and fine-tune the sparsely trained model. Finally, Pruned-YOLO v5s+Shuffle model is deployed to Jetson Nano, and the real-time performance is confirmed in melon greenhouses. The experimental results show that the proposed model has 93.2% and 98.2% mAP@0.5 for melon (Cucumis melon. L) powdery mildew and melon real leaves, respectively. Compared with YOLO v5s, the performance of our proposed model is improved 6.2% and 6.4% in the term of mAP@0.5 and precision, respectively. The model size and inference time are reduced 85% and 7.5%. In addition, the PYSS demonstrates the higher detection precision and faster inference speed in the comparison of YOLO v3, Faster R-CNN, RetinaNet, Cascade R-CNN, YOLO v4 and YOLO v5s. Being deployed to Jetson Nano, the detection results are displayed on the monitor in real time: mAP@0.5 is 96.7%, the model size is 1.1 MB, and the inference time is 13.8 ms.
Similar content being viewed by others
References
Guo, S., Zhang, C., Ren, Y. et al.: Application of comparative genomics in developing markers tightly linked to the pm-2f gene for powdery mildew resistance in melon (Cucumis melo l.). Euphytica 190(2), 157–168 (2013)
Wu, Q., Zhang, H., Pei, Z., Jing, Y.: Identification on races of powdery mildew of melon in Huaibei. Molecular Plant Breeding (2017)
Li, B., Zhao, Y., Zhu, Q., Zhang, Z., Fan, C., Amanullah, S., Gao, P., Luan, F.: Mapping of powdery mildew resistance genes in melon (cucumis melo l.) by bulked segregant analysis. Entia Horticult. 220, 160–167 (2017)
Jorkesh, A., Safaei, M., Olfati, J.: Chemical and biological products for control of powdery mildew on cucumber. Int. J. Veg. Sci. pages 1–6 (2021)
Pizolotto, C.A., Harrington, M., Brown, L., Murdock, M., Harrington, S., Marshall, J., Neher, O.T., Woodhall, J.W.: A real-time pcr assay for erysiphe betae and its effectiveness when used with different spore trapping methods. Eur. J. Plant Pathol. 162(2), 329–341 (2021)
Hong, Y.J., Hossain, M.R., Kim, H.T., Park, J.I., Nou, I.S.: Identification of two new races of podosphaera xanthii causing powdery mildew in Melon in South Korea. Plant Pathol. J. 34(3), 182–190 (2018)
Kunova, A., Pizzatti, C., Saracchi, M., Pasquali, M., Cortesi, P.: Grapevine powdery mildew: fungicides for its management and advances in molecular detection of markers associated with resistance in. Microorganisms 9(7), 1541 (2021)
Qu, Y., Wu, P., Hu, J., Chen, Y., et al.: Molecular detection of the powdery mildew resistance genes in winter wheats dh51302 and shimai 26. J. Integr. Agric. 19(4), 931–940 (2020)
Li, R., Wang, J. et al.: Artificial intelligence and wireless communications. Front. Inf. Technol. Electron. Eng. 21(10), 1413–1425 (2020)
Bai, L., Wang, Z. et al.: An oil well dataset derived from satellite-based remote sensing. Remote Sens. 13(6) (2021)
Abdolahnejad, M., Liu, P.X.: Deep learning for face image synthesis and semantic manipulations: a review and future perspectives. Artif. Intell. Rev. 53(8), 5847–5880 (2020)
Min, S., Rim, B., Sung, N.J., Hong, M.: Deep learning in physiological signal data: a survey. Sensors 20(4) (2020)
Cheng, G., Xie, X.X., Han, J.W., Guo, L., Xia, G.S.: Remote sensing image scene classification meets deep learning: challenges, methods, benchmarks, and opportunities. IEEE J. Select. Top. Appl. Earth Observ. Remote Sens. 13, 3735–3756 (2020)
Sun, S., Yang, Y., Song, H. et al.: A fast and effective video vehicle detection method leveraging feature fusion and proposal temporal link. J. Real-time Image Process. 18(4, SI), 1261–1274 (2021)
Liu, C., Zhu, H.J., Guo, W., Han, X., Chen, C., Wu, H.R.: Efdet: an efficient detection method for cucumber disease under natural complex environments. Comput. Electron. Agric. 189 (2021)
Anagnostis, A., Tagarakis, A.C., Asiminari, G., Papageorgiou, E., Kateris, D., Moshou, D., Bochtis, D.: A deep learning approach for anthracnose infected trees classification in walnut orchards. Comput. Electron. Agric. 182 (2021)
Song, C.L., Zhang, Y., Zhang, D.W.: Deep learning-based object detection improvement for tomato disease. Ieee Access 8, 56607–56614 (2020)
Wang, H.Q., Xia, W., Yu, F., Hong, R.: A high-precision lightweight smoke detection model based on se attention mechanism. In: 2nd International Conference on Consumer Electronics and Computer Engineering, 2(2), 941–944 (2022)
Wang, Z., Ma, L.: SYOLO: an efficient pedestrian detection. IOP Conf. Ser. Mater. Sci. Eng. 768, 072067 (2020)
Girshick, R., Redmon, J., Divvala, S., Farhadi, A.: You only look once: Unified, real-time object detection (2015)
Ma, N., Zhang, X., Zheng, H., Sun, J.: Shufflenet v2: Practical guidelines for efficient cnn architecture design (2018)
Lin, M., Zhang, X., Zhou, X., Sun, J.: Shufflenet: an extremely efficient convolutional neural network for mobile devices (2017)
Huang, G., Liu, S., Maaten, L., Weinberger, K.Q.: Densely connected convolutional networks (2016)
Huang, G., Liu, S., Maaten, L., Weinberger, K.Q.: Condensenet: an efficient densenet using learned group convolutions (2017)
Liu, Z., Li, J., Shen, Z., Huang, G., Yan, S., Zhang, C.: Learning efficient convolutional networks through network slimming (2017)
Redmon, J., Farhadi, A.: Yolov3: an incremental improvement (2018)
Wang, C.-Y., Bochkovskiy, A., Liao, H.-Y. M.: Yolov4: optimal speed and accuracy of object detection (2020)
Girshick, R., Ren, S., He, K., Sun, J.: Faster r-cnn: towards real-time object detection with region proposal networks (2015)
Lin, T., Goyal, P., Girshick, R., He, K., Dollár, P.: Focal loss for dense object detection (2017)
Cai, Z.W., Vasconcelos, N.: Cascade r-cnn: Delving into high quality object detection (2017)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition (2015)
Author information
Authors and Affiliations
Contributions
Conceptualization, YX, QC and ZY; methodology, YX, ZY, and SK; software, YX, QC, and YZ; validation, YX, QC, LX, QW, and YZ; formal analysis, YX and YZ; investigation, YX and YZ; resources, YX and YZ; data curation, YX, QC, XC, and QW; writing—original draft preparation, YX, QC, and YZ; writing—review and editing, YX, QC, and YZ. All the authors have read and agreed to the published version of the manuscript.
Corresponding author
Ethics declarations
Funding
This work is supported by the Science and Technology Development Plan Project of Changchun [Grant number 21ZGN28]; the Jilin Provincial Science and Technology Department International Exchange and Cooperation Project [Grant number 20200801014GH].
Conflict of interest
The authors declare no conflict of interest.
Availability of data and materials
Not applicable.
Code availability
Not applicable.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Qingyuan Chen, Shuolin Kong, Lu Xing, Qi Wang, Xue Cong, and Yang Zhou have contributed equally to this work.
Rights and permissions
Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Xu, Y., Chen, Q., Kong, S. et al. Real-time object detection method of melon leaf diseases under complex background in greenhouse. J Real-Time Image Proc 19, 985–995 (2022). https://doi.org/10.1007/s11554-022-01239-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11554-022-01239-7