Adaptive precision cutting method for rootstock grafting of melons: modeling, analysis, and validation
文献类型: 外文期刊
作者: Chen, Shan 1 ; Jiang, Kai 2 ; Zheng, Wengang 2 ; Jia, Dongdong 2 ; Zhao, Chunjiang 1 ;
作者机构: 1.Shihezi Univ, Coll Mech & Elect Engn, Shihezi 832003, Xinjiang, Peoples R China
2.Beijing Acad Agr & Forestry Sci, Res Ctr Intelligent Equipment, Beijing 100097, Peoples R China
3.Natl Engn Res Ctr Informat Technol Agr, Beijing 100097, Peoples R China
关键词: Melon; Grafting robot; Adaptive cutting; Rootstock pith cavity; Machine vision
期刊名称:COMPUTERS AND ELECTRONICS IN AGRICULTURE ( 2022影响因子:8.3; 五年影响因子:8.3 )
ISSN: 0168-1699
年卷期: 2024 年 218 卷
收录情况: SCI
摘要: The current grafting machine employs a fixed-angle cutting method, which fails to adapt to changes in the pith cavity resulting from variations in seedling age. Additionally, cutting wounded rootstock seedlings occasionally leads to grafting failures. To enhance cutting precision and adaptability in machine grafting, this study proposes a rootstock adaptive cutting method capable of achieving variable-angle precision cutting for rootstocks at different seedling ages. Firstly, a visual image recognition system is established to obtain characteristic parameters of rootstocks' external and internal pith cavities at varying seedling ages. Subsequently, a geometric model of the rootstock pith cavity is constructed to determine precise cutting parameters. The study analyzes the growth pattern and correlation between external morphological features and the internal pith cavity of rootstocks at different seedling ages. A logistic model establishes the dynamic growth process of six exterior features of rootstocks. At the same time, a polynomial function is fitted to depict the dynamic growth process of the pith cavity vertex-cotyledon intersection. Furthermore, regression models of rootstock external characteristics and pith cavity vertex-cotyledon intersection were constructed. Using the pith cavity vertex-cotyledon intersection as an intermediate bridge, a prediction model is created to determine the short axis of the seedling stem and precise cutting angle based on the adaptive method. Validation tests are conducted to evaluate the proposed models. The results demonstrate that the growth patterns of the six external characteristics of rootstocks follow an "S" curve, indicating that seedlings aged between 1 and 9 days are suitable for cutting, with the pith cavity vertex located below the cotyledon intersection, meeting the cutting requirements. Additionally, all six external features significantly correlate negatively with the pith cavity vertex-cotyledon intersection (P <= 0.01). The seedling stem short axis prediction model and precise cutting angle shows a satisfactory residual distribution. The model achieves an 86 % success rate in rootstock cutting and a 4 % occurrence of cutting through the pith cavity, demonstrating its ability to predict variations in the seedling stem short axis and precise cutting angle. The findings of this study provide a foundation for the design and implementation of adaptive cutting in grafting robots, thereby improving the adaptability and safety of machine grafting on seedlings.
- 相关文献
作者其他论文 更多>>
-
Recognition of wheat rusts in a field environment based on improved DenseNet
作者:Chang, Shenglong;Cheng, Jinpeng;Fan, Zehua;Ma, Xinming;Li, Yong;Zhao, Chunjiang;Chang, Shenglong;Yang, Guijun;Cheng, Jinpeng;Fan, Zehua;Yang, Xiaodong;Zhao, Chunjiang
关键词:Plant disease; Wheat rust; Image processing; Deep learning; Computer vision (CV); DenseNet
-
GCVC: Graph Convolution Vector Distribution Calibration for Fish Group Activity Recognition
作者:Zhao, Zhenxi;Zhao, Chunjiang;Zhao, Zhenxi;Yang, Xinting;Zhou, Chao;Zhao, Chunjiang;Zhao, Zhenxi;Yang, Xinting;Zhou, Chao;Zhao, Chunjiang;Zhao, Zhenxi;Yang, Xinting;Zhou, Chao;Zhao, Chunjiang;Liu, Jintao
关键词:Fish; Feature extraction; Activity recognition; Calibration; Adhesives; Training; Convolution; Graph convolution vector calibration; fish group activity; activity feature vector calibration; fish activity dataset
-
An Original UV Adhesive Watermelon Grafting Method, the Grafting Device, and Experimental Verification
作者:Zhang, Xin;Kong, Linghao;Lu, Hanwei;Zhang, Xin;Feng, Qingchun;Li, Tao;Jiang, Kai;Zhang, Qian
关键词:watermelon grafting; UV adhesive; fluent; VOF-DPM numerical simulation; grafting device; test
-
Long-range infrared absorption spectroscopy and fast mass spectrometry for rapid online measurements of volatile organic compounds from black tea fermentation
作者:Yang, Chongshan;Li, Guanglin;Zhao, Chunjiang;Fu, Xinglan;Yang, Chongshan;Jiao, Leizi;Wen, Xuelin;Lin, Peng;Duan, Dandan;Zhao, Chunjiang;Dong, Daming;Yang, Chongshan;Jiao, Leizi;Wen, Xuelin;Lin, Peng;Duan, Dandan;Dong, Daming;Dong, Chunwang
关键词:Black tea fermentation; Volatile organic compounds; Proton transfer reaction mass spectrometry; Fourier transform infrared spectroscopy; Principal component analysis; Extreme learning machine
-
Design and Experiment of Automatic Transport System for Planting Plate in Plant Factory
作者:Jia, Dongdong;Gao, Guohua;Jia, Dongdong;Guo, Wenzhong;Wang, Lichun;Zheng, Wengang
关键词:plant factory; automatic transport system; structure design; dynamic simulation; positioning accuracy
-
Navigation line extraction algorithm for corn spraying robot based on YOLOv8s-CornNet
作者:Guo, Peiliang;Diao, Zhihua;Ma, Shushuai;He, Zhendong;Zhao, Suna;Zhao, Chunjiang;Li, Jiangbo;Zhang, Ruirui;Yang, Ranbing;Zhang, Baohua
关键词:agricultural robotics; computer vision; deep learning; navigation line extraction; network lightweight
-
An ultra-lightweight method for individual identification of cow-back pattern images in an open image set
作者:Wang, Rong;Gao, Ronghua;Li, Qifeng;Zhao, Chunjiang;Ding, Luyu;Yu, Ligen;Ma, Weihong;Wang, Rong;Zhao, Chunjiang;Gao, Ronghua;Li, Qifeng;Zhao, Chunjiang;Ding, Luyu;Yu, Ligen;Ma, Weihong;Ru, Lin
关键词:Cow-back pattern; Cow recognition; LightCowsNet; Open image set; Deep learning