An automatic approach for detecting seedlings per hill of machine-transplanted hybrid rice utilizing machine vision

Li, Hongwei; Li, Zehua; Dong, Wenhao; Cao, Xiulong; Wen, Zhicheng; Xiao, Ronghao; Wei, Yuhao; Zeng, Hong; Xu, Ma

doi:10.1016/j.compag.2021.106178

Cited by 19 publications

(10 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To reduce damage to seedlings during the transplanting process, in addition to using the inclined transplanting manipulator, the principle of visual processing is used to avoid obstacles (Li et al, 2021). During the transplanting process, to complete intelligent obstacle avoidance, the main information collected is the height information of the seedlings and edge information of the seedlings (Singh and Moses, 2021).…”

Section: Design Scheme Of the Manipulator Test Platformmentioning

confidence: 99%

Research and Development for the Hurdle Avoidance During Transplanting Manipulation Based on Kinect Vision Processing

Chen¹,

Yuan²,

Zhao³

et al. 2022

American Journal of Biochemistry and Biotechnology

View full text Add to dashboard Cite

To avoid damage to plants during the transplanting process, a transplanting method based on the combination of the Kinect machine visual system and the low-damage transplanting end effector was proposed to reduce the damage to the stems and leaves from transplanting end effectors during the transplanting process. The inclined manipulator was designed can avoid the characteristics of the seedling itself during the transplanting process to avoid obstacles. Under the experimental condition, the oblique transplanting manipulator assisted by the Kinect vision processing system achieved 4.70% leaf damage, 16.70% stem bending, 83.45% substrate integrity, and the transplanting quality parameters were 87.36%. The completion time of a single transplant was 8.32±0.4 s, the results proved that the low-damage transplanting manipulator based on the machine vision system can effectively reduce the damage rate of the stems and leaves. This obstacle avoidance transplant method can be extended to seedlings planted in 50 plugs. Research and development for hurdle avoidance during transplanting manipulation based on Kinect vision processing can provide theoretical support for future obstacle avoidance transplantation.

show abstract

Section: Design Scheme Of the Manipulator Test Platformmentioning

confidence: 99%

Research and Development for the Hurdle Avoidance During Transplanting Manipulation Based on Kinect Vision Processing

Chen¹,

Yuan²,

Zhao³

et al. 2022

American Journal of Biochemistry and Biotechnology

View full text Add to dashboard Cite

show abstract

“…Skeleton endpoint detection was used to characterize the leaf of seedling and to judge whether the field crop had reached the three-leaf stage or not. Recently, Li et al (2021) utilized computer vision to detect rice seedling hill in the paddy field. The preferred laboratory color model along with Otsu’s method was used to extract rice seedling information, and the skeleton of the seedling hill was extracted using the thinning algorithm to effectively characterize the morphological structure of single seedling hill.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Approaches for Rice Seedling Growth Stages Detection

Tan

Liu

et al. 2022

Front. Plant Sci.

Self Cite

View full text Add to dashboard Cite

Recognizing rice seedling growth stages to timely do field operations, such as temperature control, fertilizer, irrigation, cultivation, and disease control, is of great significance of crop management, provision of standard and well-nourished seedlings for mechanical transplanting, and increase of yield. Conventionally, rice seedling growth stage is performed manually by means of visual inspection, which is not only labor-intensive and time-consuming, but also subjective and inefficient on a large-scale field. The application of machine learning algorithms on UAV images offers a high-throughput and non-invasive alternative to manual observations and its applications in agriculture and high-throughput phenotyping are increasing. This paper presented automatic approaches to detect rice seedling of three critical stages, BBCH11, BBCH12, and BBCH13. Both traditional machine learning algorithms and deep learning algorithms were investigated the discriminative ability of the three growth stages. UAV images were captured vertically downward at 3-m height from the field. A dataset consisted of images of three growth stages of rice seedlings for three cultivars, five nursing seedling densities, and different sowing dates. In the traditional machine learning algorithm, histograms of oriented gradients (HOGs) were selected as texture features and combined with the support vector machine (SVM) classifier to recognize and classify three growth stages. The best HOG-SVM model obtained the performance with 84.9, 85.9, 84.9, and 85.4% in accuracy, average precision, average recall, and F1 score, respectively. In the deep learning algorithm, the Efficientnet family and other state-of-art CNN models (VGG16, Resnet50, and Densenet121) were adopted and investigated the performance of three growth stage classifications. EfficientnetB4 achieved the best performance among other CNN models, with 99.47, 99.53, 99.39, and 99.46% in accuracy, average precision, average recall, and F1 score, respectively. Thus, the proposed method could be effective and efficient tool to detect rice seedling growth stages.

show abstract

“…Additionally, the PVH100 sweet potato tilt transplanting machine developed by Iseki in Japan can realize film transplanting and complete the oblique planting agronomic. In contrast, the research on mechanical transplanting technology in China has been relatively slow for a long time (Zhong et al, 2021;Hu et al, 2020;Yu et al, 2014;Li et al, 2021). For example, In recent years, the national sweet potato industry technology system has cooperated with Fulaiwei Agricultural Equipment Co., Ltd., and improved the 2ZL-1 semi-automatic chain clamp planter based on the chain clamp tobacco transplanter.…”

Section: Introductionmentioning

confidence: 99%

Design and experiment of a control system for sweet potato seedling feeding and planting device based on pretreatment seedling belt

Zhu²,

Fu³

et al. 2022

J. Agric. Eng.

View full text Add to dashboard Cite

Although existing sweet potato transplanters require automatic seedling feeding instead of manual seedling feeding, this causes seedling leakage and low efficiency. In this work, a control system for automatic seedling feeding of sweet potatoes was designed based on a pretreatment seedling belt. The system uses STM32 as the main controller and obtains the running speed of the machine through the encoder. The speed of the planting motor can be adjusted in real-time according to the running speed to keep the planting distance stable. The speed control model and linkage control strategy of seedling feeding and planting motors are investigated to keep the system feeding frequency and planting frequency consistent under running speed change. In order to verify the performance of this control system, a test bench was built, and some experiments were conducted. The test results show that the average error of seedling feeding motor speed is 4.04%, and that of planting motor speed is 3.28%. At medium and low operating speed levels, the stability of the seedling feeding mechanism is good, and the relative errors of automatic seedling feeding operation are 7.8% and 5.1%, respectively. The variation coefficients of plant spacing were 9.34% and 7.42%, respectively, indicating that the system could meet the seedling feeding and planting device control requirements based on the seedling belt and realize continuous automatic seedling feeding in the process of sweet potato seedling transplanting.

show abstract

An automatic approach for detecting seedlings per hill of machine-transplanted hybrid rice utilizing machine vision

Cited by 19 publications

References 41 publications

Research and Development for the Hurdle Avoidance During Transplanting Manipulation Based on Kinect Vision Processing

Research and Development for the Hurdle Avoidance During Transplanting Manipulation Based on Kinect Vision Processing

Machine Learning Approaches for Rice Seedling Growth Stages Detection

Design and experiment of a control system for sweet potato seedling feeding and planting device based on pretreatment seedling belt

Contact Info

Product

Resources

About