黄成龙,李曜辰,骆树康,杨万能,朱龙付.基于结构光三维点云的棉花幼苗叶片性状解析方法[J].农业机械学报,2019,50(8):243-248,288.
HUANG Chenglong,LI Yaochen,LUO Shukang,YANG Wanneng,ZHU Longfu.Cotton Seedling Leaf Traits Extraction Method from 3D Point Cloud Based on Structured Light Imaging[J].Transactions of the Chinese Society for Agricultural Machinery,2019,50(8):243-248,288.
摘要点击次数: 1706
全文下载次数: 834
基于结构光三维点云的棉花幼苗叶片性状解析方法   [下载全文]
Cotton Seedling Leaf Traits Extraction Method from 3D Point Cloud Based on Structured Light Imaging   [Download Pdf][in English]
投稿时间:2019-02-14  
DOI:10.6041/j.issn.1000-1298.2019.08.026
中文关键词:  棉花  幼苗叶片  结构光成像  性状解析  点云处理  三维表型检测
基金项目:中央高校基本科研业务费专项资金项目(2662018JC004)和国家自然科学基金项目(31600287)
作者单位
黄成龙 华中农业大学 
李曜辰 华中农业大学 
骆树康 华中农业大学 
杨万能 华中农业大学 
朱龙付 华中农业大学 
中文摘要:针对传统的棉花叶片表型测量方法主观、低效,对复杂性状如卷叶程度、黄叶占比等很难量化的问题,提出一种基于结构光三维成像的棉花幼苗叶片性状解析方法。首先,采用结构光扫描仪获取棉花幼苗的三维点云数据;然后,利用直通滤波、超体聚类、条件欧氏距离算法,实现叶片点云的识别与分割;最后,基于分割的叶片点云,采用三角面片化、随机采样一致性、Lab颜色分割等处理,实现叶片面积、周长、生长角度、卷曲度、黄叶占比等参数的快速、准确、无损提取。对40株棉花幼苗进行三维结构光成像试验,结果表明,3D叶片面积、周长测量的平均绝对误差分别为2.59%、2.85%,具有较高的测量精度,还证明叶片卷曲度和黄叶占比能显著区分病叶和正常叶。
HUANG Chenglong  LI Yaochen  LUO Shukang  YANG Wanneng  ZHU Longfu
Huazhong Agricultural University,Huazhong Agricultural University,Huazhong Agricultural University,Huazhong Agricultural University and Huazhong Agricultural University
Key Words:cotton  seedling leaf  structured light imaging  traits analysis  point cloud processing  three dimensional phenotypic traits detection
Abstract:Cotton is an important agricultural crop in China, which is related to national economy and people’s life. The production, consumption and import of cotton in China always keep the front place in the world. Cotton leaves are the main organs controlling photosynthesis and transpiration, and the seedling leaves have significant influence on cotton yield and disease resistance. Therefore, accurate quantification of cotton seedling leaf traits is necessary and helpful for the cotton breeding, disease resistance research and functional gene mapping. However, the traditional method for the leaf traits investigation is generally manual measurement, which is labor intensive, subjective, and even destructive. To solve the problem, a novel method was demonstrated to extract cotton seedling leaf traits from 3D point cloud based on structured light imaging. In the study, the 3D point cloud data, including color information was acquired by the structured light scanner. Specific point cloud processing pipeline was developed to identify each leaf, by applying pass through filtering, super voxel and conditional Euclidean clustering algorithms. Based on the segmented leaf point clouds, the leaf traits, including leaf area, leaf perimeter, leaf angle, leaf rolling degree and leaf yellow ratio were extracted accurately by using triangular patches generation, random sampling consensus, and Lab color space segmentation algorithms. To evaluate this method, 40 cotton plants treated by verticillium wilt virus were measured in seedling stage, and totally 175 leaf point clouds were obtained. Totally 75 leaves were randomly selected to be cut off for manual validation, and the leaf area and perimeter were compared with manual measurements. The results showed that the mean absolute percentage error of leaf area and perimeter was 2.59% and 2.85%, respectively, the R2 values of leaf area and perimeter was 0.9973 and 0.9822, respectively. The results proved that the automatic measurement had a high accordance with manual measurements, which proved the high accuracy of this method. In addition, the left 100 leaves were divided into infected leaves and healthy leaves by manual observation, meanwhile the leaf traits were extracted with segmented point cloud data to calculate the P value by single factor analysis of variance. The measured P values were 0.099, 0.242, 0.346, 0.531, 0.002 and 0, respectively, and the results proved that the traits of leaf rolling degree, and leaf yellow ratio were able to distinguish the infected leaves from healthy leaves evidently. In conclusion, the study demonstrated an effective novel method for accurate and non destructive measurement of cotton seedling leaf traits, which would be helpful for the cotton breeding, disease resistance research and functional gene mapping research. 

Transactions of the Chinese Society for Agriculture Machinery (CSAM), in charged of China Association for Science and Technology (CAST), sponsored by CSAM and Chinese Academy of Agricultural Mechanization Science(CAAMS), started publication in 1957. It is the earliest interdisciplinary journal in Chinese which combines agricultural and engineering. It always closely grasps the development direction of agriculture engineering disciplines and the published papers represent the highest academic level of agriculture engineering in China. Currently, nearly 8,000 papers have been already published. There are around 3,000 papers contributed to the journal each year, but only around 600 of them will be accepted. Transactions of CSAM focuses on a wide range of agricultural machinery, irrigation, electronics, robotics, agro-products engineering, biological energy, agricultural structures and environment and more. Subjects in Transactions of the CSAM have been embodied by many internationally well-known index systems, such as: EI Compendex, CA, CSA, etc.

   下载PDF阅读器