面向植物抗旱性研究的多源表型信息采集和分析技术

利用表型信息采集系统获取不同生长环境下的植物形态结构和生理生化数据,研究植物体对不同胁迫的反应,从而进行抗性育种和筛选优质良种.本文构建了一套由双CCD相机、热成像仪、水分控制模块、称量模块、光源等组成的多源表型信息采集系统,采用YOLO v3目标检测算法和图像处理算法提取了植物投影叶面积、株高、叶片数量、冠层温度等表...

Multi-source Phenotypic Information Collection and Analysis Techniques for Drought Resistance of Plants

The phenotypic information acquisition system can be used to obtain the morphological structure, physiological and biochemical data of plants under different growth environments, so as to study the response of plants to different stresses, and carry out resistance breeding and selection of high-quality varieties. A multi-source phenotypic information acquisition system composed of dual CCD camera, infrared thermal image, water control module, weighing module, light source, etc., was constructed. The phenotypic parameters were extracted by using the target detection algorithm of YOLO v3 and image processing algorithm, such as projected leaf area, plant height, number of leaves and canopy temperature of plants. Using Salix suchowensis Cheng as the research object, drought stress test was carried out and correlation analysis was carried out. The results showed that the plant height, projected leaf area, daily water consumption and water use efficiency of dustpan under drought stress were significantly lower than those under normal water treatment over time. Through correlation analysis, it can be concluded that the projected leaf area of Salix suchowensis Cheng presented a positive linear correlation with its fresh weight and dry weight, and the determination coefficients were 0.89 and 0.78, respectively. There was a positive correlation between plant fresh weight and plant dry weight, and the determination coefficient was 0.76. There was a positive correlation between the cutting days and the average canopy temperature difference, and the determination coefficient was 0.79. The difference of daily water consumption was positively correlated with the number of days after cutting, and the determination coefficient was 0.84. The canopy temperature difference was positively correlated with the daily water consumption difference, and the determination coefficient was 0.85. By collecting plant phenotypic parameter information from multiple sources and carrying out association analysis, a feasible technique was provided for real-time and continuous assessment of plant growth-related parameters under drought stress, which can guide the application of water in resistant species and the rapid screening of drought-resistant genotypes.