| In crop breeding,cultivation and other related research it is necessary to measure phenotypic trait parameter information such as leaf area,plant height and biomass of crops.In the traditional corn breeding and cultivation process,the measurement of phenotypic parameters such as leaf area,Plant height and biomass is done by manual operation,which is time-consuming,laborious and inefficient.Therefore,the design and development of a phenotypic trait formultiple com seedlings can be carried out.The detection device,which realizes the rapid detection of phenotypic information of corn seedlings,is of great significance in the field of breeding.In this project,a machine vision-based phenotypic trait detection device for corn seedlings was designed and constructed to achieve rapid detection of phenotypic traits in com seedlings.At the same time,the relevant control circuits were constructed and the relevant control programs were written.The image acquisition software and image processing software were written.The com seedlings are used as research objects to test the phenotypic traits of com seedlings and verify the working performance of the device.At the same time,the phenotypic information of corn seedlings was obtained and used to construct a predictive model for the phenotypic traits of corn seedling The main tasks of the project were as follows:1.In order to realize the detection function a phenotypic trait detection device for com seedlings was designed and components were assembled according to functional needs,and the control circuit based on Arduino control board was built,and the control program was wrote,which could automatically obtain the images of cron seedlings from different perspectives according to the test requirements;2.Based on Vision C++6.0 and Mil 9.0 software,the image acquisition software and image processing software were developed.The interactive interface was used to control the detection device to realize automatic collection and processing of corn seedling images and obtain relevant phenotypic data.3.The phenotypic test was carried out with com seedlings as the research object.The image of corn seedlings was colleeted on the detection device and the working performance of the device was tested.The image was analyzed and processed to obtain the phenotypic data of the corn seedling image.The data on leaf area plant height and fresh weight of maize seedlings were measured by artificial destructiveness.4.Regression analysis was carried out on the phenotypic data of corn seedlings,and the prediction models of phenotypic parareters such as leaf area,fresh weight and plant height were constructed.The accuracy of the model was verified.At the same time,multiple corn seedlings were continuously monitored to explore the growth pattern of corn seedlings.The overall length,width and height of the phenotypic trait detection device designed and constructed by this subject was 1250*1210*2000mm.The effective running area of the two-dimensional mobile platform was 1100*800mm and the maximum lifting height of the seedling potting lifting device was 200mm.The test results showed that when the moving speed of the camera in the X direction and the Y direction were 830 mm·s-1 and 32 mm·s-1,respectively,the total time of one test was 545s and the average time of single seedlings was 34s.The phenotypic trait deteeting device could automatically and quickly collect three views of the corn seedlings.The detecting device was simple in control and convenient to use,and it had high detection efficiency and better realized the detecting function.The results of data analysis indicated that the three-phenotypic information of maize seedlings could reflect the actual phenotypic information of maize seedlings,so maize seedlings could be continuously monitored based on the image phenotype information of maize seedlings.The results of continuous monitoring experiments showed that the growth of maize seedlings had certain regularity and the growth of maize seedlings could be analyzed according to this,which provided a feasible schebe for related breeding experiments. |
PDF Full Text Request