| Agriculture is one of the main factors restricting China's economic development.With the in-depth study of " agriculture,rural areas and farmers",in order to realize the modernization of agricultural production,the application of high technology in agriculture is increasing day by day,which has attracted more and more attention in the field of information and intelligent precision agriculture.Gansu is one of the main planting areas of grape in China.Reducing sugar of grape is not only an important nutritional component,but also an important flavor substance,and its composition and content level are the key factors to determine the sweet and sour flavor of grape.In order to study the changes of reducing sugar content in grape leaves during its growth process,the grape leaves in the vineyard were taken as the monitoring object,and an image detection system was designed for recognizing and classifying reducing sugar content in grape leaves based on LabVIEW.On the basis of the spectral chemical determination experiments of reducing sugar in collected grape leaves during growing period,Vision Assistant was used to complete the leaf image processing,and MATLAB software was used to analyze and extract the image features of grape leaves.Finally,the online determination system was realized for reducing sugar content in grape leaves.The thesis mainly completed the following aspects of the research:1.Based on the limitations of grape vineyard and the functional requirements of the test system,the overall framework of online reducing sugar detection system was designed,and then the software and hardware were selected;2.A program for collecting and saving grape image was compiled by using vision acquisition software and vision development module software under LabVIEW software platform,after the image acquisition card and USB industrial camera were chosen,and the collecting and saving experiments of grape leaves image were implemented in real time through all grape growth season;3.The virtual programs for preprocessing and background segmentation of grape leaf images were written by using Vision Assistant under the NI-MAX software,andthen the MATLAB script node was called by LabVIEW script and formula module,which used to extract the characteristic parameters of color and texture in leaf image.Support vector machine(SVM)algorithm was finally used to construct a classifier model of sugar content classification with the samples of 487 grape leaves.The experimental results showed that the correct classification and recognition of the sugar content is as high as 87.35 % in this system;4.The human-computer interaction interface of leaf reducing sugar content was designed through LabVIEW modular programming language,which can be operated on the front panel conveniently by user,and the sugar content of the leaves can also be detected in real time by the administrators.The experimental test time of the system was from 2 to 5 minutes;5.Using LabVIEW application builder,the final LabVIEW installation package of image detection system was generated,which can be installed on the terminals without LabVIEW,and which reduced the limitation of the detection systemand made it more convenience for popularization and application;The experimental results showed that the system has higher precision and can work steadily,which provides a rapid and real-time detection method for agricultural and horticultural crops.It is of great significance to improve the economic benefits of agriculture and has more valuable reference and application to the measurement of other nutritional elements of crops in agriculture field. |
PDF Full Text Request