Research On Intelligent Management System Of Grafted Cucumber Seedlings

Grafting as an efficient technique for disease resistance and yield increase, is able to solve the succession cropping obstacle, and the grafted seedlings can not only overcome the soil-disseminated diseases, also can promote the growth of the plants, increase the yield, and enhance the stress resistance. This paper was supported by National High Technology Research and Development Program(863 plan) “ Research on Key Technology and Complete Sets of Equipment of Full Automatic Grafting Seedlings ”(2012AA10A506)”. Taking cucumber grafted seedlings as the research object, optimal environmental parameters provide the control standard for intelligent management system; different grafting methods and rapid rooting techniques provide the theoretical foundation and basis for the high quality and high survival rate of grafted seedlings.Combining the growth indexes and image technology, studied the quality grading standard and grading model of grafted cucumber seedlings, and built the intelligent management system of cucumber grafted seedlings. The main content are as follows:1.Extract the optimal environmental information from the cucumber grafted seedlingsUsing segmentation management, grafted cucumber management is divided into two stages, respectively for healing period and survival period, taking 3 days as the critical value, get optimal environment of each stage. During healing period, grafted seedlings need to avoid the light, so we chose 4 levels of temperature(T) and 2 levels of relative humidity(RH, hereinafter referred to as the humidity) as variables, survival rate as an index value, establish experimental design table. The results showed that humidity of 95% and temperature 22~28℃ as the optimal management conditions. Based on the optimal condition of the healing period, the survival period selected 3 factors to studing, 3 levels of temperature(T), 3 levels of humidity(RH) and 3 levels of light(L) as variables, survival rate and of false survival rate as index values. Experimental design table and analysis of the test results by using design-expert software and response surface methodology. To establish the quadratic regression equation between survival rate and variables. Considering impact of environment on the survival rate, false survival rate and actual situation, the best environmental conditions of survival period are temperature 26±2℃, humidity 85±3% and light 3000 Lx.2.Different grafting methods and rapid rooting techniquesThe results show that the survival rate of root-cutting grafted seedlings reaches 99.31% and the survival rate of close joining grafted seedlings reaches 96.53%.What’s more, root-cutting grafted seedlings are basically in the same size, which is convenient for the mass production of grafting machine. Analyzing the plant height, stem diameter, leaf area and root / shoot ratio, we know that the grafted seedlings were better than those of the self-rooted seedlings, and the root-cutting grafted seedlings measurement indexes are significantly higher than that of self-rooted seedlings. Analyzing the root length, total root surface area, total root surface area, total root volume and root activity, we know that the grafted seedlings were better than those of the self-rooted seedlings, and the root-cutting grafted seedlings measurement indexes are significantly higher than that of close joining seedlings and self-rooted seedlings. Grafting increased the net photosynthetic rate of cucumber leaves, and the net photosynthetic rate of root-cutting grafted seedlings being the maximum. The plant growth regulator concentrations and matrix combinations which were suitable for the regeneration of the root of the root were screened out, respectively were 100 ppm IBA and 2JZ+1ST.3.Establish quality grading standards and quality grading model for grafted seedlingsThe surviving grafted seedlings transferred to the daily management after completing quality grading. In this study, the plant height, stem diameter ratio, leaf width, incision joint area and disease grade as the grading indexes, using dynamic clustering algorithm to establish the quality grading standard of grafted cucumber seedlings. In order to overcome the disadvantages of human eyes at grading the degree of plant diseases, this paper presents a method to grade the degree of plant diseases based on improved GA and clustering algorithm.By using image processing and data mining methods, quality grading recognition model of grafted seedlings was established. Firstly, the image features were extracted by principal component analysis(PCA), independent component analysis(ICA), histogram principal component analysis(CH-PCA), gray level co-occurrence matrix(GSM), invariant moment(IM). Then using four kinds of modeling methods to build models,respectively for discriminant analysis(DA), BP neural network(BPNN), support vector machine(SVM) and relevance vector machine(RVM), the results of test samples for models validation show that CH-PCA method for image feature extraction parameters with RBFNN modeling methods and ICA method for image feature extraction parameters with RBF kernel function of SVM modeling both reach 92.2%, the highest recognition accuracy of modeling.4.Establish healing intelligent management system of grafted seedlingsThis intelligent management system of grafted seedlings can achieve three functions: continuous monitoring of growth conditions, grafted seedling environment automatic control and automatic grading of quality. The controller of control system was based on the optimal environment parameter values of grafted seedlings, by analyzing the overall demand of healing intelligent control system, hardware and software designs are conducted respectively for data acquisition node module, data center node module and the control terminal node module, and works, features, functionality and flow of each module neutron node were introduced in detail, and collection function, transfer function, the control function are tested respectively. PC system software using Labview programming and modular programming techniques to achieve the received data, display, storage, and parameter setting, historical data management and 24-hour monitoring of the entire life cycle of grafted seedlings and other functions. The results show that the parameters measured by this system and information transformed by this system are accurate, and peripheral mechanism operating in the specified mode, and the collected images by the growth monitoring system are clear, greatly improve the scientific management level and production efficiency of grafted seedlings, provide the methods and basis for the key technology of the factory grafted seedlings.