Study On Automated Observation Technology Of Characteristic Parameters Of The Main Crops Growth

Agro-meteorological observations are the basis of agro-meteorological servicesand research, and observations of crop growth is an important part of it, and byobserving crop growth to promptly take appropriate management measures, such asEvaluation of the influence of meteorological conditions and forecasting crop yield, toensure that the normal growth of crops. Present in our country agro-meteorologicalobservations mainly rely on artificial observation, modernization is lagging behind.Technology is far from being able to meet the high precision of modern agriculture andhigh limitation of observational requirements. There is an urgent need to improve theautomation of agro-meteorological observation capabilities.China Meteorological Administration, Meteorological Observation Center,presided over the public service sectors of the Ministry of Science and Technology(WMO) research projects, agro-meteorological observation automation system, to carryout related research work. The industry projects through developing a method ofcombining image-based mutation and gradient characteristics, to achieve the automaticobservation of the characteristic parameters of crop(wheat, corn, rice, cotton)，such asgrowth period, coverage, growth, density, etc. In this paper by using trial data ofautomation observation carry out research of the online crop coverage.Coverage is one of the important growth parameters of crop growth, reflecting thecombined effects of environment on crop. In the paper, through the developedequipment, automatic and manual comparison test data of Crop coverage, that is forwheat and corn for two years in three locations, carry out quality control of the cropcoverage, analysis of the variation of coverage of corn and wheat, and correlationbetween coverage and other crop growth parameters. Giving the following conclusions:(1) Since the equipment of comparative test is permanently installed in the fields.Crop image automatically collected are not clear because of being impacted byprecipitation, high winds, fog, snow and other severe weather, which will lead to falsecalculation of coverage. Crop images also is influenced by the changes in light intensity in the day, calculation of coverage will produce significant fluctuations, Thus, need todo real-time quality control of image acquisition. According to the field microclimateobservations, images captured under abnormal disasters, weather conditions should beremoved. Image processing techniques cannot avoid the light, noise and other factors,leading to coverage fluctuation, it also should be controlled: The coverage of thesummer corn within a day should not be greater than6.4%, the coverage of winterwheat one day change should be in8.6%.(2) The coverage of the summer corn showed a single peak curve trend throughoutthe developmental period, coverage in the corn tasseling period around reached its peak,with an average of about87.5%. Coverage of winter wheat showed a bimodal curvetrend throughout the developmental period, respectively, before the start of the winterperiod reached the first peak, approximately45.3%, after the end of the winter period,the winter wheat to restore growth, around jointing stage to booting stage reached thesecond peak, approximately68.2%. Coverage of the summer corn and winter wheat inthe milk stage rapidly dropped to a minimum value.(3) Through analysis and testing, correlation coefficient of coverage and LAI of thesummer corn is0.94, of coverage and LAI the winter wheat is0.78, all through a verysignificant test. Coverage and LAI of the two crops is an exponential curve relationship.Through testing model, the relative error of summer corn LAI values estimated by theexponential model and the measured LAI is of6.04%, that is of the winter wheat is of7.64%. Exponential model has higher simulation accuracy. The results lay thefoundation for estimating LAI through real-time coverage of observations.(4) Summer corn coverage and plant height, growth amount is a linear relationship,correlation coefficients above0.8. Explain that changes of corn plant height and growthamount to affect the changes of plant cover; Winter wheat Coverage and plant heighthas a low correlation. Relationship of winter wheat Coverage and growth amount isdifferent of three trial sites. Reason may be as follows: wheat covering algorithm shouldbe improved, artificial observation data might exist errors, or the change of a singleelement had no significant effect on wheat coverage, subject to further experimentalstudy of in-depth analysis.