Research On Environmental Regulation Method And External Features Detection Of Grafting Seedlings

One of the factors affects the grafting efficiency and success rate is the seedlings’condition. Currently seedlings are cultivated in greenhouse environment. Due to incompletecontrol of temperature, humidity, light and other environmental conditions, seedlings externalfeatures do not fully meet the grafting requirements，cause some wasted in the process ofgrafting. Thus, real-time features and growth conditions detection for seedlings growth incontrolled environmental conditions has some theoretical and practical significance.The primary task of nurturing standardization seedlings is seedling growth conditionsjudgement and environmental control methods. In this paper, external feature detectionmethod and cultivation system of plug rootstock seedlings were studied. In natural lightconditions of daytime, detection method of seedling external features and new plug seedlingsgrowing equipment combines detection method were studied. The main contents and resultsare as follows:(1) The requirements of grafting seedlings, the current situation, problems and trends ofnursery equipment and cultivation system were determined by research. The requirements ofgrafting seedlings external features and environmental conditions were get by queries andresearch data. The problems of environment conditions and greenhouse seedlings werediscovered from field testing. The research directions are clear through investgating themodern agricultural high-tech and greenhouse nursery research.(2) A platform for plug seedling cultivation with detection function was designed. Theplatform includes plug seedling growth detection devices and environmental control devices.According to the requirements of standard seedlings parameters, a three-dimensional devicewhich can detect the status of seedling growth automatically was designed. Free movement ofx, y, z in three directions makes it possible that the camera can catch images from top viewand side view. Environmental control devices include the box, plug drip device, fill lightdevice, plug bracket and LED lights. There is a controller on the box. LED lights are mountedon the inner side wall. Plug bracket are mounted inside the box stratification. Plugs are placedon the bracket. Drip irrigation system and fill light device is installed in the frame above theplug. A multilayer nursery unit with drip and light photography is proposed. There are many layers to improve the seedling raise efficiency and save cost. Drip irrigation combined withlight photography, not only to ensure adequate and uniformity watering, water resourcessaving, but also ensure seedlings receiving light evenly, improve the survival rate anduniformity of seedlings.(3) An extraction method of leaf and hypocotyl grafting connection position (referencepoint) is present. White pumpkin seeds were used to study extraction method of the referencepoint according to external characteristics of seedlings. First, by contrasting acquired image inRGB channels and Lab channels, color channel a*was selected for thresholding. And then,morphological operation was applied to remove noise of the target binary image. After that,the binary image was scanned line by line from top to bottom to get the number of pixels ineach row and draw a scatter plot. Then points after the maximum point were selected foranalyzing. Finally, three methods of hand access point, curve fitting and discrete point rangespecified access points were employed to extract the reference point. The results show that themethod of specified range access points to extract the location of the reference point is themost stable method, Select the manner specified range of access points used to extract thereference point position for visual feature extraction seedlings.(4) By research on rootstock seedlings, an automatic detection method for externalfeatures of grafting seedlings based on mathematical modeling was studied. The detectingitems included growth status (straight or curved, bending direction), cotyledons parameters(cotyledon flare angle, cotyledons flare spans), hypocotyls parameters (curvature, hypocotyllength, hypocotyl shaft), and other external parameters. First, image preprocessing was usedto extract the binary image. Then, a reference point and its position were determined bystatistic of horizontal pixels. Next, growth status was decided by a set inclination angle of theminimal bounding rectangle of the hypocotyl and its width. After that, the cotyledons spanwas calculated by the distance of the two cotyledons endpoint, and the cotyledons angle wascomputed by the angle between two lines that fitting with the bottom of flat cotyledons.Finally, the stem length and the coarse strains were obtained by doing slope compensations totwo sections of stem separately which was divided at the point with maximum curvature.