Monitoring Of Greenhouse Vegetable Growth Status Based On Machine Vision

Monitoring the growth condition of Greenhouse vegetables has the important significance for the improvement of greenhouse vegetables production and maximize resource conservation. As the traditional method with human eye to watching the growing condition of vegetables has the problems of low inefficient and observations being too subjective, the technology based on machine vision is proposed in this paper to analyze the growing condition of vegetables in greenhouses, which gets the growing condition of vegetables through the analysis of the results, so as to realize the monitoring of growth of vegetables.In this paper, CCD camera is used to obtain images of the greenhouse vegetables, also the corresponding image processing algorithm is designed, which is able to calculate leaf area per plant vegetables, and leaf area parameter is used to analyse and judge the growth condition of vegetables. The concrete research content is as follows:1) The design of greenhouse vegetables image acquisition plan: according to the needs of this subject, industrial cameras, lens and the light source are chosen reasonably. Also, suitable image acquisition scheme for this subject is designed, and image of greenhouse vegetables is collected, which realizes the first step in the growth status and monitoring.2) The design of vegetables motion blur image restoration algorithm: through the analysis of the degradation and restoration of motion blurred image model, it is concluded that the image degradation of point spread function(PSF). According to each restoration algorithm for blurred image restoration after the effect of different, eventually, wiener filtering recovery blurred image is accepted. Image restoration by simulation and the experiment demonstrates verify the accuracy of the motion blur parameter.3) The segmentation of vegetables image: through the analysis of several common image segmentation algorithm, and compared the segmentation results of the algorithm, under L*a*b* color space for K-means clustering segmentation algorithm is putting forward finally, which not only achieves the target area(vegetable) and background region effective separation(soil), and also displays result in a colour image segmentation, retains the vegetable leaf information, improves the precision of identification.4) The evaluation of vegetables growth: using vegetable leaf area as a measure of its growth. By analysing a variety of common leaf area measurement method, the use of pixel points is putting forward instead of leaf area algorithm. Determining its growth status through the comparisons of vegetable leaf area.5) The developed of vegetables growing monitoring system software: designing image processing algorithms and monitoring system in C# development environment, which realizes the monitoring of vegetable growth.