Design Of Seedling Box Environment Control System Based On Image Processing

Nowadays,people's demand for green vegetables is becoming stronger,the demand for the quality of vegetables is becoming more and more high.During the growth of vegetables,the development of seedling stage plays a decisive role in the quality of vegetables.The traditional vegetable seedling has strong dependence on the external environment and farmers' planting experience,and there are many problems such as poor facilities,occupied land,unsound environmental control methods and so on.In view of the current research on seedling environmental control,the environmental parameters were not adjusted according to the actual growth of seedlings.A environmental control system of seedling box based on image processing is proposed.The stem diameter is used as a characteristic parameter to characterize the growth stage of the seedlings.The suitable temperature value of seedlings at the current growth stage is deduced by measuring the diameter of stem and diameter of stem.,and the micro controller controls the actuator to achieve the environment control in the seedling box.The main work of this paper will be summarized as following 6 aspects:(1)The seedling box environment control system scheme is designed.The system scheme includes the design of the collection scheme of the seedling growth state information,the design of the environment control scheme of the seedling box and the design of the human-computer interaction platform.Select the image processing methods to obtain seedling stem diameter;The temperature of seedling box has the characteristics of nonlinearity and hysteresis.It is difficult to establish mathematical model.Therefore,the fuzzy reasoning method is used to inference the temperature setting value of the suitable seedling growth stage based on the change of stem diameter and stem diameter,and using PID method to achieve the control.Other environmental factors,such as humidity and illumination,are controlled in the suitable range.meanwhile,the user interface is designed to storage,inquiry,analysis and manual control of theseedling box environment.(2)The hardware of seedling box environment control system is designed.According to the needs of processing and storing the image,the raspberry pie is selected as the micro controller.Image acquisition need to minimize the degree of occlusion among the seedlings,therefore,a dual axis steering gear camera is chosen as the image acquisition module.The seedlings images are collected by the rotation of the camera around 32 degrees~40 degrees.According to the requirement of seedling growth,the semiconductor refrigeration and heating module is selected as the executing mechanism of the temperature control.The HMI serial screen is selected to accomplish the human-computer interaction function.(3)The design and Simulation of the algorithm for measuring the diameter of the seedling stem based on image processing.The image characteristics of the seedlings is analyzed,In the Windows7+Python3 environment,the OpenCV visual library is used to design and simulate the stem diameter measurement algorithm.First of all,the camera angle deviates from the front of Seedling plate and the image has the distortion,so the camera is calibrated to complete the distortion correction of the seedling image.Due to the noise and background interference of seedling image,the image is denoised and enhanced.In view of the single threshold is difficult to apply to the multiple seedling images,the Otsu automatic threshold segmentation method is used to remove the seedling background.In view of the cross adhesion of the stem in the seedling image,the method of extracting the skeleton line of the seedling was used to divide the cross adhesive stem in the seedling image.The diameter of the stem is calibrated to complete the transformation of the pixel value of the stem diameter to the actual value.(4)The design and Simulation of the environment control system of the seedling box.Design fuzzy inference system,stem diameter D and stem diameter variation ?D is the input of fuzzy inference system,the output of the fuzzy inference system is the temperature setting value.Temperature control using PID method to achieve the control of the temperature inside the seedling box.The control system model of Simulink in MATLAB is built and simulated to verify the feasibility of the fuzzy inference combined with the PID control method.(5)The human-computer interaction interface of seedling box environment control system is designed.Design the man-machine interaction interface,build the HMI serial interface and PC interface to monitor and control the environment information in the seedling box.And the storage of environmental information is achieved by database technology,which provides a basis for later seedling environment data analysis.(6)Experiment of tomato seedlings and result analysis.Compare the experimental results of the seedling box environmental control system with the traditional nutrition bowl seedling experiment.The test results show that the system can through image processing methods of real-time measurement of seedling stem diameter,and the relative error of measurement remain in the range of 5%;the system can stabilize the temperature in the seedling box at about 20?~30?,the absolute error of 1?,the humidity can be stabilized at around 60%~70%RH,the system can meet the seedling requirements.Reducing the dependence on the external environment and farmers' planting experience.The system can Embody scientific planting pattern and improve economic benefit,and promote the cultivation of high quality seedlings.