Design And Test Of Portable Nondestructive Detector For Cadmium Content In Lettuce Leaves

With the improvement of people’s living standards,the demand for vegetables is increasing day by day.In recent years,factors such as sewage irrigation and irrational application of agricultural materials have caused heavy metal pollution in the soil and water environment to become more and more serious,and vegetables have also been polluted to varying degrees.Cadmium is a very toxic heavy metal element.Eating cadmium-containing vegetables will endanger human health.Therefore,rapid and effective detection of heavy metal cadmium pollution in vegetables is of great significance for the safe production of vegetables and the protection of human health.Traditional chemical detection methods have a wide range of applications and good accuracy,but they also have disadvantages such as complex pretreatment and destructiveness.Spectroscopy has the advantages of high resolution and multiple bands.There have been some theoretical studies on the use of spectroscopy to detect heavy metal cadmium pollution in vegetables,but there are few reports on portable vegetable leaf cadmium content detection equipment.In this study,the heavy metal cadmium in lettuce leaves was used as the detection object,and the sensitive characteristic band was screened out by Principal Component Analysis(PCA),and a sensitive characteristic band was constructed based on the sensitive characteristic band without pretreatment,fast detection speed,small size and convenient the heavy metal cadmium content detection instrument carried can be suitable for mobile on-site detection.The research content and conclusions are as follows:(1)Lettuce samples were cultivated in a soil-free environment using perlite bagging mode,equipped with 4 cadmium stress gradients(0mg/kg,1mg/kg,3mg/kg and 5mg/kg)reagents to obtain the content of different gradients.A sample of cadmium lettuce was used to determine the cadmium content of lettuce leaves by chemometric method.The hyperspectral image of lettuce leaves was collected by a visible light-near infrared hyperspectral imaging system.The entire leaf was selected as the region of interest(ROI),and the average spectrum in the ROI was taken as Spectral data of the sample.Analyzed the correlation between chlorophyll content,cell structure,intracellular water content and other factors and spectral characteristics,and combined the average spectrum of cadmium-containing lettuce leaf samples under 4 gradients to discuss the above correlation in depth,indicating the use of spectroscopy to detect lettuce leaves the cadmium content is feasible.(2)Using the principal component analysis method,based on the weight coefficient analysis,6 characteristic bands of 500nm,550nm,680nm,760nm,810nm,940nm were screened out,and the partial least squares method(Partial least squares regression,PLSR)was used to fit the lettuce Leaf cadmium quantitative prediction model.The correlation coefficient R_C of the training set of this model is 0.9720,the root mean square error RMSE_C of the training set is 0.3142mg/kg,the correlation coefficient R_P of the test set is 0.9604,and the root mean square error RMSE_P of the test set is0.3922mg/kg.The model achieved good prediction results,indicating that it is feasible to detect the cadmium content of lettuce leaves by using 6 characteristic bands.(3)According to the selected characteristic bands,the software and hardware of the instrument are designed.The hardware parts of the designed lettuce leaf cadmium content detection instrument include:light source part,signal processing part,display part,power supply part and control part.The size of the instrument is 50mm.×70mm×60mm.Choose LED as the light source,photodiode as the receiving device,and 6 light sources with 6 photodiode receiving devices.According to the sensitivity of the photodiodes at different wavelengths,adjust the gain coefficient of each amplifier circuit to make the amplified signal output of each channel swing.The amplitude is the same and is within the lower limit of the voltage collected by the A/D circuit,so that the collected signal has the highest sensitivity and improves the accuracy of the instrument.The software design adopts a modular design idea,encapsulates each functional block into an independent function,reduces the coupling between each functional block,and facilitates subsequent modifications and maintenance upgrades.(4)A self-made instrument is used for data collection,and a linear prediction model is established.The correlation coefficient R_C of the model training set is 0.8875,the root mean square error RMSE_C of the training set is 0.4975mg/kg,the correlation coefficient R_P of the test set is 0.8656,and the test set is average the root square error RMSE_P is 0.5526mg/kg,and the model predicts better.Finally,the lettuce leaf heavy metal cadmium content spectroscopic non-destructive tester was tested,and 30 groups of cadmium-containing lettuce leaves that were not related to the modeling were selected for testing.Compared with the standard physical and chemical values,the absolute measurement error was-0.69～0.66mg/kg,and the average absolute error was0.26mg/kg,the root mean square error is 0.32mg/kg,and the average percentage error is 0.25.The results show that the instrument can realize the non-destructive detection of cadmium content in lettuce leaves.The detection instrument designed in this paper can realize the rapid and nondestructive detection of heavy metal cadmium in lettuce leaves.From theoretical analysis to equipment development,the idea is clear and relatively complete,which can provide reference for the research and development of the same type of instrument.