Study On Ripening On Cerasus Humilis Based On Electronic Nose Technology

Cerasus humilis(Bge.)Sok.,also known as calcium fruit,is rich in mineral elements,vitamins and anthocyanins,which has high nutritional value and broad market prospects.In the production,there is often the use of ripening agent treatment to promote the fruit in advance of the market,so as to obtain higher benefits.In this study,cerasus humilis,Nongda No.6 was used as experimental material,and ethephon was sprayed to accelerate the ripening of cerasus humilis when cerasus humilis were 70% mature and 80% mature respectively,and the natural mature cerasus humilis were used as control group.In order to identify whether the cerasus humilis adopt ripening methods,a self-developed electronic nose test system equipped with 12(TGS2612(Japan Figaro),GSBT11(Korea Ogam),WSP2110(China Weisheng),MP135(China Weisheng),MS1100(Korea Ogam),MP901(China Weisheng),TGS2611(Japan Figaro),TGS2620(Japan Figaro),TGS2602(Japan Figaro),TGS2610(Japan Figaro),TGS2600(Japan Figaro),TGS2603(Japan Figaro))sensors were used to collect odor information data.And random forest(RF),support vector machine(SVM)and k-Nearest Neighbors(KNN)pattern recognition algorithms were used for analysis,which can identify different sample types,so as to provide the basis for the quality detection of cerasus humilis fruit in the market,and provide a theoretical reference for the better application of electronic nose in the quality detection of cerasus humilis.The main results are as follows:1.Study on the changes of nutritional components in natural ripening and ripening cerasus humilis based on physical and chemical experiments.The contents of titratable acid,soluble protein,soluble sugar,vitamin C and tannin in natural ripe and ripening cerasus humilis fruits were determined.The results showed that the contents of soluble sugar and soluble protein in seven ripening and eight ripening were significantly lower than those in natural ripening.Moreover,the vitamin C content of the ripened fruit was lower than that of the natural ripened fruit.The difference between the vitamin C content of the ripened fruit and the natural ripened fruit reached a significant level.In addition,the contents of titratable acid and tannin in the ripening treatment were higher than those in the natural ripening treatment.Therefore,from the perspective of nutrition,the nutritional value of natural ripening cerasus humilis fruit is higher than that of ripening cerasus humilis fruit2.Study on rapid recognition of ripening treatment in intact fruit of cerasus humilis based on electronic nose detection technology.Electronic nose detection technology was used to collect the data of seven ripening cerasus humilis,eight ripening cerasus humilis and natural ripening cerasus humilis.By extracting the maximum value to obtain the eigenvalue of the response signal,the principal component analysis(PCA)was used to reduce the dimension of the data,and the random forest(RF),support vector machine(SVM)and k-Nearest Neighbors(KNN)were used to analyze the data,so as to explore the feasibility of e-nose technology for ripening recognition of cerasus humilis.The experimental results showed that the classification recognition rate of the data set reduced by principal component analysis after random forest,support vector machine and k-Nearest Neighbors processing was higher than that of the original data set,which is higher on average,after dimensionality reduction,and the recognition accuracies was 90.74%,88.25% and 87.81%,respectively.Among them,random forest had the highest recognition rate of 90.74%.Compared with the k-Nearest Neighbors method,the classification effect were not ideal and the recognition effect was poor.The random forest,support vector machine and k-Nearest Neighbors method can effectively identify ripening cerasus humilis after dimension reduction.The electronic nose technology can quickly identify the ripening cerasus humilis,and the classification effect were ideal.3.Rapid recognition of ripening treatment in beating cerasus humilis based on electronic nose detection technology.After beating and processing the complete cerasus humilis,an electronic nose equipped with12 metal oxide semiconductor(MOS)sensors were used for data collection.The maximum value is extracted by MATLAB to obtain the eigenvalue of the response signal,and the principal component analysis was used to reduce the dimension of the data.Random forest(RF),support vector machine(SVM)and k-Nearest Neighbors(KNN)were used to analyze the data to explore the feasibility of e-nose technology for ripening recognition of beating cerasus humilis,and the results of beating cerasus humilis and complete fruit cerasus humilis were compared and analyzed.The experimental results showed that the classification recognition rate of random forest,support vector machine and k-Nearest Neighbors is higher than that of the original data set,and after dimensionality reduction,the recognition accuracy was 95.35%,92.49% and 90.57%,respectively.Compared with support vector machine and k-Nearest Neighbors method,random forest classifier has the highest recognition rate of 95.35%.Whether the whole fruit or the beating fruit was used,it can better identify whether the cerasus humilis fruit was ripening or not,but the precision of beating fruit was higher than that of the whole fruit,which verifies that the electronic nose can quickly,accurately,nondestructive and efficiently identify the cerasus humilis fruit processed by ripening means,so as to provide the basis for the quality detection of the market cerasus humilis fruit.