Limits...
Identification of pesticide varieties by testing microalgae using Visible/Near Infrared Hyperspectral Imaging technology.

Shao Y, Jiang L, Zhou H, Pan J, He Y - Sci Rep (2016)

Bottom Line: In our study, the feasibility of using visible/near infrared hyperspectral imaging technology to detect the changes of the internal components of Chlorella pyrenoidosa so as to determine the varieties of pesticides (such as butachlor, atrazine and glyphosate) at three concentrations (0.6 mg/L, 3 mg/L, 15 mg/L) was investigated.The RC-LDA model, which achieved an average correct classification rate of 97.0% was more superior than FW-PLSDA (72.2%) and CARS-PLSDA (84.0%), and it proved that visible/near infrared hyperspectral imaging could be a rapid and reliable technique to identify pesticide varieties.It also proved that microalgae can be a very promising medium to indicate characteristics of pesticides.

View Article: PubMed Central - PubMed

Affiliation: College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China.

ABSTRACT
In our study, the feasibility of using visible/near infrared hyperspectral imaging technology to detect the changes of the internal components of Chlorella pyrenoidosa so as to determine the varieties of pesticides (such as butachlor, atrazine and glyphosate) at three concentrations (0.6 mg/L, 3 mg/L, 15 mg/L) was investigated. Three models (partial least squares discriminant analysis combined with full wavelengths, FW-PLSDA; partial least squares discriminant analysis combined with competitive adaptive reweighted sampling algorithm, CARS-PLSDA; linear discrimination analysis combined with regression coefficients, RC-LDA) were built by the hyperspectral data of Chlorella pyrenoidosa to find which model can produce the most optimal result. The RC-LDA model, which achieved an average correct classification rate of 97.0% was more superior than FW-PLSDA (72.2%) and CARS-PLSDA (84.0%), and it proved that visible/near infrared hyperspectral imaging could be a rapid and reliable technique to identify pesticide varieties. It also proved that microalgae can be a very promising medium to indicate characteristics of pesticides.

No MeSH data available.


Related in: MedlinePlus

The effective wavelengths selected by regression coefficients.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4829843&req=5

f5: The effective wavelengths selected by regression coefficients.

Mentions: Although the identification model built by CARS-PLSDA seems better than other models, the disadvantage of requiring 71 variables (wavelengths) to establish a model is still unacceptable in some cases for taking much more calculation time, and on the other hand, its identification accuracy was not high enough. Therefore, an optimal model which consumes less time and has higher identification accuracy seems much more attractive and necessary. The regression coefficient of the variables (wavelengths) plays a rather important role in the PLS regression27. The absolute values of the peaks indicate the contribution of wavelengths at these positions in regression model. Wavelengths with high regression coefficient were selected for further analysis, but wavelengths with small coefficients were excluded for little contribution to improving the productivity of the model. In the study, the calibration model was built using all hyperspectral preprocessing data at a concentration of 15 mg/L on day 5, and four effective variables (474, 512, 650 and 692 nm) were selected by regression coefficients which is shown in Fig. 5. Compared to the wavelengths selected by CARS, the number of variables selected by regression coefficients was obvious less and more redundant information was excluded. Furthermore, more useful information was also picked out. In Fig. 5, the wavelength at 474 nm was associated with carotenoids b28, and 512, 650 and 692 nm might be related to chlorophyll192930. LDA combined with effective variables selected by regression coefficients was applied in our investigation to build another model to identify the pesticides.


Identification of pesticide varieties by testing microalgae using Visible/Near Infrared Hyperspectral Imaging technology.

Shao Y, Jiang L, Zhou H, Pan J, He Y - Sci Rep (2016)

The effective wavelengths selected by regression coefficients.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4829843&req=5

f5: The effective wavelengths selected by regression coefficients.
Mentions: Although the identification model built by CARS-PLSDA seems better than other models, the disadvantage of requiring 71 variables (wavelengths) to establish a model is still unacceptable in some cases for taking much more calculation time, and on the other hand, its identification accuracy was not high enough. Therefore, an optimal model which consumes less time and has higher identification accuracy seems much more attractive and necessary. The regression coefficient of the variables (wavelengths) plays a rather important role in the PLS regression27. The absolute values of the peaks indicate the contribution of wavelengths at these positions in regression model. Wavelengths with high regression coefficient were selected for further analysis, but wavelengths with small coefficients were excluded for little contribution to improving the productivity of the model. In the study, the calibration model was built using all hyperspectral preprocessing data at a concentration of 15 mg/L on day 5, and four effective variables (474, 512, 650 and 692 nm) were selected by regression coefficients which is shown in Fig. 5. Compared to the wavelengths selected by CARS, the number of variables selected by regression coefficients was obvious less and more redundant information was excluded. Furthermore, more useful information was also picked out. In Fig. 5, the wavelength at 474 nm was associated with carotenoids b28, and 512, 650 and 692 nm might be related to chlorophyll192930. LDA combined with effective variables selected by regression coefficients was applied in our investigation to build another model to identify the pesticides.

Bottom Line: In our study, the feasibility of using visible/near infrared hyperspectral imaging technology to detect the changes of the internal components of Chlorella pyrenoidosa so as to determine the varieties of pesticides (such as butachlor, atrazine and glyphosate) at three concentrations (0.6 mg/L, 3 mg/L, 15 mg/L) was investigated.The RC-LDA model, which achieved an average correct classification rate of 97.0% was more superior than FW-PLSDA (72.2%) and CARS-PLSDA (84.0%), and it proved that visible/near infrared hyperspectral imaging could be a rapid and reliable technique to identify pesticide varieties.It also proved that microalgae can be a very promising medium to indicate characteristics of pesticides.

View Article: PubMed Central - PubMed

Affiliation: College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China.

ABSTRACT
In our study, the feasibility of using visible/near infrared hyperspectral imaging technology to detect the changes of the internal components of Chlorella pyrenoidosa so as to determine the varieties of pesticides (such as butachlor, atrazine and glyphosate) at three concentrations (0.6 mg/L, 3 mg/L, 15 mg/L) was investigated. Three models (partial least squares discriminant analysis combined with full wavelengths, FW-PLSDA; partial least squares discriminant analysis combined with competitive adaptive reweighted sampling algorithm, CARS-PLSDA; linear discrimination analysis combined with regression coefficients, RC-LDA) were built by the hyperspectral data of Chlorella pyrenoidosa to find which model can produce the most optimal result. The RC-LDA model, which achieved an average correct classification rate of 97.0% was more superior than FW-PLSDA (72.2%) and CARS-PLSDA (84.0%), and it proved that visible/near infrared hyperspectral imaging could be a rapid and reliable technique to identify pesticide varieties. It also proved that microalgae can be a very promising medium to indicate characteristics of pesticides.

No MeSH data available.


Related in: MedlinePlus