Statistical Analysis of Photopyroelectric Signals using Histogram and Kernel Density Estimation for differentiation of Maize Seeds

J. E. Rojas-Lima, A. Domínguez-Pacheco, C. Hernández-Aguilar, A. Cruz-Orea

Research output: Contribution to journalArticleResearchpeer-review

1 Citation (Scopus)

Abstract

© 2016, Springer Science+Business Media New York. Considering the necessity of photothermal alternative approaches for characterizing nonhomogeneous materials like maize seeds, the objective of this research work was to analyze statistically the amplitude variations of photopyroelectric signals, by means of nonparametric techniques such as the histogram and the kernel density estimator, and the probability density function of the amplitude variations of two genotypes of maize seeds with different pigmentations and structural components: crystalline and floury. To determine if the probability density function had a known parametric form, the histogram was determined which did not present a known parametric form, so the kernel density estimator using the Gaussian kernel, with an efficiency of 95 % in density estimation, was used to obtain the probability density function. The results obtained indicated that maize seeds could be differentiated in terms of the statistical values for floury and crystalline seeds such as the mean (93.11, 159.21), variance (1.64 × 10 3, 1.48 × 10 3) , and standard deviation (40.54, 38.47) obtained from the amplitude variations of photopyroelectric signals in the case of the histogram approach. For the case of the kernel density estimator, seeds can be differentiated in terms of kernel bandwidth or smoothing constant h of 9.85 and 6.09 for floury and crystalline seeds, respectively.
Original languageAmerican English
JournalInternational Journal of Thermophysics
DOIs
StatePublished - 1 Sep 2016

Fingerprint

histograms
statistical analysis
Seed
seeds
Statistical methods
probability density functions
estimators
Probability density function
Crystalline materials
smoothing
standard deviation
bandwidth
Bandwidth
Industry

Cite this

@article{81fd9c606c3744aa8cc97be74e3f2670,
title = "Statistical Analysis of Photopyroelectric Signals using Histogram and Kernel Density Estimation for differentiation of Maize Seeds",
abstract = "{\circledC} 2016, Springer Science+Business Media New York. Considering the necessity of photothermal alternative approaches for characterizing nonhomogeneous materials like maize seeds, the objective of this research work was to analyze statistically the amplitude variations of photopyroelectric signals, by means of nonparametric techniques such as the histogram and the kernel density estimator, and the probability density function of the amplitude variations of two genotypes of maize seeds with different pigmentations and structural components: crystalline and floury. To determine if the probability density function had a known parametric form, the histogram was determined which did not present a known parametric form, so the kernel density estimator using the Gaussian kernel, with an efficiency of 95 {\%} in density estimation, was used to obtain the probability density function. The results obtained indicated that maize seeds could be differentiated in terms of the statistical values for floury and crystalline seeds such as the mean (93.11, 159.21), variance (1.64 × 10 3, 1.48 × 10 3) , and standard deviation (40.54, 38.47) obtained from the amplitude variations of photopyroelectric signals in the case of the histogram approach. For the case of the kernel density estimator, seeds can be differentiated in terms of kernel bandwidth or smoothing constant h of 9.85 and 6.09 for floury and crystalline seeds, respectively.",
author = "Rojas-Lima, {J. E.} and A. Dom{\'i}nguez-Pacheco and C. Hern{\'a}ndez-Aguilar and A. Cruz-Orea",
year = "2016",
month = "9",
day = "1",
doi = "10.1007/s10765-016-2097-2",
language = "American English",
journal = "International Journal of Thermophysics",
issn = "0195-928X",
publisher = "Springer New York",

}

TY - JOUR

T1 - Statistical Analysis of Photopyroelectric Signals using Histogram and Kernel Density Estimation for differentiation of Maize Seeds

AU - Rojas-Lima, J. E.

AU - Domínguez-Pacheco, A.

AU - Hernández-Aguilar, C.

AU - Cruz-Orea, A.

PY - 2016/9/1

Y1 - 2016/9/1

N2 - © 2016, Springer Science+Business Media New York. Considering the necessity of photothermal alternative approaches for characterizing nonhomogeneous materials like maize seeds, the objective of this research work was to analyze statistically the amplitude variations of photopyroelectric signals, by means of nonparametric techniques such as the histogram and the kernel density estimator, and the probability density function of the amplitude variations of two genotypes of maize seeds with different pigmentations and structural components: crystalline and floury. To determine if the probability density function had a known parametric form, the histogram was determined which did not present a known parametric form, so the kernel density estimator using the Gaussian kernel, with an efficiency of 95 % in density estimation, was used to obtain the probability density function. The results obtained indicated that maize seeds could be differentiated in terms of the statistical values for floury and crystalline seeds such as the mean (93.11, 159.21), variance (1.64 × 10 3, 1.48 × 10 3) , and standard deviation (40.54, 38.47) obtained from the amplitude variations of photopyroelectric signals in the case of the histogram approach. For the case of the kernel density estimator, seeds can be differentiated in terms of kernel bandwidth or smoothing constant h of 9.85 and 6.09 for floury and crystalline seeds, respectively.

AB - © 2016, Springer Science+Business Media New York. Considering the necessity of photothermal alternative approaches for characterizing nonhomogeneous materials like maize seeds, the objective of this research work was to analyze statistically the amplitude variations of photopyroelectric signals, by means of nonparametric techniques such as the histogram and the kernel density estimator, and the probability density function of the amplitude variations of two genotypes of maize seeds with different pigmentations and structural components: crystalline and floury. To determine if the probability density function had a known parametric form, the histogram was determined which did not present a known parametric form, so the kernel density estimator using the Gaussian kernel, with an efficiency of 95 % in density estimation, was used to obtain the probability density function. The results obtained indicated that maize seeds could be differentiated in terms of the statistical values for floury and crystalline seeds such as the mean (93.11, 159.21), variance (1.64 × 10 3, 1.48 × 10 3) , and standard deviation (40.54, 38.47) obtained from the amplitude variations of photopyroelectric signals in the case of the histogram approach. For the case of the kernel density estimator, seeds can be differentiated in terms of kernel bandwidth or smoothing constant h of 9.85 and 6.09 for floury and crystalline seeds, respectively.

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84982743946&origin=inward

UR - https://www.scopus.com/inward/citedby.uri?partnerID=HzOxMe3b&scp=84982743946&origin=inward

U2 - 10.1007/s10765-016-2097-2

DO - 10.1007/s10765-016-2097-2

M3 - Article

JO - International Journal of Thermophysics

JF - International Journal of Thermophysics

SN - 0195-928X

ER -