The Experts below are selected from a list of 105 Experts worldwide ranked by ideXlab platform
Francisco Dauria Marques - One of the best experts on this subject based on the ideXlab platform.
-
from sensor data to Munsell Color System machine learning algorithm applied to tropical soil Color classification via nix pro sensor
Geoderma, 2020Co-Authors: Marcelo Mancini, David C Weindorf, Maria Eduarda Carvalho Monteiro, Alvaro Jose Gomes De Faria, Anita Fernanda Dos Santos Teixeira, Wellington De Lima, Francielle Roberta Dias De Lima, Thais Santos Branco Dijair, Francisco Dauria MarquesAbstract:Abstract Soil Color has historically drawn humans’ attention, although its definition is somewhat subjective. It is correlated with several soil attributes and it allows for inferences about several soil aspects. New proximal sensors, such as the Nix™ Pro Color sensor, can determine soil Color values, but its correlation with the widely used Munsell soil Color chart (MSCC) has yet to be investigated. This work aimed to train machine learning models using the Random Forest (RF) algorithm to predict each notation of MSCC chips from data extracted by the Nix™ Pro sensor, test the model’s accuracy by evaluating whether it can identify MSCC chips using a brand-new and a dirty MSCC, and compare model predictions with soil Color classifications made by the human eye. Additionally, MSCC data obtained via Nix™ was compared to Munsell renotation data to assess the Color detection accuracy of the sensor. Prediction models were calibrated by scanning every MSCC chip (437 in total) in triplicate. All validation samples were excluded from model calibration. Accuracy of the predictions of MSCC notation reached overall accuracy and Kappa index values of 0.93 for the brand-new MSCC and of 0.70 for the dirty MSCC. Soil Color classification by human eye had little agreement with the predicted MSCC notation, as expected due to the variable conditions affecting soil Color conventional determination in the field. Color difference was calculated by the Euclidian distance (ΔE*ab) between three Color stimuli in the CIELAB Color space. The mean ΔE*ab between Nix™-provided data and renotation data was 2.9, demonstrating high Color detection accuracy. The Nix™ Pro Color sensor allows for assessment of accurate Color data. When applied together with machine learning algorithms, Nix™ Pro provides a reliable determination of soil Color classification equivalent to MSCC in an easily reproducible, rapid, inexpensive and non-subjective way.
David C Weindorf - One of the best experts on this subject based on the ideXlab platform.
-
from sensor data to Munsell Color System machine learning algorithm applied to tropical soil Color classification via nix pro sensor
Geoderma, 2020Co-Authors: Marcelo Mancini, David C Weindorf, Maria Eduarda Carvalho Monteiro, Alvaro Jose Gomes De Faria, Anita Fernanda Dos Santos Teixeira, Wellington De Lima, Francielle Roberta Dias De Lima, Thais Santos Branco Dijair, Francisco Dauria MarquesAbstract:Abstract Soil Color has historically drawn humans’ attention, although its definition is somewhat subjective. It is correlated with several soil attributes and it allows for inferences about several soil aspects. New proximal sensors, such as the Nix™ Pro Color sensor, can determine soil Color values, but its correlation with the widely used Munsell soil Color chart (MSCC) has yet to be investigated. This work aimed to train machine learning models using the Random Forest (RF) algorithm to predict each notation of MSCC chips from data extracted by the Nix™ Pro sensor, test the model’s accuracy by evaluating whether it can identify MSCC chips using a brand-new and a dirty MSCC, and compare model predictions with soil Color classifications made by the human eye. Additionally, MSCC data obtained via Nix™ was compared to Munsell renotation data to assess the Color detection accuracy of the sensor. Prediction models were calibrated by scanning every MSCC chip (437 in total) in triplicate. All validation samples were excluded from model calibration. Accuracy of the predictions of MSCC notation reached overall accuracy and Kappa index values of 0.93 for the brand-new MSCC and of 0.70 for the dirty MSCC. Soil Color classification by human eye had little agreement with the predicted MSCC notation, as expected due to the variable conditions affecting soil Color conventional determination in the field. Color difference was calculated by the Euclidian distance (ΔE*ab) between three Color stimuli in the CIELAB Color space. The mean ΔE*ab between Nix™-provided data and renotation data was 2.9, demonstrating high Color detection accuracy. The Nix™ Pro Color sensor allows for assessment of accurate Color data. When applied together with machine learning algorithms, Nix™ Pro provides a reliable determination of soil Color classification equivalent to MSCC in an easily reproducible, rapid, inexpensive and non-subjective way.
Marcos Fogagnoli - One of the best experts on this subject based on the ideXlab platform.
-
Desenvolvimento de metodologia para analise dos atributos da cor atraves do processamento digital de imagens
2017Co-Authors: Marcos FogagnoliAbstract:Resumo: Na busca de um método que possa definir, a partir de informações rápidas e de baixo custo, um mapa de aplicação de insumos agrícolas, objetivou-se estabelecer qual a capacidade de interpretação das cores pelos computadores de uso comercial hoje utilizados, com o intuito de relacionar o estado nutricional das plantas com a cor que apresentam. Partindo de um sistema de classificação de cores reconhecido internacionalmente (Munsell Color System) que, através de atributos matemáticos, é capaz de descrever as cores, verificou-se qual a relação entre os valores dos atributos da cor indicados pelo Sistema de Munsell e os valores de seus atributos identificados pelo computador após a digitalização. Estudou-se dois processos: o primeiro, consistiu na digitalização de 417 cores selecionadas diretamente do catálogo de Munsell que, através de programas computacionais, tiveram seus atributos identificados e comparados aos atributos indicados pelo catálogo de cores. Verificou-se que o processo de digitalização interpreta as cores como se fossem mais escuras do que de fato são, sendo que essa distorção diminui à medida que a cor se torna mais luminosa. Verificou-se também que esse sistema, apesar da distorção percebida, mantém boa capacidade de diferenciação entre as cores, ou seja, é capaz de distinguir numericamente cores aparentemente casadas. O segundo consistiu na digitalização de fotos das mesmas 417 cores, obtidas a três distâncias diferentes ( 50 cm, 75 cm e 100 cm) sob condições semelhantes de iluminação (brilho solar com variação máxima de 2 horas), identificação dos atributos, através de programas computacionais e comparação com o valor real e com os valores do processo anterior. Observou-se que as cores interpretadas pelo computador após o processo fotográfico e a digitalização, apresentaram-se mais escuras que no processo anterior, porém isto não mostrou relação com a distância da obtenção da foto. Foi observado que o processo fotográfico impõe alguma distorção na interpretação das cores. Isto fica evidenciado nas análises dos gráficos de distribuição de freqüência dos parâmetros das cores, que apresentam maior amplitude para as imagens obtidas das fotos do que nas imagens obtidas diretamente do catálogo de Munsell, o que torna menor a capacidade de diferenciação de cores semelhantes. Concluiu-se que existe uma estreita relação entre os valores dos atributos das cores reais e os valores dos atributos das cores obtidas a partir de digitalização.Abstract: The objective of this study was to establish a correlation between the Colors interpretation by the conventional computers and the plant nutritional status in a way to define, using fast and low cost information, the use of fertilizers into commercial crops. Munsell Color System, a Color classification known internationally was used to define each Color numeric value. Thought out this System it was examined the relationship between the numeric values of the real Color and the numeric value found by the computer after photos being scanned. Two processes were studied: On the first one 417 Colors was scanned directly from the Munsell Color System book. Using the computer System each Color received a numeric value. These numeric values were compared to the numeric values found on Munsell’s book. On the second study photos of the same 417 Colors were taken from three different distances (50 cm, 75 cm and 100 cm) using the same luminosity (sunshine variation of 2 hours maximum) received a numeric value and these values were compared to the original values (Munsell’s Book) and the ones obtained on the previous study. Results of the first study showed that after being scanned the Colors appeared darker than the original ones. This difference is reduced as the Colors get lighter. It was shown also that the computer System is precise to differentiate the Colors that seem to be the very similar. Results from the second study showed that the photographic and scanner process also turned the Colors darker, but it was not found any effect of the distance. The photographs process increases the deviation of the Colors and this was confirmed on the graphics of the numeric distribution parameter. It was shown wider amplitude for the figures obtained from photos than the ones obtained directly from Munsell’s Book. The Color differentiation was smaller on the second study. From the two studies, it can be concluded that there is a good correlation between the numeric values of Colors scanned and the real ones
-
Desenvolvimento de metodologia para analise dos atributos da cor atraves do processamento digital de imagens
Universidade Estadual de Campinas. Faculdade de Engenharia Agricola, 2000Co-Authors: Marcos FogagnoliAbstract:Na busca de um método que possa definir, a partir de informações rápidas e de baixo custo, um mapa de aplicação de insumos agrícolas, objetivou-se estabelecer qual a capacidade de interpretação das cores pelos computadores de uso comercial hoje utilizados, com o intuito de relacionar o estado nutricional das plantas com a cor que apresentam. Partindo de um sistema de classificação de cores reconhecido internacionalmente (Munsell Color System) que, através de atributos matemáticos, é capaz de descrever as cores, verificou-se qual a relação entre os valores dos atributos da cor indicados pelo Sistema de Munsell e os valores de seus atributos identificados pelo computador após a digitalização. Estudou-se dois processos: o primeiro, consistiu na digitalização de 417 cores selecionadas diretamente do catálogo de Munsell que, através de programas computacionais, tiveram seus atributos identificados e comparados aos atributos indicados pelo catálogo de cores. Verificou-se que o processo de digitalização interpreta as cores como se fossem mais escuras do que de fato são, sendo que essa distorção diminui à medida que a cor se torna mais luminosa. Verificou-se também que esse sistema, apesar da distorção percebida, mantém boa capacidade de diferenciação entre as cores, ou seja, é capaz de distinguir numericamente cores aparentemente casadas. O segundo consistiu na digitalização de fotos das mesmas 417 cores, obtidas a três distâncias diferentes ( 50 cm, 75 cm e 100 cm) sob condições semelhantes de iluminação (brilho solar com variação máxima de 2 horas), identificação dos atributos, através de programas computacionais e comparação com o valor real e com os valores do processo anterior. Observou-se que as cores interpretadas pelo computador após o processo fotográfico e a digitalização, apresentaram-se mais escuras que no processo anterior, porém isto não mostrou relação com a distância da obtenção da foto. Foi observado que o processo fotográfico impõe alguma distorção na interpretação das cores. Isto fica evidenciado nas análises dos gráficos de distribuição de freqüência dos parâmetros das cores, que apresentam maior amplitude para as imagens obtidas das fotos do que nas imagens obtidas diretamente do catálogo de Munsell, o que torna menor a capacidade de diferenciação de cores semelhantes. Concluiu-se que existe uma estreita relação entre os valores dos atributos das cores reais e os valores dos atributos das cores obtidas a partir de digitalização.The objective of this study was to establish a correlation between the Colors interpretation by the conventional computers and the plant nutritional status in a way to define, using fast and low cost information, the use of fertilizers into commercial crops. Munsell Color System, a Color classification known internationally was used to define each Color numeric value. Thought out this System it was examined the relationship between the numeric values of the real Color and the numeric value found by the computer after photos being scanned. Two processes were studied: On the first one 417 Colors was scanned directly from the Munsell Color System book. Using the computer System each Color received a numeric value. These numeric values were compared to the numeric values found on Munsells book. On the second study photos of the same 417 Colors were taken from three different distances (50 cm, 75 cm and 100 cm) using the same luminosity (sunshine variation of 2 hours maximum) received a numeric value and these values were compared to the original values (Munsells Book) and the ones obtained on the previous study. Results of the first study showed that after being scanned the Colors appeared darker than the original ones. This difference is reduced as the Colors get lighter. It was shown also that the computer System is precise to differentiate the Colors that seem to be the very similar. Results from the second study showed that the photographic and scanner process also turned the Colors darker, but it was not found any effect of the distance. The photographs process increases the deviation of the Colors and this was confirmed on the graphics of the numeric distribution parameter. It was shown wider amplitude for the figures obtained from photos than the ones obtained directly from Munsells Book. The Color differentiation was smaller on the second study. From the two studies, it can be concluded that there is a good correlation between the numeric values of Colors scanned and the real ones
Marcelo Mancini - One of the best experts on this subject based on the ideXlab platform.
-
from sensor data to Munsell Color System machine learning algorithm applied to tropical soil Color classification via nix pro sensor
Geoderma, 2020Co-Authors: Marcelo Mancini, David C Weindorf, Maria Eduarda Carvalho Monteiro, Alvaro Jose Gomes De Faria, Anita Fernanda Dos Santos Teixeira, Wellington De Lima, Francielle Roberta Dias De Lima, Thais Santos Branco Dijair, Francisco Dauria MarquesAbstract:Abstract Soil Color has historically drawn humans’ attention, although its definition is somewhat subjective. It is correlated with several soil attributes and it allows for inferences about several soil aspects. New proximal sensors, such as the Nix™ Pro Color sensor, can determine soil Color values, but its correlation with the widely used Munsell soil Color chart (MSCC) has yet to be investigated. This work aimed to train machine learning models using the Random Forest (RF) algorithm to predict each notation of MSCC chips from data extracted by the Nix™ Pro sensor, test the model’s accuracy by evaluating whether it can identify MSCC chips using a brand-new and a dirty MSCC, and compare model predictions with soil Color classifications made by the human eye. Additionally, MSCC data obtained via Nix™ was compared to Munsell renotation data to assess the Color detection accuracy of the sensor. Prediction models were calibrated by scanning every MSCC chip (437 in total) in triplicate. All validation samples were excluded from model calibration. Accuracy of the predictions of MSCC notation reached overall accuracy and Kappa index values of 0.93 for the brand-new MSCC and of 0.70 for the dirty MSCC. Soil Color classification by human eye had little agreement with the predicted MSCC notation, as expected due to the variable conditions affecting soil Color conventional determination in the field. Color difference was calculated by the Euclidian distance (ΔE*ab) between three Color stimuli in the CIELAB Color space. The mean ΔE*ab between Nix™-provided data and renotation data was 2.9, demonstrating high Color detection accuracy. The Nix™ Pro Color sensor allows for assessment of accurate Color data. When applied together with machine learning algorithms, Nix™ Pro provides a reliable determination of soil Color classification equivalent to MSCC in an easily reproducible, rapid, inexpensive and non-subjective way.
Maria Eduarda Carvalho Monteiro - One of the best experts on this subject based on the ideXlab platform.
-
from sensor data to Munsell Color System machine learning algorithm applied to tropical soil Color classification via nix pro sensor
Geoderma, 2020Co-Authors: Marcelo Mancini, David C Weindorf, Maria Eduarda Carvalho Monteiro, Alvaro Jose Gomes De Faria, Anita Fernanda Dos Santos Teixeira, Wellington De Lima, Francielle Roberta Dias De Lima, Thais Santos Branco Dijair, Francisco Dauria MarquesAbstract:Abstract Soil Color has historically drawn humans’ attention, although its definition is somewhat subjective. It is correlated with several soil attributes and it allows for inferences about several soil aspects. New proximal sensors, such as the Nix™ Pro Color sensor, can determine soil Color values, but its correlation with the widely used Munsell soil Color chart (MSCC) has yet to be investigated. This work aimed to train machine learning models using the Random Forest (RF) algorithm to predict each notation of MSCC chips from data extracted by the Nix™ Pro sensor, test the model’s accuracy by evaluating whether it can identify MSCC chips using a brand-new and a dirty MSCC, and compare model predictions with soil Color classifications made by the human eye. Additionally, MSCC data obtained via Nix™ was compared to Munsell renotation data to assess the Color detection accuracy of the sensor. Prediction models were calibrated by scanning every MSCC chip (437 in total) in triplicate. All validation samples were excluded from model calibration. Accuracy of the predictions of MSCC notation reached overall accuracy and Kappa index values of 0.93 for the brand-new MSCC and of 0.70 for the dirty MSCC. Soil Color classification by human eye had little agreement with the predicted MSCC notation, as expected due to the variable conditions affecting soil Color conventional determination in the field. Color difference was calculated by the Euclidian distance (ΔE*ab) between three Color stimuli in the CIELAB Color space. The mean ΔE*ab between Nix™-provided data and renotation data was 2.9, demonstrating high Color detection accuracy. The Nix™ Pro Color sensor allows for assessment of accurate Color data. When applied together with machine learning algorithms, Nix™ Pro provides a reliable determination of soil Color classification equivalent to MSCC in an easily reproducible, rapid, inexpensive and non-subjective way.
