Digital RGB photography and visible-range spectroscopy for soil composition analysis

Cameron, Clare; Gaskin, G.J.; Choisy, Bastien; Coull, Malcolm; Black, H. I. J.

doi:10.1016/j.geoderma.2017.11.020

Cited by 40 publications

(20 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A support vector regression model is built using the Matlab Statistics and Machine Learning Toolbox [19] in order to predict the soil moisture from the soil digital images. The values of the model parameters are the default values set by the toolbox.…”

Section: Support Vector Regression Resultsmentioning

confidence: 99%

Machine learning methods for soil moisture prediction in vineyards using digital images

Hajjar

Esta

et al. 2020

E3S Web Conf.

View full text Add to dashboard Cite

In this paper, we propose to estimate the moisture of vineyard soils from digital photography using machine learning methods. Two nonlinear regression models are implemented: a multilayer perceptron (MLP) and a support vector regression (SVR). Pixels coded with RGB colour model extracted from soil digital images along with the associated known soil moisture levels are used to train both models in order to predict moisture content from newly acquired images. The study is conducted on samples of six soil types collected from Chateau Kefraya terroirs in Lebanon. Both methods succeeded in forecasting moisture giving high correlation values between the measured moisture and the predicted moisture when tested on unknown data. However, the method based on SVR outperformed the one based on MLP yielding Pearson correlation coefficient values ranging from 0.89 to 0.99. Moreover, it is a simple and noninvasive method that can be adopted easily to detect vineyards soil moisture.

show abstract

Section: Support Vector Regression Resultsmentioning

confidence: 99%

Machine learning methods for soil moisture prediction in vineyards using digital images

Hajjar

Esta

et al. 2020

E3S Web Conf.

View full text Add to dashboard Cite

show abstract

“…Soil color is an easily measured attribute and has been used to estimate other attributes (Aitkenhead et al., 2018; Gholizadeh et al., 2020; Morais et al., 2020; Saad Hajjar et al., 2020; Viscarra Rossel et al., 2008), allowing the mapping of large areas. Color is a characteristic expressed by the perception of light reflected from objects to human eyes (Scarinci & Marineli, 2014).…”

Section: Introductionmentioning

confidence: 99%

“…Color varies in the landscape and depth according to drainage and infiltration conditions, parent material, soil aeration, and mineralogy (Aitkenhead et al., 2018; Morais et al., 2020; Ramos et al., 2020; Viscarra Rossel, Minasny, et al., 2006). In other words, it directly reflects the factors and processes of soil formation.…”

Section: Introductionmentioning

confidence: 99%

Soil mineralogical attributes estimated by color as accessed by proximal sensors and machine learning

Baldo

Marques

Fernandes

et al. 2021

Soil Science Soc of Amer J

View full text Add to dashboard Cite

Detailed mapping is essential for land use and management planning. The mappings require a robust database. Costs and time associated with obtaining the database are high and, therefore, it is not always possbile to obtain it. Soil color is a pedoindicator attribute that can be easily characterized. This study aimed to use soil color, based on the RGB (red–green–blue) system and obtained by diffuse reflectance spectroscopy (DRS) and mobile proximal sensor (MPS) to estimate mineralogical attributes using machine learning techniques for the Western Plateau of São Paulo. A total of 600 samples were collected throughout the study area. The samples were analyzed by DRS and then photographed. The color data were obtained by the RGB system after analysis in a computer program. The samples were subjected to laboratory analysis to quantify the contents of crystalline and noncrystalline Fe, hematite, goethite, kaolinite, and gibbsite. The database was subjected to the random forest machine learning algorithm and geostatistics. The use of random forest allowed estimating soil mineralogical attributes based on the RGB system by DRS and MPS. Detailed maps of mineralogical attributes could be constructed using the RGB system by the DRS and MPS techniques. The MPS technique can be used to characterize soil color, reducing the costs associated with analysis and the time required for data collection.

show abstract

“…Besides the visual method, several researchers have demonstrated the great potential of digital cameras including smartphone cameras for rapid measurement of soil color (Viscarra Rossel, Fouad, and Walter 2008;Fan et al 2017;Aitkenhead et al 2018;Schmidt and Ahn 2019). The number of smartphone users in the world was 1.86 billion in 2015 and is predicted to be approximately 2.9 billion by 2020 (Fan et al 2017).…”

Section: Introductionmentioning

confidence: 99%

Comparison of visual and instrumental measurements of soil color with different low-cost colorimeters

Moritsuka

Kawamura

Tsujimoto

et al. 2019

Soil Science and Plant Nutrition

View full text Add to dashboard Cite

Soil color has been conventionally measured using a Munsell soil color chart. Recently launched colorimeters can also measure the color of an object at a reasonable cost. This study was undertaken to evaluate to what extent such low-cost colorimeters (< 500 USD) can be useful for soil color analysis in the laboratory as compared with conventional colorimeters (> 3000 USD) and a Munsell soil color chart (about 200 USD). Sixty two air-dried soil samples collected from rice fields in Madagascar were subjected to two pretreatments for homogenization (2-mm sieving or additional hand-grinding) and instrumental analysis using four low-cost colorimeters (CS-10, Cube, Nix Pro, and Color Muse) and four conventional colorimeters (SPAD-503, CR-20, CR-400, and CR-410). The color of 2-mm sieved samples was also measured visually using a color chart. The effects of pretreatments and the analytical conditions were evaluated by the repeatability and stability of the measurement, the comparability of the soil color data obtained, and the time required for analysis. Overall, instrumental measurement was much more repeatable than visual observation. Both the repeatability and stability of the low-cost colorimeters tended to be lower than those of the conventional colorimeters. Among the low-cost colorimeters examined, soil color parameters (L*, a*, and b*) obtained with Nix Pro were most comparable (r > 0.97 for all parameters of 2-mm sieved samples) with those obtained with SPAD-503, which is an instrument designed specifically for soil color analysis. Additional hand-grinding pretreatment improved the repeatability of the instrumental analysis and reduced the subsample weight to two grams. However, this pretreatment also increased the L* value (lightness) of the samples, decreased the comparability with the data from 2-mm sieved samples, and prolonged the time required to complete the whole analysis. Among the various methods tested, 2-mm sieving of air-dried samples followed by the color measurement with Nix Pro several times per sample was considered the most cost-effective approach for measuring soil color in the laboratory.

show abstract

Digital RGB photography and visible-range spectroscopy for soil composition analysis

Cited by 40 publications

References 29 publications

Machine learning methods for soil moisture prediction in vineyards using digital images

Machine learning methods for soil moisture prediction in vineyards using digital images

Soil mineralogical attributes estimated by color as accessed by proximal sensors and machine learning

Comparison of visual and instrumental measurements of soil color with different low-cost colorimeters

Contact Info

Product

Resources

About