Munsell Soil Colour Classification Using Smartphones through a Neuro-Based Multiclass Solution

Abstract

Colour is a property widely used in many fields to extract information in several ways. In soil science, colour provides information regarding the chemical and physical characteristics of soil, such as genesis, composition, and fertility, amongst others. Thus, accurate estimation of soil colour is essential for many disciplines. To achieve this, experts traditionally rely on comparing Munsell colour charts with soil samples, which is a laborious process. In this study, we proposed using artificial neural networks to catalogue soil colour with a two-step classification. Firstly, the hue variable is estimated, and then the remaining two coordinates, value and chroma. Our experiments were conducted using three different, common cameras (one digital camera and two mobile phones). The results of our tests showed a 20% improvement in classification accuracy using the lowest-quality camera and an average accuracy of over 90%.

Soil Color Comparison Using Munsell Soil Color Chart and Calibrated Smartphone Camera

“…Soil color is one of the characteristics that can provide this information. Precisely assessing soil color is crucial since it yields vital data for soil scientists (Liu et al, 2020;Pegalajar et al, 2018Pegalajar et al, , 2020Pegalajar et al, , 2023. Soil color can provide valuable insights into soil development, composition, age of soil and rock surfaces, and variables that restrict plant growth.…”

“…Soil color can provide valuable insights into soil development, composition, age of soil and rock surfaces, and variables that restrict plant growth. Hence, soil color serves as a significant soil indicator and attribute that can be employed to characterize, categorize, and distinguish soils (Fan et al, 2017;Pegalajar et al, 2018Pegalajar et al, , 2023Priandana et al, 2014). Furthermore, soil color plays a crucial role in determining a range of soil activities and conditions, including oxidation-reduction processes, iron solubility, organic matter content, soil development, and the formation of distinct soil materials like concretions, nodules, plintites, and horizons.…”

“…The need for soil data in several fields, such as precision agriculture and dynamic environmental monitoring, is driving the advancement of soil sensors that can provide more accurate information, including data on soil color. The utilization of MSCC is prevalent due to its relative simplicity, as it solely involves a qualitative comparison of soil color in the field using MSCC (Fan et al, 2017;Pegalajar et al, 2023;Priandana et al, 2014). The Munsell chart comprises 238 standardized color rectangular pieces (Pegalajar et al, 2018(Pegalajar et al, , 2023, organized based on three coordinates: hue (H), Value or brightness level (V), and Chroma or color intensity (C) (Liu et al, 2020;Pegalajar et al, 2023;Sinclair et al, 2024).…”

“…The primary issue that has arisen thus far is the ambiguity resulting from the subjectivity of the observer (Bloch et al, 2021;Fan et al, 2017;Milotta et al, 2020;Pegalajar et al, 2023;Priandana et al, 2014). This issue is worsened by the extensive range of hues in MSCC, some of which bear resemblances.…”

“…These techniques involve the conversion of RGB values into XYZ space and HVC (Han et al, 2016) or L*a*b color space (Kirillova et al, 2015). Another instance can be observed in the study conducted by Han et al, (2016) and Pegalajar et al, (2023), where the researchers categorize soil using color sensors in smartphone photographs. Issues undeniably accompany the utilization of photos produced by mobile phones.…”