Determination of Munsell Soil Colour Using Smartphones.

Area of Science:

• Agricultural Science
• Soil Science
• Digital Imaging

Background:

• Soil colour is crucial for assessing soil health and properties in agriculture.
• Munsell Soil Colour Charts (MSCC) are standard but subjective and error-prone.
• Digital soil colour determination offers potential for improved accuracy.

Purpose of the Study:

• To accurately determine Munsell soil colour from images captured by smartphones.
• To address colour discrepancies between smartphone images and sensor-based readings.
• To enhance the reliability of digital soil colour analysis for agricultural applications.

Main Methods:

• Utilized popular smartphones to capture images of the Munsell Soil Colour Book (MSCB).
• Compared smartphone-derived colours with readings from a Nix Pro-2 sensor.
• Investigated various colour models and distance functions to establish a colour-intensity relationship.
• Developed a method to adjust pixel intensity in smartphone images for accurate colour determination.

Main Results:

• Significant colour reading discrepancies were observed between smartphone and Nix Pro-2 sensor.
• The proposed method, involving pixel intensity adjustment, achieved 74% accuracy in Munsell soil colour determination.
• This represents a substantial improvement from the baseline accuracy of 9% for top 5 predictions without adjustment.

Conclusions:

• Smartphone-based digital soil colour analysis can be significantly improved through image processing techniques.
• The developed colour-intensity relationship enhances the accuracy of Munsell soil colour determination from MSCB images.
• This advancement offers a more reliable and accessible tool for soil health monitoring in agriculture.