|Chung, Sun-Ok -|
|Cho, Ki-Hyun -|
|Kong, Jae-Woong -|
|Jung, Ki-Youl -|
Submitted to: International Federation of Automatic Control (IFAC) Symposium
Publication Type: Proceedings
Publication Acceptance Date: September 8, 2010
Publication Date: December 6, 2010
Citation: Chung, S., Cho, K., Kong, J., Sudduth, K.A., Jung, K. 2010. Soil texture classification algorithm using RGB characteristics of soil images. In: Proc Third IFAC International Conf. on Agricontrol. December 6-8, 2010, Kyoto, Japan. Available: http://www.ifac-papersonline.net/ersonline.net/. Interpretive Summary: Soil texture is an important physical property that varies across landscapes and affects processes important to crop production and the environment. The usual method of quantifying soil texture is based on laboratory analysis of the proportions of sand, silt, and clay in a soil sample. This requires considerable time, labor, and expense, and is not practical for the intense data collection needed to map texture variations at the sub-field scale. The purpose of this research was to investigate the ability of a more efficient method, based on images collected with a digital camera, to quantify soil texture variations. Soil samples representing important rice production areas in Korea were collected, images of these samples were obtained, and they were also subjected to standard laboratory texture analysis. Statistics calculated from the images were related to percent sand, silt, and clay in the samples. Texture classification (e.g., silt loam vs. loam) by image processing was identical to that by laboratory analysis for 48 percent of the samples. Although this result shows some potential, improved accuracy is needed before the method could be used in practice. This research may benefit scientists and engineers working on similar projects by providing information to guide future work.
Technical Abstract: Soil texture has an important influence on agriculture, affecting crop selection, movement of nutrients and water, soil electrical conductivity, and crop growth. Soil texture has traditionally been determined in the laboratory using pipette and hydrometer methods that require a considerable amount of time, labor, and expense. Recently, in-situ soil texture classification systems using optical diffuse reflectometry or mechanical resistance have been reported, especially for precision agriculture where more data is needed than in conventional agriculture. This paper is a part of overall research to develop a soil texture classification system using image processing. Application of image processing was motivated by simple traditional approaches such as visual inspection and the “hand-feel” method. In this paper, the potential of soil texture classification using RGB histograms was investigated. Seven sites representing major Korean paddy soil series were selected, 4-6 core samples up to a 50-cm depth were obtained from each site, and each sample was segmented by 5-cm intervals. For each segmented soil sample, four surface images were taken using a miniaturized CCD camera, and texture fractions were determined by the pipette method. Scatter plots showed linear patterns between silt content and histogram variables such as brightness, skewness, and mode - brightness. When 5 percent averaged silt content was linearly regressed with mode - brightness, R2, RMSEC, and RMSEP were 0.96, 2.2 percent, and 6.3 percent, respectively. When soils were classified using USDA criteria, the laboratory method and the in-situ image processing method produced the same results for 48 percent of the samples.