J. Desbiolles

Evaluation of Color Indices for Improved Segmentation of Plant Images

This article presents both a geometric methodology and an empirical experiment with real plant images to evaluate the performance of different color indices for plant segmentation from two-dimensional digital images. Six well-known color indices given in the literature were selected for the evaluation: the normalized green, the difference between normalized green and normalized red, the normalized difference index, the excessive green index, the modified excessive green index, and the hue. Segmentation performance was assessed on the normalized green-red plane (the rg plane) by using Type I and Type II errors, commonly used in statistics to estimate the errors associated with null hypothesis testing. The two errors were applied in this study to quantifying the misclassified pixel percentages for plant and background, respectively. The rg plane is shown to facilitate visualizing geometrically the behavior of the color indices. An experiment with 240 digital images of four plant types under different lighting and background conditions was carried out to empirically investigate their segmentation performance. Segmentation results showed that the hue achieved the least amount of Type II error with a small loss of plant pixels, which confirms expectations from the geometric analysis on the rg plane.

Analyzing the mixing performance of a rotary spader using digital image processing and discrete element modelling (DEM)

Abstract In order to incorporate a range of surface applied materials into the soil profile to ameliorate crop production constraints, rotary spaders have recently been adopted in Australia. However, no fundamental studies have been carried out to quantify the mixing ability of rotary spaders. The discrete element method (DEM) has the potential to model soil to rotary spader interaction in terms of soil movement. In this study a full scale rotary spader was tested in the field and then simulated using DEM. The DEM model predicted top soil burial to a similar depth in the soil profile as was measured in the field. A method to compare and quantify soil movement was also developed using digital image processing. The forward soil movement which was not examined in the field was also simulated using DEM and it was found that higher forward speed increases the bite lengths, and hence results a heterogeneous top soil mixing. It was also found that in order to carry out a proper analysis, both cross section and forward soil movement must be considered while quantifying the mixing ability of the rotary spader. The results of this study proves that (1) DEM can be used to model soil to rotary spader interaction to analyse different operation conditions without time consuming and costly field tests which can only be performed certain times of the year and (2) digital image processing can be used as a method to quantify and validate simulations of field experiments.