The role of textures to improve the detection accuracy of Rumex obtusifolius in robotic systems

Abstract

Hiremath S, Van Der Heijden G, Van Evert FK & Stein A (2012) The role of textures to improve the detection accuracy of Rumex obtusifolius in robotic systems. Weed Research.52, 430–440. Summary Rumex obtusifolius is a common weed that is difficult to control in organic farming systems. Among the proposed non‐chemical treatment methods, robotic systems to mechanically remove the weed have potential for its automatic detection and removal. This article considers a recently developed robot with a real‐time vision system capable of detecting R. obtusifolius in a pasture with the objective of improving its detection accuracy. We show that the texture measure used by the current system is equivalent to local variance and has limited value for detecting R. obtusifolius in a complex background like pasture. To improve the system, two different sets of visual texture features corresponding to Gray level Co‐occurence Matrix (GLCM) and Laws’ filter masks were investigated. Through feature selection, we determined that GLCM features of contrast, entropy and correlation were the best among the two sets of features and were 25% more accurate in estimating the taproot location than the current system. We incorporated these texture features in a new segmentation algorithm and demonstrated its robustness by testing it on a data set of 92 images with high complexity in terms of variation in illumination and weed size. The new segmentation algorithm had a detection accuracy of 90%, with an average error of 141 mm in the estimation of the location of the taproot of R. obtusifolius, compared with 308 mm with the former algorithm.

Evaluation of bag-of-features (BoF) technique for weed management in sugarcane production

“…4). The results suggest that some of the problems resulting from partial occlusion and overlapping of leaves, such as reported by Lamm et al, (2002), Golzarian and Frick et al, (2011) and Hiremath et al, (2012) have been reduced.…”

Development and testing of image processing algorithm to estimate weed infestation level in corn fields

“…The correlation between the level of weed infestation estimated by algorithm and that attested to by the experts was high, with a RMSE value of 0.44 and 90% accuracy, which represents a good fit of the data for all three categories considered. A similar level of accuracy in the identification of a specific weed (Rumex obtusefolius) in an area of pasture was obtained by Hiremath et al (2012), but here the problem consists of a complex system of various kinds of weeds and the similar leaf shapes of the crop. This is an extremely challenging problem, especially since various environmental factors affect the images, such as differences in lighting, position, scale and overlapping of leaves, all of which can contribute noise and hinder the rating process, as discussed in the literature review presented by Peteinatos et al (2014).…”