Real-time vision-based detection of Rumex obtusifolius in grassland

• Published in: Weed research 2009-04, Vol.49 (2), p.164-174
• Main Authors: VAN EVERT, F K, POLDER, G, VAN DER HEIJDEN, G W A M, KEMPENAAR, C, LOTZ, L A P
• Format: Article
• Language: English
• Subjects: Biological and medical sciences; broad-leaved dock; color; crispus; crop; digital images; Fourier analysis; Fundamental and applied biological sciences. Psychology; image analysis; Parasitic plants. Weeds; pasture weed; Phytopathology. Animal pests. Plant and forest protection; robot; soil; swards; system; weed detection; weed-control; Weeds
• ISSN: 0043-1737; 1365-3180
• DOI: 10.1111/j.1365-3180.2008.00682.x

van Evert FK, Polder G, van der Heijden GWAM, Kempenaar C & Lotz LAP (2009). Real‐time vision‐based detection of Rumex obtusifolius in grassland. Weed Research49, 164–174. Summary Rumex obtusifolius is a common grassland weed that is hard to control in a non‐chemical way. The objective of our research was to automate the detection of R. obtusifolius as a step towards fully automated mechanical control of the weed. We have developed a vision‐based system that uses textural analysis to detect R. obtusifolius against a grass background. Image sections are classified as grass or weed using 2‐D Fourier analysis. We conducted two experiments. In the first (laboratory) experiment, we collected 28 images containing R. obtusifolius and 28 images containing only grass. Between 23 and 25 of 28 images were correctly classified (82–89%) as showing R. obtusifolius; all grass images were correctly classified as such. In the second (field) experiment, a self‐propelled platform was used to obtain five sequences of images of R. obtusifolius plants. We used the parameters that gave the best classification results in the first experiment. We found, after changing one of the algorithm’s parameters in response to prevailing light conditions, that we were able to detect R. obtusifolius in each image of each sequence. The algorithm scans a ground area of 1.5 m2 in 30 ms. We conclude that the algorithm developed is sufficiently fast and robust to eventually serve as a basis for a practical robot to detect and control R. obtusifolius in grassland.