Learning crop models for vision-based guidance of agricultural robots
Authors:
- 1 September 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 9, 1158-1163
Abstract
This paper describes a vision-based method of guiding autonomous vehicles within crop rows in agricultural fields where the crop rows are challenging to detect or their appearance is not known a-priori. The location of the crop rows is estimated with an SVM regression algorithm using colour, texture and 3D structure descriptors from a forward facing stereo camera pair. Our system rapidly learns a model online with minimal user input, and then uses this model to track crop rows. Results demonstrate our method is able to learn and track a wide variety of crops with an RMS error of less than 3cm. We also present online control results demonstrating our system autonomously steering a robot for 3km.Keywords
This publication has 15 references indexed in Scilit:
- Robotics for Sustainable Broad-Acre Agriculture2015
- Bayesian classification and unsupervised learning for isolating weeds in row cropsPattern Analysis and Applications, 2012
- Tracking-Learning-DetectionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2011
- Struck: Structured output tracking with kernels2011
- Evaluating Color Descriptors for Object and Scene RecognitionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2009
- Development of a stereovision sensing system for 3D crop row structure mapping and tractor guidanceBiosystems Engineering, 2008
- A vision based row detection system for sugar beetComputers and Electronics in Agriculture, 2008
- Crop/weed discrimination in perspective agronomic imagesComputers and Electronics in Agriculture, 2008
- Automatic guidance of agricultural vehicles using a laser sensorComputers and Electronics in Agriculture, 2000
- Color Indices for Weed Identification Under Various Soil, Residue, and Lighting ConditionsTransactions of the ASAE, 1995