Machine vision for precise control of weeds
Derechos de accesoopenAccess
MetadataShow full item record
Cita bibliográficaBlasco, J., Benlloch, J.V., Agusti, M., Molto, E. (1999). Machine vision for precise control of weeds. Precision Agriculture and Biological Quality, 3543, 336-343.
Consumers demand for natural quality products and concern about the ecological impact of agriculture is growing in all European countries. For the farmers to follow the evolution of the market, new procedures have to be introduced in agriculture to obtain satisfactory production levels, keeping high quality standards, without damaging the environment. Image processing techniques have been traditionally used in the industry, where controlling most of the environmental variables (mainly lighting, background and speed) uses to be easy. Although these techniques can be also useful in agriculture, working outdoors is much more complicated, mainly due to the variability of the natural objects and the environmental conditions. European Project AIR-CT93-1299 (PATCHWORK) was aimed at reducing or eliminating the use of chemicals by automatically detecting the position and/or density of weeds using computer vision and applying an herbicide treatment, which could be chemical or mechanical. This paper describes the work carried out in developing image. analysis procedures for two different purposes: In horticultural, row crops, the aim was to develop a real-time machine vision system that provides the position of weeds to a moving robot that will apply an electric discharge to them, thus eliminating the use of herbicides. - In cereals, the objective was to create weed density maps that will help an especial sprayer boom, incorporating a GPS sensor, to dose the herbicide at 4 concentrations, corresponding to 4 infestation levels during operation. The first system is based on a Bayesian algorithm for segmenting the images, which requires to be previously trained by an expert, who selects areas of different images, trying to represent the colour variability of the plants, the soil and the weeds. After segmentation, pixels belonging to class soil are correctly classified and morphological operations are applied to discriminate between plants and weeds. The system is able to properly locate more than 90% of weeds with very little confusion with the crop (1 %) in lettuce cultures. Current processing time is under 500 ms. The second vision system uses a normalised difference index (green and red channels) to enhance the contrast of the field images. Then, growing techniques are applied to discriminate between vegetation and background. Once plant pixels are identified, weeds are distinguished from the crop by estimating the position of the row and employing shape analysis techniques. The performance of the method showed that more than 85% of weeds were properly detected.