A survey of image processing techniques for plant extraction and segmentation in the field

A review on weed detection using ground-based machine vision and image processing techniques

http://oa.upm.es/13699/2/INVE_MEM_2011_115574.pdf

Original paper: Automatic segmentation of relevant textures in agricultural images

http://oa.upm.es/13898/2/INVE_MEM_2011_115592.pdf

Original paper: Real-time image processing for crop/weed discrimination in maize fields

In Precision Agriculture (PA) automatic image segmentation for plant identification is an important issue to be addressed. Emerging technologies in optical imaging sensors play an important role in PA. In maize fields, site-specific treatments, with chemical products or mechanical manipulations, are applied for weeds elimination. Maize is an irrigated crop, also unprotected from rainfall. After a strong rain, soil materials (particularly clays) mixed with water impregnate the vegetative cover. The green spectral component associated to the plants is masked by the dominant red spectral component coming from soil materials. This makes methods based on the greenness identification fail under such situations. We propose a new method based on Support Vector Machines for identifying plants with green spectral components masked and unmasked. The method is also valid for post-treatment evaluation, where loss of greenness in weeds is identified with the effectiveness of the treatment and in crops with damage or masking. The performance of the method allows to verify its viability for automatic tasks in agriculture based on image processing.

Qianyu Wang

Papers

Mean-shift-based color segmentation of images containing green vegetation