Plant diseases have a significant impact in reducing the quantity and quality of agricultural products. Viruses as one of most important plant pathogenic are causing considerable damage on crops. In cucurbits such as cucumber, various viruses that cause disease, which is often produce mosaic symptoms on leaves or fruit. One of the most important viruses that infected cucurbits is virus infection or mosaic disease by the scientific name of cucumber mosaic virus, which is one of the most important plant pathogenic viruses globally and have a broad range of hosts. The economic losses of this disease in Iran is high and in cucumber fields, the yield reduction caused by this disease is estimated to be a third. The use of insecticides to protect the crop against diseases and pests not only increases the product price, but also increases the risk of pesticide residues in agricultural products. Early detection of disease or vector (aphids) is a key point in the management and control of diseases. Symptoms of the disease at an early stage are chlorotic spots (bright green spots on dark green substrate or vice versa) on the leaf surfaces. Due to temperature and humidity conditions in greenhouses, quick decision to control diseases and vectors is essential, in order to prevent the spread of disease in greenhouses So in this project, a mobile system was designed and evaluated for detecting mosaic symptoms in cucumber and squash plants grown in greenhouse using photography and image processing in visible spectrum. First, some cucumber and squash plants were grown under controlled conditions, in order to avoid getting infected by other plant diseases and after inoculation of the plants with the cucumber mosaic virus and emergence of the symptoms, the plants were photographed with a camera (Canon s110) in natural light, using the mobile system in a greenhouse. Then, for separating the canopy from the background (soil and other parts of the greenhouse), the images were processed by GMR algorithm of MATLAB; and 31 color features were extracted from all images. Artificial neural network (ANN) method was used for left; MARGIN: 0in 0in 8pt" dir=rtl align=right Key words: Image processing, Artificial neural network, Feature extraction, Plant disease, Chlorotic spots.
