METHOD OF NON-INVASIVE DETERMINATION OF FUNGAL DISEASES OF COMMON GARDEN STRAWBERRY

Abstract

The relevance of early undamaging diagnosis of fungal, viral and bacterial diseases of common garden strawberry is proved. External symptoms of fungal diseases are given. On the basis of literature data, the existing optical methods of early diagnostics of cultivated plants are analyzed. It is established that the methods of pixel calculation of the image in the space of color channels of red, green and blue colors are more preferable than other optical methods for detection of strawberry fungal diseases. This results from the fact that fungal diseases create specific color spots and their distribution on the surface of a plant can be easily identified in the colorimetric CIE Lab system. The work presents a number of approaches to the new method of early diagnostics of common garden strawberry fungal diseases with use of technical means and software developed for the smartphone. Implementation of the method does not require big expenses as it is done in the form of the software application in the smartphone on the basis of the Android operating system. This application will enable to obtain high quality images of a leaf of a plant, it will also provide segmentation and calculation of the quantity and the specific area of color spots on a contour of a leaf. Moreover, it will work with the database of model images of plants with fungal diseases. Classification of fungal diseases and forecasting of their development will be carried out by means of artificial neural network. The proposed method will allow to determine diseases of common garden strawberry leaves, to predict their development and to establish possible borders of distribution on the chosen plantation.