Fly detection using image processing

CUSTOMER
DESCRIPTION
Monitoring pests and diseases is a vital step in the efficient management of a modern farm. To this end, NERAtech was selected as a reality for the development of an appropriate tool for the analysis and identification of the presence of harmful insects within olive tree crops. The activity carried out by NERAtech can be divided into two: the development of the recognition algorithm and the creation of a portal for the management of all the collected data.
As regards the first phase, i.e. the development of the recognition algorithm, NERAtech took care of identifying the best procedure in order to be able to discriminate the presence of the oil fly from other insects that may be present on the SpyFly trap. Therefore, algorithms have been developed based on the OpenCV library through the use of languages such as Python and C++. These languages are compliant with the operating systems on board the SBC to be used in the field for the acquisition and analysis of images. These algorithms were written with logic devoted to using in machine learning architecture.
To improve the interaction between the system and the end user, a web application (Smart-HUB) has been developed according to the customer’s requests, capable of collecting all the data coming from the various control units installed within the fields, both traps or sensors of various kinds, filing them in a database, processing them and finally making them usable according to different types to the end user.