With the inspiration of helping the farmer to grow his crop in the optimal way, Pessl Instruments GmbH, from Weiz, Austria, developed optical high-resolution camera system, together with a computer vision software which is able to recognize insects, fruits on trees and growth of crop. Pessl Instruments develops decision support system which is consisting from remote monitoring of insect traps and remote monitoring of fields and crops. Optical high-resolution camera system can be installed on the field, to remotely monitor fields and crops, or it can be embedded inside insect trap, to remotely monitor insect pressure at fields. Pessl Instruments provides different variations of insect traps, to cover a broad specter of insect species. All of the photos and data from computer vision software is displayed online, on a web portal called ng.fieldclimate.com. For automatic recognition of objects from the photos, we use a computer vision software, which is based on deep learning methods. We use a subset of machine learning algorithms, which uses deep artificial neural networks as models and does not require future engineering. To successfully train our algorithms for fruit recognition, we acquired 1620 photos of apples in different phenological stages (other fruit trees will follow). In the next step we made 6899 manual annotations and divided them in 4 phenological stages. For insect recognition we acquired 2472 photos form 8 different species (Lobesia Botrana, Ceratitas Capitata, Drosophila Suzukii, Halyomorpha Halys, Diabrotica Vergifera, Helicoverpa Armigera, Eupoecilia Ambiguella, Bactocera Oleae) and made 18042 manual annotations. For running the computer vision software, we use special production server with three GPUs and two CPUs. With this configuration a full training process for the automatic recognition algorithm with 200.000 steps takes 2.1 days. From this software we retrieve data about numbers of insects by species caught in the insect trap, refreshed daily and also represented in a chart showing population growth. In application for recognizing fruits on the trees, machine learning software is also capable to measure fruit diameters on all recognized fruits, refreshed daily. With this information we can display a chart of fruit growth during the growing season and in the future, we will develop yield prognosis model.