In a number of branches of agricultural production, including agriculture, land reclamation, etc., there are problems, the solution of which requires the use of artificial intelligence. In particular, the assessment of the reclamation state of agricultural fields in large areas is a very time-consuming task, even with the use of unmanned aerial vehicles. To automate these intelligent approaches, it is effective to use artificial neural networks (INS) implemented in the form of computer programs. The use of software as a service (SaaS) is a modern approach to computer support of various intelligent production processes, including agricultural. Agriculture is a promising industry for the introduction of such technologies. The aim of the study is to develop a methodology and create a cloud-based SaaS system for identifying defective areas of agricultural fields based on INS. The development of neural network technologies and cloud services makes it possible to process a large amount of information in the cloud and provide user access to computing power. The article describes the methodology of building a service architecture for recognizing problem areas of cultivated agricultural fields, data preparation, network training, development of client and server parts. Such implementation is possible with the use of such technologies and tools as CUDA, CNN, PyTorch. As a result, the strengths and weaknesses of their use for solving the problem of image recognition on the example of problem areas of agricultural fields were identified. It has been established that classification-type INS are capable of solving problems of recognizing the reclamation state of fields, and modern information technologies make it possible to transfer calculations to the cloud, while the cloud service can be monetized as a SaaS model.

Keywords: agriculture, color images, SaaS system, artificial neural network, image classification

Artificial intelligence methods can be used to solve the problems of agricultural production. Assessing the condition of crops in large areas, even with the use of unmanned aerial vehicles, is a time-consuming task. The peculiarities of the task of such an assessment are the multifactorial nature of the analyzed structures, which require the use of a systematic approach at all stages of the study from the formation of a database of color images to the intelligent solution of problems of their analysis. The results of the analysis of the U-net architecture of the INS and its limitations for the problem of image segmentation are presented. The purpose of the study is to substantiate the architecture of the segmentation artificial neural network (INS) to identify problem areas of agricultural fields. The hypothesis of the segmentation network advantage was tested on the DeepLabV3 ResNet50 architecture. Numerical experiments have established that the increase in the accuracy of segmentation of images of agricultural fields is constrained by the limited resolution and accuracy of manual markup dataset. The built architectures can be used as an algorithmic core for creating SaaS systems, while the performance of the used configuration of the INS can be crucial.

Keywords: color images, segmentation task, agropole plots, deep neural network, INS architecture, convolutional layers