Processing and analysis of botanical micro-and macroobjects using computer vision technologies

Abstract

This article discusses the problems of automation of processing and analysis of botanical micro-and macroobjects. The problem is a comprehensive interdisciplinary research aimed at recognizing and searching for botanical micro-and macroobjects using computer vision technologies. Computer vision is an area of artificial intelligence that deals with the analysis and extraction of information from images (in this case, the fern dispute). The purpose of this study is to solve a number of applied and fundamental problems in the biosystematics of botanical objects and the study of microevolutionary processes. The results of the research are stored in a specially developed database “Spore morphology of ferns of the Pteridaceae family”. The database implements a set of queries relevant to biometrics tasks. The database is based on images and biometric data on the morphology of fern spores obtained in the form of measurements of systematically important signs of spores, as well as data on the shape and surface of the studied objects. The diagnostic characters of taxa of the Pteridaceae family have been standardized for solving the research problems. In the course of the study, fundamentally new knowledge was obtained that will supplement the natural classification of the Pteridaceae family, as well as test hypotheses about the relationship of a number of taxa represented on the continents of Eurasia and Africa, for an objective reconstruction of events of past geological periods. Processing and analysis of botanical micro-and macro-objects and the development of specialized software are carried out in the Python programming language using an integrated development environment for Python - PyCharm. When developing the graphical interface of the program, we used the sets of Python libraries PyQt5 and a visual tool for designing and creating graphical user interfaces (GUI) from Qt components - Qt Designer.