Publication Details
Abstract
This article discusses the development of a mobile application for automated identification of medicinal plants and quality assessment of medicinal plant materials using artificial intelligence and computer vision. This research is relevant due to the need to implement objective digital control methods in the procurement and processing of raw materials and the digitalization of the pharmaceutical industry. Neural network architectures for plant image recognition are analyzed, including MobileNet, EfficientNet, and YOLO family models. A functional architecture of the mobile application is proposed that enables field recognition of medicinal plants, analysis of the morphological purity of raw materials, and the automatic generation of quality control protocols. The development of a training set and the implementation of a mobile solution based on TensorFlow Lite are considered. The development could contribute to the automation of plant identification, increased accuracy of pharmacognostic analysis, and the development of digital technologies in the pharmaceutical industry.