Article · January 31, 2020

Применение систем распознавания образов для автоматизации процесса классификации объектов растениеводческого производства и выявления отклонений в развитии растений

Gihong Kim, Alexandr Demyanenko, V. Tyukanko

Read on vestnik.satbayev.universityDOI 10.51301/vest.su.2020.v137.i1.03

Abstract

To automate many agricultural processes, classification (sorting) of plants by species and by other quantitative and qualitative characteristics is important. Examples include automatic weeding or spot spreading of fertilizers or herbicides. In addition, the timely and accurate determination of deviations in the development of plants, including the diagnosis of their diseases, plays an important role in preventing a decrease in the yield of agricultural products. To solve these problems, it is necessary to automate as much as possible all processes in agriculture by methods based on machine learning and, in particular, on methods of pattern recognition. Currently, deep learning is widely used in image processing. The article provides examples of the successful application of pattern recognition methods in agriculture and discusses various methods of machine learning and deep feature extraction that adapt deep learning models to the problem under consideration.

Smart Agriculture and AIAdvanced Neural Network ApplicationsIndustrial Vision Systems and Defect Detectionimage recognitionplant taxonomyplant disease diagnosis

Abstract

More by Alexandr Demyanenko