Open AccessObject Detection in Images Using Deep Neural Networks for Agricultural Machinery
Author(s) -
Nikita Andreyanov,
Anatoly Sytnik,
Mikhail P. Shleymovich
Publication year - 2022
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/988/3/032002
Subject(s) - tractor , field (mathematics) , computer science , artificial neural network , plough , artificial intelligence , object (grammar) , computer vision , object detection , agricultural machinery , agriculture , pattern recognition (psychology) , engineering , mathematics , automotive engineering , ecology , agronomy , pure mathematics , biology
The article deals with the creation of intelligent tractor driver support systems based on computer vision technologies for analyzing the direction of movement and detecting obstacles when performing specified operations, such as plowing, harrowing, weeding, and fertilizing. Electric power poles, trees, rocks, bird nests, animals, people, and field roads are identified as obstacles. The solution of functional problems in the system is based on the extraction of information from images using methods for detecting and recognizing objects in images. The analysis of existing approaches to solving the problems under consideration is carried out and it is shown that the use of deep neural networks is effective. The practical use of the methods based on the chosen approach is based on the performance of the computing system, the availability of sufficient training data and the optimality of the training method. It is shown that these factors are important when implementing an intelligent tractor driver support system.