Automation of Information Processing in the Intelligent System of Marine Monitoring
V.A. Pyatakovich, Aleksey Vladimirovich Nikolaev, E. A. Kostikov, O. V. Khotinskii · Journal of Machinery Manufacture and Reliability · 2021
Abstract— A solution to the problem of sampling to automate sea target (ST) classification by features is presented. A new method of generation of training samples is proposed, which preserves the topological properties of the original sample in the generated subsample and does not require loading the whole sample in the computer memory, which makes it possible to reduce the sample volume and lower the requirements for resources of the intelligent ST classification system operating in an autonomous mode.