Open AccessUSING OF PROJECTIVE TRANSFORMATION IN CREATING NEURAL NETWORK DATA SET
Author(s) -
Viktoriia Kuzmina,
А. В. Хамухин,
Alexandra Koova
Publication year - 2018
Publication title -
voprosy radioèlektroniki
Language(s) - English
Resource type - Journals
eISSN - 2686-7680
pISSN - 2218-5453
DOI - 10.21778/2218-5453-2018-8-73-78
Subject(s) - computer science , license , artificial neural network , data set , set (abstract data type) , distortion (music) , artificial intelligence , transformation (genetics) , generator (circuit theory) , automation , convolutional neural network , pattern recognition (psychology) , data mining , computer vision , engineering , computer network , mechanical engineering , amplifier , biochemistry , chemistry , power (physics) , physics , bandwidth (computing) , quantum mechanics , gene , programming language , operating system
An experience of the neural network data set creation automation, which is used for license plate recognition, has been presented. The main problem of neural network training with data, obtained by nature filming is that collecting require amount of data takes a long time, beside this neural network does not effectively recognizer rare license plate formats after training. The main objective of the work is to improve recognition quality and training speed of the neural network. To achieve this objective, training data set is formed from automatically generated license plate images. Projective transformation are used for filming distortion imitation. The data set, generated in this way, includes all license plate standards, and the rare kinds percentage is enough to effectively recognize them. Using of the presented generator allows not only to significantly accelerate training data set creation, but also to improve rarely used standards of license plates recognition quality.