Информационные технологии интеллектуальной поддержки принятия решений, Информационные технологии интеллектуальной поддержки принятия решений 2018

Размер шрифта: 
Implementation convolution neural network model method for solve real time objects detection task with geospatial approach use
А. А. Махмутов

Изменена: 2018-06-20

Аннотация


In this paper, considering the implementation of the model of a convolutional neural network for solving real-time object detection problems using the approach of geoinformation technologies for processing and mapping of detected objects as geospatial units. Also this article covers issues related to the peculiarities of training, software and hardware implementation of the model of the convolutional neural network. The relevance of this article is to identify the behavior of detectable objects in different conditions on the investigated spatial sections using the convolutional network as deep learning method, therefore, it allows identifying predicative models, the use of which is necessary for solving the problems for urbanized entities infrastructure control.

Ключевые слова


Real Time Nearest Remote Sensing; Deep Learning; Convolutional Neural Network; GIS; Machine Learning; Recognition; Object Detection

Литература


1.         A. A. Makhmutov. “Use Hamming neural network for remote sensing data classification in GRASS”. 2016

2.         Makhmutov A.A., Khristodulo O.I., Gatiyatullin D.I., Musin I.A. Estimation classification accuracy of algorithms based at artificial intelligence methods for remote sensing data of South Ural area Proceedings of the 4-th international conference ITIDS’2016. Ufa State Aviation Technical University 2016; 2: 19-24.

3.         O.I. Khristoduloa, A.A. Makhmutova, T.V. Sazonovab. Use algorithm based at Hamming neural network method for natural objects classification. XIIth International Symposium «Intelligent Systems», INTELS’16, 5-7 October 2016, Moscow, Russia. Procedia Computer Science 103 (2017) 388 – 395.

4.         A. Krizhevsky. Learning multiple layers of features from tiny images. Master’s thesis, Department of Computer Science, University of Toronto, 2009

5.         L. Yao, A. Torabi, K. Cho, N. Ballas, C. J. Pal, H. Larochelle and A. C. Courville. Describing videos by exploiting temporal structure. In ICCV, 2015

6.         M. Hoai and F. De la Torre.  Max-margin early event detectors. In CVPR, 2012

7.         T. Lan, T.-C. Chen, and S. Savarese.  A hierarchical representation for future action prediction. In ECCV, 2014

8.         Hengshuang Zhao, Jianping Shi, Xiaojuan Qi, Xiaogang Wang, Jiaya Jia. Pyramid Scene Parsing Network, The Chinese University of Hong Kong, SenseTime Group Limited. arXiv:1612.01105v2