THE NEURAL NETWORK MODEL FOR MULTIPARAMETER OBJECTS RECOGNITION
Abstract
The article contains the description of object detection and recognition method. The approach bases on the use of convolu-tional neural network with high-order neurons and the hypotheses removal.
About the Authors
Oksana Stanislavovna Mezentseva
North-Caucasus Federal University
Russian Federation
Russian Federation
Nikita Alekseevich Lagunov
North-Caucasus Federal University
Russian Federation
Russian Federation
Dmitry Viktorovich Mezentsev
North-Caucasus Federal University
Russian Federation
Russian Federation
References
1. Erhan D. Scalable Object Detection using Deep Neural Networks / D. Erphan, C. Szegedy, A. Toshev, D. Anguelov // Computer Vision and Pattern Recognition. Columbus, 2014. pp. 2155-2162.
2. Sermanet P. OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks / P. Ser-manet, D. Eigen, X. Zhang, M. Mathieu, R. Fergus, Y. LeCun // Computer Vision and Pattern Recognition. Columbus, 2013. pp. 1082-1090.
3. Girshick R. Rich feature hierarchies for accurate object detection and semantic segmentation / R. Girshick, J. Donahue, T. Darrell, J. Malik // Computer Vision and Pattern Recognition. Columbus, 2014. pp. 580-587.
4. Cheng M.-M. BING: Binarized Normed Gradients for Objectness Estimation at 300fps / M.-M. Cheng, Z. Zhang, W. Y. Lin, P. Torr // Computer Vision and Pattern Recognition. Puerto-Rico, 2014. pp. 260-275.
5. Немков Р. М. Экспериментальное исследование и анализ влияния базовых параметров сверточных нейронных сетей на качество их обучения / Р. М. Немков, О. С. Мезенцева // Вестник Северо-Кавказского федерального университета. Ставрополь. 2013. №3 (36). C. 21-26.
Review
For citations:
Mezentseva O.S., Lagunov N.A., Mezentsev D.V. THE NEURAL NETWORK MODEL FOR MULTIPARAMETER OBJECTS RECOGNITION. Modern Science and Innovations. 2016;(4):33-38. (In Russ.)
Views:
67
Email this article 