Optimal Recognition Model Based on Convolutional Neural Networks and Fuzzy Gravitational Search Algorithm Method

In this paper we propose the optimization of a convolutional neural network (CNN) using the Fuzzy Gravitational Search Algorithm method (FGSA). The FGSA is inspired in extension of the Gravitational Search Algorithm (GSA) using fuzzy logic and this method is used to obtain the number of images per block that will enter in the training phase. The optimized CNN is applied for pattern recognition using the 10 handwritten numbers of the MINIST database. The model of the CNN model presented in this paper can be applied for any recognition or image classification application. In addition, the recognition rate achieved with the CNN optimized by the FGSA was compared against the results obtained with the non-optimized CNN.

[1]  Patrice Y. Simard,et al.  High Performance Convolutional Neural Networks for Document Processing , 2006 .

[2]  Yoshua Bengio,et al.  Convolutional networks for images, speech, and time series , 1998 .

[3]  Oscar Castillo,et al.  Genetic optimization of modular neural networks with fuzzy response integration for human recognition , 2012, Inf. Sci..

[4]  Oscar Castillo,et al.  Ant colony optimization for the design of Modular Neural Networks in pattern recognition , 2016, 2016 International Joint Conference on Neural Networks (IJCNN).

[5]  Oscar Castillo,et al.  Face Recognition with a Sobel Edge Detector and the Choquet Integral as Integration Method in a Modular Neural Networks , 2015, Design of Intelligent Systems Based on Fuzzy Logic, Neural Networks and Nature-Inspired Optimization.

[6]  Hossein Nezamabadi-pour,et al.  GSA: A Gravitational Search Algorithm , 2009, Inf. Sci..

[7]  O. P. Verma,et al.  Newtonian Gravitational Edge Detection Using Gravitational Search Algorithm , 2012, 2012 International Conference on Communication Systems and Network Technologies.

[8]  Oscar Castillo,et al.  An improved sobel edge detection method based on generalized type-2 fuzzy logic , 2014, Soft Computing.

[9]  Siti Zaiton Mohd Hashim,et al.  Training feedforward neural networks using hybrid particle swarm optimization and gravitational search algorithm , 2012, Appl. Math. Comput..

[10]  Urs A. Muller,et al.  Learning long-range vision for autonomous off-road driving , 2009 .

[11]  Oscar Castillo,et al.  Interval Type-2 TSK Fuzzy Logic Systems Using Hybrid Learning Algorithm , 2005, The 14th IEEE International Conference on Fuzzy Systems, 2005. FUZZ '05..

[12]  Tao Wang,et al.  End-to-end text recognition with convolutional neural networks , 2012, Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012).

[13]  Marco Zennaro,et al.  Large-scale privacy protection in Google Street View , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[14]  Yihong Gong,et al.  Linear spatial pyramid matching using sparse coding for image classification , 2009, CVPR.

[15]  Oscar Castillo,et al.  Modelling, Simulation and Control of Non-linear Dynamical Systems: An Intelligent Approach Using Soft Computing and Fractal Theory , 2001 .

[16]  Yoshua Bengio,et al.  Greedy Layer-Wise Training of Deep Networks , 2006, NIPS.

[17]  Patricia Melin,et al.  Modular Neural Networks Architecture Optimization with a New Evolutionary Method Using a Fuzzy Combination Particle Swarm Optimization and Genetic Algorithms , 2009, Bio-inspired Hybrid Intelligent Systems for Image Analysis and Pattern Recognition.

[18]  Oscar Castillo,et al.  Optimization by Cuckoo Search of Interval Type-2 Fuzzy Logic Systems for Edge Detection , 2014, WCSC.

[19]  Fevrier Valdez,et al.  Fuzzy logic in the gravitational search algorithm for the optimization of modular neural networks in pattern recognition , 2015, Expert Syst. Appl..

[20]  S. Mirjalili,et al.  A new hybrid PSOGSA algorithm for function optimization , 2010, 2010 International Conference on Computer and Information Application.

[21]  Oscar Castillo,et al.  Fuzzy Adaptation for Particle Swarm Optimization for Modular Neural Networks Applied to Iris Recognition , 2017, NAFIPS.

[22]  Antonio Rodríguez Díaz,et al.  Simulation of the bird age-structured population growth based on an interval type-2 fuzzy cellular structure , 2011, Inf. Sci..

[23]  Oscar Castillo,et al.  Edge-Detection Method for Image Processing Based on Generalized Type-2 Fuzzy Logic , 2014, IEEE Transactions on Fuzzy Systems.

[24]  Salwani Abdullah,et al.  Gravitational search algorithm with heuristic search for clustering problems , 2011, 2011 3rd Conference on Data Mining and Optimization (DMO).

[25]  P. Melin,et al.  Optimization of modular granular neural networks using a firefly algorithm for human recognition , 2017, Eng. Appl. Artif. Intell..

[26]  Phil Kim,et al.  MATLAB Deep Learning , 2017, Apress.

[27]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[28]  Patricia Melin,et al.  General type-2 fuzzy edge detector applied on face recognition system using neural networks , 2016, 2016 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE).

[29]  Lawrence D. Jackel,et al.  Handwritten Digit Recognition with a Back-Propagation Network , 1989, NIPS.

[30]  Yefeng Zheng,et al.  Deep Learning and Convolutional Neural Networks for Medical Image Computing , 2017, Advances in Computer Vision and Pattern Recognition.

[31]  Nohé R. Cázarez-Castro,et al.  Designing Type-1 and Type-2 Fuzzy Logic Controllers via Fuzzy Lyapunov Synthesis for nonsmooth mechanical systems , 2012, Eng. Appl. Artif. Intell..

[32]  Patricia Melin,et al.  New Hybrid Intelligent Systems for Diagnosis and Risk Evaluation of Arterial Hypertension , 2017 .

[33]  Patricia Melin,et al.  A hybrid model based on modular neural networks and fuzzy systems for classification of blood pressure and hypertension risk diagnosis , 2018, Expert Syst. Appl..

[34]  Baoxin Li,et al.  Convolutional Neural Networks in Visual Computing: A Concise Guide , 2017 .

[35]  Marc'Aurelio Ranzato,et al.  Unsupervised Learning of Invariant Feature Hierarchies with Applications to Object Recognition , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[36]  Patricia Melin,et al.  Multi-objective optimization for modular granular neural networks applied to pattern recognition , 2017, Inf. Sci..

[37]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[38]  Oscar Castillo,et al.  Intelligent control of complex electrochemical systems with a neuro-fuzzy-genetic approach , 2001, IEEE Trans. Ind. Electron..

[39]  Patricia Melin,et al.  A hybrid modular neural network architecture with fuzzy Sugeno integration for time series forecasting , 2007, Appl. Soft Comput..

[40]  Patricia Melin,et al.  Fuzzy logic in the gravitational search algorithm enhanced using fuzzy logic with dynamic alpha parameter value adaptation for the optimization of modular neural networks in echocardiogram recognition , 2015, Appl. Soft Comput..

[41]  Oscar Castillo,et al.  A new gravitational search algorithm using fuzzy logic to parameter adaptation , 2013, 2013 IEEE Congress on Evolutionary Computation.