Abstract
In order to solve the poor accuracy problem which caused by the gradient descent easily fail into local optimum during the training process and the noise interference in process of feature extracting. This paper presents an integrated optimization method of simulated annealing (SA) and Gaussian convolution based on Convolutional Neural Network (CNN). Firstly, the improved algorithm extract some features from the central feature of a model as priori information, and find the optimal solution as initial weights of full-connection layer by simulating annealing, so as to accelerate the weight updating and convergence rate. Secondly, using the Gaussian convolution method, this paper can smooth image to reduce noise disturbing. Finally, the improved integrated optimization method is applied to the MNIST and CIFAR-10 databases. Simulation results show that the accuracy rate of the integrated network is improved through the contrastive analysis of different algorithms.
Similar content being viewed by others
References
Sui TT, Wang XF (2016) Convolutional neural networks with candidate location and multi-feature fusion. Acta Automat Sin 42(6):875–882
Xiao S, Pan T, Ren FJ (2016) 2016. Facial expression recognition using ROI-KNN deep convolutional neural networks. Acta Automat Sin 42(6):883–891
Liu Z, Hu J, Weng L et al (2018) Rotated region based CNN for ship detection[C]// IEEE international conference on image processing. IEEE:900–904
Tang H, Xiao B, Li W et al (2017) Pixel convolutional neural network for multi-focus image fusion[J]. Inf Sci 433
Zheng L, Yang Y, Tian QSIFT, Meets CNN (2018) A decade survey of instance retrieval.[J]. IEEE transactions on pattern analysis & machine. Intelligence 40(5):1224–1244
Szegedy C, Liu W, Jia Y, et al. (2015) Going deeper with convolutions. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA. 1–9
He K, Zhang X, Ren S, Sun J (2016) Deep Residual Learning for Image Recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, United States, 770–778
Qiu T, Chen N, Li K, Atiquzzaman M, Zhao W (2018) How can heterogeneous internet of things build our future: a survey. IEEE Commun Surveys Tutor 99:1–1
Qiu T, Qiao R, Wu D (2018) EABS: an event-aware backpressure scheduling scheme for emergency internet of things. IEEE Trans Mob Comput 17(1):72–84
Sun W, Shao S, Yan R (2016) 2016. Induction motor fault diagnosis based on deep neural network of sparse auto-encoder. J Mech Eng 52(9):65–71
Chuanpeng LI, Qin P, Zhang J (2017) Research on image Denoising based on deep convolutional neural network. Comput Eng 43(3):253–260
Ayumi V, Rere LMR, Fanany MI et al (2017) Optimization of convolutional neural network using microcanonical annealing algorithm. International conference on advanced computer science and information systems. IEEE:506–511
Rere LMR, Fanany MI, Arymurthy AM (2016) Metaheuristic algorithms for convolution neural network. Hindawi Publishing Corp
Kim J, Kim J, Jang GJ et al (2016) Fast learning method for convolutional neural networks using extreme learning machine and its application to lane detection. Neural Netw 87(2016):109–121
Dauphin YN, Vries HD, Bengio Y (2015) Equilibrated adaptive learning rates for non-convex optimization 35(3): 285–290
Chang L, Deng XM, Zhou MQ et al (2016) Convolutional neural networks in image understanding. Acta Auto Sin 42(9):1300–1312
Cortes C, Mohri M, Rostamizadeh A (2012) 2012. Algorithms for learning kernels based on centered alignment. J Mach Learn Res 13(2):795–828
Li D (2012) The MNIST database of handwritten digit images for machine learning research [best of the web]. IEEE Signal Process Mag 29(6):141–142
Acknowledgements
This work was supported by the National Science Foundation Council of China (nos. 61771006), the Programs for Science and Technology Development of He’nan Province, China (nos. 162102210401, 162102210022), the Open Fund Project of Key Laboratory of Spectral Imaging Technology of Chinese Academy of Sciences (nos. LSIT201711D).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Hu, Zt., Zhou, L., Jin, B. et al. Applying Improved Convolutional Neural Network in Image Classification. Mobile Netw Appl 25, 133–141 (2020). https://doi.org/10.1007/s11036-018-1196-7
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11036-018-1196-7