GKFC-CNN: Modified Gaussian Kernel Fuzzy C-means and Convolutional Neural Network for Apple Segmentation and Recognition

Xiaowei  Wang; Shoulin  Yin; Ke  Sun; Hang  Li; Jie  Liu; Shahid  Karim

doi:10.6180/jase.202009_23(3).0020

GKFC-CNN: Modified Gaussian Kernel Fuzzy C-means and Convolutional Neural Network for Apple Segmentation and Recognition

Computer Science and Information Engineering

Xiaowei Wang¹, Shoulin Yin This email address is being protected from spambots. You need JavaScript enabled to view it.¹, Ke Sun This email address is being protected from spambots. You need JavaScript enabled to view it.¹, Hang Li¹, Jie Liu¹ and Shahid Karim²

¹Software College, Shenyang Normal University, Shenyang, P.C 110034 - China
²Department of Computer Science, ILMA University, Karachi, Pakistan

Received: October 10, 2019
Accepted: January 5, 2020
Publication Date: September 1, 2020

Download Citation: ||https://doi.org/10.6180/jase.202009_23(3).0020

ABSTRACT

In this paper, we propose a new apple segmentation and recognition method based on improved Gaussian kernel combining fuzzy c-means and convolutional neural network. The importance of determining data distribution characteristics is analyzed. The convolution neural network with good self-learning ability is used to supervised learn the image and extract the image features. Meanwhile, the modified fuzzy c-means is used for feature clustering analysis. We modify the selection of radial width to improve Gaussian kernel function and use it for support vector machine, which will classify the extracted features. Finally, experiments on Fuji apple images demonstrate that the robustness stability and accuracy of the proposed algorithm is better than other state-of-the-art representative methods.

Keywords: Image segmentation, Fuzzy c-means, Gaussian kernel, Convolutional neural network, Support vector ma-chine, Apple recognition

REFERENCES

[1]Yin S, Zhang Y, Karim S. “Large Scale Remote Sensing Image Segmentation Based on Fuzzy Region Competition and Gaussian Mixture Model”, IEEE Access, Vol. 6, pp:26069-26080. (2018) doi: 10.1109/ACCESS.2018.2834960
[2]Teng L, Li H, Yin S. “Modified pyramid dual tree direction filter‐based image denoising via curvature scale and nonlocal mean multigrade remnant filter”, International Journal of Communication Systems, (3):e3486. (2017) doi: 10.1002/dac.3486
[3]Lin Teng, Hang Li, Shoulin Yin, Yang Sun. “Improved krill group-based region growing algorithm for image segmentation”, International Journal of Image and Data Fusion. (2019). doi: 10.1080/19479832.2019.1604574
[4]Wang P, Dong X, Qiang Z, et al. “The Auto-Obtaining of the Knowledge about Developing Hailstone Weather”, International Conference on Fuzzy Systems and Knowledge Discovery. IEEE, pp:394-399 (2007). doi: 10.1109/FSKD.2007.564
[5]Fei Z, Guo J. “A new hybrid image segmentation method for fingerprint identification”, IEEE International Conference on Computer Science and Automation Engineering. IEEE, pp:382-386 (2011). doi: 10.1109/CSAE.2011.5952702
[6]Lin Teng, Hang Li. “CSDK: A Chi-square Distribution-Kernel method for Image De-noising Under the IoT Big Data Environment”, International Journal of Distributed Sensor Networks. Vol. 15, No. 5 (2019). DOI: 10.1177/1550147719847133
[7]Lin Teng, Hang Li, Shoulin Yin, Shahid Karim&Yang Sun. “An active contour model based on hybrid energy and fisher criterion for image segmentation”, International Journal of Image and Data Fusion. 2019. doi: 10.1080/19479832.2019.1649309
[8]Shoulin Yin, Lei Meng and Jie Liu. “A New Apple Segmentation and Recognition Method Based on Modified Fuzzy C-means and Hough Transform”, Journal of Applied Science and Engineering. Vol. 22, No. 2, pp. 349-354, 2019.
[9]Kuang H, Liu C, Chan L L H, et al. “Multi-class Fruit Detection Based on Image Region Selection and Improved Object Proposals”, Neurocomputing, vol 283, 241-255, 2018. doi: 10.1016/j.neucom.2017.12.057
[10]Jeng-Shyang Pan, Junbao Li and Zheming Lu. “Adaptive Quasiconformal Kernel Discriminant Analysis”, Neurocomputing, Vol. 71, No. 13-15, pp. 2754-2760, 2008. doi: 10.1016/j.neucom.2007.10.003
[11]Hang Li, Lin Teng and Shoulin Yin. “A New Bidirectional Research Chord Method Based on Bacterial Foraging Algorithm”, Journal of Computers (Taiwan). v 29, n 3, p 210-219, June (2018). doi: 10.3966/199115992018062903020
[12]Yin Shoulin, Liu Jie, Teng Lin. “A new krill herd algorithm based on SVM method for road feature extraction”, Journal of Information Hiding and Multimedia Signal Processing, v 9, n 4, p 997-1005, July (2018).
[13]Yin Shoulin, Liu Jie, Li Hang. “A Self-Supervised Learning Method for Shadow Detection in Remote Sensing Imagery”, 3D Research, v 9, n 4, December 1, (2018). doi: 10.1007/s13319-018-0204-9
[14]Moreno R, Graña M, Ramik D M, et al. “Image Segmentation on the Spherical Coordinate Representation of the RGB Color Space”, Iet Image Processing, Vol. 6, No. 9, pp:1275-1283 (2018). doi: 10.1049/iet-ipr.2011.0634
[15]Zhang K, Zhang L, Lam K M, et al. “A Level Set Approach to Image Segmentation With Intensity Inhomogeneity”, IEEE Transactions on Cybernetics, Vol. 46, No. 2, pp:546-557 (2017). doi: 10.1109/TCYB.2015.2409119
[16]Li, Zhikui Chen, Laurence T. Yang, Liang Zhao, Qingchen. Zhang. “A Privacy-preserving High-order Neuro-Fuzzy c-Means Algorithm with Cloud Computing”, Neurocomputing, vol. 256, pp. 82-89 (2017). doi: 10.1016/j.neucom.2016.08.135
[17]Keshtkar F, Gueaieb W, Masud M. “Knowledge-based Image Segmentation Using Swarm Intelligence Techniques”, International Journal of Innovative Computing & Applications, 4:75-91. (2012) doi: 10.1504/ijica.2012.046779
[18]Yin S, Zhang Y. “Singular value decomposition-based anisotropic diffusion for fusion of infrared and visible images” International Journal of Image and Data Fusion, vol 9, issue 4, pp:146-1 (2018) doi: 10.1080/19479832.2018.1487886
[19]Jian M, Jung C. “Interactive Image Segmentation Using Adaptive Constraint Propagation”, IEEE Transactions on Image Processing, Vol. 25, No. 3, pp:1301-1311 (2016). doi: 10.1109/TIP.2016.2518480
[20]Medeiros R S, Scharcanski J, Wong A. “Image segmentation via multi-scale stochastic regional texture appearance models”, Computer Vision & Image Understanding, Vol. 142, pp:23-36 (2016). doi: 10.1016/j.cviu.2015.06.001
[21]Hussain M K , Sufyan M , Ali Q M , et al. “Kernel Possibilistic Fuzzy c-Means Clustering with Local Information for Image Segmentation”, International Journal of Fuzzy Systems, 21, no. 1, pp. 321–332, 2019. doi: 10.1007/s40815-018-0537-9