A HYBRID IMPLEMENTATION OF MULTICLASS RECOGNITION ALGORITHM FOR CLASSIFICATION OF CRABS AND LOBSTERS

TitleA HYBRID IMPLEMENTATION OF MULTICLASS RECOGNITION ALGORITHM FOR CLASSIFICATION OF CRABS AND LOBSTERS
Publication TypeJournal Article
Year of Publication2018
AuthorsPRATHUSHA P., JYOTHI S., MAMATHA D.M.
JournalNeural, Parallel, and Scientific Computations
Volume26
Issue1
Start Page75
Pagination20
Date Published2018
ISSN1056-2176
KeywordsGLCM, KNN, ROI/NROI
Abstract

Crabs and lobsters have a major share in the production of sea food. Among the marine foods available along the coasts of south India, crabs occupy the major role for fetching economy to the farmers in fishing Industry. The problem faced by many fishing farmers is the manual sorting of the sea food available. Because of the huge variations in the price of the marine fauna it will be beneficial to farmers if they are classified and packaged. The proposed technique in this paper uses a hybrid implementation of multi class crab recognition system. The crab features are extracted using reduced gray level co-occurrence matrix (GLCM) with less feature set in contrast to traditional huge feature set and a multiclass training vector is created. Further, Crab images are classified using KNN classification. Test crab samples (images) features are matched with the stored database by various matching techniques such as Euclidean distance, cosine and city block distances. The experiments are carried on images collected from various coasts of south India and result shows that hybrid multiclass approach using KNN classifier has better recognition accuracy. It uses hybrid approach of GLCM with less features , Segmentation with ROI/NROI technique and KNN classifier and K-fold cross validation. Dimensionality Reduction is applied which is a significant improvement in multiclass recognition process. This paper successfully proposes a hybrid multi-class recognition algorithm which uses less feature set compared to traditional feature set. Further it reduces the feature set to minimal set and achieved good accuracy for multiclass problems. The proposed technique is tested with KNN classifier with various distance measures like Euclidian, cosine and city block. The novelty of the proposed multiclass recognition algorithm lies in training with minimal feature set.

URLhttps://acadsol.eu/npsc/articles/26/1/5.pdf
DOI10.12732/npsc.v26i1.5
Refereed DesignationRefereed
Full Text

REFERENCES
[1] P. Prathusha, S. Jyothi (2015), A comparative study on certain class of noise
reduction filters classification of marine fauna, International Journal on Computing,
Communications and Systems (IJCCS), ISSN:2277-6699 (December-2015),Vol4, Issue- 2, December.
[2] S. Jyothi, P. Prathusha, K. Himabindu, (2016),, A novel step wise algorithm for
removal of high density salt and pepper noise,, International Journal of Computational
Science, Mathematics and Engineering, Special Issue on Computational
Science, Mathematics and Biology, ISSN: 2349-8439, Impact Factor: 2. 651.
DOI: 10. 18645/IJCSME. SPC. 0012.
[3] P. Prathusha, S. Jyothi, A study on Hybrid noise reduction filters Classification
of Marine fauna Effect of thickness and testing parameters on Tensile strength
Variablility of Electrospunnano fibrous mat, at Global Congress on Computing
and Media Technologies 2015, Satayabama university, chennai from November
25 th to november 27 th 2015.
[4] R. ObulaKonda Reddy et. al, An Effective GLCM and Binary Pattern Schemes
Based Classification for Rotation Invariant Fabric Textures, International Journal
of Recent Trends in Electrical & Electronics Engg., Volume 3, Issue 1 Dec.
2013. ISSN: 2231-6612.
[5] KhinNyeinNyeinHlaing, First Order statistics and GLCM based feature extraction
for recognition of myanmar paper currency, Proceedings of 31st The IIER
International Conference, Bangkok, Thailand, 2nd Aug. 2015, ISBN: 978-93-
85465-65-9.
[6] IshakTaman, Classification System for Wood Recognition Using K-Nearest Neighbor
with Optimized Features from Binary Gravitational Algorithm, International
Conference Recent treads in Engineering & Technology (ICRET’2014) Feb 13-14,
2014 Batam (Indonesia)http://dx. doi. org/10. 15242/IIE. E0214508.
[7] M. Harshavardhan, GLCMarchitechture for Image Extraction, International Journal
of Advanced Research in Electronics and Communication Engineering (IJARECE)
Volume 3, Issue 1, January 2014 ISSN: 2278 – 909X.
[8] Girisha et. al, Texture Feature Extraction of Video Frames using GLCM, International
Journal of Engineering Trends and Technology (IJETT) Volume 4 Issue
6- June 2013 ISSN: 2231-5381 http://www. ijettjournal. org
[9] Keesook J. Han and Ahmed H. Tewfik, Expert Computer vision based crab
recognition system., DOI: 10. 1109/ICIP. 1996. 560961IEEE Xplore: 06 August 2002.
[10] P. Mohanaiah et. al Image Texture Feature Extraction Using GLCM Approach
International Journal of Scientific and Research Publications, Volume 3, Issue 5,
May 2013 ISSN 2250-3153www. ijsrp. org
[11] Mahfuzah Mustafa et. al, GLCM Texture Classification for EEG Spectrogram Image,
2010 IEEE EMBS Conference on Biomedical Engineering & Sciences (IECBES
2010), Kuala Lumpur, Malaysia, 30th November - 2nd December 2010.
[12] AbdolvahabEhsanirad and Sharath Kumar Y. H., Leaf recognition for plant classification
using GLCM and PCA methods, Oriental Journal of Computer Science
&TechnologyVol. 3(1), 31-36 (2010).
[13] Krzysztof Okarma and Jaros lawFastowicz, No-Reference Quality Assessment of
3D PrintsBased on the GLCM Analysis, 978-1-5090-1866-6/16/$31. 00 c 2016 IEEE
[14] Gunjan Mukherjee et. al, Study on the potential of combined GLCM features towards
medicinal plant classification, proceedings of 2nd International Conference
on Control, Instrumentation, Energy & Communication (CIEC) 2016 978-1-5090-
0035-7/16/$31. 00c 2016IEEE.
[15] Maryam Imani et. al, GLCM,Gabor and Morphology pprofiles fusion for Hyperspectral
Image Classification, proceedingsof 24th Iranian Conference on Electrical
Engineering(ICEE) 2016.
[16] SomnathDey and DebasisSamanta, Iris Data Indexing Method Using Gabor Energy
Features IEEE TRANSACTIONS ON INFORMATION FORENSICS AND
SECURITY, VOL. 7, NO. 4, AUGUST 2012
[17] Kulkarni, S. B., et al. ”Iris Recognition using Fusion of Gray Level Co-occurrence
Matrix and Gray Level Run Length Matrix,,. ” (2013): 241-246.
[18] Kulkarni, Shrinivasrao B., et al. ”GLCM-based multiclass iris recognition using
FKNN and KNN. ” International Journal of Image and Graphics 14. 03 (2014):
1450010.
[19] Kulkarnia, S. B., U. P. Kulkarni, and R. S. Hegadi. ”Analysis of Iris Recognition
using Normalized and Un-normalized Iris images. ” International Journal of
Information Processing 7. 3 (2013): 26-33.
[20] Celik, Elif Tuba. ”Iris Recognition—Selecting a Fuzzy Region of Interest in
Standard Eye Images. ” Soft Computing Applications. Springer, Cham, 2016. 793-804.
[21] Bremananth, R., B. Nithya, and R. Saipriya. ”Wood species recognition using
GLCM and correlation. ” Advances in Recent Technologies in Communication
and Computing, 2009. ARTCom’09. International Conference on. IEEE, 2009.
[22] Khalid, Marzuki, et al. ”Design of an intelligent wood species recognition system.
” International Journal of Simulation System, Science and Technology 9. 3 (2008): 9-19.
[23] Ding, Chris HQ, and Inna Dubchak. ”Multi-class protein fold recognition using
support vector machines and neural networks. ” Bioinformatics 17. 4 (2001): 349-358.
[24] In press, A random pick multi dimensionality algorithm for multiclass problems,
[25] A. R. Yadav,R. Yadav, R. Anand,M. Dewal and S. Gupta,, Hardwood species
classification with DWT based hybrid texture feature extraction techniques,,
Sadhana,vol. 40,pp. 227-2312,2015.