ABSTRACT
In an underwater view, wavelength-reliant light absorption in addition to scattering worsens the visibility of images that lead to lower contrast and also distorted color casts. To handle these problems, a proficient approach is proposed for the enhancement and also for the classification of UI. Originally, the underwater input image RGB is modified using Enhanced ABC algorithm. The subsequent phase is to extract the features. The extracted attributes are inputted to the Modified PCA approach, in this phase, the dimensionality of the features is reduced. Then, the classification operation is performed by utilizing ANFIS classifier. At last, the classified enhanced deeper water images along with the enhanced shallow water images are analyzed during the testing phase. The performance analysis is made for the proposed classifiers and existing techniques such as NN, SVM, and KNN. In addition, the performance of the proposed EABC model is compared over ABC, GA, and PSO in terms of correlation, Spearman rank correlation, sharpness, EME, Mutual information and NMI. The proposed classified method obtain the accuracy (0.9473), sensitivity (0.9230), specificity (0.9677) however, the existing methods provide only 0.8771 accuracy. Similarly, the proposed Enhanced ABC methods provide the improved performance while considering the other optimization algorithm.
Abbreviations: ABC: artificial bee colony; ANFIS: adaptive neuro-fuzzy inference system; ANN: artificial neural network; CM: covariance matrix; DWT: discrete wavelet transform; EABC: enhanced artificial bee colony; EME: enhanced measurement error; FDR: false discovery rate; FLS: forward-looking sonar; FPR: false positive rate; FS: forward scattering; GA: genetic algorithm; KNN: k-nearest neighbors; MCC: Mathew’s correlation coefficient; MOE: measure of entropy; NMI: normalized mutual information; NN: neural network; NPV: negative prediction value; PCA: principal component analysis; PD: probability distribution; PSO: particle swarm optimization; RGB: red green blue; SD: standard deviation; SRC: Spearman rank correlation; SURF: speeded up robust feature; SVM: support vector machine; SWT: stationary wavelet transform; UIE: underwater image enhancement; WVD: Wigner-Ville distribution
Disclosure statement
No potential conflict of interest was reported by the author(s).
Additional information
Notes on contributors
R. Prasath
R. Prasath has completed his BE(ECE) from Sona College of Technology, Salem Affliated to Anna University in the year 2009. He completed his MTech Multimedia Technology in the year 2011 from Karunya Institute of Science and Technology (Deemed to be University), Coimbatore. He currently pursuing his (PhD) as Research Scholar from Meenakshi Academy of Higher Education and Research (Deemed to be University), Chennai. He has published two International Conference and two National Conference. He has published two International Journals. His area of interest is Computer Graphics, Image Processing, Computer Vision, Multimedia and Cryptography.
T. Kumanan
Dr T. Kumanan received ME and PhD Degrees in 2005 and 2014 from Anna University, Chennai. He is working as Professor in Department of CSE at Meenakshi Academy of Higher Education (Deemed to be University), Chennai. He has published 10 paper in International level Conferences and Two papers in National level Conferences. He has published 15 paper in Internal National Journal. His areas of interests are in Computer Networks, Mobile Computing, Cryptography and Network Security, Image Processing and High Speed Network. He is member of ISTE and CSI.