A Multi-class SVM Based Content Based Image Retrieval System Using Hybrid Optimization Techniques

Abstract

Due to the increasing usage of multimedia and storage devices accessible, searching for large image databases has become imperative. Furthermore, the handiness of high-speed internet has escalated the exchange of images by users enormously. Content-Based Image Retrieval is proposed in this work, taking features based on Exact Legendre Moments, HVS color quantization with dc coefficient and statistical properties such as variance, mean, and skew of Conjugate Symmetric Sequency Complex Hadamard Transform (CS-SCHT). In most of the machine learning tasks, the quality of the learning process depends on dimensionality. High dimensional datasets can influence the classification outcome and training time. To overcome this problem, we use DE (Differential Evolution) to generate the optimal feature subsets. The features scaled by weights derived from the firefly algorithm, which fed to Multi-Class SVM. The fitness function taken for the firefly algorithm is the classification error of SVM. By minimizing fitness function, optimum weights are obtained. When these optimal weights are applied to SVM, the proposed algorithm exhibits better precision, recall, and accuracy when compared to some of the existing algorithms in the literature.

Content‐based image retrieval using Gaussian–Hermite moments and firefly and grey wolf optimization

“…When these optimized weights are applied, accuracies are maximized. In hunting, the global maxima is represented by catching the prey using Equations (11)(12)(13) in the hunting of grey wolves.…”

Analysis of Classroom Learning Behaviors Based on Internet of Things and Image Processing

“…Chao et al [10] extracted the differential entropies of EEG signals, trained deep belief network (DBN) and DBNhidden Markov model (HMM), and achieved high classification accuracy of two types of emotions. Aditya and Tibarewala [11] preprocessed the EEG signals from ten channels through discrete wavelet transform (DWT), extracted energy entropy as features, and classified them with support vector machine (SVM) and k-nearest neighbors (KNN) algorithm; experimental results show that band was classified more accurately than the other low-frequency channels [12][13][14][15][16].…”

A Hotspot Discovery Method Based on Improved FIHC Clustering Algorithm

“…The experimental data are the data of Sina blog used in literature [7], which contain 4236 relevant pages published by Sina blog, the specific time is from April 1 to April 30, 2014. The microblog texts were selected and the topics were manually annotated, and 10 topic categories were obtained which were manually counted, as shown in Tab.…”

“…Fan et al, and Li et al, respectively, improved the single-Pass clustering algorithm, solved the problems of the original algorithm, and effectively improved the discovery efficiency of microblog hotspot [4,5]. Yi et al, and some other scholars used different clustering algorithms and simulation mathematical models, such as firefly clustering algorithm, OLDA model, SEPPM model, impulse time series behavior dynamic model, to discover and simulate the network hot events, and got some results [6][7][8][9][10][11][12][13][14][15][16][17][18].…”