Deep Learning Using Symmetry, FAST Scores, Shape-Based Filtering and Spatial Mapping Integrated with CNN for Large Scale Image Retrieval

Abstract

This article presents symmetry of sampling, scoring, scaling, filtering and suppression over deep convolutional neural networks in combination with a novel content-based image retrieval scheme to retrieve highly accurate results. For this, fusion of ResNet generated signatures is performed with the innovative image features. In the first step, symmetric sampling is performed on the images from the neighborhood key points. Thereafter, the rotated sampling patterns and pairwise comparisons are performed, which return image smoothing by applying standard deviation. These values of smoothed intensity are calculated as per local gradients. Box filtering adjusts the results of approximation of Gaussian with standard deviation to the lowest scale and suppressed by non-maximal technique. The resulting feature sets are scaled at various levels with parameterized smoothened images. The principal component analysis (PCA) reduced feature vectors are combined with the ResNet generated feature. Spatial color coordinates are integrated with convolutional neural network (CNN) extracted features to comprehensively represent the color channels. The proposed method is experimentally applied on challenging datasets including Cifar-100 (10), Cifar-10 (10), ALOT (250), Corel-10000 (10), Corel-1000 (10) and Fashion (15). The presented method shows remarkable results on texture datasets ALOT with 250 categories and fashion (15). The proposed method reports significant results on Cifar-10 and Cifar-100 benchmarks. Moreover, outstanding results are obtained for the Corel-1000 dataset in comparison with state-of-the-art methods.

Deep Learning Using Isotroping, Laplacing, Eigenvalues Interpolative Binding, and Convolved Determinants with Normed Mapping for Large-Scale Image Retrieval

“…This fast KLT calculates eigenvectors efficiently when this is treated with small samples. In [ 38 ], deep learned features are computed by finding the symmetry in FAST scores with neighborhood, smoothing, and standard deviation. Feature scaling, reduction, and filtering is applied to resize the features for a variety of datasets.…”

“…The corel-10000 database [ 38 ] comprises several image classes. The corel-10000 dataset is comprised of 100 classes where each class consists of 100 images.…”

“…The presented approach shows AR rates for all categories of Corel-1000 dataset in Figure 13b. Various sample images for Corel-1000 database [38].…”

“…Sample images from this dataset are shown in Figure 15. Various sample images for corel-10000 dataset [38].…”

Image Retrieval Method Based on Image Feature Fusion and Discrete Cosine Transform

“…LeNetF6 [28] extracted the fully connected F6 of the LeNet network [29] as the feature; shape-based filtering. SBF-SMI-CNN [30] integrated spatial mapping with CNN. The dimension reduction-based methods [31] use multilinear principal component analysis (MPCA) to reduce the dimensionality of image features.…”