Hybrid Bag-of-Visual-Words and FeatureWiz Selection for Content-Based Visual Information Retrieval

Abstract

Recently, content-based image retrieval (CBIR) based on bag-of-visual-words (BoVW) model has been one of the most promising and increasingly active research areas. In this paper, we propose a new CBIR framework based on the visual words fusion of multiple feature descriptors to achieve an improved retrieval performance, where interest points are separately extracted from an image using features from accelerated segment test (FAST) and speeded-up robust features (SURF). The extracted keypoints are then fused together in a single keypoint feature vector and the improved RootSIFT algorithm is applied to describe the region surrounding each keypoint. Afterward, the FeatureWiz algorithm is employed to reduce features and select the best features for the BoVW learning model. To create the codebook, K-means clustering is applied to quantize visual features into a smaller set of visual words. Finally, the feature vectors extracted from the BoVW model are fed into a support vector machines (SVMs) classifier for image retrieval. An inverted index technique based on cosine distance metric is applied to sort the retrieved images to the similarity of the query image. Experiments on three benchmark datasets (Corel-1000, Caltech-10 and Oxford Flower-17) show that the presented CBIR technique can deliver comparable results to other state-of-the-art techniques, by achieving average accuracies of 92.94%, 98.40% and 84.94% on these datasets, respectively.

Content-Based Image Retrieval Using BRISK and SURF as Bag-of-Visual-Words for Naïve Bayes Classifier

“…Bakheet et al [7] proposed a CBIR based on keypoint feature fusion and hybrid of BoVW. They utilized FeatureWiz Selection for feature selection process.…”

“…. ∑ ℊ(𝑃 𝑖, 𝑃 𝑗 ) (𝑃 𝑖 ,𝑃 𝑗 )∈ℒ (7) where L is the number of ℒ, ℊ 𝑥 is the gradient in the x-direction, and ℊ 𝑦 is the gradient in the y-direction.…”

“…CBIR approaches are becoming effective and efficient for managing and retrieving images from massive image collections in comparison to conventional text-based image retrieval (TBIR) techniques. The CBIR-based approaches took the place of the conventional TBIR-based approaches, which had several drawbacks, including a poor representation of the image's salient objects, a reliance on a particular language, a high computational cost, and a lengthy manual annotation process in dealing with large image databases [1].…”

Optimization of CNN-based image retrieval process based on Fourier transform and entropy

“…Local feature based methods typically use convolutional neural networks to extract local features in images, such as regions and corners. Then, methods such as Bag of Visual Words (BoW) [11] are used to encode and cluster these local features, and finally, similarity measurement is used for retrieval. The advantage of these methods is that they can preserve the detailed and structural information of the image, but due to the need to encode and cluster local features, the computational complexity is high, and there may be issues with local feature matching.…”