Every day, a tremendous amount of image data is generated as a result of recent advances in imaging and computing technology. Several content-based image retrieval (CBIR) approaches have been introduced for searching image collections. These methods, however, involve greater computing and storage resources. Cloud servers can address this issue by offering a large amount of computational power at a low cost. However, cloud servers are not completely trustworthy, and data owners are concerned about the privacy of their personal information. In this research, we propose and implement a secure CBIR (SCBIR) strategy for searching and retrieving cipher text image databases. In the proposed scheme, the extract aggregated feature vectors to represent the related image collection and use a safe Asymmetric Scalar-Product-Preserving Encryption (ASPE) approach to encrypt these vectors while still allowing for similarity computation. To improve search time, all encrypted features are recursively clustered using the k-means method to create a tree index. The results reveal that SCBIR is faster at indexing and retrieving than earlier systems, with superior retrieval precision and scalability. In addition, our paper introduces the watermark to discover any illegal distributions of the images that are received by unlawful data users. Particularly, the cloud server integrates a unique watermark directly into the encrypted images before sending them to the data users. As a result, if an unapproved image copy is revealed, the watermark can be extracted and the unauthorized data users who spread the image can be identified. The performance of the proposed scheme is proved, while its performance is demonstrated through experimental results.
Due to the recent improvements in imaging and computing technologies, a massive quantity of image data is generated every day. For searching image collection, several content-based image retrieval (CBIR) methods have been introduced. However, these methods need more computing and storage resources. Cloud servers can fill this gap by providing huge computational power at a cheap price. However, cloud servers are not fully trusted, thus image owners have legal concerns about the privacy of their private data. In this paper, we proposed and implemented a privacy-preserving CBIR (PP-CBIR) scheme that allows searching and retrieving image databases in a cipher text format. Specifically, we extract aggregated feature vectors to represent the corresponding image collection and employ the asymmetric scalar-product-preserving encryption scheme (ASPE) method to protect these vectors while allowing for similarity computation between these encrypted vectors. To enhance search time, all encrypted features are clustered by the k-means algorithm recursively to construct a tree index. Results show that PP-CBIR has faster indexing and retrieving with good retrieval precision and scalability than previous schemes.
Advancements in internet accessibility and the affordability of digital picture sensors have led to the proliferation of extensive image databases utilized across a multitude of applications. Addressing the semantic gap between low- level attributes and human visual perception has become pivotal in refining Content Based Image Retrieval (CBIR) methodologies, especially within this context. As this field is intensely researched, numerous efficient algorithms for CBIR systems have surfaced, precipitating significant progress in the artificial intelligence field. In this study, we propose employing a hard voting ensemble approach on features derived from three robust deep learning architectures: Inception, Exception, and Mobilenet. This is aimed at bridging the divide between low-level image features and human visual perception. The Euclidean method is adopted to determine the similarity metric between the query image and the features database. The outcome was a noticeable improvement in image retrieval accuracy. We applied our approach to a practical dataset named CBIR 50, which encompasses categories such as mobile phones, cars, cameras, and cats. The effectiveness of our method was thereby validated. Our approach outshone existing CBIR algorithms with superior accuracy (ACC), precision (PREC), recall (REC), and F1-score (F1-S), proving to be a noteworthy addition to the field of CBIR. Our proposed methodology could be potentially extended to various other sectors, including medical imaging and surveillance systems, where image retrieval accuracy is of paramount importance.
Content-Based Image Retrieval (CBIR) is an automatic process of retrieving images that are the most similar to a query image based on their visual content such as colour and texture features. However, CBIR faces the technical challenge known as the semantic gap between high level conceptual meaning and the low-level image based features. This paper presents a new method that addresses the semantic gap issue by exploiting cluster shapes. The method first extracts local colours and textures using Discrete Cosine Transform (DCT) coefficients. The Expectation-Maximization Gaussian Mixture Model (EM/GMM) clustering algorithm is then applied to the local feature vectors to obtain clusters of various shapes. To compare dissimilarity between two images, the method uses a dissimilarity measure based on the principle of Kullback-Leibler divergence to compare pair-wise dissimilarity of cluster shapes. The paper further investigates two respective scenarios when the number of clusters is fixed and adaptively determined according to cluster quality. Experiments are conducted on publicly available WANG and Caltech6 databases. The results demonstrate that the proposed retrieval mechanism based on cluster shapes increases the image discrimination, and when the number of clusters is fixed to a large number, the precision of image retrieval is better than that when the relatively small number of clusters is adaptively determined.