scispace - formally typeset
Search or ask a question

Showing papers on "Content-based image retrieval published in 2020"


Journal ArticleDOI
TL;DR: A content-based image retrieval (CBIR) system has been proposed to extract a feature vector from an image and to effectively retrieve content- based images.
Abstract: Measures of components in digital images are expanded and to locate a specific image in the light of substance from a huge database is sometimes troublesome. In this paper, a content-based image retrieval (CBIR) system has been proposed to extract a feature vector from an image and to effectively retrieve content-based images. In this work, two types of image feature descriptor extraction methods, namely Oriented Fast and Rotated BRIEF (ORB) and scale-invariant feature transform (SIFT) are considered. ORB detector uses a fast key points and descriptor use a BRIEF descriptor. SIFT be used for analysis of images based on various orientation and scale. K-means clustering algorithm is used over both descriptors from which the mean of every cluster is obtained. Locality-preserving projection dimensionality reduction algorithm is used to reduce the dimensions of an image feature vector. At the time of retrieval, the image feature vectors are stored in the image database and matched with testing data feature vector for CBIR. The execution of the proposed work is assessed by utilizing a decision tree, random forest, and MLP classifiers. Two, public databases, namely Wang database and corel database, have been considered for the experimentation work. Combination of ORB and SIFT feature vectors are tested for images in Wang database and corel database which accomplishes a highest precision rate of 99.53% and 86.20% for coral database and Wang database, respectively.

97 citations


Journal ArticleDOI
TL;DR: This paper demonstrates the extraction of vast robust and important features from the images database and the storage of these features in the repository in the form of feature vectors with better precision and recall values compared to other state-of-the-art CBIR systems.
Abstract: Due to the recent technology development, the multimedia complexity is noticeably increased and new research areas are opened relying on similar multimedia content retrieval. Content-based image retrieval (CBIR) systems are used for the retrieval of images related to the Query Image (QI) from huge databases. The CBIR systems available today have confined efficiency as they extract only limited feature sets. This paper demonstrates the extraction of vast robust and important features from the images database and the storage of these features in the repository in the form of feature vectors. The feature repository contains color signature, the shape features and texture features. Here, features are extracted from specific QI. Accordingly, an innovative similarity evaluation with a metaheuristic algorithm (genetic algorithm with simulating annealing) has been attained between the QI features and those belonging to the database images. For an image entered as QI from a database, the distance metrics are used to search the related images, which is the main idea of CBIR. The proposed CBIR techniques are described and constructed based on RGB color with neutrosophic clustering algorithm and Canny edge method to extract shape features, YCbCr color with discrete wavelet transform and Canny edge histogram to extract color features, and gray-level co-occurrence matrix to extract texture features. The combination of these methods increases the image retrieval framework performance for content-based retrieval. Furthermore, the results’ precision–recall value is calculated to evaluate the system’s efficiency. The CBIR system proposed demonstrates better precision and recall values compared to other state-of-the-art CBIR systems.

96 citations


Journal ArticleDOI
TL;DR: Zhang et al. as mentioned in this paper analyzed the three core issues of remote sensing image retrieval and provided a comprehensive review on existing methods, focusing on the feature extraction issue and how to use powerful deep representations to address this task.
Abstract: Remote sensing (RS) image retrieval is of great significant for geological information mining. Over the past two decades, a large amount of research on this task has been carried out, which mainly focuses on the following three core issues: feature extraction, similarity metric, and relevance feedback. Due to the complexity and multiformity of ground objects in high-resolution remote sensing (HRRS) images, there is still room for improvement in the current retrieval approaches. In this article, we analyze the three core issues of RS image retrieval and provide a comprehensive review on existing methods. Furthermore, for the goal to advance the state-of-the-art in HRRS image retrieval, we focus on the feature extraction issue and delve how to use powerful deep representations to address this task. We conduct systematic investigation on evaluating correlative factors that may affect the performance of deep features. By optimizing each factor, we acquire remarkable retrieval results on publicly available HRRS datasets. Finally, we explain the experimental phenomenon in detail and draw conclusions according to our analysis. Our work can serve as a guiding role for the research of content-based RS image retrieval.

95 citations


Journal ArticleDOI
TL;DR: A method that uses phylogenetic diversity indexes to characterize images for creating a model to classify histopathological breast images into four classes - invasive carcinoma, in situ carcinomas, normal tissue, and benign lesion is proposed.

67 citations


Journal ArticleDOI
TL;DR: A new Content-Based Image Retrieval technique to fuse the color and texture features to extract local features as a feature vector and outperform the existing research in terms of average precision and recall values is presented.
Abstract: Due to recent development in technology, the complexity of multimedia is significantly increased and the retrieval of similar multimedia content is an open research problem. In the service of multimedia service, the requirement of Multimedia Indexing Technology is increasing to retrieve and search for interesting data from huge Internet. Since the traditional retrieval method, which is using textual index, has limitation to handle the multimedia data in current Internet, alternatively, the more efficient representation method is needed. Content-Based Image Retrieval (CBIR) is a process that provides a framework for image search and low-level visual features are commonly used to retrieve the images from the image database. The basic requirement in any image retrieval process is to sort the images with a close similarity in term of visual appearance. The color, shape, and texture are the examples of low-level image features. The feature combination that is also known as feature fusion is applied in CBIR to increase the performance, a single feature is not robust to the transformations that are in the image datasets. This paper represents a new Content-Based Image Retrieval (CBIR) technique to fuse the color and texture features to extract local features as our feature vector. The features are created for each image and stored as a feature vector in the database. The proposed research is divided into three phases that feature extraction, similarities match, and performance evaluation. Color Moments (CM) are used for Color features and extract the Texture features, used Gabor Wavelet and Discrete Wavelet transform. To enhance the power of feature vector representation, Color and Edge Directivity Descriptor (CEDD) is also included in the feature vector. We selected this combination, as these features are reported intuitive, compact and robust for image representation. We evaluated the performance of our proposed research by using the Corel, Corel-1500, and Ground Truth (GT) images dataset. The average precision and recall measures are used to evaluate the performance of the proposed research. The proposed approach is efficient in term of feature extraction and the efficiency and effectiveness of the proposed research outperform the existing research in term of average precision and recall values.

45 citations


Journal ArticleDOI
TL;DR: Experimental results show that BiCBIR system outperforms the available state-of-the-art image retrieval systems.
Abstract: Large amount of multi-media content, generated by various image capturing devices, is shared and downloaded by millions of users across the globe, every second. High computation cost is inured in providing visually similar results to the user’s query. Annotation based image retrieval is not efficient since annotations vary in terms of languages while pixel wise matching of images is not preferred since the orientation, scale, image capturing style, angle, storage pattern etc. bring huge amount of variations in the images. Content Based Image Retrieval (CBIR) system is frequently used in such cases since it computes similarity between query image and images of reference dataset efficiently. A Bi-layer Content Based Image Retrieval (BiCBIR) system has been proposed in this paper which consists of two modules: first module extracts the features of dataset images in terms of color, texture and shape. Second module consists of two layers: initially all images are compared with query image for shape and texture feature space and indexes of M most similar images to the query image are retrieved. Next, M images retrieved from previous layer are matched with query image for shape and color feature space and F images similar to the query image are returned as a output. Experimental results show that BiCBIR system outperforms the available state-of-the-art image retrieval systems.

29 citations


Journal ArticleDOI
TL;DR: A CBIR system based on transfer learning from a CNN trained on a vast image database, thus exploiting the generic image representation that it has already learned, and different strategies to exploit the updated CNN for returning a novel set of images that are expected to be relevant to the user’s needs are suggested.
Abstract: Given the great success of Convolutional Neural Network (CNN) for image representation and classification tasks, we argue that Content-Based Image Retrieval (CBIR) systems could also leverage on CNN capabilities, mainly when Relevance Feedback (RF) mechanisms are employed. On the one hand, to improve the performances of CBIRs, that are strictly related to the effectiveness of the descriptors used to represent an image, as they aim at providing the user with images similar to an initial query image. On the other hand, to reduce the semantic gap between the similarity perceived by the user and the similarity computed by the machine, by exploiting an RF mechanism where the user labels the returned images as being relevant or not concerning her interests. Consequently, in this work, we propose a CBIR system based on transfer learning from a CNN trained on a vast image database, thus exploiting the generic image representation that it has already learned. Then, the pre-trained CNN is also fine-tuned exploiting the RF supplied by the user to reduce the semantic gap. In particular, after the user’s feedback, we propose to tune and then re-train the CNN according to the labelled set of relevant and non-relevant images. Then, we suggest different strategies to exploit the updated CNN for returning a novel set of images that are expected to be relevant to the user’s needs. Experimental results on different data sets show the effectiveness of the proposed mechanisms in improving the representation power of the CNN with respect to the user concept of image similarity. Moreover, the pros and cons of the different approaches can be clearly pointed out, thus providing clear guidelines for the implementation in production environments.

28 citations


Journal ArticleDOI
TL;DR: A novel CBIR system that achieves a coarse-to-fine progressive RS image description and retrieval in the partially decoded Joint Photographic Experts Group (JPEG) 2000 compressed domain and experimental results point out that the proposed system is much faster while providing a similar retrieval accuracy than the standard CBIR systems.
Abstract: Due to the dramatically increased volume of remote sensing (RS) image archives, images are usually stored in a compressed format to reduce the storage size. Existing content-based RS image retrieval (CBIR) systems require as input fully decoded images, thus resulting in a computationally demanding task in the case of large-scale CBIR problems. To overcome this limitation, in this article, we present a novel CBIR system that achieves a coarse-to-fine progressive RS image description and retrieval in the partially decoded Joint Photographic Experts Group (JPEG) 2000 compressed domain. The proposed system initially: 1) decodes the code blocks associated only to the coarse wavelet resolution and 2) discards the most irrelevant images to the query image based on the similarities computed on the coarse resolution wavelet features of the query and archive images. Then, the code blocks associated with the subsequent resolution of the remaining images are decoded and the most irrelevant images are discarded by computing similarities considering the image features associated with both resolutions. This is achieved by using the pyramid match kernel similarity measure that assigns higher weights to the features associated with the finer wavelet resolution than to those related to the coarse wavelet resolution. These processes are iterated until the codestreams associated with the highest wavelet resolution are decoded. Then, the final retrieval is performed on a very small set of completely decoded images. Experimental results obtained on two benchmark archives of aerial images point out that the proposed system is much faster while providing a similar retrieval accuracy than the standard CBIR systems.

25 citations



Journal ArticleDOI
TL;DR: The algorithm introduced outperforms the existing techniques improving the segmentation ratio and recognition accuracy of lung cancer which can be validated using experimental analysis.
Abstract: The past few decades have witnessed a steep increase in image data analysis for lung cancer, leading to huge repositories in the area of research in the medical sector. Content Based medical Image Retrieval (CBMIR) methods for lung cancer have been tried with the objective of facilitating access to image data. Many research works have been developed in content based medical image retrieval. But the techniques have the drawback of low efficiency and high computation cost. Image segmentation, extraction and classification methods of various kinds was taken upusing traditional methodswhich involves extraction of a specific region of interest and given to medical experts for diagnosis. The extracted region of interest region provides information useful for the for diagnosis of the disease. But the segmentation methods have some limitations such as flat valleys, noise sensitive and computational expensive which lead to reduction in the entire system performance. This is addressed by animproved watershed histogram thresholding using the probabilistic neural networks (IWHT-PNN) approach. The algorithm introduced outperforms the existing techniques improving the segmentation ratio and recognition accuracy of lung cancer which can be validated using experimental analysis.

25 citations


Journal ArticleDOI
TL;DR: This paper proposes to use features derived from pre-trained network models from a deep-learning convolution network trained for a large image classification problem to produce vastly superior results for a variety of databases, and it outperforms many contemporary CBIR systems.
Abstract: In a Content Based Image Retrieval (CBIR) System, the task is to retrieve similar images from a large database given a query image. The usual procedure is to extract some useful features from the query image, and retrieve images which have similar set of features. For this purpose, a suitable similarity measure is chosen, and images with high similarity scores are retrieved. Naturally the choice of these features play a very important role in the success of this system, and high level features are required to reduce the semantic gap. In this paper, we propose to use features derived from pre-trained network models from a deep-learning convolution network trained for a large image classification problem. This approach appears to produce vastly superior results for a variety of databases, and it outperforms many contemporary CBIR systems. We analyse the retrieval time of the method, and also propose a pre-clustering of the database based on the above-mentioned features which yields comparable results in a much shorter time in most of the cases.

Journal ArticleDOI
TL;DR: Wang et al. as mentioned in this paper proposed a similarity search for encrypted images in secure cloud computing (called SEI), where the feature descriptors extracted by the convolutional neural network (CNN) model are used to improve search accuracy.
Abstract: With the emergence of intelligent terminals, the Content-Based Image Retrieval (CBIR) technique has attracted much attention from many areas (i.e., cloud computing, social networking services, etc.). Although existing privacy-preserving CBIR schemes can guarantee image privacy while supporting image retrieval, these schemes still have inherent defects (i.e., low search accuracy, low search efficiency, key leakage, etc.). To address these challenging issues, in this paper we provide a similarity Search for Encrypted Images in secure cloud computing (called SEI). First, the feature descriptors extracted by the Convolutional Neural Network (CNN) model are used to improve search accuracy. Next, an encrypted hierarchical index tree by using K-means clustering based on Affinity Propagation (AP) clustering is devised, which can improve search efficiency. Then, a limited key-leakage k-Nearest Neighbor (kNN) algorithm is proposed to protect key from being completely leaked to untrusted image users. Finally, SEI is extended to further prevent image users' search information from being exposed to the cloud server. Our formal security analysis proves that SEI can protect image privacy as well as key privacy. Our empirical experiments using a real-world dataset illustrate the higher search accuracy and efficiency of SEI.

Journal ArticleDOI
TL;DR: A new multi-level colored directional motif histogram (MLCDMH) for devising a content-based image retrieval scheme which reflects the local structural and directional motif features in detail and also reduces the computation overhead.
Abstract: Color features and local geometrical structures are the two basic image features which are sufficient to convey the image semantics. Both of these features show diverse nature on the different regions of a natural image. Traditional local motif patterns are standard tools to emphasize these local visual image features. These motif-based schemes consider either structural orientations or limited directional patterns which are not sufficient to realize the detailed local geometrical properties of an image. To address these issues, we have proposed a new multi-level colored directional motif histogram (MLCDMH) for devising a content-based image retrieval scheme. The proposed scheme extracts local structural features at three different levels. Initially, MLCDMH scheme extracts directional structural patterns from a $$3 \times 3$$ pixel grids of an image. This reflects the $$9^9$$ different structural arrangements using 28 directional patterns. Next, we have used a weighted neighboring similarity (WNS) scheme to exploit the uniqueness of each motif pixel in its local surrounding. The WNS scheme will compute the importance of each directional motif pattern in its $$3 \times 3$$ local neighborhood. In the last level, we have fused all directional motif images into a single directional difference matrix which reflects the local structural and directional motif features in detail and also reduces the computation overhead. The MLCDMH considers all possible permutations and rotations of the motif patterns to generate rotational invariant structural features. The image retrieval performance of this proposed scheme has been evaluated using different Corel/natural, object, texture and heterogeneous image datasets. The results of the retrieval experiments have shown satisfactory improvement over other motif- and non-motif-based CBIR approaches.

Journal ArticleDOI
TL;DR: The proposed scheme is shown to outperform conventional privacy-preserving CBIR schemes including state-of-the-art ones in terms of mean average precision (mAP) scores.
Abstract: In this article, we propose a novel content-based image-retrieval (CBIR) scheme using compressible encrypted images called “encryption-then-compression (EtC) images.” The proposed scheme allows us not only to directly retrieve images from visually protected images but to also apply EtC images that can be compressed by using the JPEG standard for the first time. In addition, the sensitive management of secret keys is not required in our framework. The proposed retrieval scheme is carried out on the basis of weighted searching images with MPEG-7-powered localized descriptors (weighted SIMPLE descriptors) combining scalable color descriptor (SCD) or color and edge directivity descriptor (CEDD). Weighted SIMPLE descriptors are extended, and CEDD is also modified to avoid the influence of image encryption. In an experiment, the proposed scheme is demonstrated to have almost no degradation in retrieval performance compared with conventional content-based retrieval methods with plain images under the use of two datasets. In addition, the proposed scheme is shown to outperform conventional privacy-preserving CBIR schemes including state-of-the-art ones in terms of mean average precision (mAP) scores.

Journal ArticleDOI
TL;DR: A novel multiple query retrieval approach, named weight-learner, which relies on visual feature discrimination to estimate the distances between the query images and images in the database, and aims to produce optimal feature relevance weights with respect to the user query.
Abstract: We propose a novel multiple query retrieval approach, named weight-learner, which relies on visual feature discrimination to estimate the distances between the query images and images in the database. For each query image, this discrimination consists of learning, in an unsupervised manner, the optimal relevance weight for each visual feature/descriptor. These feature relevance weights are designed to reduce the semantic gap between the extracted visual features and the user’s high-level semantics. We mathematically formulate the proposed solution through the minimization of some objective functions. This optimization aims to produce optimal feature relevance weights with respect to the user query. The proposed approach is assessed using an image collection from the Corel database.

Journal ArticleDOI
Guangyi Xie1, Baolong Guo1, Zhe Huang1, Yan Zheng1, Yunyi Yan1 
TL;DR: The proposed method first applies the texton template to detect and extract the consistent zone of an image, and calculates the dominant color descriptor feature on the pixels in this consistent zone, and the combination of the dominant dolor descriptor and the Hu moments is used for content-based image retrieval.
Abstract: The rapidly increasing number of digital images requires effective retrieval. Meanwhile, the dominant color descriptor has been widely used in image processing. Due to the influence of lighting and other factors, the same color in nature may have some different changes. The human eye is usually more sensitive to zones of consistent color, often identifying objects by zones of consistency. Therefore, the proposed method in this paper first applies the texton template to detect and extract the consistent zone of an image, and calculates the dominant color descriptor feature on the pixels in this consistent zone. Besides, the translation and rotation invariance of the Hu moments feature is applied to extract the shape information in the same consistent zone of the image. Finally, the combination of the dominant dolor descriptor and the Hu moments is used for content-based image retrieval. The algorithm proposed in this paper is tested on three data sets: Corel-1k, Corel-5k and Corel-10k, and the experimental results show that it is superior to the current content-based image retrieval methods.

Journal ArticleDOI
TL;DR: This article presents a SIARS using deep learning (DL) and multiple share creation schemes, which involves Adagrad based convolutional neural network (AG-CNN) based feature extractor to extract the useful set of features from the input images.
Abstract: Due to the advanced growth in multimedia data and Cloud Computing (CC), Secure Image Archival and Retrieval System (SIARS) on cloud has gained more interest in recent times. Content based image retrieval (CBIR) systems generally retrieve the images relevant to the query image (QI) from massive databases. However, the secure image retrieval process is needed to ensure data confidentiality and secure data transmission between cloud storage and users. Existing secure image retrieval models faces difficulties like minimum retrieval performance, which fails to adapt with the large-scale IR in cloud platform. To resolve this issue, this article presents a SIARS using deep learning (DL) and multiple share creation schemes. The proposed SIARS model involves Adagrad based convolutional neural network (AG-CNN) based feature extractor to extract the useful set of features from the input images. At the same time, secure multiple share creation (SMSC) schemes are executed to generate multiple shares of the input images. The resultant shares and the feature vectors are stored in the cloud database with the respective image identification number. Upon retrieval, the user provides a query image and reconstructs the received shared image to attain the related images from the database. An elaborate experimentation analysis is carried out on benchmark Corel10K dataset and the results are examined in terms of retrieval efficiency and image quality. The attained results ensured the superior performance of the SIARS model on all the applied test images.

Proceedings ArticleDOI
10 Jun 2020
TL;DR: In this conference paper, a content-based image retrieval system that uses an innovative type of neural network known as autoencoder is discussed and a basic system to understand it is developed.
Abstract: Image retrieval technology is a very fast-growing digital technology for researchers in the field of computer science from a very long period. It is a system for retrieving digital images from a large database. The well-known organizations that are using this system are Google and Pinterest. In this conference paper, a content-based image retrieval system that uses an innovative type of neural network known as autoencoder is discussed and developed a basic system to understand it. The methodology that has been used is an unsupervised method which is a machine learning algorithm in which the system retrieves images without searching about its name, labels, and tags. This system retrieves images just by its visual information. This approach of image retrieval is known as Content-Based Image Retrieval (CBIR).

Journal ArticleDOI
01 Feb 2020
TL;DR: A new variant of multi-trend structure descriptor (MTSD) is contributed for efficient content based image retrieval and achieves the state-of-the-art performance for natural, textural and biomedical image retrieval.
Abstract: This paper contributes a new variant of multi-trend structure descriptor (MTSD) for efficient content based image retrieval. The proposed variant of MTSD encodes color/edge orientation/texture quantized values versus orientation of equal, small and large trends instead of color/edge orientation/texture quantized values versus equal, small and large trends. In addition, it also encodes color/edge orientation/texture quantized values versus average location of distribution of pixel values for equal, small and large trends at each orientation. To reduce the time cost of the proposed variant of MTSD with the preservation of its accuracy, the image is decomposed into fine level using discrete Haar wavelet transform and the fine level for the decomposition of an image is determined empirically. Comprehensive experiments are conducted using the benchmark Corel-1k, Corel-5k, Corel-10k, Caltech-101, LIDC-IDRI-CT, VIA/I-ELCAP-CT and OASIS-MRI image datasets and the results evident that the proposed variant of MTSD achieves the state-of-the-art performance for natural, textural and biomedical image retrieval. Precision and recall are the measures used to measure the accuracy. Euclidean similarity measure is used to calculate the similarity information between query and target images.

Journal ArticleDOI
TL;DR: The proposed method uses sparse complementary features for vigorous image representation, optimal feature selection based on locality-preserving projection, fuzzy c-means clustering, and soft label support vector machine for robust image classification for effective and efficient CBIR.
Abstract: Content-based image retrieval (CBIR) states the procedure of recovering images having similar visual content against a query image from image datasets. In CBIR, the selection of redundant and irrelevant features from images results in the semantic gap issue, which occurs during feature representation and machine learning process. The robust image representation for effective and efficient image retrieval mainly depends upon robust feature selection and classification, which also reduces the semantic gap problem of CBIR. This paper proposed an innovative method for effective and efficient CBIR. The method uses sparse complementary features for vigorous image representation, optimal feature selection based on locality-preserving projection, fuzzy c-means clustering, and soft label support vector machine for robust image classification. In CBIR, smaller and larger sizes of codebook improve the recall and precision (accuracy) of the system, respectively. Due to this reason, the proposed method introduces complementary features based on a larger size codebook, which is assembled using two small sizes of codebooks to increase CBIR performance. The three well-known image datasets (i.e. Corel-1000, Corel-1500, and Holidays) are used to assess the performance of the proposed method. The experimental evaluation highlights promising results as compared to recent methods of CBIR.

Journal ArticleDOI
TL;DR: A new combination of descriptors for the effective retrieval of common imaging signs, which play a significant role in the identification of cancerous lung nodules and many other lung diseases, is proposed.
Abstract: The main problem in content-based image retrieval (CBIR) systems is the semantic gap which needs to be reduced for efficient retrieval. The common imaging signs (CISs) which appear in the patient’s lung CT scan play a significant role in the identification of cancerous lung nodules and many other lung diseases. In this paper, we propose a new combination of descriptors for the effective retrieval of these imaging signs. First, we construct a feature database by combining local ternary pattern (LTP), local phase quantization (LPQ), and discrete wavelet transform. Next, joint mutual information (JMI)–based feature selection is deployed to reduce the redundancy and to select an optimal feature set for CISs retrieval. To this end, similarity measurement is performed by combining visual and semantic information in equal proportion to construct a balanced graph and the shortest path is computed for learning contextual similarity to obtain final similarity between each query and database image. The proposed system is evaluated on a publicly available database of lung CT imaging signs (LISS), and results are retrieved based on visual feature similarity comparison and graph-based similarity comparison. The proposed system achieves a mean average precision (MAP) of 60% and 0.48 AUC of precision-recall (P-R) graph using only visual features similarity comparison. These results further improve on graph-based similarity measure with a MAP of 70% and 0.58 AUC which shows the superiority of our proposed scheme.

Journal ArticleDOI
TL;DR: This contribution presents a novel features detector by locating the interest points by applying non-maximum suppression to productive sum of derivative of pixels computed from differential of corner scores and has significant precision rates, recall rates, average retrieval precision & recall, mean average precision & recalls for many image semantic groups of the challenging datasets.
Abstract: Intelligent and efficient image retrieval from versatile image datasets is an inevitable requirement of the current era. Primitive image signatures are vital to reflect the visual attributes for content based image retrieval (CBIR). Algorithmically descriptive and well identified visual contents form the image signatures to correctly index and retrieve similar results. Hence feature vectors should contain ample image information with color, shape, objects, spatial information perspectives to distinguish image category as a qualifying candidate. This contribution presents a novel features detector by locating the interest points by applying non-maximum suppression to productive sum of derivative of pixels computed from differential of corner scores. The interest points are described by applying scale space interpolation to scale space division produced from Hessian blob detector resulted after Gaussian smoothing. The computed shape and object information is fused with color features extracted from the spatially arranged L2 normalized coefficients. High variance coefficients are selected for object based feature vectors to reduce the massive data which in fuse form transformed to bag-of-words (BoW) for efficient retrieval and ranking. To check the competitiveness of the presented approach it is experimented on nine well-known image datasets Caltech-101, ImageNet, Corel-10000, 17-Flowers, Columbia object image library (COIL), Corel-1000, Caltech-256, tropical fruits and Amsterdam library of textures (ALOT) belong to shape, color, texture, and spatial & complex objects categories. Extensive experimentation is conducted for seven benchmark descriptors including maximally stable extremal region (MSER), speeded up robust features (SURF), difference of Gaussian (DoG), red green blue local binary pattern (RGBLBP), histogram of oriented gradients (HOG), scale invariant feature transform (SIFT), and local binary pattern (LBP). Remarkable outcomes reported that the presented technique has significant precision rates, recall rates, average retrieval precision & recall, mean average precision & recall rates for many image semantic groups of the challenging datasets. Results comparison is presented with research techniques and reported improved results.

Journal ArticleDOI
13 Apr 2020-Symmetry
TL;DR: 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.
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.

Journal ArticleDOI
TL;DR: This paper aims to categorize and evaluate the existing BoVW model based formulations for the task of content based image retrieval, and proposes certain promising directions for future research on the basis of the existing models and the demand from real-world.
Abstract: The exponential growth of digital image data poses numerous open problems to computer vision researchers. In this regard, designing an efficient and more accurate mechanism that finds and retrieve desired images from large repositories is of greater importance. To this end, various types of content based image retrieval (CBIR) systems have been developed. A typical CBIR system enables the search and retrieval of desired images from large databases that are similar to a given query image by means of automatically extracted visual features from image pixels. In CBIR domain, the bag of visual words (BoVW) model is one of the most widely used feature representation scheme and there exist a number of image retrieval frameworks based on BoVW model. It has been observed that most of them demonstrated promising results for the task of medium and large scale image retrieval. However, image retrieval literature lacks a comparative evaluation of these extended BoVW formulations. To this end, this paper aims to categorize and evaluate the existing BoVW model based formulations for the task of content based image retrieval. The commonly used datasets and the evaluation metrics to assess the retrieval effectiveness of these existing models are discussed. Moreover, quantitative evaluation of state of the art image retrieval systems based on BoVW model is also provided. Finally, certain promising directions for future research are proposed on the basis of the existing models and the demand from real-world.

Journal ArticleDOI
01 Sep 2020
TL;DR: CBIR algorithm is proposed with fusion of color and texture descriptors such as binary Gabor pattern (BGP), segmentation based fractal texture analysis (SFTA), edge histogram descriptor (EHD), Gabor wavelet texture features and fuzzy histogram-based descriptors.
Abstract: In communication, medium images are used for various applications such as social websites, education, bio-medical field and industrial applications. Indexing and retrieval of such large image database pose a big problem. The content-based image retrieval (CBIR) approaches are used to select information from the input images using feature descriptors. In this article, CBIR algorithm is proposed with fusion of color and texture descriptors such as binary Gabor pattern (BGP), segmentation based fractal texture analysis (SFTA), edge histogram descriptor (EHD), Gabor wavelet texture features and fuzzy histogram-based descriptors. Laplacian score has employed to reduce feature vector dimensions. Retrieval rate is computed using different distance metrics. Experimentation was performed with ten different classes of 1000 images using Wang database. The simulation results indicate improvement in terms of precision compared with the existing techniques.

Journal ArticleDOI
TL;DR: This research paper used a set of lung computed tomography scanned images as inputs, obtained from lung image archives and applied image processing techniques such as feature extraction, segmentation and Content-based image retrieval technique to compare lung image features.
Abstract: Cancer is a sickness brought about by an uncontrolled division of eccentric cells in any part of human body. It is in the top of few places in the killer disease list and pervades in the entire world, but still on the rise. Most of the cases an early detection of lung cancer is cumbersome. This research paper is aimed to present an effective and an efficient way of computer-assisted detection method for lung cancer. In this research we used a set of lung computed tomography scanned images as inputs, obtained from lung image archives and applied image processing techniques such as feature extraction, segmentation. In this approach, a proper combination of Adaptive thresholding segmentation algorithm has been used for segmenting input images, a well-known Support Vector Machine image classification algorithm has been used for lung tumor classification and Content-based image retrieval technique has been used to compare lung image features such as contract, intensity, texture and shape. A set of patient personal data is included to get more accurate and correct prediction results, and it is dealt with data mining approach. The proposed segmentation method shows improved prediction results.

Journal ArticleDOI
01 Feb 2020
TL;DR: The experimental results demonstrate that the proposed approach works well for CBIR and can classify specific types of material surfaces in images with a reasonably high level of accuracy as well as outperform other existing algorithms.
Abstract: Image retrieval is the process of searching for digital images from a large database. There exist two distinctive research groups, which employ the content‐based and description‐based approaches, respectively. However, research in the content‐based domain is currently dominant in the field, while the other approaches are not as widely utilized. Although there are a number of different techniques that are available for image retrieval, the development of more effective methods is still necessary. In recent years, previous research has shown that biologically inspired metaheuristic algorithms have great potential for use in solving problems in many science and engineering domains. The artificial bee colony (ABC) algorithm is one of the more promising biologically inspired metaheuristic approaches used to find optimal solutions as it has the advantages of convenient implementation and efficient performance. In this article, a new efficient method based on a combination of the gray‐level cooccurrence matrix (GLCM) with the ABC, referred to as “GLCM‐ABC,” is proposed for use in content‐based image retrieval (CBIR). The experimental results demonstrate that the proposed approach works well for CBIR and can classify specific types of material surfaces in images with a reasonably high level of accuracy as well as outperform other existing algorithms.

Journal ArticleDOI
TL;DR: This work is the first one that developing scheme under the assumption that all the entities involved in privacy-preserving image retrieval system are semi-trusted and ensures the confidentiality of key and privacy of query information at the same time.
Abstract: Owing to the rapid development of cloud services and personal privacy demand, secure cloud storage services and search over encrypted datasets have become an important issue Recently, the leaking of images such as identification and driver's licenses catches much attention The trend towards secure computation has been widely discussed, especially asymmetric scalar-product-preserving encryption (ASPE) and homomorphic encryption (HE) Although ASPE have ability to encrypt and determine the similarity between ciphertexts efficiently, it is not a practical methodology due to its assumption that the users are fully trusted in real world and it also may have key leakage problem Contrary to ASPE, HE can execute addition and multiplication in the encrypted domain and solve key leakage problem Hence, in this paper, we combine the opinions of HE and ASPE to propose new privacy-preserving content-based image retrieval with key confidentiality scheme against the attacks from data owner, cloud server and users Our privacy preserving image retrieval scheme is developed under strong threat model that is close to real world Furthermore, to the best of our knowledge, our work is the first one that developing scheme under the assumption that all the entities involved in privacy-preserving image retrieval system are semi-trusted Our scheme ensures the confidentiality of key and privacy of query information at the same time In addition, we provide a lightweight verification to check whether the search query is fake or not Finally, the experiment results show that the computation overheads and search precision are acceptable at the same time

Proceedings ArticleDOI
28 Feb 2020
TL;DR: The tentative effects illustrate that the proposed model derived enhanced retrieval accurateness at the optimum level as well as significantly more effective and secure data from the Mobile Ad-hoc Network (MANET).
Abstract: In the current era, content based image retrieval based on pattern recognition and classification using machine learning paradigm is an innovative way. In order to retrieve high resolution security for satellite images Support Vector Machine (SVM) a machine learning paradigm is helpful for learning process and for pattern recognition and classification; ensemble methods give better machine learning results. In this paper, SVM based on random subspace and boosting ensemble learning is proposed for very high resolution security for satellite image retrieval and secure from the Mobile Ad-hoc Network (MANET). To taught SVM ensemble model is use to recognize the images that almost all similar informative for dynamic learning. Bias-weighting systems implement to direct the ensemble model to pay a lot of attention on the positive illustration than the negative ones. The UC Merced security for satellite image from dataset is used for experimental work. Accuracy and error rate are found to be precise. The tentative effects illustrate that the proposed model derived enhanced retrieval accurateness at the optimum level as well as significantly more effective and secure data from the MANET. The comparisons for the existing approaches are evaluated by using precision and recall measurements.

Proceedings ArticleDOI
01 Feb 2020
TL;DR: A framework for content based fine-grained image retrieval (CB-FGIR) by using CNN is proposed and achieves superior retrieval results than other handcrafted and state of art methods.
Abstract: Content based image retrieval (CBIR) is the problem of retrieving similar images from a database to a given query by use of its visual information only. It has been a hot topic for years. Current CBIR methods rely on the fact that the database consists of large inter-class variance but in real scenario a user for example wants to retrieve same sub-category images to the query, in that case inter-class variance is quite small. Retrieving similar images from database of small inter-class variance is quite difficult from that of large inter-class variance. Convolutional neural networks (CNN) has shown tremendous results in image tasks such as classification, detection, retrieval, segmentation and more. In this paper we proposed a framework for content based fine-grained image retrieval (CB-FGIR) by using CNN. Oxford flower-17 dataset is used to test the proposed framework. Five splits of the dataset is used to evaluate the CB-FGIR framework and achieves superior retrieval results than other handcrafted and state of art methods Keywords— CNN, Retrieval, CB-FGIR, Deep learning.