A Survey on Content-based Visual Information Retrieval

The combination of convolution neural networks (CNN) and deep learning generated a stunning result in a variety of image processing applications. For separating comparable images, CNN-based techniques to isolate image features from the last layer and the use of a single CNN structure could be used. The Content-Based Image Retrieval system is used to learn highlight extraction and efficient similarity examination (CBIR). Highlight extraction, like similarity tests, plays an important role in CBIR. The research is carried out using datasets such as the UC Merced Land Use Dataset. Using a pre-trained model that has been adjusted for the retrieval task and has been trained on a large number of photographs. For the retrieval process, pre-trained CNN models are used to generate image highlight descriptors. By using move learning and retrieval of highlight vectors using various similarity measures, this technique manages component extraction from the two completely connected layers present in the VGG-16 network. The proposed engineering has a fantastic presentation in terms of extracting features as well as learning features without any prior knowledge of the images. With vgg19 , we're going to be able to extend our work with CNN. Investigation of configuration and bunching for signs of improved execution. The outcomes were measured and execution correlation was completed using various execution metrics. In both completely connected layers, cosine similarity and Euclidean distance work better.

International Journal of Reconfigurable and Embedded Systems (IJRES), 2023

Content-based image retrieval (CBIR) uses the content features for retrieving and searching the images in a given large database. Earlier, different hand feature descriptor designs are researched based on cues that are visual such as shape, colour, and texture used to represent these images. Although, deep learning technologies have widely been applied as an alternative to designing engineering that is dominant for over a decade. The features are automatically learnt through the data. This research work proposes integrated dual deep convolutional neural network (IDD-CNN), IDD-CNN comprises two distinctive CNN, first CNN exploits the features and further custom CNN is designed for exploiting the custom features. Moreover, a novel directed graph is designed that comprises the two blocks i.e. learning block and memory block which helps in finding the similarity among images; since this research considers the large dataset, an optimal strategy is introduced for compact features. Moreover, IDD-CNN is evaluated considering the two distinctive benchmark datasets the oxford dataset considering mean average precision (mAP) metrics and comparative analysis shows IDD-CNN outperforms the other existing model.

With the development of the Internet, and the availability of image capturing devices such as image scanners, the size of digital image collection is rising quickly. A well-organized image searching, browsing and retrieval tools are essential for users from different domains, including remote sensing, crime prevention, publishing, medicine, architecture, etc. Therefore, several general purpose image retrieval systems have been exploited. In CBIR, images are indexed by their visual content. Content based image retrieval consists of two parts: feature extraction/indexing and retrieval part. The techniques which are used to extract features of an image are called feature extraction techniques. The choice of features plays a significant role in image retrieval. Some of the features used are color, texture and shape. Combination of these features provides better performance than single feature. Here we have proposed a combination of statistical features and deep neural network which has improved the performance of the system which is shown in the results.

SN Computer Science

Advancements in the sector of computer and multimedia technology and introduction of the World Wide Web have increased the volume of image databases and collections, for example medical imageries, digital libraries, art galleries which in total contain millions of images. The retrieval process of images from such huge database by traditional methods such as Text Based Image Retrieval, Color Histogram and Chi Square Distance may take a lot of time to get the desired images. It is necessity to develop an effective image retrieval system which can handle these huge amounts of images at once. The main purpose is to build a robust system that builds, executes and responds to data in an efficient manner. A Content-Based Image Retrieval (CBIR) system has been developed as an efficient image retrieval tool where user can provide their query to the system to allow it to retrieve user's desired image from the image collection. Moreover, the emergence of web development and transmission networks and also the number of images which are available to users continue to grow. We propose an effective deep learning framework based on Convolution Neural Networks (CNN) and Support Vector Machine (SVM) for fast image retrieval. Proposed architecture extracts features using CNN and classification using SVM. The results demonstrate the robustness of the system.

International Journal of Applied Pattern Recognition, 2021

In the last few decades, content-based image retrieval is considered as one of the most vivid research topics in the field of information retrieval. The limitation of current content-based image retrieval systems is that low-level features are highly ineffective to represent the semantic contents of the image. Most of the research work in content-based image retrieval is focused on bridging the semantic gap between the low-level features and high-level semantic concepts of image. This paper presents a thorough study of different techniques for the reduction of semantic gap. The existing techniques are broadly categorised as: 1) image annotation techniques to define the high-level concepts in image; 2) relevance feedback techniques to integrate user's perception; 3) machine learning and deep learning techniques to associate low-level features with high-level concepts. In addition, the general architecture of semantic-based image retrieval system has been discussed in this survey. This paper also highlights the current and future applications of content-based image retrieval. The paper concludes with promising future research directions.

Indonesian Journal of Electrical Engineering and Computer Science

The image retrieval focuses on finding images that are similar from a dataset that is of a large scale against an image of a query. Earlier, different hand feature descriptor designs are researched based on cues that are visual such as their shape, colour, and texture. used to represent these images. Although, deep learning technologies have widely been applied as an alternative to designing engineering that is dominant for over a decade. The features are automatically learnt through the data. This research work proposes integrated dual deepconvolutional neural networks (IDD-CNN), IDD-CNN comprises two distinctive CNN, first CNN exploits the features and further custom CNN is designed for exploiting the custom features. Moreover, a novel directed graph is designed that comprises the two blocks i.e., learning block and memory block which helps in finding the similarity among images; since this research considers the large dataset, an optimal strategy is introduced for compact feature...

Cluster Computing, 2018

Content-based image retrieval (CBIR) uses image content features to search and retrieve digital images from a large database. A variety of visual feature extraction techniques have been employed to implement the searching purpose. Due to the computation time requirement, some good algorithms are not been used. The retrieval performance of a content-based image retrieval system crucially depends on the feature representation and similarity measurements. The ultimate aim of the proposed method is to provide an efficient algorithm to deal with the above mentioned problem definition. Here the deep belief network (DBN) method of deep learning is used to extract the features and classification and is an emerging research area, because of the generation of large volume of data. The proposed method is tested through simulation in comparison and the results show a huge positive deviation towards its performance.

IEEE Access

In the era of massive data production through the internet and social media, the volume of images generated is immense. Storing and retrieving relevant images efficiently pose significant challenges. Content-based image retrieval (CBIR) has emerged as a prevalent method for retrieving relevant images based on query images from large image collections. CBIR relies on three fundamental elements: the selection, extraction, and representation of features. This paper delves into a comprehensive survey of these crucial aspects. This paper begins by investigating the significance and wide-ranging applications of CBIR. It subsequently delves into an intricate analysis of feature selection, encompassing attributes such as color, texture, shape, and descriptors. Following this, the paper navigates through sections dedicated to feature extraction techniques and their subsequent representation. Furthermore, this paper includes an assessment of recent research articles and the methodologies they employ within the realm of CBIR. Significantly, CBIR has witnessed a notable expansion to incorporate deep learning techniques in recent times. The survey presents an overview of these recent methods and their integration into CBIR frameworks. This paper concludes by offering an extensive outline of 215 articles, encompassing a wide range of analyses conducted within the field of CBIR. Finally, this paper also outlines potential research directions for the future. It sheds light on areas where CBIR can continue to evolve and enhance its capabilities. INDEX TERMS Content-based image retrieval, text-based image retrieval, deep learning, image processing, image retrieval, color feature, texture feature, shape feature, key point descriptors, speed up robust feature.

2019 XXVII International Conference on Information, Communication and Automation Technologies (ICAT), 2019

In this work we will show a novel method for improving the performance of content-based image retrieval using a deep neural network. The main focus of the method is training the distance function using the deep neural network and transfer learning. An existing pretrained network for image classification is used as a basis. One of the method's benefits is the fact that the existing network is not retrained and features for the distance function are the neural network's layers which were trained and stored before. The method is tested with the publicly available VGG19 deep neural network. Obtained results are comparable or in some cases better than the state of the art methods with a similar approach.

IRJET, 2022

The retrieval of useful data from large collections of data has gotten a lot of attention in recent years. There are a variety of search systems available for this purpose, but they should be able to identify the most relevant search results based on the user's query that meet the user's requirements. There are a variety of approaches for retrieving this data. Text materials are frequently regarded by traditional search engines, but information retrieval with respect to images are overlooked in the paper. In most cases, images in HTML web pages are used to find more relevant images by comparing their linguistic and visual contents. Relevant photos may also be found using visual characteristics in standard text-based search engines by entering a textual query. For quick access and retrieval of relevant multimedia material, a variety of methods and search engines are available. Most of them rely on textual data in conjunction with visual material. As a result, this study presents a technique for producing search results based on the features of pictures in web pages.