Image Search Engines Research Articles

A Portable Linux-based Firmware for NVMe Computational Storage Devices

Abstract: Over the years, interest in computational storage devices has been growing steadily. This is largely due to the rise of data-intensive applications, such as machine learning, online video distribution, astrophysics, and genomics. Moving compute operations closer to the data provides benefits in terms of scaling possibilities and energy efficiency. The development of computational storage devices has been limited by the need for specialized and complex hardware. In this work we propose a portable Linux-based firmware framework for the development of NVMe computational storage devices. Our firmware runs on a variety of hardware platforms ranging from expensive FPGA solutions to inexpensive off-the-shelf single board computers. The firmware leverages the vast Linux software ecosystem to facilitate the development and prototyping of novel computational storage devices. We benchmark our firmware on multiple hardware platforms and demonstrate its versatility through several computational examples including a content-aware disk image search engine based on natural language processing and AI-driven image recognition.

The Contemporary Art of Image Search: Iterative User Intent Expansion via Vision-Language Model

Image search is an essential and user-friendly method to explore vast galleries of digital images. However, existing image search methods heavily rely on proximity measurements like tag matching or image similarity, requiring precise user inputs for satisfactory results. To meet the growing demand for a contemporary image search engine that enables accurate comprehension of users' search intentions, we introduce an innovative user intent expansion framework. Our framework leverages visual-language models to parse and compose multi-modal user inputs to provide more accurate and satisfying results. It comprises two-stage processes: 1) a parsing stage that incorporates a language parsing module with large language models to enhance the comprehension of textual inputs, along with a visual parsing module that integrates an interactive segmentation module to swiftly identify detailed visual elements within images; and 2) a logic composition stage that combines multiple user search intents into a unified logic expression for more sophisticated operations in complex searching scenarios. Moreover, the intent expansion framework enables users to perform flexible contextualized interactions with the search results to further specify or adjust their detailed search intents iteratively. We implemented the framework into an image search system for NFT (non-fungible token) search and conducted a user study to evaluate its usability and novel properties. The results indicate that the proposed framework significantly improves users' image search experience. Particularly the parsing and contextualized interactions prove useful in allowing users to express their search intents more accurately and engage in a more enjoyable iterative search experience.

Gauging the Google gaze

This study explores the visual representation of Great Yarmouth, a British coastal town caught between the urban and the rural, as seen through the quasi-monopolistic image search engine Google Images. The research examines levels of pluralistic or biased place representations to consider how rankings employed by Google Images algorithms represent Great Yarmouth’s identity. The study adopts a visual culture perspective that recognises the role of images in place making and combines digital methods with an image type analysis to investigate how online representations reflect and create the town’s identities. The data shows that Google Images’ preference for representing Yarmouth as a sunny seaside town indicates that the search engine prioritises marketable assets above its connections with its hinterland, its diversity of people, and the cultural activities it has to offer. This, the authors state, is a place far away from Tuan’s (1979) idea of a place that is given meaning and identity from the perspective of people. Instead, Google Images’ representations of Great Yarmouth are an example of a created form of place making as commodification. The article concludes that the inscribed bias and unbalanced search priority criteria employed by the search engine impact upon the diversity of the semi-peripheral town.

Deep Learning based Smart Image Search Engine

Abstract: This paper introduces a new reverse search engine integration into content-based image retrieval (CBIR) systems that employs convolutional neural networks (CNNs) for feature extraction. It generates global descriptors using pre-trained CNN architectures such as ResNet50, InceptionV3, and InceptionResNetV2. It retrieves visually similar images without depending on linguistic annotations. Comparative analysis against existing methods, such as Gabor Wavelet, CNN-SVM, Metaheuristic Algorithm, etc., has been tested, and it proves the superiority of the proposed algorithm, the Cartoon Texture Algorithm, in CBIR. As the Internet sees an exponential growth of different data types, the importance of CBIR continues to grow. In order to efficiently retrieve images, solely relying on image features while ignoring metadata is exactly what we need. As such, this paper is a reminder of the need for CBIR in this changing world. They showed that CBIR continues to be quite effective in the age of the Internet. Their proposed model for CBIR, which integrates ResNet-50-based feature extraction, a neural network model trained on different image datasets, and clustering techniques to make retrieval fast, provides a significant improvement in accuracy and efficiency for content-dependent image retrieval. This methodology is likely to be very useful as we work with the increasingly huge data of vision and beyond on the Internet. It provides a good basis for an effective image search and retrieval system

Improving the Precision of Image Search Engines with the Psychological Intention Diagram

With the increase in the amount of images online, the whole Internet is becoming an image database. Since there are so many available images, it is difficult for users to find the desired images. Unlike text search engines, image search engines cannot fully recognize the visual meaning of an image. In addition, it is difficult to obtain the desired images from the keywords provided by the user, since a keyword may contain multiple meanings. To solve these problems, this paper proposes a psychological intention diagram of past users, if inquiring using a keyword, to predict the images that these users want. Based upon the novel psychological diagram, this paper proposes a search engine that analyzes images in the sequential probing of the current user if he/she inquires after the same keywords as previous users. Moreover, this paper also constructs a psychological intention diagram of the designers of the web pages containing the keyword. This type of psychological intention diagram is used when a query is not issued by past users. To the best of our knowledge, this paper is the first one considering the psychological viewpoint of users and web designers in guiding the retrieval of the search engine. The experimental results show that the proposed image search engine has high precision; therefore, the method of providing images can help users to find their desired image more easily.

Searching for similar images using Nash game and machine learning

The storage of large amounts of digital data, as well as the processing of digital images, are currently expanding significantly across a range of application areas. As a result, effective management of big images databases is necessary, which calls for the employment of automated and cutting-edge indexing techniques. One method used for this is Content-Based Image Retrieval (CBIR), which tries to index and query the picture database using visual aspects of the image rather than its semantic features. In this article, we propose to explore a digital search engine for similar images, based on multiple image representations and clustering, improved by game theory and machine learning methods.

Analysis and Validation of Image Search Engines in Histopathology.

Searching for similar images in archives of histology and histopathology images is a crucial task that may aid in patient tissue comparison for various purposes, ranging from triaging and diagnosis to prognosis and prediction. Whole slide images (WSIs) are highly detailed digital representations of tissue specimens mounted on glass slides. Matching WSI to WSI can serve as the critical method for patient tissue comparison. In this paper, we report extensive analysis and validation of four search methods bag of visual words (BoVW), Yottixel, SISH, RetCCL, and some of their potential variants. We analyze their algorithms and structures and assess their performance. For this evaluation, we utilized four internal datasets (1269 patients) and three public datasets (1207 patients), totaling more than 200, 000 patches from 38 different classes/subtypes across five primary sites. Certain search engines, for example, BoVW, exhibit notable efficiency and speed but suffer from low accuracy. Conversely, search engines like Yottixel demonstrate efficiency and speed, providing moderately accurate results. Recent proposals, including SISH, display inefficiency and yield inconsistent outcomes, while alternatives like RetCCL prove inadequate in both accuracy and efficiency. Further research is imperative to address the dual aspects of accuracy and minimal storage requirements in histopathological image search.

I-Cubid: a nonlinear cubic graph-based approach to visualize and in-depth browse Flickr image results.

The existing image search engines allow web users to explore images from the grids. The traditional interaction is linear and lookup-based. Notably, scanning web search results is horizontal-vertical and cannot support in-depth browsing. This research emphasizes the significance of a multidimensional exploration scheme over traditional grid layouts in visually exploring web image search results. This research aims to antecedent the implications of visualization and related in-depth browsing via a multidimensional cubic graph representation over a search engine result page (SERP). Furthermore, this research uncovers usability issues in the traditional grid and 3-dimensional web image search space. We provide multidimensional cubic visualization and nonlinear in-depth browsing of web image search results. The proposed approach employs textual annotations and descriptions to represent results in cubic graphs that further support in-depth browsing via a search user interface (SUI) design. It allows nonlinear navigation in web image search results and enables exploration, browsing, visualization, previewing/viewing, and accessing images in a nonlinear, interactive, and usable way. The usability tests and detailed statistical significance analysis confirm the efficacy of cubic presentation over grid layouts. The investigation reveals improvement in overall user satisfaction, screen design, information & terminology, and system capability in exploring web image search results.

QBox: Partial Transfer Learning With Active Querying for Object Detection.

Object detection requires plentiful data annotated with bounding boxes for model training. However, in many applications, it is difficult or even impossible to acquire a large set of labeled examples for the target task due to the privacy concern or lack of reliable annotators. On the other hand, due to the high-quality image search engines, such as Flickr and Google, it is relatively easy to obtain resource-rich unlabeled datasets, whose categories are a superset of those of target data. In this article, to improve the target model with cost-effective supervision from source data, we propose a partial transfer learning approach QBox to actively query labels for bounding boxes of source images. Specifically, we design two criteria, i.e., informativeness and transferability, to measure the potential utility of a bounding box for improving the target model. Based on these criteria, QBox actively queries the labels of the most useful boxes from the source domain and, thus, requires fewer training examples to save the labeling cost. Furthermore, the proposed query strategy allows annotators to simply labeling a specific region, instead of the whole image, and, thus, significantly reduces the labeling difficulty. Extensive experiments are performed on various partial transfer benchmarks and a real COVID-19 detection task. The results validate that QBox improves the detection accuracy with lower labeling cost compared to state-of-the-art query strategies for object detection.

Image to Image Search Engine

In this research, we propose a concept for a search engine that enables users to input a picture from their local database and obtain data from the internet about it. With the exception of the fact that an image is submitted here as a query rather than text-based keywords, this is very similar to the conventional keyword search utilised by the majority of search engines. Because the image is the search query, finding the information corresponding to the uploaded image requires analysis and matching the image's content. This complicates the search process. This is most apt for searching information about images of dogs or any animal or thing that is identifiable.

Image retrieval effectiveness of Bing Images, Google Images and Yahoo Image Search in the scientific field of tourism and COVID-19

The year 2020 brought a big concern for the global community because of COVID-19, which affected every sector of society, and tourism is no exception. Researchers across the globe are publishing their studies related to different dimensions of tourism in the context of COVID-19, and images have formed an essential component of their research. In tourism, images related to COVID-19 can open new dimensions for scholars. The main aim of the research is to measure the retrieval effectiveness of three image search engines (ISEs), that is, Bing Images, Google Images and Yahoo Image Search, concerning images related to COVID-19 and tourism. The study attempts to identify the capability of the ISEs to retrieve the desired and actual images related to COVID-19 and tourism. The PubMed Central (PMC) Database was consulted to retrieve the desired images and develop a testbed. The advanced search feature of PMC Database was explored by typing the search terms ‘COVID-19’ and ‘Tourism’ using ‘AND’ operator to make the search more comprehensive. Both the terms were searched against the ‘Figure/Table’ caption to retrieve papers carrying images related to COVID-19 and tourism. Queries were executed across the select ISEs, that is, Bing Images, Google Images and Yahoo Image Search. Retrieved images were individually analysed against the original image from the articles to determine the Precision, Relative Recall, F-Measure and Fallout Ratio. The format of the images in JPG/JPEG, besides checking the original image rank in the retrieved lot, was also ascertained. Bing Images scores more in terms of Mean Precision, followed by Google Images and Yahoo Image Search. For the Relative Recall measure, Google Images scores high, followed by Bing Images and Yahoo Image Search, respectively. Regarding F-Measure and Fallout Ratio, Bing Images outperforms Google Images and Yahoo Image Search. In retrieving the sought format of JPG/JPEG, Google Images performs best, followed by Yahoo Image Search and Bing Images. Google Images produces the original image at the first rank on more than one occasion. In contrast, Bing Images retrieves the original image at the first rank in two instances. Yahoo Images performs poorly over this metric as it does not retrieve any original image at the first rank on any other instance. The study cannot be generalised as the scope is only limited to the images indexed by PMC. Furthermore, the retrieval effectiveness of only three ISEs is measured. The study is the first to measure the retrieval effectiveness of ISEs in retrieving images related to the COVID-19 pandemic and tourism. The study can be extended across other image-indexing databases pertinent to tourism studies, and the retrieval effectiveness of other ISEs can also be considered.

Use of Convolutional Neural Networks to Evaluate Auricular Reconstruction Outcomes for Microtia.

The objective of this study is to determine whether machine learning may be used for objective assessment of aesthetic outcomes of auricular reconstructive surgery. Images of normal and reconstructed auricles were obtained from internet image search engines. Convolutional neural networks were constructed to identify auricles in 2D images in an auto-segmentation task and to evaluate whether an ear was normal versus reconstructed in a binary classification task. Images were then assigned a percent score for "normal" ear appearance based on confidence of the classification. Images of 1115 ears (600 normal and 515 reconstructed) were obtained. The auto-segmentation task identified auricles with 95.30% accuracy compared to manually segmented auricles. The binary classification task achieved 89.22% accuracy in identifying reconstructed ears. When the confidence of the classification was used to assign percent scores to "normal" appearance, the reconstructed ears were classified to a range of 2% (least like normal ears) to 98% (most like normal ears). Image-based analysis using machine learning can offer objective assessment without the bias of the patient or the surgeon. This methodology could be adapted to be used by surgeons to assess quality of operative outcome in clinical and research settings. 4 Laryngoscope, 133:2413-2416, 2023.

A unified approach of detecting misleading images via tracing its instances on web and analyzing its past context for the verification of multimedia content.

The verification of multimedia content over social media is one of the challenging and crucial issues in the current scenario and gaining prominence in an age where user-generated content and online social web-platforms are the leading sources in shaping and propagating news stories. As these sources allow users to share their opinions without restriction, opportunistic users often post misleading/unreliable content on social media such as Twitter, Facebook, etc. At present, to lure users toward the news story, the text is often attached with some multimedia content (images/videos/audios). Verifying these contents to maintain the credibility and reliability of social media information is of paramount importance. Motivated by this, we proposed a generalized system that supports the automatic classification of images into credible or misleading. In this paper, we investigated machine learning-based as well as deep learning-based approaches utilized to verify misleading multimedia content, where the available image traces are used to identify the credibility of the content. The experiment is performed on the real-world dataset (Media-eval-2015 dataset) collected from Twitter. It also demonstrates the efficiency of our proposed approach and features using both Machine and Deep Learning Model (Bi-directional LSTM). The experiment result reveals that the Microsoft BING image search engine is quite effective in retrieving titles and performs better than our study's Google image search engine. It also shows that gathering clues from attached multimedia content (image) is more effective than detecting only posted content-based features.

Reverse Image Querying

Abstract: As days go by, the volume of graphic i.e. visual data keeps increasing exponentially. This presents a more recent and complex set of problems for handling and making use of such vast, disorganised and unregulated data. Multiple organisations have launched many commercial products to tap into this market. There are multiple methods and functions for reverse image querying but most of them are either incomplete or compute-intensive to be a viable option for smaller use cases. In this study, we have implemented and improvised a rudimentary yet efficient solution by using the cosine similarity model to extract similar images for the given input image from the user. The model used is explained comprehensively together with visual representations. We have also explored multiple use cases that can be used as commercial products. The resulting web application can be utilised to act as a reverse image search engine as a standalone application or can be embedded as a submodule in a larger application.

Advanced Reverse Image Search and Profile Creation using Machine Learning

The Reverse image search is most widely used image search engine due to its extensive database that contains billions of images uploaded over the web. It is best to use the image search when your aim is to find identical images against your queried image. The reverse image search by image is idle option for people getting down similar images in different quality, size, or formats. This online facility leads us to explore image search results with a single click. A convolutional neural network is a type of artificial neural network used for image processing.

Has CEO Gender Bias Really Been Fixed? Adversarial Attacking and Improving Gender Fairness in Image Search

Gender bias is one of the most common and well-studied demographic biases in information retrieval, and in general in AI systems. After discovering and reporting that gender bias for certain professions could change searchers' worldviews, mainstreaming image search engines, such as Google, quickly took action to correct and fix such a bias. However, given the nature of these systems, viz., being opaque, it is unclear if they addressed unequal gender representation and gender stereotypes in image search results systematically and in a sustainable way. In this paper, we propose adversarial attack queries composed of professions and countries (e.g., 'CEO United States') to investigate whether gender bias is thoroughly mitigated by image search engines. Our experiments on Google, Baidu, Naver, and Yandex Image Search show that the proposed attack can trigger high levels of gender bias in image search results very effectively. To defend against such attacks and mitigate gender bias, we design and implement three novel re-ranking algorithms -- epsilon-greedy algorithm, relevance-aware swapping algorithm, and fairness-greedy algorithm, to re-rank returned images for given image queries. Experiments on both simulated (three typical gender distributions) and real-world datasets demonstrate the proposed algorithms can mitigate gender bias effectively.

Comparative Analysis of Tineye and Google Reverse Image Search Engines

Comparative Analysis of Tineye and Google Reverse Image Search Engines

Editorial Commentary

Editorial Commentary

МАНДРІВНА ФІЛОСОФІЯ ГРИГОРІЯ СКОВОРОДИ У ВІДОБРАЖЕННІ НЕЙРОМЕРЕЖ

Hryhorii Skovoroda is the most prominent figure in the history of Ukrainian philosophy. His works also had been elaborated in detail. The article substantiates the relevance of further studies of the image and creativity of the philosopher, their importance for the image of modern Ukrainian studies. The author stated that a new look at the social reality of the 21st century is impossible without taking into account the development of digital technologies. The interaction of the participants of the modern Internet goes beyond the scope of improved communication. The use of neural networks for socio-philosophical studies of visual culture is a new stage in the development of visual research methods. In the article, we had been tried to combine the theories of visual analysis with empirical sociophilosophical research. It was concluded that the use of a neural network has a considerable heuristic potential, complements, strengthens and enriches scientific analytics on the problems of Skovorodology. The textbook doctrine of Hryhorii Skovoroda about the three worlds inspired us to explore the world of texts, the world of images, and the world of symbols. H. Skovoroda’s author’s texts named skovorodians in Ukrainian language had been analyzed in the study. Emphasis was placed on the availability and distribution of full-text works in the Internet network. Key words and concepts in the scientific works of Ukrainian researchers in various fields of knowledge were analyzed. The world of symbols that make up the sacralized space of the culture of the modern world had been processed with the help of the Google image search engine. The work examines how the general features of the biography and personality of the philosopher and writer, as well as the philosophical and literary heritage of the thinker, are reflected in the symbolic world. In the theories of symbolism of H. S. Skovoroda, we relied on the developments of D. Chyzhevsky, T. Lutoy, M. Ryk. The world of images had been explored using the Midjourney neural network query generation. It was concluded that adressing to the neural network leads the research of Hryhoriy Skovoroda's artistic images beyond the boundaries of aesthetics and philosophy of arts into the existence of "living Ukrainian philosophy". Carrying out an interpretive analysis of images generated by a neural network, the authors are faced with the question: is the generated image a new source of primary information for analysis? This article describes the search for an answer to this question.

‘WHEN YOU REALISE YOUR DAD IS CRISTIANO RONALDO’: CELEBRITY SHARENTING AND CHILDREN'S DIGITAL IDENTITY

Sharing of parenthood experiences can be performed by ordinary users, to influencers and celebrities. This paper seeks to understand the digital identity of a child of a global celebrity through their presentation on social media, and to analyse qualitatively the visual formats of children’s representation in a platformised digital environment. It focuses on the representation of Cristiano Ronaldo’s children, especially his oldest son (Junior), through his Instagram account, his wife’s and his mother’s, in a two-year period. After identifying the main forms of representation of the children, we searched for images of Cristiano Junior across image search engines, social media platforms, sites and repositories; then, we used reverse image search for the identified memes. Ronaldo’s children are represented at home and family moments, playing and amidst luxury; 10-year-old Junior is presented as the ‘natural successor’ to Ronaldo’s talent for football, and also takes part in social or business events in public places. Visual representations created by Ronaldo’s sharenting are specially appropriated into memes that mock the football celebrity’s narcissism; or the fact that the child doesn’t know how to pose for selfies. Cristiano’s children’s digital identity and Junior’s, in particular, is populated with images as shared by parents and wider family, as well as constructed through participatory culture. An archaeological approach to visual social media content is not without difficulties, but proved to be fruitful to unpack the tensions of this wide visibility and the meanings it creates for the (digital) identity of the child.