Scalable Retrieval Research Articles

Scalable Retrieval of Similar Landscapes in Optical Satellite Imagery Using Unsupervised Representation Learning

Global Earth observation is becoming increasingly important in understanding and addressing critical aspects of life on our planet, including environmental issues, natural disasters, sustainable development, and others. Finding similarities in landscapes may provide useful information regarding applying contiguous policies, by making similar decisions or learning from best practices for events and occurrences that previously occurred in similar landscapes in the past. However, current applications of similar landscape retrieval are limited by a moderate performance and the need for time-consuming and costly annotations. We propose splitting the similar landscape retrieval task into a set of smaller tasks that aim to identify individual concepts inherent to satellite images. Our approach relies on several models trained using unsupervised representation learning on Google Earth images to identify these concepts. We show the efficacy of matching individual concepts for retrieving landscape(s) similar to a user-selected satellite image of the geographical territory of the Republic of Cyprus. Our results demonstrate the benefits of breaking up the landscape similarity task into individual concepts closely related to remote sensing, instead of applying a single model targeting all underlying concepts.

SCALABLE INFORMATION RETRIEVAL SYSTEM IN SEMANTIC WEB BY QUERY EXPANSION AND ONTOLOGICAL BASED LSA RANKING SIMILARITY MEASUREMENT

In recent days, semantic web presents a key role in intelligent retrieval of information system that resolves vocabulary mismatch problem by query expansion process. However, achieving the scalable information retrieval (IR) in semantic web is a challenging issue in a large dataset. The semantic IR problem is addressed by an ontological-based semantic similarity measurement using natural language processing. The two novel algorithms namely syntactic correlation coefficient (SCC) and mapping-based K-nearest neighbour (M-KNN) for semantic similarity measurement is proposed which improves the accuracy of relevant result. The ontological constructs with word sense disambiguation (WSD) algorithm for document repository improves the conceptual relationships, reduces the ambiguities in ontology and improves scalability by intensely analysing the semantic relationship as well as dynamically reconstructing the ontology when numbers of documents are updated. Ranking is done with latent semantic analysis (LSA) after semantic similarity analysis, which improves the retrieved result and reduces the complexity in relevancy. The performance of the system is analysed with respect to different metrics such as processing time, F-measure (0.97), time complexity, precision (0.95), recall (0.98) and space complexity.

Towards Private and Scalable Cross-Media Retrieval

Cross-media retrieval (CMR) is an attractive networked application where a server responds to queries with retrieval results of different modalities. Different from traditional information retrieval, CMR relies on a more enriched set of machine learning techniques to produce semantic models projecting multimodal data into a common space. A larger training dataset usually gives more accurate models, leading to a better retrieval result. Despite very promising with potential underpinnings in network analytics and multimedia applications, applying CMR in such contexts also faces severe privacy challenges, due to the fact that various data scattering among multiple parties may be sensitive and not allowed to be shared publicly. Studies jointly considering cross-media analytics, privacy protection, collaborative learning, and distributed networking contexts, are relatively sparse. In this work, we propose the first practical system for privacy-preserving cross-media retrieval by utilizing trusted processors. Our scheme enables secure aggregation of the data from distinct parties, and secure canonical correlation analysis (CCA) over collaborated data to obtain semantic models. Verification mechanisms are designed to defend against active attacks from a malicious adversary. Furthermore, to deal with large data sets, we provide a set of optimization methods to accomodate to limited trusted memory and improve the efficiency of training process in CMR. We consider issues such as data block splitting to manage memory overhead, ordering of operations as well as parameters reuse and release to simplify I/O, and parallel computation to speed up dual operations. Our experiments over both synthetic and real datasets show that our solution is very efficient in practice, outperforms the existing solutions, and performs comparably with the original CMR system.

WITHDRAWN: Adaptive multi-graph hashing for scalable multimedia retrieval

Abstract Hashing has become a promising technique to be applied to the large-scale visual retrieval tasks. Multi-view data has multiple views, providing more comprehensive information. Existing hashing methods measure the similarity among the data points by the pre-defined graph Laplacian in each view separately, and thus ignore the underlying common structures across different views. In this paper, we propose a novel hashing method, called adaptive multi-graph hashing (AMGH) to address the above issue. AMGH uses the adaptive weights among the data points, and the weights of each view to characterize the common structure of multi-view data. In addition, AMGH further minimizes the quantization errors. We develop an efficient alternating algorithm to solve the formulated optimization problem. The experiments on three large-scale datasets demonstrate the superiority of the proposed method over the existing multi-view hashing methods.

Hardware Accelerated Semantic Declarative Memory Systems through CUDA and MapReduce

Declarative memory enables cognitive agents to effectively store and retrieve factual memory in real-time. Increasing the capacity of a real-time agent's declarative memory increases an agent's ability to interact intelligently with its environment but requires a scalable retrieval system. This work represents an extension of the Accelerated Declarative Memory (ADM) system, referred to as Hardware Accelerated Declarative Memory (HADM), to execute retrievals on a GPU. HADM also presents improvements over ADM's CPU execution and considers critical behavior for indefinitely running declarative memories. The negative effects of a constant maximum associative strength are considered, and mitigating solutions are proposed. HADM utilizes a GPU to process the entire semantic network in parallel during retrievals, yielding significantly faster declarative retrievals. The resulting GPU-accelerated retrievals show an average speedup of approximately 70 times over the previous Service Oriented Architecture Declarative Memory (soaDM) implementation and an average speedup of approximately 5 times over ADM. HADM is the first GPU-accelerated declarative memory system in existence.

Learning to Hash With Optimized Anchor Embedding for Scalable Retrieval.

Sparse representation and image hashing are powerful tools for data representation and image retrieval respectively. The combinations of these two tools for scalable image retrieval, i.e., sparse hashing (SH) methods, have been proposed in recent years and the preliminary results are promising. The core of those methods is a scheme that can efficiently embed the (high-dimensional) image features into a low-dimensional Hamming space, while preserving the similarity between features. Existing SH methods mostly focus on finding better sparse representations of images in the hash space. We argue that the anchor set utilized in sparse representation is also crucial, which was unfortunately underestimated by the prior art. To this end, we propose a novel SH method that optimizes the integration of the anchors, such that the features can be better embedded and binarized, termed as Sparse Hashing with Optimized Anchor Embedding. The central idea is to push the anchors far from the axis while preserving their relative positions so as to generate similar hashcodes for neighboring features. We formulate this idea as an orthogonality constrained maximization problem and an efficient and novel optimization framework is systematically exploited. Extensive experiments on five benchmark image data sets demonstrate that our method outperforms several state-of-the-art related methods.

Learning Short Binary Codes for Large-scale Image Retrieval.

Large-scale visual information retrieval has become an active research area in this big data era. Recently, hashing/binary coding algorithms prove to be effective for scalable retrieval applications. Most existing hashing methods require relatively long binary codes (i.e., over hundreds of bits, sometimes even thousands of bits) to achieve reasonable retrieval accuracies. However, for some realistic and unique applications, such as on wearable or mobile devices, only short binary codes can be used for efficient image retrieval due to the limitation of computational resources or bandwidth on these devices. In this paper, we propose a novel unsupervised hashing approach called min-cost ranking (MCR) specifically for learning powerful short binary codes (i.e., usually the code length shorter than 100 b) for scalable image retrieval tasks. By exploring the discriminative ability of each dimension of data, MCR can generate one bit binary code for each dimension and simultaneously rank the discriminative separability of each bit according to the proposed cost function. Only top-ranked bits with minimum cost-values are then selected and grouped together to compose the final salient binary codes. Extensive experimental results on large-scale retrieval demonstrate that MCR can achieve comparative performance as the state-of-the-art hashing algorithms but with significantly shorter codes, leading to much faster large-scale retrieval.

An ontology-based intelligent data query system in manufacturing networks

This paper investigates the development of an intelligent data query framework through the use of semantic web technologies for manufacturing purposes. The primary objectives of the ontology-based data query were to develop an efficient and scalable data interoperability and retrieval system; in order to find the most relevant query results with minimum message cost, most hits per query and least response time. This document explains the idea of ontology and the application of the same in the manufacturing domain. A computer simulation software was developed based on a real case study of distributed networks of manufacturing workshops. In this research, a semantic query algorithm was developed where query results are returned by investigating the semantic richness of each workshop. Results were compared with a semantic-free search mechanism based on key performance indicators. The results show the validity of the proposed model for efficient data query when compared to random search.

Visual place recognition with repetitive structures.

Repeated structures such as building facades, fences or road markings often represent a significant challenge for place recognition. Repeated structures are notoriously hard for establishing correspondences using multi-view geometry. They violate the feature independence assumed in the bag-of-visual-words representation which often leads to over-counting evidence and significant degradation of retrieval performance. In this work we show that repeated structures are not a nuisance but, when appropriately represented, they form an important distinguishing feature for many places. We describe a representation of repeated structures suitable for scalable retrieval and geometric verification. The retrieval is based on robust detection of repeated image structures and a suitable modification of weights in the bag-of-visual-word model. We also demonstrate that the explicit detection of repeated patterns is beneficial for robust visual word matching for geometric verification. Place recognition results are shown on datasets of street-level imagery from Pittsburgh and San Francisco demonstrating significant gains in recognition performance compared to the standard bag-of-visual-words baseline as well as the more recently proposed burstiness weighting and Fisher vector encoding.

Regional content-based image retrieval for solar images: Traditional versus modern methods

Abstract This work presents an extensive evaluation between conventional (distance-based) and modern (search-engine) information retrieval techniques in the context of finding similar Solar image regions within the Solar Dynamics Observatory (SDO) mission image repository. We compare pre-computed image descriptors (image features) extracted from the SDO mission images in two very different ways: (1) similarity retrieval using multiple distance-based metrics and (2) retrieval using Lucene, a general purpose scalable retrieval engine. By transforming image descriptors into histogram-like signatures and into Lucene-compatible text strings, we are able to effectively evaluate the retrieval capabilities of both methodologies. Using the image descriptors alongside a labeled image dataset, we present an extensive evaluation under the criteria of performance, scalability and retrieval precision of experimental retrieval systems in order to determine which implementation would be ideal for a production level system. In our analysis we performed key transformations to our sample datasets to properly evaluate rotation invariance and scalability. At the end of this work we conclude which technique is the most robust and would yield the best performing system after an extensive experimental evaluation, we also point out the strengths and weaknesses of each approach and theorize on potential improvements.

Scalable mobile image retrieval by exploring contextual saliency.

Nowadays, it is very convenient to capture photos by a smart phone. As using, the smart phone is a convenient way to share what users experienced anytime and anywhere through social networks, it is very possible that we capture multiple photos to make sure the content is well photographed. In this paper, an effective scalable mobile image retrieval approach is proposed by exploring contextual salient information for the input query image. Our goal is to explore the high-level semantic information of an image by finding the contextual saliency from multiple relevant photos rather than solely using the input image. Thus, the proposed mobile image retrieval approach first determines the relevant photos according to visual similarity, then mines salient features by exploring contextual saliency from multiple relevant images, and finally determines contributions of salient features for scalable retrieval. Compared with the existing mobile-based image retrieval approaches, our approach requires less bandwidth and has better retrieval performance. We can carry out retrieval with <200-B data, which is <5% of existing approaches. Most importantly, when the bandwidth is limited, we can rank the transmitted features according to their contributions to retrieval. Experimental results show the effectiveness of the proposed approach.

BSIFT: toward data-independent codebook for large scale image search.

Bag-of-Words (BoWs) model based on Scale Invariant Feature Transform (SIFT) has been widely used in large-scale image retrieval applications. Feature quantization by vector quantization plays a crucial role in BoW model, which generates visual words from the high- dimensional SIFT features, so as to adapt to the inverted file structure for the scalable retrieval. Traditional feature quantization approaches suffer several issues, such as necessity of visual codebook training, limited reliability, and update inefficiency. To avoid the above problems, in this paper, a novel feature quantization scheme is proposed to efficiently quantize each SIFT descriptor to a descriptive and discriminative bit-vector, which is called binary SIFT (BSIFT). Our quantizer is independent of image collections. In addition, by taking the first 32 bits out from BSIFT as code word, the generated BSIFT naturally lends itself to adapt to the classic inverted file structure for image indexing. Moreover, the quantization error is reduced by feature filtering, code word expansion, and query sensitive mask shielding. Without any explicit codebook for quantization, our approach can be readily applied in image search in some resource-limited scenarios. We evaluate the proposed algorithm for large scale image search on two public image data sets. Experimental results demonstrate the index efficiency and retrieval accuracy of our approach.

Learning salient visual word for scalable mobile image retrieval

Owing to the portable and excellent phone camera, people now prefer to take photos and share them in social networks with their friends. If a user wants to obtain relevant information about an image, content based image retrieval method can be utilized. Taking the limited bandwidth and instability of wireless channel into account, in this paper we propose an effective scalable mobile image retrieval approach by exploiting the advantage of mobile end that people usually take multiple photos of an object in different viewpoints and focuses. The proposed algorithm first determines the truly relevant photos according to visual similarity in mobile end, then learns salient visual words by exploring saliency from these relevant images, and finally determines the contribution order of salient visual words to carry out scalable retrieval. Moreover, to improve the retrieval performance, soft spatial verification is proposed to re-rank the results. Compared to the existing approaches of mobile image retrieval, our approach transmits less data and reduces the computational cost of spatial verification. Most importantly, when the bandwidth is limited, we can transmit only a part of features according their contributions to retrieval. Experimental results show the effectiveness of the proposed approach.

Reconfigurable content-based image retrieval on peer-to-peer networks

Performing content–based image retrieval (CBIR) from Internet scale databases connected through peer–to–peer (P2P) networks, abbreviated as P2P–CBIR, helps to explore the large–scale image database distributed over connected peers. For P2P topology, the flexible decentralised unstructured P2P network topology is adopted, based on which informed–like searching approach enables flexible routing control when peers join/leave. The scalable retrieval control can be developed based on the P2P–CBIR framework, which can adapt the query scope and progressively refine the accuracy during the retrieval process. By reconfiguring system at each regular interval times, we can effectively reduce trivial peer routing and retrieval operations due to imprecise configurations. Experiments demonstrated that the average recall rate of the proposed P2P–CBIR method with reconfiguration is higher than that without about 10.73%, while the latter outperforms previous methods, i.e., firework query model (FQM) and breadth–first search (BFS), about 27% and 57%, respectively, under the same time–to–live (TTL) setting.

Unsupervised learning of electrocorticography motifs with binary descriptors of wavelet features and hierarchical clustering.

We describe a novel method for data mining spectro-spatiotemporal network motifs from electrocorticographic (ECoG) data. The method utilizes wavelet feature extraction from ECoG data, generation of compact binary vectors from these features, and binary vector hierarchical clustering. The potential utility of this method in the discovery of recurring neural patterns is demonstrated in an example showing clustering of ictal and post-ictal gamma activity patterns. The method allows for the efficient and scalable retrieval and clustering of neural motifs occurring in massive amounts of neural data, such as in prolonged EEG/ECoG recordings and in brain computer interfaces.

Large scale document image retrieval by automatic word annotation

In this paper, we present a practical and scalable retrieval framework for large-scale document image collections, for an Indian language script that does not have a robust OCR. OCR-based methods face difficulties in character segmentation and recognition, especially for the complex Indian language scripts. We realize that character recognition is only an intermediate step toward actually labeling words. Hence, we re-pose the problem as one of directly performing word annotation. This new approach has better recognition performance, as well as easier segmentation requirements. However, the number of classes in word annotation is much larger than those for character recognition, making such a classification scheme expensive to train and test. To address this issue, we present a novel framework that replaces naive classification with a carefully designed mixture of indexing and classification schemes. This enables us to build a search system over a large collection of 1,000 books of Telugu, consisting of 120K document images or 36M individual words. This is the largest searchable document image collection for a script without an OCR that we are aware of. Our retrieval system performs significantly well with a mean average precision of 0.8.

WORDGRAPH: Keyword-in-Context Visualization for NETSPEAK's Wildcard Search

The WORDGRAPH helps writers in visually choosing phrases while writing a text. It checks for the commonness of phrases and allows for the retrieval of alternatives by means of wildcard queries. To support such queries, we implement a scalable retrieval engine, which returns high-quality results within milliseconds using a probabilistic retrieval strategy. The results are displayed as WORDGRAPH visualization or as a textual list. The graphical interface provides an effective means for interactive exploration of search results using filter techniques, query expansion, and navigation. Our observations indicate that, of three investigated retrieval tasks, the textual interface is sufficient for the phrase verification task, wherein both interfaces support context-sensitive word choice, and the WORDGRAPH best supports the exploration of a phrase's context or the underlying corpus. Our user study confirms these observations and shows that WORDGRAPH is generally the preferred interface over the textual result list for queries containing multiple wildcards.

Optimising context data dissemination and storage in distributed pervasive computing systems

Context management systems are expected to administrate large volumes of spatial and non-spatial information in geographical disperse domains. In particular, when these systems cover wide areas such as cities, countries or even the entire planet, the design of scalable storage, retrieval and propagation mechanisms is paramount. This paper elaborates on mechanisms that address advanced requirements, including support for distributed context databases management; efficient query handling; innovative management of mobile physical objects and optimization strategies for distributed context data dissemination. These mechanisms establish a robust spatially-enhanced distributed context management framework that has already been designed and carefully implemented and thoroughly evaluated.

Scalable Retrieval and Mining With Optimal Peer-to-Peer Configuration

We proposed to utilize the scalable peer-to-peer network to perform the content-based image retrieval and mining, i.e, P2P-CBIRM. The decentralized unstructured P2P model with certain overheads, i.e., peer clustering and update procedures, is adopted to compromise with the structured one while still reserving flexible routing control when peers join/leave or network fails. The peer CBIRM engine is designed to utilize multi-instance query with multi-feature types to effectively reduce network traffic while maintaining high retrieval accuracy. It helps to enhance the knowledge discovery and image data mining capability. The proposed P2P-CBIRM system provides the scalable retrieval and mining function that the query scope and retrieval accuracy can be adaptively and progressively controlled. To improve the query efficiency (recall-rate/query-scope), it effectively utilizes both: 1) forwarding query message (forward phase) to reduce the query scope and 2) transmitting retrieval results (backward phase) such that activated peers keep filtering high similarity images on the link-path toward the query peer. Experiments show that the query efficiency of the scalable retrieval approach is better than previous methods, i.e., firework query model and breadth-first search. It provides a scalable knowledge discovery platform for efficient image data mining applications. We also proposed to optimally configure the P2P-CBIRM system such that, under a certain number of online users, it would yield the highest recall rate. Simulations demonstrate that, with the optimal configuration, recall rates can be improved to 2.5 to 3 times larger while the network traffic of each peer is reduced to 30% of the original, under the same number of on-line users.

PowerDB-IR --- Scalable Information Retrieval and Storage with a Cluster of Databases

Our objective is a scalable infrastructure for information retrieval (IR) with up-to-date retrieval results in the presence of updates. Timely processing of updates is important with novel applicat...