Binary Hash Research Articles

Heterogeneous Hashing Network for Face Retrieval Across Image and Video Domains

In this paper, we present a heterogeneous hashing network to generate effective and compact hash representations of both face images and face videos for face retrieval across image and video domains. The network contains an image branch and a video branch to project face images and videos into a common space, respectively. Then, the non-linear hash functions are learned in the common space to obtain the corresponding binary hash representations. The network is trained with three loss functions: 1) the Fisher loss; 2) the softmax loss; and 3) the triplet ranking loss. The Fisher loss uses the difference form of within-class and between-class scatter and is appropriate for the mini-batch-based optimization method. The Fisher loss together with the softmax loss is exploited to enhance the discriminative power of the common space. The triplet ranking loss is enforced on the final binary hash representations to improve retrieval performance. Experiments on a large-scale face video dataset and two challenging TV-series datasets demonstrate the effectiveness of the proposed method.

Discrete Latent Factor Model for Cross-Modal Hashing.

Due to its storage and retrieval efficiency, cross-modal hashing (CMH) has been widely used for cross-modal similarity search in many multimedia applications. According to the training strategy, existing CMH methods can be mainly divided into two categories: relaxation-based continuous methods and discrete methods. In general, the training of relaxation-based continuous methods is faster than that of discrete methods, but the accuracy of relaxation-based continuous methods is not satisfactory. On the contrary, the accuracy of discrete methods is typically better than that of the relaxation-based continuous methods, but the training of discrete methods is very time-consuming. In this paper, we propose a novel CMH method, called Discrete Latent Factor model-based cross-modal Hashing (DLFH), for cross modal similarity search. DLFH is a discrete method which can directly learn the binary hash codes for CMH. At the same time, the training of DLFH is efficient. Experiments show that the DLFH can achieve significantly better accuracy than existing methods, and the training time of DLFH is comparable to that of the relaxation-based continuous methods which are much faster than the existing discrete methods.

Relative similarity preserving bitwise weights generated by an adaptive mechanism

Due to its high query speed and low storage cost, binary hashing has been widely used in approximate nearest neighbors (ANN) search. However, the binary bits are generally considered to be equal, which causes data points with different codes to share the same Hamming distance to the query sample. To solve the above distance measure ambiguity, bitwise weights methods were proposed. Unfortunately, in most of the existing methods, the bitwise weights and the binary codes are learnt separately in two stages, and their performances cannot be further improved. In this paper, to effectively address the above issues, we propose an adaptive mechanism that jointly generate the bitwise weights and the binary codes by preserving different types of similarity relationship. As a result, the binary codes are utilized to obtain the initial retrieval results, and they are further re-ranked by the weighted Hamming distance. This ANN search mechanism is termed AR-Rank in this paper. First, this joint mechanism allows the bitwise weights and the binary codes to be used as mutual feedback during the training stage, and they are well adapted to one other when the algorithm converges. Furthermore, the bitwise weights are required to preserve the relative similarity which is consistent with the nature of ANN search task. Thus, the data points can be accurately re-sorted based on the weighted Hamming distances. Evaluations on three datasets demonstrate that the proposed AR-Rank retrieval system outperforms nine state-of-the-art methods.

Multimedia retrieval by deep hashing with multilevel similarity learning

Deep multimodal hashing has received increasing research attention in recent years due to its superior performance for large-scale multimedia retrieval. However, limited e orts have been made to explore the complex multilevel semantic structure for deep multimodal hashing. In this paper, we propose a novel deep multimodal hashing method, termed as Deep Hashing with Multilevel Similarity Learning (DHMSL), for learning compact and discriminative hash codes, which explores multilevel semantic similarity correlations of multimedia data. In DHMSL, multilevel similarity correlation is explored to learn the unified binary hash codes by exploiting the local structure and semantic label information simultaneously. Meanwhile, the bit balance and quantization constraints are taken into account to further make the unified hash codes compact. With the unified binary codes learned, two deep neural networks are jointly trained to simultaneously learn feature representations and two sets of nonlinear hash functions. Specifically, the well-designed loss functions are introduced to minimize the prediction errors of the feature representations as well as the errors between the unified binary codes and outputs of the networks. Extensive experiments on two widely-used multimodal datasets demonstrate that the proposed method can achieve the state-of-the-art performance for both image-query-text and text-query-image tasks.

A feature extraction method for lung nodules based on a multichannel principal component analysis network (PCANet)

Feature extraction of lung nodules is very important in the diagnosis of lung cancer and is the premise of feature description, target matching, recognition and benign and malignant diagnosis. The main contribution of this work is the development of a new end-to-end feature extraction method that learns effective feature representation from images to effectively establish the direct relationship between multiple features and tissue features (that is, benign or malignant). The architecture consists of three seamlessly connected functional layers. RGB multichannel can automatically extract ROI sequence images involving lung parenchyma from the lung imaging sequence. The feature extraction layers, using Principal Component Analysis Network - random binary hash (PCANet-RBH), a) extract high-level semantic features of the R/G/B channel by cascading PCA and fuse the extracted normal color patterns, and b) generate multiple binary patterns via RBH to produce richer features with color information. The connected spatial pyramid pooling (SPP) layer can extract the location features of the lung nodules and map the feature matrix to the low-dimensional space and then establish a correspondence between the image features and the organizational identity. We validate the performance of the proposed method using the public dataset LIDC. The experimental results show that the fusion features extracted by our method have high and stable classification accuracy (accuracy:93.25 ± 0.53, sensitivity:93.12 ± 0.62 specificity:91.37 ± 0.62), which is significantly better than the traditional algorithm for lung nodule feature extraction. Moreover, RGB-PCANet has a short training time, which can meet the requirements of real-time diagnosis of lung cancer. In general, the advantage of our framework is that it provides a better and more comprehensive method to establish a direct relationship between image high-level semantic features, color features, location features and tissue features, making it an attractive clinical diagnostic tool for lung cancer.

One-Factor Cancellable Palmprint Recognition Scheme Based on OIOM and Minimum Signature Hash

The traditional Cancellable biometrics scheme needs another key or token to generate the revocable template, which usually suffers from the token-stolen problem. To solve this problem, a one-factor cancellable palmprint biometric recognition scheme based on Orthogonal Index of Maximum (OIOM) hash and Minimum Signature Hash (MSH) is proposed to generate pseudonymous identifier. Firstly, to improve the efficiency and effectiveness, a parallel structure is designed to obtain Orthogonal Gaussian random projection(GRP)matrices, which is employed to generate the OIOM hash code. Secondly, a random binary string is XOR the binary OIOM hash code to construct the helper data and it is stored in the database. Meanwhile, this string is hashed by MSH to get the final pseudonymous identifier. Lastly, during matching stage, based on helper data and a palmprint images, another pseudonymous identifier is generated to recognition. This implies that, just one factor, a palmprint of user, is needed in the matching stage, and therefore the privacy of user is preserved. To evaluate the proposed scheme, PolyU database and touchless TJU database are used in experiments. The noninvertible, renewability, unlinkability and several security attacks of the proposed scheme are analyzed. The experiment results and analysis show that the proposed scheme has strong security on the basis of maintaining palmprint recognition performance.

Medical Image Retrieval Based on Convolutional Neural Network and Supervised Hashing

In recent years, with extensive application in image retrieval and other tasks, a convolutional neural network (CNN) has achieved outstanding performance. In this paper, a new content-based medical image retrieval (CBMIR) framework using CNN and hash coding is proposed. The new framework adopts a Siamese network in which pairs of images are used as inputs, and a model is learned to make images belonging to the same class have similar features by using weight sharing and a contrastive loss function. In each branch of the network, CNN is adapted to extract features, followed by hash mapping, which is used to reduce the dimensionality of feature vectors. In the training process, a new loss function is designed to make the feature vectors more distinguishable, and a regularization term is added to encourage the real value outputs to approximate the desired binary values. In the retrieval phase, the compact binary hash code of the query image is achieved from the trained network and is subsequently compared with the hash codes of the database images. We experimented on two medical image datasets: the cancer imaging archive-computed tomography (TCIA-CT) and the vision and image analysis group/international early lung cancer action program (VIA/I-ELCAP). The results indicate that our method is superior to existing hash methods and CNN methods. Compared with the traditional hashing method, feature extraction based on CNN has advantages. The proposed algorithm combining a Siamese network with the hash method is superior to the classical CNN-based methods. The application of a new loss function can effectively improve retrieval accuracy.

Learning binary hash codes for finger vein image retrieval

High dimensional real-valued features have been shown to be effective for finger vein identification, but result in high computational cost especially in a large-scale finger vein database. Therefore compact binary codes are generally used to obtain fast query speed as well as reduce storage requirements. However, only a few strategies are available for finger vein retrieval in Hamming space, and existing binary representation of finger veins can become unstable and undiscriminating due to finger vein variations induced by translation, rotation and illumination. In this paper, we propose a binary hash codes learning algorithm to map finger vein images in the original feature space to Hamming space. First, to obtain the discriminative finger vein image features, a novel finger vein image representation method called Nonlinearly Subspace Coding (NSC) is proposed. The codebook is a union of low-dimensional linear subspaces instead of visual words. For a given local finger vein texton, its top-k nearest subspaces are found and the texton is nonlinearly mapped into these subspaces. Then, a finger vein Binary Hash Codes (BHC) learning method is proposed by jointly considering the discriminability and the stability of the binary code. Experimental results on a two-session public finger vein database and a large fused finger vein database demonstrate the effectiveness and efficiency of our binary hash coding learning algorithm for large-scale finger vein retrieval.

Fixed-length asymmetric binary hashing for fingerprint verification through GMM-SVM based representations

Fingerprint minutiae information is an unordered and variable-sized collection of minutiae locations and orientations. Advanced template protection algorithms which require a fixed-length binary template cannot operate on minutiae points. In this paper, we propose a novel framework that provides practical solutions that can be used in developing secure fingerprint verification systems. The framework, by using a GMM-SVM fingerprint representation scheme, first generates fixed-length feature vectors from minutiae point sets. The fixed-length representation enables the application of modern cryptographic alternatives based on homomorphic encryption to minutiae template protection. Our framework then utilizes an asymmetric locality sensitive hashing (ALSH) in order to convert the generated fixed-length but real valued GMM-SVM feature vector to a binary bit string. This binarization step transforms the matching process to calculating Hamming distance between binary vectors and expedites fingerprint matching. The verification performance of the framework is evaluated on FVC2002DB1A and DB2A databases.

Learning Discrete Hashing Towards Efficient Fashion Recommendation

In our daily life, how to match clothing well is always a troublesome problem especially when we are shopping online to select a pair of matched pieces of clothing from tens of thousands available selections. To help common customers overcome selection issues, recent studies in the recommender system area have started to infer the fashion matching results automatically. The traditional fashion recommendation is normally achieved by considering visual similarity of clothing items or/and item co-purchase history from existing shopping transactions. Due to the high complexity of visual features and the lack of historical item purchase records, most of the existing work is unlikely to make an efficient and accurate recommendation. To address the problem, in this paper, we propose a new model called Discrete Supervised Fashion Coordinates Hashing. Its main objective is to learn meaningful yet compact high-level features of clothing items, which are represented as binary hash codes. In detail, this learning process is supervised by a clothing matching matrix, which is initially constructed based on limited known matching pairs and subsequently on the self-augmented ones. The proposed model jointly learns the intrinsic matching patterns from the matching matrix and the binary representations from the clothing items’ images, where the visual feature of each clothing item is discretized into a fixed-length binary vector. The binary representation learning significantly reduces the memory cost and accelerates the recommendation speed. The experiments compared with several state-of-the-art approaches have evidenced the superior performance of the proposed approach on efficient fashion recommendation.

Unsupervised adaptive hashing based on feature clustering

Abstract An attractive method for image retrieval is binary hashing, which aims to reduce the dimensionality and generate similarity-preserving binary codes. To map the high-dimensional data into a low-dimensional subspace, majority of current unsupervised hashing approaches reduce the dimensionality by principal component analysis (PCA). However, PCA will yield unbalanced variances of projection directions and cause inconvenience in the quantization step. Besides, preserving the original similarity in existing unsupervised hashing methods remains as an NP-hard problem. For addressing these problems, we explore a novel hashing method based on feature clustering to simultaneously generate low-dimensional data with balanced variance and preserve the similarity in Euclidean space. Furthermore, we also propose an adaptive quantization approach to displace the fixed threshold quantization. Our novel method, dubbed as Feature Clustering Hashing (FCH), has shown its superiority to state-of-the-art methods on three benchmark datasets.

PCANet based nonlocal means method for speckle noise removal in ultrasound images.

Speckle reduction remains a critical issue for ultrasound image processing and analysis. The nonlocal means (NLM) filter has recently attached much attention due to its competitive despeckling performance. However, the existing NLM methods usually determine the similarity between two patches by directly utilizing the gray-level information of the noisy image, which renders it difficult to represent the structural similarity of ultrasound images effectively. To address this problem, the NLM method based on the simple deep learning baseline named PCANet is proposed by introducing the intrinsic features of image patches extracted by this network rather than the pixel intensities into the pixel similarity computation. In this approach, the improved two-stage PCANet is proposed by using Parametric Rectified Linear Unit (PReLU) activation function instead of the binary hashing and block histograms in the original PCANet. This model is firstly trained on the ultrasound database to learn the convolution kernels. Then, the trained PCANet is utilized to extract the intrinsic features from the image patches in the pre-denoised version of the noisy image to be despeckled. These obtained features are concatenated together to determine the structural similarity between image patches in the NLM method, based on which the weighted mean of all pixels in a search window is computed to produce the final despeckled image. Extensive experiments have been conducted on a variety of images to demonstrate the superiority of the proposed method over several well-known despeckling algorithm and the PCANet based NLM method using ReLU function and sigmoid function. Visual inspection indicates that the proposed method outperforms the compared methods in reducing speckle noise and preserving image details. The quantitative comparisons show that among all the evaluated methods, our method produces the best structural similarity index metrics (SSIM) values for the synthetic image, as well as the highest equivalent number of looks (ENL) value for the simulated image and the clinical ultrasound images.

Binary Hashing CNN Features for Action Recognition

Binary Hashing CNN Features for Action Recognition

Deep Learning Hash for Wireless Multimedia Image Content Security

With the explosive growth of the wireless multimedia data on the wireless Internet, a large number of illegal images have been widely disseminated in wireless networks, which seriously endangers the content security of wireless networks. However, how to identify and classify illegal images quickly, accurately, and in real time is a key challenge for wireless multimedia networks. To avoid illegal images circulating on the Internet, each image needs to be detected, extracted features, and compared with the image in the feature library to verify the legitimacy of the image. An improved image deep learning hash (IDLH) method to learn compact binary codes for image search is proposed in this paper. Specifically, there are three major processes of IDLH: the feature extraction, deep secondary search, and image classification. IDLH performs image retrieval by the deep neural networks (DNN) as well as image classification with the binary hash codes. Different from other deep learning-hash methods that often entail heavy computations by using a conventional classifier, exemplified by K nearest neighbor (K-NN) and support vector machines (SVM), our method learns classifiers using binary hash codes, which can be learned synchronously in training. Finally, comprehensive experiments are conducted to evaluate IDLH method by using CIFAR-10 and Caltech 256 image library datasets, and the results show that the retrieval performance of IDLH method can effectively identify illegal images.

Deep learning with multi-scale feature fusion in remote sensing for automatic oceanic eddy detection

Oceanic eddies are ubiquitous in global oceans and play a major role in ocean energy transfer and nutrients distribution, thus being significant for understanding ocean current circulation and marine climate change. They are characterized by a combination of high-speed vertical rotations and horizontal movements, leading to irregular three-dimensional spiral structures. While the ability to detect eddies automatically and remotely is crucial to monitoring important spatial–temporal dynamics, existing methods are inaccurate because eddies are highly dynamic and the underlying physical processes are not well understood. Typically, remote sensing is used to detect eddies based on physical parameters, geometrics or other handcrafted features. In this paper, we show how Deep Learning may be used to reliably extract higher-level features and then fuse multi-scale features to identify eddies, regardless of their structures and scales. We learn eddy features using two principal component analysis convolutional layers, then perform a non-linear transformation of the features through a binary hashing layer and block-wise histograms. To handle the difficult problem of spatial variability across synthetic aperture radar (SAR) images, we introduce a spatial pyramid model to allow multi-scale features fusion. Finally, a linear support vector machine classifier recognizes the eddies. Our method, dubbed DeepEddy, is benchmarked against a dataset of 20,000 SAR image samples, achieving a 97.8 ± 1% accuracy of detection.

Nonlinear Asymmetric Multi-Valued Hashing.

Most existing hashing methods resort to binary codes for large scale similarity search, owing to the high efficiency of computation and storage. However, binary codes lack enough capability in similarity preservation, resulting in less desirable performance. To address this issue, we propose Nonlinear Asymmetric Multi-Valued Hashing (NAMVH) supported by two distinct non-binary embeddings. Specifically, a real-valued embedding is used for representing the newly-coming query by an ideally nonlinear transformation. Besides, a multi-integer-embedding is employed for compressing the whole database, which is modeled by Binary Sparse Representation (BSR) with fixed sparsity. With these two non-binary embeddings, NAMVH preserves more precise similarities between data points and enables access to the incremental extension with database samples evolving dynamically. To perform meaningful asymmetric similarity computation for efficient semantic search, these embeddings are jointly learnt by preserving the pairwise label-based similarity. Technically, this results in a mixed integer programming problem, which is efficiently solved by a well-designed alternative optimization method. Extensive experiments on seven large scale datasets demonstrate that our approach not only outperforms the existing binary hashing methods in search accuracy, but also retains their query and storage efficiency.

Bagging trees with Siamese-twin neural network hashing versus unhashed features for unsupervised image retrieval

The goal of this paper is twofold. Firstly, a Siamese-twin random projection neural network (ST-RPNN) is proposed for unsupervised binary hashing of images and compared with state-of-the-art techniques. Secondly, a comparison between Hamming-distance-based retrieval and a proposed bagging trees retrieval (BT-retrieval) algorithm operating directly on the PCA features is made with respect to performance, storage and search time. The ST-RPNN is made of two identical random projection neural networks and is trained to produce similar binary codes for similar input image pairs and different binary codes otherwise. The learning process is divided into two steps: a fast sparse neurons selection algorithm followed by an unsupervised bagging trees algorithm to extract the compact required-length code. Moreover, a BT-retrieval algorithm is proposed in this paper as a fast retrieval tool that ranks the database with respect to a query without distance calculations. Furthermore, (BT-PCA) is a novel extension where the BT-retrieval is applied directly on the PCA features with a significantly lower time search than Hamming-distance-based approach. The proposed technique is compared with ten unsupervised image binary hashing techniques on the COREL1K dataset and the CIFAR10 dataset. The proposed technique obtained better precision–recall results than all compared techniques on the COREL1K dataset, and better than eight of them on the CIFAR10 dataset.

ACPC: Efficient revocation of pseudonym certificates using activation codes

Vehicular communication (V2X) technologies allow vehicles to exchange information about the road conditions and their own status, and thereby enhance transportation safety and efficiency. For broader deployment, however, such technologies are expected to address security and privacy concerns, preventing abuse by users and by the system’s entities. In particular, the system is expected to enable the revocation of malicious vehicles, e.g., in case they send invalid information to their peers or to the roadside infrastructure; it should also prevent the system from being misused for tracking honest vehicles. Both features are enabled by Vehicular Public Key Infrastructure (VPKI) solutions such as Security Credential Management Systems (SCMS), one of the leading candidates for protecting V2X communication in the United States. Unfortunately, though, SCMS’s original revocation mechanism can lead to large Certification Revocation Lists (CRLs), which in turn impacts the bandwidth usage and processing overhead of the system. In this article, we propose a novel design called Activation Codes for Pseudonym Certificates (ACPC), which can be integrated into SCMS to address this issue. Our proposal is based on activation codes, short bit-strings without which certificates previously issued to a vehicle cannot be used by the latter, which are periodically distributed to non-revoked vehicles using an efficient broadcast mechanism. As a result, the identifiers of the corresponding certificates do no need to remain on the CRL for a long time, reducing the CRLs’ size and streamlining their distribution and verification of any vehicle’s revocation status. Besides describing ACPC in detail, we also compare it to similar-purpose solutions such as Issue First Activate Later (IFAL) and Binary Hash Tree based Certificate Access Management (BCAM). This analysis shows that our proposal not only improves privacy (e.g., in terms of resilience against colluding system authorities), but also leads to processing and bandwidth overheads that are orders of magnitude smaller than those observed in the state of the art.

Hadamard Coding for Supervised Discrete Hashing.

In this paper, we propose a learning-based supervised discrete hashing method. Binary hashing is widely used for large-scale image retrieval as well as video and document searches because the compact binary code representation is essential for data storage and reasonable for query searches using bit-operations. The recently proposed supervised discrete hashing (SDH) method efficiently solves mixed-integer programming problems by alternating optimization and the discrete cyclic coordinate descent (DCC) method. Based on some preliminary experiments, we show that the SDH method can be simplified without performance degradation. We analyze the simplified model and provide a mathematically exact solution thereof; we reveal that the exact binary code is provided by a "Hadamard matrix." Therefore, we named our method Hadamard codedsupervised discrete hashing (HC-SDH). In contrast to SDH, our model does not require an alternating optimization algorithm and does not depend on initial values. HC-SDH is also easier to implement than iterative quantization (ITQ). Experimental results involving a large-scale database show that Hadamard coding outperforms conventional SDH in terms of precision, recall, and computational time. On the large datasets SUN-397 and ImageNet, HC-SDH provides a superior mean average of precision (mAP) and top-accuracy compared to the conventional SDH methods with the same code length and FastHash. The training time of HC-SDH is 170 times faster than conventional SDH and the testing time including the encoding time is seven times faster than FastHash which encodes using a binary-tree.

Deep hashing network for material defect image classification

Common non‐destructive material testing technology has some well‐known problems such as slow detection, low detection accuracy, and low level of information obtained. To solve these problems, this study applied recent advances in convolution neural networks to propose an effective deep learning network using casting datasets. The approach achieves non‐destructive material testing with automatic, intelligent detection technology. For most existing deep learning networks, an image is eventually transformed into a multidimensional visual feature vector for comparison and classification. However, such vectors may not optimally improve detection precision and speed, and can lead to significant storage problems. A deep hashing network is proposed in which images are mapped into compact binary codes. There are three key components: (i) a sub‐network with multiple convolution‐pooling layers to capture image representations; (ii) a hashing layer to generate compact binary hash codes; (iii) an encoder module to divide the image feature vector from the output of the sub‐network above into multiple branches, each encoded into one hash bit. Extensive experiments using a casting dataset show promising performance compared with the state‐of‐the‐art approach.