Search

Haze severely degrades the definition of images captured in outdoor scenes. The goal of image dehazing is to restore clear images from hazy ones. This problem has been significantly advanced by using deep neural networks. The performance gains mainly depend on large capacity models, which inevitably increases memory consumption and is not benefit to deployment on mobile devices. In contrast, we propose an effective image dehazing method based on a multi-scale recursive network which does not simply stack deep neural networks to improve dehazing performance. The proposed network consists of both internal and external recursions and some residual blocks. In addition, an auxiliary network is developed to collaboratively train with the primary network and guide the training process of the primary network, which is termed as the auxiliary loss. To better train the proposed network, we develop the smooth L 1 -norm-based content loss, perceptual loss, and auxiliary loss to regularize the proposed network. Extensive experiments demonstrate that the multi-scale recursive network achieves favorable performances against state-of-the-art image dehazing methods. [ABSTRACT FROM AUTHOR]

Local binary feature learning has attracted a lot of researches in image recognition due to its vital effectiveness. Generally, in the traditional local feature learning methods, a projection is learned to map the patches of image into binary features and then a codebook is generated by clustering the binary features with K-means clustering. However, these local feature learning methods, such as compact binary face descriptor and discriminative binary descriptor, ignore the category specific distributions of the original features during the feature learning process and use the real-valued clustering approach to generate the codebook, the discriminant of feature is degraded and the merits of binary feature are lost. To tack these problems, in this paper, we propose a novel category-preserving binary feature learning and binary codebook leaning (CPBFL-BCL) method for finger vein recognition. In CPBFL-BCL, the discrimination of learned binary features is generated by criteria of fisher discriminant analysis and category manifold preserving regularity during the feature learning process, and a novel binary clustering method based on K-means clustering is designed to generate binary codebook. Experimental results on recognition and retrieval tasks using two public finger vein databases are presented and demonstrate the effectiveness and efficiency of the proposed method over the state-of-the-art finger vein methods and a finger vein retrieval method. [ABSTRACT FROM AUTHOR]

ECG classification is important to the diagnosis of cardiovascular disease. This paper develops a robust and accurate algorithm for automatic detection of heart arrhythmias from ECG signals recorded with one lead. A novel model based on the convolutional neural network is proposed to extract low-level and high-level features of short term ECG. In addition, Information-Theoretic Metric Learning is utilized as a final classification model to boost the discrimination abilities of the network trained features. The experimental results over the MIT-BIH arrhythmia database show that the model achieves a comparable performance with most of the state-of-the-art methods and Information-Theoretic Metric Learning further improves the performance. Besides the good accuracy achieved, the proposed method balances different criteria. [ABSTRACT FROM AUTHOR]

In higher education, the initial studying period of each course plays a crucial role for students, and seriously influences the subsequent learning activities. However, given the large size of a course’s students at universities, it has become impossible for teachers to keep track of the performance of individual students. In this circumstance, an academic early warning system is desirable, which automatically detects students with difficulties in learning (i.e., at-risk students) prior to a course starting. However, previous studies are not well suited to this purpose for two reasons: 1) they have mainly concentrated on e-learning platforms, e.g., massive open online courses (MOOCs), and relied on the data about students’ online activities, which is hardly accessed in traditional teaching scenarios; and 2) they have only made performance prediction when a course is in progress or even close to the end. In this paper, for traditional classroom-teaching scenarios, we investigate the task of pre-course student performance prediction, which refers to detecting at-risk students for each course before its commencement. To better represent a student sample and utilize the correlations among courses, we cast the problem as a multi-instance multi-label (MIML) problem. Besides, given the problem of data scarcity, we propose a novel multi-task learning method, i.e., MIML-Circle, to predict the performance of students from different specialties in a unified framework. Extensive experiments are conducted on five real-world datasets, and the results demonstrate the superiority of our approach over the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Automated segmentation of choroidal neovascularization (CNV) in optical coherence tomography (OCT) images plays an important role for the treatment of CNV disease. This paper proposes multi-scale convolutional neural networks with structure prior to segment CNV from OCT data. The proposed framework consists of two stages. In the first stage, the structure prior learning method based on sparse representation-based classification and the local potential function is developed to capture the global spatial structure and local similarity structure prior. The obtained prior can be used to improve the distinctiveness between CNV and background patches. In the second stage, multi-scale CNN model with incorporation of the learned structure prior is constructed for CNV segmentation. In this stage, multi-scale analysis is used to capture effective contextual information, which is robust to varying sizes of CNV. The proposed method was evaluated on 15 spectral domain OCT data with CNV. The experimental results demonstrate the effectiveness of proposed method. [ABSTRACT FROM AUTHOR]

Finger vein image retrieval is a biometric identification technology that has recently attracted a lot of attention. It has the potential to reduce the search space and has attracted a considerable amount of research effort recently. It is a challenging problem owing to the large number of images in biometric databases and the lack of efficient retrieval schemes. We apply a hierarchical vocabulary tree model-based image retrieval approach because of its good scalability and high efficiency.However, there is a large accumulative quantization error in the vocabulary tree (VT)model that may degrade the retrieval precision. To solve this problem, we improve the vector quantization coding in the VT model by introducing a non-negative locality-constrained constraint: the non-negative locality-constrained vocabulary tree-based image retrieval model. The proposed method can effectively improve coding performance and the discriminative power of local features. Extensive experiments on a large fused finger vein database demonstrate the superiority of our encoding method. Experimental results also show that our retrieval strategy achieves better performance than other state-of-the-art methods, while maintaining low time complexity. [ABSTRACT FROM AUTHOR]

Multi-view features are often used in video hashing for near-duplicate video retrieval because of their mutual assistance and complementarity. However, most methods only consider the local available information in multiple features, such as individual or pairwise structural relations, which do not fully utilize the dependent nature of multiple features. We thus propose a global-view hashing (GVH) framework to address the above-mentioned issue; our framework harnesses the global relations among samples characterized by multiple features. In the proposed framework, multiple features of all videos are jointly used to explore a common Hamming space, where the hash functions are obtained by comprehensively utilizing the relations from not only intra-view but also inter-view objects. In addition, the hash function obtained from the proposed GVH can learn multi-bit hash codes in a single iteration. Compared to existing video hashing schemes, the GVH not only globally considers the relations to obtain a more precise retrieval with short-length hash codes but also achieves multi-bit learning in a single iteration. We conduct extensive experiments on the CC_WEB_VIDEO and UQ_VIDEO datasets, and the experimental results show that our proposed method outperforms the state-of-the-art methods. As a side contribution, we will release the codes to facilitate other research. [ABSTRACT FROM AUTHOR]

Research on biometrics for high security applications has not attracted as much attention as civilian or forensic applications. Limited research and deficient analysis so far has led to a lack of general solutions and leaves this as a challenging issue. This work provides a systematic analysis and identification of the problems to be solved in order to meet the performance requirements for high security applications, a double low problem. A hybrid ensemble framework is proposed to solve this problem. Setting an adequately high threshold for each matcher can guarantee a zero false acceptance rate (FAR) and then use the hybrid ensemble framework makes the false reject rate (FRR) as low as possible. Three experiments are performed to verify the effectiveness and generalization of the framework. First, two fingerprint verification algorithms are fused. In this test only 10.55% of fingerprints are falsely rejected with zero false acceptance rate, this is significantly lower than other state of the art methods. Second, in face verification, the framework also results in a large reduction in incorrect classification. Finally, assessing the performance of the framework on a combination of face and gait verification using a heterogeneous database show this framework can achieve both 0% false rejection and 0% false acceptance simultaneously. [ABSTRACT FROM AUTHOR]

Finger-knuckle-print pattern is considered to be one of the most promising biometric techniques for personal identification. In this paper, we propose a new region of interest (ROI) extraction method based on the probabilistic region growing algorithm, which is robust to the finger displacement and rotation. Firstly, contrast enhancement is performed for the finger-knuckle-print images. And then we detect and correct the skewed images. Finally, ROI is extracted using the new probabilistic region growing algorithm. Experimental results show that the proposed method can achieve better performance compared with other existing methods. [ABSTRACT FROM AUTHOR]

SIFT is a novel and promising method for iris recognition. However, some shortages exist in many related methods, such as difficulty of feature extraction, feature loss, and noise point introduction. In this paper, a new method named SIFT-based iris recognition with normalization and enhancement is proposed for achieving better performance. In Comparison with other SIFT-based iris recognition algorithms, the proposed method can overcome the difficulties of extreme point extraction and exclude the noise points without feature loss. Experimental results demonstrate that the normalization and enhancement steps are crucial for SIFT-based iris recognition, and the proposed method can achieve satisfactory recognition performance. [ABSTRACT FROM AUTHOR]