Search

Many recent variational optical flow methods are not robust for illumination variance, and they only consider local image relation in terms of illumination. In this paper, we propose a new efficient illumination-invariance total variation optical flow method called the weighted regularization transform, which uses and optimizes the Weber’s Law. Our method exploits unequal probability as the weight that has non-local information to estimate stable optical flow despite illumination changes. The proposed method uses a coarse-to-fine pyramid model to reduce the influence on the data term from illumination. Then, an energy optimization procedure is introduced to constrain the minimization of the data term with the non-local regularization. Experimentation with the proposed method has been performed on three optical flow datasets and a face liveness detection database, which have challenging illumination variations, and the results demonstrate that the proposed method is quite robust with respect to variations in illumination.

The single-image super-resolution techniques (SISR) have been significantly promoted by deep networks. However, the storage and computation complexities of deep models increase dramatically alongside with the reconstruction performance. This paper proposes a densely connected recursive network (DCRN) to trade off the performance and complexity. We introduce an enhanced dense unit by removing the batch normalization (BN) layers and employing the squeeze-and-excitation (SE) structure. A recursive architecture is also adopted to control the parameters of deep networks. Moreover, a de-convolution based residual learning method is proposed to accelerate the residual feature extraction process. The experimental results validate the efficiency of the proposed approach.

Illumination problem is still a bottleneck of robust face recognition system, which demands extracting illumination invariant features. In this field, existing works only consider the variations caused by lighting direction or magnitude (denoted as homogeneous lighting), but the effect of spectral wavelength is always ignored and thus existing illumination invariant descriptors have its limitation on processing face images under different spectral wavelengths (denoted as heterogeneous lighting). We propose a novel gradient based descriptor, namely Logarithm Gradient Histogram (LGH), which takes the illumination direction, magnitude and the spectral wavelength together into consideration, so that it can handle both homogeneous and heterogeneous lightings. Our proposal contributes in three-folds: (1) we incorporate LMSN-LoG filter to eliminate the lighting effect of each image and extract two illumination invariant components, namely logarithm gradient orientation (LGO) and logarithm gradient magnitude (LGM); (2) we propose an effective post-processing strategy to make our model tolerant to noise and generate a histogram representation to integrate both LGO and LGM; (3) we present solid theoretical analysis on the illumination invariant properties of our proposed descriptors. Extensive experimental results on CMU-PIE, Extended YaleB, FRGC and HFB databases are reported to verify the effectiveness of our proposed model. HighlightsTwo illumination invariant components, namely logarithm gradient orientation (LGO) and logarithm gradient magnitude (LGM), are extracted.An effective post-processing strategy is proposed to integrate both LGO and LGM, generating the logarithm gradient histogram (LGH).Solid theoretical analysis on the illumination invariant properties of the proposed descriptors is presented.Competitive results are reported, both in homogeneous and heterogeneous lighting conditions.

Age estimation based on unconstrained face images remains a challenging problem in computer vision and pattern recognition. We address this by ensembling part-based features and designing a cost sensitive loss to overcome the general imbalance data in age estimation. Specifically, we treat age estimation as a multi-class classification problem and mainly make two contributions: (i) We present a Part-based Convolutional Network(PCN) for age estimation to extract regional features instead of the holistic ones, which can preserve more discriminative features in facial regions like forehead, canthus, cheek, jaw, etc; (ii) A balanced loss for multi-class classification is designed to handle the extremely imbalance age distribution. The loss pays more attention to the hard examples, and can automatically adjust the weights of samples according to their contribution. State-of-the-art performance was achieved on FG-NET, MORPH and CACD databases, validating the effectiveness of the proposed approach.

It is generally known that the illumination could seriously affect the performance of face analysis algorithms. Moreover, in most practical applications, the illumination is usually uncontrolled. A number of methods have been put forward to tackle the problem of illumination variations in face images, but they always only work on facial region and need to segment faces in advance. Furthermore, many illumination processing methods only demonstrate on grayscale images and require strict alignment of face images, resulting in limited applications in the real world. In this paper, we propose a face image illumination processing method based on the Generative Adversarial Network (GAN) with dual triplet loss. Through considering the inter-domain similarity and intra-domain difference between the generated images and the real images, we put forward the dual triplet loss. At the same time, we introduce the self-similarity constraint of the images in the target illumination field. Experiments on the CMU Multi-PIE face datasets demonstrate that the proposed method preserve the facial details well when relighting. The experiment of 3D face reconstruction also verifies the effectiveness of the proposed method.

Visual versus near infrared (VIS-NIR) face image matching uses an NIR face image as the probe and conventional VIS face images as enrollment. It takes advantage of the NIR face technology in tackling illumination changes and low-light condition and can cater for more applications where the enrollment is done using VIS face images such as ID card photos. Existing VIS-NIR techniques assume that during classifier learning, the VIS images of each target people have their NIR counterparts. However, since corresponding VIS-NIR image pairs of the same people are not always available, which is often the case, so those methods cannot be applied. To address this problem, we propose a transductive method named transductive heterogeneous face matching (THFM) to adapt the VIS-NIR matching learned from training with available image pairs to all people in the target set. In addition, we propose a simple feature representation for effective VIS-NIR matching, which can be computed in three steps, namely Log-DoG filtering, local encoding, and uniform feature normalization, to reduce heterogeneities between VIS and NIR images. The transduction approach can reduce the domain difference due to heterogeneous data and learn the discriminative model for target people simultaneously. To the best of our knowledge, it is the first attempt to formulate the VIS-NIR matching using transduction to address the generalization problem for matching. Experimental results validate the effectiveness of our proposed method on the heterogeneous face biometric databases.

Face recognition is an increasingly popular technology for user authentication. However, face recognition is susceptible to spoofing attacks. Therefore, a reliable way to detect malicious attacks is crucial to the robustness of the face recognition system. This paper describes a new approach to utilizing light field camera for defending spoofing face attacks, like (warped) printed 2D facial photos and high-definition tablet images. The light field camera is a sensor that can record the directions as well as the colors of incident rays. Needing only one snapshot, multiple refocused images can be generated. In the proposed method, three kinds of features extracted from a pair of refocused images are extracted to discriminate fake faces and real faces. To verify the performance, we build a light field photograph databases and conduct experiments. Experimental results reveal that the employed features can achieve remarkable anti-spoofing accuracy under different types of spoofing attacks.

Person re-identification (re-id) aims at matching person images across multiple surveillance cameras. Currently, most re-id systems highly rely on color cues, which are only effective in good illumination conditions, but fail in low lighting conditions. However, for security issues, it is very important to conduct surveillance in low lighting conditions. To remedy this problem, we propose using depth cameras to perform surveillance in dark places, while using traditional RGB cameras in bright places. Such a heterogeneous camera network brings a challenge to match images across depth and RGB cameras. In this paper, we mine the correlation between two heterogeneous cues (depth and RGB) on both feature-level and transformation-level. As such, depth-based features and RGB-based features are transformed to the same space, which alleviates the problem of cross-modality matching between depth and RGB cameras. Experimental results on two benchmark heterogeneous person re-id datasets show the effectiveness of our method.

Partitioning nonlinearly separable datasets is a basic problem that is associated with data clustering. In this paper, a novel approach that is termed graph-based multiprototype competitive learning (GMPCL) is proposed to handle this problem. A graph-based method is employed to produce an initial, coarse clustering. After that, a multiprototype competitive learning is introduced to refine the coarse clustering and discover clusters of an arbitrary shape. The GMPCL algorithm is further extended to deal with high-dimensional data clustering, i.e., the fast graph-based multiprototype competitive learning (FGMPCL) algorithm. An experimental comparison has been performed by the exploitation of both synthetic and real-world datasets to validate the effectiveness of the proposed methods. Additionally, we apply our GMPCL/FGMPCL to two computer-vision tasks, namely, automatic color image segmentation and video clustering. Experimental results show that GMPCL/FGMPCL provide an effective and efficient tool with application to computer vision.

Kernel-based clustering is one of the most popular methods for partitioning nonlinearly separable datasets. However, exhaustive search for the global optimum is NP-hard. Iterative procedure such as k-means can be used to seek one of the local minima. Unfortunately, it is easily trapped into degenerate local minima when the prototypes of clusters are ill-initialized. In this paper, we restate the optimization problem of kernel-based clustering in an online learning framework, whereby a conscience mechanism is easily integrated to tackle the ill-initialization problem and faster convergence rate is achieved. Thus, we propose a novel approach termed conscience online learning (COLL). For each randomly taken data point, our method selects the winning prototype based on the conscience mechanism to bias the ill-initialized prototype to avoid degenerate local minima and efficiently updates the winner by the online learning rule. Therefore, it can more efficiently obtain smaller distortion error than k-means with the same initialization. The rationale of the proposed COLL method is experimentally analyzed. Then, we apply the COLL method to the applications of digit clustering and video clustering. The experimental results demonstrate the significant improvement over existing kernel-based clustering methods.

The affinity propagation (AP) clustering algorithm has received much attention in the past few years. AP is appealing because it is efficient, insensitive to initialization, and it produces clusters at a lower error rate than other exemplar-based methods. However, its single-exemplar model becomes inadequate when applied to model multisubclasses in some situations such as scene analysis and character recognition. To remedy this deficiency, we have extended the single-exemplar model to a multi-exemplar one to create a new multi-exemplar affinity propagation (MEAP) algorithm. This new model automatically determines the number of exemplars in each cluster associated with a super exemplar to approximate the subclasses in the category. Solving the model is NP--hard and we tackle it with the max-sum belief propagation to produce neighborhood maximum clusters, with no need to specify beforehand the number of clusters, multi-exemplars, and superexemplars. Also, utilizing the sparsity in the data, we are able to reduce substantially the computational time and storage. Experimental studies have shown MEAP's significant improvements over other algorithms on unsupervised image categorization and the clustering of handwritten digits.

In the last decade, illumination problem has been the bottleneck of robust face recognition system. Extracting illumination invariant features becomes more and more significant to solve this issue. However, existing works in this field only consider the variations caused by lighting direction or magnitude (denoted as homogeneous lighting), while the spectral wavelength is always ignored in most of the developed illumination invariant descriptors. In this paper, we claim that the spectral wavelength is important, and we propose a novel gradient based descriptor, namely Logarithm Gradient Histogram (LGH), which takes the illumination direction, magnitude and even the spectral wavelength together into consideration (denoted as heterogeneous lighting). Our proposal contributes in the following three-folds: (1) we incorporate homogeneous filtering to alleviate the illumination effect for each image and extract two illumination invariant components, namely logarithm gradient orientation (LGO) and logarithm gradient magnitude (LGM); (2) we propose an effective postprocessing strategy to guarantee the fault-tolerant ability and generate a histogram representation to integrate both LGO and LGM; (3) we present thorough theoretical analysis on the illumination invariant properties for our proposed method. Experimental results on CMU-PIE, Extended YaleB and HFB databases are reported to verify the effectiveness of our proposed method.

The visual-near infrared (VIS-NIR) face matching, sharing the illumination-invariant property of NIR face image and remaining the use of existing VIS face images as enrollment, has been a popular issue in recent years. However, existing techniques assume that there are sufficient pairwise VIS and NIR images for each person during training, which is not realistic in VIS-NIR matching problem, as no NIR images are available for people who have already been registered in the existing face recognition system and only a handful of pairwise VIS and NIR face images captured from new people are available. To address this problem, we formulate the VIS-NIR matching as a transductive learning problem, which is a first attempt to our best knowledge. Moreover, we propose a transductive method named Transductive Heterogeneous Face Matching (THFM) by alleviating the domains difference and learning the discriminative model for target simultaneously, making it possible to take the query/probe NIR images into account in a transductive way. Experimental results validate the effectiveness of our approach on the heterogeneous face biometric database.

Due to their role in eliciting protective Th1 cell-mediated immune responses in definitive hosts lung stage schistosomula are in the focus of intensive research. In vitro culture approaches in the past exhibited significant differences in gene expression profiles between lung stage schistosomula isolated from hosts and those cultured conventionally. Therefore, new approaches to culture schistosomula are of broad interest. In the present study, co-culture systems of schistosomula of Schistosoma japonicum and different vertebrate host cells were tested. Among these, human hepatic venous endothelial cells (ED25) turned out to be very suitable and interesting feeder cells. Compared with controls cultured in vitro or co-cultured with other cells, schistosomula co-cultured with ED25 cells shared more similarities in morphology and tegumental structures with schistosomula directly obtained from infected mice as microscopically determined. According to results from a suppression subtractive hybridization approach to compare transcriptional differences of co-cultured and host group or control group parasites, four candidate transcripts encoding cathepsin L precursor, heat shock protein 70, glyceraldehyde 3-phosphate dehydrogenase, and programmed cell death protein 10 were shown to be differently expressed among the three groups by real-time PCR. Sodium dodecyl sulfate polyacrylamide gel electrophoresis analysis finally confirmed not only congruent protein patterns but also interesting differences among the compared schistosomula groups. The co-culture system between schistosomula of S. japonicum and ED25 cells established in the present study improved existing cultivation attempts. Although some differences to host-derived schistosomula were still observed, co-culture with ED25 cells positively influenced parasite morphology and gene expression in a more host-like manner.

In the past decade, illumination problem has been the bottleneck of developing robust face recognition systems. Extracting illumination invariant features, especially the gradient based descriptor [13], is an effective tool to solve this issue. In this paper, we propose a novel gradient based descriptor, namely Complete Gradient Face (CGF), to compensate the limitations in [13] and contribute in three folds: (1) we incorporate homogeneous filtering to alleviate the illumination effect and enhance facial information based on the Lambertian assumption; (2) we demonstrate the gradient magnitude in logarithm domain is insensitive to lighting change; (3) we propose a histogram based feature descriptor to integrate both magnitude and orientation information. Experiments on CMU-PIE and Extended YaleB are conducted to verify the effectiveness of our proposed method.

Oncomelania hupensis is the intermediate host of Schistosoma japonicum. In the present study, we investigated the effects of protein extracts from head–foot or gland tissue of O. hupensis on mother sporocysts of S. japonicum cultured in vitro. In the presence of head–foot protein extract of snails from the native province Hunan, in-vitro-transformed mother sporocysts presented not only a longer survival time and stronger motility, but also a bigger size than parasites cultured with protein extracts of glands of the same snail or head–foot tissue of a non-native snail from the Hubei province. Using suppression subtractive hybridization, two subtractive libraries were constructed on the basis of RNA of sporocysts cultured with or without native snail head–foot protein extract. A number of 31 transcripts were found to be up-regulated. Sequence analyses revealed that they represented genes involved among others in metabolic process, electron transport chain, response to chemical stimulus, and oxidation–reduction processes. Opposite to that 20 down-regulated transcripts were among others related to pseudouridine synthesis, RNA processing, and ribosome biogenesis. The differential expression of three of these transcripts, encoding cytochrome c oxidase subunit 2 (Cox2), NADH-ubiquinone oxidoreductase (ND1), and dyskeratosis congenita 1 protein (DKC1), were confirmed by real-time PCR. The promoted development and the differential gene expression of cultured sporocysts under the influence of head–foot protein extract of native O. hupensis implied not only its ability to improve in vitro culture conditions for intramolluscan stages, it may also represent a priming result with respect to the identification and characterization of factors involved in the parasite–host interplay between S. japonicum and O. hupensis.

As an emerging field of speech recognition, dialect identification plays an important role for promoting applications of speech recognition technology. Since the communications among Mainland China, Hong Kong and Taiwan are becoming frequently, it is particularly necessary to identify their dialects. This paper makes contributions to this issue in the following threefolds: 1) we build a speech corpus for main dialects of the three areas; 2) we use the popular GMM based method to extensively evaluate the main dialects between Mainland China and Hong Kong and the ones between Mainland China and Taiwan, and we find the differences between Mainland China Mandarin and Taiwan Mandarin are much smaller than those between Mandarin and Cantonese, resulting in unsatisfactory results in the latter case; 3) we propose an improved method based on the analysis of GMM, namely, maximum KL distance based Gaussian component selection (MKLD-GCS) in order to improve the performance of dialect identification between Mainland China Mandarin and Taiwan Mandarin. Experimental results show that our proposed method obtains better identification performance than related methods.

Kernel-based clustering is one of the most popular methods for partitioning nonlinearly separable dataset. However, exhaustive search for the global optimum is NP-hard. Iterative procedure such as k-means can be used to seek one of the local minima. Unfortunately, it is easily trapped into degenerate local minima when the prototypes of clusters are ill-initialized. In this paper, we restate the optimization problem of kernel-based clustering in an on-line learning framework, whereby a conscience mechanism is easily integrated to tackle the ill-initialization problem and faster convergence rate is achieved. Thus, we propose a novel approach termed conscience on-line learning (COLL). For each randomly taken data point, our method selects the winning prototype based on the conscience mechanism to bias the ill-initialized prototype to avoid degenerate local minima, and efficiently updates the winner by the on-line learning rule. Therefore, it can more efficiently obtain smaller distortion error than k-means with the same initialization. Experimental results on synthetic and large-scale real-world datasets, as well as that in the application of video clustering, have demonstrated the significant improvement over existing kernel clustering methods.

To develop a method for primary culture of cells from Oncomelania hupensis liver, and to observe the distribution of succinate dehydrogenase (SDH) and lactate dehydrogenase (LDH) in the cultured cells.O. hupensis was anatomized to separate the liver. Livers were soaked in 0.2% benzalkonium bromide and washed by physiological saline containing antibiotics in turns. Cells from the liver were harvested by mechanical mulling and filtering. The isolated cells were then incubated with methods of the combination culture and standing suspension culture, respectively. The culture medium for the cells was a mixture of Medium 199 (50 ml), 0.3% lactoalbumin hydrolysate dissolved in a balanced salt solution (BBS, 30 ml), and fetal calf serum (FCS, 20 ml) containing a moderate amount of antibiotics (100 IU/ml penicillin, 100 microg/ml streptomycin and 50 microg/ml kanamycin) at pH 7.2-7.4 under the temperature of 26.5 degrees C. The cells were stained by using Giemsa and Pearson methods (for SDH and LDH respectively) to observe the shape of cultured cells and enzyme distribution in cells. The living and stained cells were microscopically observed.Under microscope, the attached cells incubated with method of the combination culture showed round, elliptic, triangular and irregular shapes, with more round and elliptic cells. The size was approximately (4-16) microm x (6-20) microm in average. The clustered cells with an unclear nucleus and abundant and lucid cytoplasm were smaller than diffused cells with a large, obvious nuclei and less cytoplasm. Degeneration was observed after culturing for 5-7 days. The cultured cells could be divided into two types based on the color shown after Giemsa staining. The first type cells showed blue cytoplasm and mauve nuclei while the second type cells were opposite. There were blue granules in different sizes and shade in the cytoplasm after SDH and LDH staining. It was difficult for the cells to attach the wall of the culture flask using method of the standing suspension culture. The shape of the cultured cells were almost round with unclear nuclei, and the size was about (4-6) microm x (6-8) microm in average. The cells incubated with the standing suspension method were found to be contaminated after culturing for 3 days.The combination culture method is suitable for primary culture of the cells from O. hupensis liver and the cells show activities of both SDH and LDH in cytoplasm.

To investigate the effects of KN-93, a CaMKII selective inhibitor on cell proliferation and the expression of p53 or p21 protein in human hepatic stellate cells.Human hepatic stellate cells (LX-2) were incubated with various concentrations (0-50 micromol/L) of KN-93 or its inactive derivative, KN-92. Cell proliferation was measured by CCK-8 assay, and the expression of two cell cycle regulators, p53 and p21, was determined by SDS-PAGE and Western blotting.KN-93 (5-50 micromol/L) decreased the proliferation of human hepatic stellate cells in a dose-dependent manner from 81.76% (81.76% +/- 2.58% vs 96.63% +/- 2.69%, P0.05) to 27.15% (27.15% +/- 2.86% vs 96.59% +/- 2.44%, P0.01) after 24 h treatment. Incubation of 10 micromol/L KN-93 induced the cell growth reduction in a time-dependent manner from 78.27% at 8 h to 11.48% at 48 h. However, KN-92, an inactive derivative of KN-93, did not inhibit cell proliferation effectively. Moreover, analysis of cell cycle regulator expression revealed that KN-93 rather than KN-92 reduced the expression of p53 and p21.KN-93 has potent inhibitory effect on proliferation of LX-2 cells by modulating the expression of two special cell cycle regulators, p53 and p21.

A numerical calculation method is used to predict the variation in the characteristics of a fuel spray injected into high-temperature crossflowing airstreams, namely, the variation in Sauter mean diameter, droplet size distribution parameter N of the Rosin-Rammler distribution, and evaporation percentage of the spray

The effects of three parameters: air velocity, nozzle pressure drop and injector orifice diameter, on the spray characteristics of a plain orifice injector under uniform and non-uniform cross flowing air stream have been studied experimentally. For uniform cross flowing air stream, the results show that the effects of these parameters are interrelated. The exponents of these terms in a correlation are not constants. Based on a very large amount of experimental data, the following correlation has been derived for Sauter Mean Diameter. SMD = 8.28 • 10 4 V ¯ a A • Δ P ¯ f B • d ¯ C where: A = −1.59 −0.0044ΔP̄f −0.01 d̄ B = −0.13 −0.025 d̄ +0.34 Ma C = 0.36 −0.55 Ma −0.0032ΔP̄f (Va ≤ 140 M/s ; ΔPf ≤ 11 Kg·f/cm2 ; d ≤ 2.5 mm) For small orifice diameters, the drop size distribution parameter, N (Rosin-Rammler distribution ), decreases until a minimum then increases with air velocity. For large orifice diameters, it decreases with air velocity. N always decreases with the increases of nozzle pressure drop or orifice diameter. For non-uniform cross flowing air stream, atomizations under four velocity profiles with same averaged velocity and with a velocity recess of same shape but at different radial positions have been tested. The atomization data were compared with that of uniform cross flowing air stream. Two types of comparison were made based on: a) the undisturbed velocity, b) the averaged velocity, equals to the velocity of uniform cross flowing air stream. For former situation the atomization for non-uniform cross flowing air stream tested is always poorer. The influence from the velocity recess will be maximum at certain nozzle pressure drop. The experimental evidence obtained has shown that cross flow atomization is a combination of pressure atomization (at low air flow velocity) and airblast atomization (at high air flow velocity).