Search

Effectively imaging within volumetric scattering media is of great importance and challenging especially in macroscopic applications. Recent works have demonstrated the ability to image through scattering media or within the weak volumetric scattering media using spatial distribution or temporal characteristics of the scattered field. Here, we focus on imaging Lambertian objects embedded in highly scattering media, where signal photons are dramatically attenuated during propagation and highly coupled with background photons. We address these challenges by providing a time-to-space boundary migration model (BMM) of the scattered field to convert the scattered measurements in spectral form to the scene information in the temporal domain using all of the optical signals. The experiments are conducted under two typical scattering scenarios: 2D and 3D Lambertian objects embedded in the polyethylene foam and the fog, which demonstrate the effectiveness of the proposed algorithm. It outperforms related works including time gating in terms of reconstruction precision and scattering strength. Even though the proportion of signal photons is only 0.75%, Lambertian objects located at more than 25 transport mean free paths (TMFPs), corresponding to the round-trip scattering length of more than 50 TMFPs, can be reconstructed. Also, the proposed method provides low reconstruction complexity and millisecond-scale runtime, which significantly benefits its application.

Calcium imaging is inherently susceptible to detection noise especially when imaging with high frame rate or under low excitation dosage. We developed DeepCAD, a self-supervised learning method for spatiotemporal enhancement of calcium imaging without requiring any high signal-to-noise ratio (SNR) observations. Using this method, detection noise can be effectively suppressed and the imaging SNR can be improved more than tenfold, which massively improves the accuracy of neuron extraction and spike inference and facilitate the functional analysis of neural circuits.

Calcium imaging has transformed neuroscience research by providing a methodology for monitoring the activity of neural circuits with single-cell resolution. However, calcium imaging is inherently susceptible to detection noise, especially when imaging with high frame rate or under low excitation dosage. Here we developed DeepCAD, a self-supervised deep-learning method for spatiotemporal enhancement of calcium imaging data that does not require any high signal-to-noise ratio (SNR) observations. DeepCAD suppresses detection noise and improves the SNR more than tenfold, which reinforces the accuracy of neuron extraction and spike inference and facilitates the functional analysis of neural circuits. DeepCAD is a self-supervised deep-learning approach for denoising calcium imaging data. DeepCAD improved SNR and facilitates neuron extraction and spike inference.

Light-field (LF) camera holds great promise for passive/general depth estimation benefited from high angular resolution, yet suffering small baseline for distanced region. While stereo solution with large baseline is superior to handle distant scenarios, the problem of limited angular resolution becomes bothering for near objects. Aiming for all-in-depth solution, we propose a cross-baseline LF camera using a commercial LF camera and a monocular camera, which naturally form a 'stereo camera' enabling compensated baseline for LF camera. The idea is simple yet non-trivial, due to the significant angular resolution gap and baseline gap between LF and stereo cameras. Fusing two depth maps from LF and stereo modules in spatial domain is fluky, which relies on the imprecisely predicted depth to distinguish close or distance range, and determine the weights for fusion. Alternatively, taking the unified representation for both LF and monocular sub-aperture view in epipolar plane image (EPI) domain, we show that for each pixel, the minimum variance along different shearing degrees in EPI domain estimates its depth with the highest fidelity. By minimizing the minimum variance, the depth error is minimized accordingly. The insight is that the calculated minimum variance in EPI domain owns higher fidelity than the predicted depth in spatial domain. Extensive experiments demonstrate the superiority of our cross-baseline LF camera in providing high-quality all-in-depth map from 0.2m to 100m.

During magnetic resonance imaging (MRI), the strong response to the signal is usually displayed as structural edges and textures, which is important for distinguishing different tissues and lesions. In the current superresolution (SR) methods with the usage of deep learning, some low-level structural information tends to gradually disappear as the network deepens, resulting in excessive smoothness in high-frequency regions. This phenomenon is particularly noticeable in MRI with poor brightness contrast and small gray dynamic range. Although the generative adversarial network (GAN) can repair structured textures well in natural images, it is likely to learn patterns that do not exist in the images, which poses risks to the reconstruction of medical images. Therefore, we propose an enhanced gradient guiding network (EG2N) to alleviate these problems. On the one hand, for improving the contrast and suppress the noise effectively, we use a multi-scale wavelet enhancement for preprocessing, where the enhanced gradient map is considered as the structural prior. On the other hand, blindly using dense connections in the feed-forward network will bring about redundancy, so structural features from an additional branch are added to specific layers as a supplement to high-level features and constrain optimization. We add a feedback mechanism to promote cross-layer flow between low-level and high-level features. In addition, the perceptual loss is added to avoid distortion caused by excessive smoothing. The experimental results show that our method achieves the best visual results and excellent performance compared with state-of-the-art methods on most popular MR images SR benchmarks.

In the current popular visual tasks, one single model is usually used to output the final results. But no model is perfect, in this paper, we propose a simple and general multi-model method. To combine the advantages of multiple models, we design a familiarity prediction network to output the model's familiarity of images, then select the optimal model based on the familiarity value. Since the loss value is a single value, the familiarity value of any task can be reflected in the loss value, so the output of the familiarity prediction network can be regarded as an estimate of the loss value. The accuracy of the multi-model exceeds any single model that composes the multi-model. By limiting the number of feature layers input into the familiarity network, the sacrifice of computation and detection speed is acceptable. Our method is general and task-agnostic, it not only performs well on classification tasks but also on object detection tasks and other vision tasks.

Pulmonary nodule detection is a challenging problem in biomedical imaging. Most existing approaches exploit the multi-scale features at a layer level to detect nodule. However, the effect of features at a layer level is limited. This study proposes an improved architecture unit, which we term the 3D DP2 block. Just as its name implies, it is improved by the idea of the 3D dual-path network. It combines multiscale features not only in a layer-wise manner but also at a granular level, which means it can combine global features with local and increases the scales of receptive fields. Moreover, we adopt the coordination-guided convolutional layers (CoordConvs) and design a loss function inspired by the loss of Fast R-CNN. The proposed 3D DP2 block can be easily plugged into the backbone CNN architectures such as the U-Net model without additional parameters introduced while increasing the model accuracy. Our 3D DP2 block based on U-Net is validated on a public LUNA16 dataset. It improves the nodule detection accuracy compared with the baseline model. This demonstrates that pulmonary nodule detection can highly benefit from the multi-scale features at a granular level. And the proposed 3D DP2Net should be useful to other medical detection problems.

Current deep learning methods for fluorescence image restoration necessitate numerous well-aligned image pairs for training. We develop an unsupervised learning framework for high-fidelity fluorescence image restoration without the laborious work of image annotation and registration.

In this paper, we propose a novel method called Residual Steps Network (RSN). RSN aggregates features with the same spatial size (Intra-level features) efficiently to obtain delicate local representations, which retain rich low-level spatial information and result in precise keypoint localization. Additionally, we observe the output features contribute differently to final performance. To tackle this problem, we propose an efficient attention mechanism - Pose Refine Machine (PRM) to make a trade-off between local and global representations in output features and further refine the keypoint locations. Our approach won the 1st place of COCO Keypoint Challenge 2019 and achieves state-of-the-art results on both COCO and MPII benchmarks, without using extra training data and pretrained model. Our single model achieves 78.6 on COCO test-dev, 93.0 on MPII test dataset. Ensembled models achieve 79.2 on COCO test-dev, 77.1 on COCO test-challenge dataset. The source code is publicly available for further research at https://github.com/caiyuanhao1998/RSN/.

Image hashing has attracted increasing popularity in recent years. Some off-the-shelf image hashing methods are able to generate more compact and robust hashes for fast indexing and content-based similarity retrieval. However, the ability to infer original image contents from their real-valued image hashes has seldom been examined. Inherited from cryptographic hashing for image privacy protection, general image hashing is supposed to be a non-revertible function. Should there be a way to revert (or perceptually reconstruct) images from the corresponding real-valued image hashes? This paper explores the feasibility of perceptually image hashing reversion, and fill this gap by proposing a deep learning based framework, entitled RevHashNet. Given real-valued image hashes from certain image hashing methods, the proposed RevHashNet can automatically reconstruct perceptually similar images with respect to the original ones with high visual quality. Experiments and simulations on real image datasets support the de-hashing effectiveness of the proposed RevHashNet.

The joint photographic experts group (JPEG) is the most popular format of digital image. In this paper, a novel reversible data hiding scheme based on JPEG image is proposed. In JPEG image, quantified DCT (discrete cosine transform) coefficients with different frequencies will yield different capacities and embedding distortions. To reduce the total distortion for the marked image, we select coefficients from frequencies yielding less distortions for embedding, and then an advanced block selection strategy is applied to always modify the block yielding less simulated distortion firstly until the given payloads are completely embedded. Experimental results show that the proposed method can keep good visual quality with small bitstream expansion.

Single-molecule localization microscopy (SMLM) can bypass the diffraction limit of optical microscopes and greatly improve the resolution in fluorescence microscopy. By introducing the point spread function (PSF) engineering technique, we can customize depth varying PSF to achieve higher axial resolution. However, most existing 3D single-molecule localization algorithms require excited fluorescent molecules to be sparse and captured at high signal-to-noise ratios, which results in a long acquisition time and precludes SMLM’s further applications in many potential fields. To address this problem, we propose a novel 3D single-molecular localization method based on a multi-channel neural network based on U-Net. By leveraging the deep network’s great advantages in feature extraction, the proposed network can reliably discriminate dense fluorescent molecules with overlapped PSFs and corrupted by sensor noise. Both simulated and real experiments demonstrate its superior performance in PSF engineered microscopes with short exposure and dense excitations, which holds great potential in fast 3D super-resolution microscopy.

In this paper, we propose a novel method for 4D light-field (LF) depth estimation exploiting the special linear structure of an epipolar plane image (EPI) and locally linear embedding (LLE). Without high computational complexity, depth maps are locally estimated by locating the optimal slope of each line segmentation on the EPIs, which are projected by the corresponding scene points. For each pixel to be processed, we build and then minimize the matching cost that aggregates the intensity pixel value, gradient pixel value, spatial consistency, as well as reliability measure to select the optimal slope from a predefined set of directions. Next, a subangle estimation method is proposed to further refine the obtained optimal slope of each pixel. Furthermore, based on a local reliability measure, all the pixels are classified into reliable and unreliable pixels. For the unreliable pixels, LLE is employed to propagate the missing pixels by the reliable pixels based on the assumption of manifold preserving property maintained by natural images. We demonstrate the effectiveness of our approach on a number of synthetic LF examples and real-world LF data sets, and show that our experimental results can achieve higher performance than the typical and recent state-of-the-art LF stereo matching methods.

To improve imaging speed, multifocal excitation is widely adopted as a parallel strategy in laser-scanning microscopy. Specifically, axial multifocal microscopy is popular in neuroscience as it enables functional imaging of neurons in multiple depths simultaneously. However, previous phase searching algorithms for axial multi-foci generation generally generate foci of uniform intensities, which cannot compensate the scattering-induced power loss in deep tissue and causes inhomogeneous excitation. Here, we propose a novel adaptive optimization-based phase-searching method (AdaPS) to generate axial multi-foci with arbitrary intensity modulations for scattering-induced loss compensation. By adopting Adaptive Moment Estimation (Adam) as the searching algorithm, our method could escape from unsatisfactory local minima and stably converge to the optimal phase pattern with errors at least an order of magnitude lower. We validate AdaPS through both numerical simulations and experiments and demonstrate that AdaPS could provide uniform multi-depth imaging in scattering phantom and enable high-fidelity multi-depth recordings of neural network dynamics in mouse brain in vivo.

Due to the trade off between model complexity and estimation accuracy, current human pose estimators either are of low accuracy or requires long running time. Such dilemma is especially severe in real time multi-person pose estimation. To address this issue, we design a deep network of reduced parameter size and high estimation accuracy, via introducing deconvolution layers instead of widely used fully-connected configuration. Specifically, our model consists of two successive parts: Detection network and matching network. The former outputs keypoint heatmap and person locations, and then the latter produces the final pose estimation using multiple deconvolutional layers. Benefiting from the simple structure and explicit utilization of previously neglected spatial structure in heatmap, the matching network is of specially high efficiency and at high precision. Experiments on the challenging COCO dataset demonstrate our method can almost cut off the running parameters of matching network, while achieving higher accuracy than existing methods.

The development of deep learning and open access to a substantial collection of imaging data together provide a potential solution for computational image transformation, which is gradually changing the landscape of optical imaging and biomedical research. However, current implementations of deep learning usually operate in a supervised manner, and their reliance on laborious and error-prone data annotation procedures remains a barrier to more general applicability. Here, we propose an unsupervised image transformation to facilitate the utilization of deep learning for optical microscopy, even in some cases in which supervised models cannot be applied. Through the introduction of a saliency constraint, the unsupervised model, named Unsupervised content-preserving Transformation for Optical Microscopy (UTOM), can learn the mapping between two image domains without requiring paired training data while avoiding distortions of the image content. UTOM shows promising performance in a wide range of biomedical image transformation tasks, including in silico histological staining, fluorescence image restoration, and virtual fluorescence labeling. Quantitative evaluations reveal that UTOM achieves stable and high-fidelity image transformations across different imaging conditions and modalities. We anticipate that our framework will encourage a paradigm shift in training neural networks and enable more applications of artificial intelligence in biomedical imaging.

The face liveness detection module is one of the most important parts in the state-of-the-art face recognition system. In this paper, we present an efficient method to further improve its accuracy by leveraging multiple feature descriptors. Firstly, a data-driven feature descriptor is proposed based on the Karhunen-Loeve Transform (KLT) learned from both client and imposter face images. Moreover, the Completed Local Binary Pattern (CLBP) algorithm is utilized to represent the local structure and the high-middle spectra components of 2D Fourier transform are also utilized to reflect the global structure. These features are fed into the support vector machine (SVM) to learn a classifier for face liveness detection. Experimental results on NUAA illustrate that our proposed method outperforms most of the widely utilized feature descriptors.

Retinex theory is developed mainly to decompose an image into the illumination and reflectance components by analyzing local image derivatives. In this theory, larger derivatives are attributed to the changes in reflectance, while smaller derivatives are emerged in the smooth illumination. In this paper, we utilize exponentiated local derivatives (with an exponent {\gamma}) of an observed image to generate its structure map and texture map. The structure map is produced by been amplified with {\gamma} > 1, while the texture map is generated by been shrank with {\gamma} < 1. To this end, we design exponential filters for the local derivatives, and present their capability on extracting accurate structure and texture maps, influenced by the choices of exponents {\gamma}. The extracted structure and texture maps are employed to regularize the illumination and reflectance components in Retinex decomposition. A novel Structure and Texture Aware Retinex (STAR) model is further proposed for illumination and reflectance decomposition of a single image. We solve the STAR model by an alternating optimization algorithm. Each sub-problem is transformed into a vectorized least squares regression, with closed-form solutions. Comprehensive experiments on commonly tested datasets demonstrate that, the proposed STAR model produce better quantitative and qualitative performance than previous competing methods, on illumination and reflectance decomposition, low-light image enhancement, and color correction. The code is publicly available at https://github.com/csjunxu/STAR., Comment: 16 pages, 13 figures, 3 tables, accepted by TIP

The orientation and wobbling behavior of the fluorescent dipoles are of great significance in revealing the structure and state of cells. Due to the poor optical sectioning capability of wide-field microscopy, the polarization modulation signals are susceptible to the neighboring fluorophores. The missing cone of wide field optical transfer function induces vast out-of-focus background, resulting in biased polarization orientation and decrease polarization factor. Here, we apply polarized structured illumination to achieve polarization modulation imaging with optical sectioning, and simultaneously measure the lipid polarity with two-color ratiometric imaging. Our results demonstrate a significant increase in measurement accuracy of not only the dipole orientations but also the wobbling behavior of the ensemble dipole. Compared to the conventional confocal polarization imaging, our method obtains an order-of-magnitude faster imaging speed, capturing the fast dynamics of subcellular structures in live cells.

In this paper, we propose a dense-residual attention network for skin lesion segmentation. The proposed network is end-to-end and doesn’t need any post-processing operations or pretrained weights to fine-tune. Specifically, we propose the dense-residual block in our network to deal with the problem of fixed receptive field and meanwhile ease the gradient vanishing problem (as often occurred in convolution neural networks). Moreover, an attention gate is designed to enhance the network discriminative ability and ensure the efficiency of feature learning. During the network training, we introduce a novel loss function based on the jaccard distance to tackle the class imbalance issue in medical datasets. The proposed network achieves the state-of-the-art performance on the benchmark ISIC 2017 Challenge dataset without any external training samples. Experimental results show the effectiveness of our dense-residual attention network.

Diffuse correlation spectroscopy (DCS) is a well-established method that measures rapid changes in scattered coherent light to identify blood flow and functional dynamics within tissue. While its sensitivity to minute scatterer displacements leads to a number of unique advantages, conventional DCS systems become photon-limited when attempting to probe deep into tissue, which leads to long measurement windows (∼1 sec). Here, we present a high-sensitivity DCS system with 1024 parallel detection channels integrated within a single-photon avalanche diode (SPAD) array, and demonstrate the ability to detect mm-scale perturbations up to 1 cm deep within a tissue-like phantom at up to 33 Hz sampling rate. We also show that this highly parallelized strategy can measure the human pulse at high fidelity and detect behaviorally-induced physiological variations from above the human prefrontal cortex. By greatly improving detection sensitivity and speed, highly parallelized DCS opens up new experiments for high-speed biological signal measurement.

Vertical tearing of oceanic slabs has been well documented at subduction zones. It remains unclear whether a subducting continent can tear vertically. Origin of the continental-scale Tan–Lu Fault Zone (TLFZ) in East China provides an example of the vertical tearing of the subducting continent at the convergent stage. The Dabie and Sulu orogens between the North China Plate (NCP) and the Yangtze Plate (YZP) are left-laterally offset about 400 km along the NE-striking TLFZ, but the fault zone terminates abruptly at the southeastern corner of the Dabie Orogen, suggesting unique origin of the fault zone. We present structural evidence to show that the TLFZ initiated as a sinistral fault zone that is exposed in ductile shear belts just to the east of the Dabie Orogen. Our muscovite 40Ar/39Ar dating results, coupled with existing age data, indicate that the TLFZ formed at 240–230 Ma (Middle Triassic), when the YZP continental crust was subducted beneath the NCP along both the Dabie and Sulu sutures. We also show that SE-directed extrusion of the subducted crust in the Dabie Orogen occurred during 230–209 Ma (Late Triassic) and overprinted contractional deformation in the TLFZ to the southeast of the Dabie Orogen. Syn-collisional folds and thrusts in the YZP, to the east of the TLFZ, exhibit evidence for large-scale dragging by the sinistral fault zone, whereas those in the NCP to the west are perpendicular to the TLFZ without obvious evidence for drag. Combining these lines of evidence with published data on the tectonic evolution of the two orogenic belts, we propose an indentation-induced continent-tearing model for the origin of the TLFZ. We suggest that the present southern boundary of the NCP represents its original shape, with a promontory in front of the Dabie Orogen. At the oceanic subduction stage, the overriding NCP promontory led to vertical tearing of the subducting oceanic plate along the TLFZ, which formed the eastern boundary of the promontory. Once the rigid NCP indenter collided with the passive YZP along the Dabie Orogen, the oceanic slab tear propagated into the YZP lithosphere, resulting in long-distance, low-angle subduction of the YZP underneath the NCP indenter and NNE-ward, horizontal motion of the torn YZP east of the TLFZ until the final collision at the Sulu Orogen. Thus, the indentation induced the vertical tearing of the subducting YZP along the TLFZ.

Patch based image modelings have shown great potential in image denoising. They mainly exploit the nonlocal self-similarity (NSS) of either input degraded images or clean natural ones when training models, while failing to learn the mappings between them. More seriously, these algorithms have very high time complexity and poor robustness when handling images with different noise variances and resolutions. To address these problems, in this paper, we propose very deeply cascaded residual networks (VDCRN) to build the precise relationships between the noisy images and their corresponding noise-free ones. It adopts a new residual unit with an identity skip connection (shortcut) to make training easy and improve generalization. The introduction of shortcut is helpful to avoid the problem of gradient vanishing and preserve more image details. By cascading three such residual units, we build the VDCRN to deploy deeper and larger convolutional networks. Based on such a residual network, our VDCRN achieves very fast speed and good robustness. Experimental results demonstrate that our model outperforms a lot of state-of-the-art denoising algorithms quantitively and qualitively.

We propose a novel method for multi-person 2D pose estimation. Our model zooms in the image gradually, which we refer to as the Magnify-Net, to solve the bottleneck problem of mean average precision (mAP) versus pixel error. Moreover, we squeeze the network efficiently by an inspired design that increases the mAP while saving the processing time. It is a simple, yet robust, bottom-up approach consisting of one stage. The architecture is designed to detect the part position and their association jointly via two branches of the same sequential prediction process, resulting in a remarkable performance and efficiency rise. Our method outcompetes the previous state-of-the-art results on the challenging COCO key-points task and MPII Multi-Person Dataset.

Deep neural networks have demonstrated their power in many computer vision applications. State-of-the-art deep architectures such as VGG, ResNet, and DenseNet are mostly optimized by the SGD-Momentum algorithm, which updates the weights by considering their past and current gradients. Nonetheless, SGD-Momentum suffers from the overshoot problem, which hinders the convergence of network training. Inspired by the prominent success of proportional-integral-derivative (PID) controller in automatic control, we propose a PID approach for accelerating deep network optimization. We first reveal the intrinsic connections between SGD-Momentum and PID based controller, then present the optimization algorithm which exploits the past, current, and change of gradients to update the network parameters. The proposed PID method reduces much the overshoot phenomena of SGD-Momentum, and it achieves up to 50% acceleration on popular deep network architectures with competitive accuracy, as verified by our experiments on the benchmark datasets including CIFAR10, CIFAR100, and Tiny-ImageNet.

The Reference-based Super-resolution (RefSR) super-resolves a low-resolution (LR) image given an external high-resolution (HR) reference image, where the reference image and LR image share similar viewpoint but with significant resolution gap (\(8{\times }\)). Existing RefSR methods work in a cascaded way such as patch matching followed by synthesis pipeline with two independently defined objective functions, leading to the inter-patch misalignment, grid effect and inefficient optimization. To resolve these issues, we present CrossNet, an end-to-end and fully-convolutional deep neural network using cross-scale warping. Our network contains image encoders, cross-scale warping layers, and fusion decoder: the encoder serves to extract multi-scale features from both the LR and the reference images; the cross-scale warping layers spatially aligns the reference feature map with the LR feature map; the decoder finally aggregates feature maps from both domains to synthesize the HR output. Using cross-scale warping, our network is able to perform spatial alignment at pixel-level in an end-to-end fashion, which improves the existing schemes [1, 2] both in precision (around 2 dB–4 dB) and efficiency (more than 100 times faster).

Visual object tracking is a fundamental task in computer vision with its wide range of applications. In this paper, we propose a robust algorithm based on the kernelized correlation filter framework to handle occlusions or scale variations. Our algorithm takes into account the relationships between the target object and its surrounding context, and learns a discriminative correlation filter for the estimation of the new position. Another discriminative regression model via constructing the target pyramid is introduced to estimate the optimal scale. The proposed algorithm integrated with two discriminative regression models can track complex targets with occlusion and deformation at real-time. The competitive experimental results on the dataset sequences show that the proposed tracker outperforms other state-of-the-art methods, in both the precision and the success rate.

In this paper, we propose a novel end-to-end deep neural network for region-enhanced depthmap reconstruction from a single face image. Unlike most of popular depthmap reconstruction methods, the proposed network fully takes the region information of RGB images into consideration, and thus results in more accurate correspondences. Specially, we treat the depthmap reconstruction task as an independent problem, investigating to improve facial depth regression through multi-task learning (facial landmark detection and mask detection), and then a attention region training technique to utilize the rich texture information in some semantic regions to infer local detailed depth information. Next, we describe a zigzag dilated convolution framework to sparse the kernel size of our network, which will enlarge the network’s receptive field and simultaneously avoid gridding artifacts. In training, we also propose a new method for setting the learning rates, called discrete cyclical learning rates, to improve the results. Experimental results illustrate that the proposed Sparse-Region Net (SRN) outperforms the state-of-the-art baseline methods by a large margin.

We report the identification, characterization, and gene cloning of a novel protein elicitor (PeBL1) secreted from Brevibacillus laterosporus strain A60. Through a purification process consisting of ion-exchange chromatography and high-performance liquid chromatography (HPLC), we isolated a protein that was identified by electrospray ionization quadrupole time of flight tandem mass spectrometry (ESI–Q-TOF–MS-MS). The 351-bp PeBL1 gene produces a 12,833-Da protein with 116 amino acids that contains a 30-residue signal peptide. The PeBL1 protein was expressed in Escherichia coli . The recombinant protein can induce a typical hypersensitive response (HR) and systemic resistance in Nicotiana benthamiana , like the endogenous protein. PeBL1-treated N. benthamiana exhibited strong resistance to the infection of tobacco mosaic virus-green fluorescent protein (TMV-GFP) and Pseudomonas syringae pv. tabaci compared to control N. benthamiana . In addition, PeBL1 triggered a cascade of events that resulted in defense responses in plants, including reactive oxygen species (ROS) production, extracellular-medium alkalization, phenolic-compound deposition, and expression of several defense-related genes. Real-time quantitative-PCR analysis indicated that the known defense-related genes PR-1 , PR-5 , PDF1.2 , NPR1 , and PAL were upregulated to varying degrees by PeBL1. This research not only provides insights into the mechanism by which beneficial bacteria activate plant systemic resistance, but also sheds new light on a novel strategy for biocontrol using strain A60.

Most state-of-the-art results on image classification tasks were obtained by residual neural networks, which use stochastic gradient descent (SGD) with momentum for training. In most cases, the learning rate drops by a constant factor every pre-defined number of epochs. However, it is difficult and time-consuming to estimate how many epochs to drop the learning rate. To tackle this problem, cyclical learning rate is gaining popularity in gradient-based optimization to improve the convergence speed in accelerated gradient schemes. But cyclical learning rate scheme scans a broad range of learning rate, some of which are not suitable for deep neural network training. In this paper, we propose a simple yet effective exponential decay sine wave like learning rate technique for SGD to improve its convergence speed. In the training process, the learning rate would vary in sine wave way. While the maximum value of sine wave would decay exponentially along with training epochs. An ensemble of wide residual nets with our proposed learning scheme achieves 3.01% and 16.03% errors on CIFAR-10 and CIFAR-100 respectively. Furthermore, our proposed method uses far less number of epochs than most recent learning rate strategies, accelerating neural network training tremendously.

In this paper, we propose a cascaded image deblurring scheme that adopts combined image prior followed by multi-scale separable kernel refinement. Specifically, we introduce a combined image prior made up of Aggressive Patch Recurrence (APR) prior and sparsity image prior, where the sparsity prior works as a good compensation for cases where APR prior fails to maintain the self-similarity assumption. We further propose a multi-scale separable kernel refinement method, which can effectively remove noise and ringing artifacts from kernels. Experimental results show that our cascaded deblurring scheme is capable of estimating noise-free blur kernels and achieves visually pleasant resultant images without obvious ringing artifacts when deblurring both synthetic and real-world blurry images.

We propose a new method to super-resolve images captured by a hybrid light field system that consists of a standard light field camera and a high-resolution standard camera. The high-resolution image is taken as a reference to help with super-resolving the low-resolution light field images. Our method combines an exemplar-based algorithm with the state of-the-art single image super-resolution approach and draws on the strengths of both. Both quantitative and qualitative experiments show that our proposed method substantially outperforms existing methods on standard light field datasets in the challenging large parallax setting.

In this paper, we propose a saliency prediction algorithm utilizing generative adversarial networks. The proposed system contains two parts: saliency network and adversarial networks. The saliency network is the basis for saliency prediction, which calculates an Euclidean cost function on the grayscale values between the predicted saliency map and the ground truth. In order to improve the accuracy of the algorithm, adversarial networks are subsequently utilized to extract the features of input data by coordinating the learning rates of the two sub-networks contained in the networks. Experimental results validate the high accuracy of the proposed approach compared with the state-of-the-art models on three public datasets, SALICON, MIT1003 and Cerf.

In this paper, an in-depth analysis on the statistical distribution characteristics of human tongue color that aims to propose a mathematically described tongue color space for diagnostic feature extraction is presented. Three characteristics of tongue color space, i.e., tongue color gamut that defines the range of colors, color centers of 12 tongue color categories, and color distribution of typical image features in the tongue color gamut, are elaborately investigated in this paper. Based on a large database, which contains over 9000 tongue images collected by a specially designed noncontact colorimetric imaging system using a digital camera, the tongue color gamut is established in the CIE chromaticity diagram by an innovatively proposed color gamut boundary descriptor using one-class SVM algorithm. Thereafter, centers of 12 tongue color categories are defined accordingly. Furthermore, color distributions of several typical tongue features, such as red points and petechial points, are obtained to build a relationship between the tongue color space and color distributions of various tongue features. With the obtained tongue color space, a new color feature extraction method is proposed for diagnostic classification purposes, with experimental results validating its effectiveness.

Stereo interleaving video coding, in which both left and right view frames are subsampled into half size and multiplexed into one single frame before being encoded by a traditional 2-D video encoder, is an efficient encoding scenario for stereoscopic video. Many existing stereo interleaving video coding methods subsample each frame by utilizing fixed subsampling filter coefficients. Such methods are easy to implement; however, the varying property of the frame signal is ignored. By jointly considering the influences of subsampling and compression, a rate and distortion analysis about stereo interleaving video coding is proposed. The final distortion in stereo interleaving video coding is the summation of errors caused by subsampling (causing distortion between subsampling-interpolated image and the original full resolution one) and by quantization during compression. Based on the provided rate distortion analysis, a content adaptive image subsampling (CAIS) is also proposed. In CAIS, the half-size frames are generated by the optimal subsampling filters, which are calculated based on frame contents and the targeted interpolation coefficients. Experimental results demonstrate that the proposed CAIS is able to greatly improve compression efficiency of stereo interleaving video coding.

In this paper, we present an accurate saliency detection algorithm based on depth feature for 3D images. We first calculate depth cue based on the sharp regions' positions within the depth ranges. Then, the coarse saliency map is computed based on the background and location prior. Finally, we employ the contrast information in the coarse saliency map to obtain the final result. Experimental evaluation by comparison with existed methods verifies the effectiveness of our proposed algorithm in terms of precision, recall and F-Measure.

There is an increasing need for high-quality multimedia applications based on block-based hybrid video coding. Inevitably, the frame will degrade during the process of block-wise intra/inter prediction, transformation, and quantization, especially when the bit rate is low. In this paper, we propose an efficient decompressed video enhancement algorithm based on the adjusted anchored neighborhood regression (A+) method. In our work, first, we learn offline linear regressors, i.e. projection matrices from the decompressed to original video frames in the training phase. For grouping anchored neighborhoods more accurately, we adopt MI-KSVD rather than KSVD to learn the dictionary. Moreover, we exploit the mutual coherence between dictionary atoms and training samples to find the nearest neighbors. Second, in the enhancement phase, we boost the quality of input decompressed videos offline by learned regression priors. To verify the robustness of our enhancement method, extensive experiments are conducted. As shown in our experimental results, the proposed enhancement method yields superior performance both objectively and subjectively.

Block-wise intra/inter prediction, transformation and quantization used in block-based hybrid video coding will inevitably result in blocking artifacts, especially at the low bit rate. To address this problem, this paper employs a deep convolutional neural network (CNN) to approximate the reverse function of video compression, motived by the great success of deep learning in computer vision fields recently. The proposed method establishes an end-to-end mapping, represented as the CNN, which takes the decompressed frame as input and outputs the enhanced one. Employing numerous sequences compressed by H.264 and HEVC reference software, the proposed CNN learns the connections between the lossy frame and the original one in an implicit way under different quantization parameters (QP). Figure 1 shows the architecture of our CNN and the pipeline of the network training. We build our network with convolution layers and ReLU layer and the weights and biases of all the convolution layers in our model are updated by minimizing the loss using stochastic gradient descent with the standard backpropagation. We implement the CNN as a post-loop deblocking filter and explore varying CNN parameters for different QPs. Various experimental results demonstrate that the proposed method is able to significantly improve the quality of enhanced frames in terms of both objective and subjective criterions.

Most of popular image denoising approaches exploit either the internal priors or the priors learned from external clean images to reconstruct the latent image. However, it is hard for those algorithms to construct the perfect connections between the clean images and the noisy ones. To tackle this problem, we present a deep convolutional-pairs network DCPN for image denoising in this paper. With the observation that deeper networks improve denoising performance, we propose to use deeper networks than those employed previously for low-level image processing tasks. In our method, we attempt to build end-to-end mappings directly from a noisy image to its corresponding noise-free image by using deep convolutional layers in pair applied to image patches. Because of those mappings trained from large data, the process of denoising is much faster than other methods. DCPN is composed of three convolutional-pairs layers and one transitional layer. Two convolutional-pairs layers are used for encoding and the other one is used for decoding. Numerical experiments show that the proposed method outperforms many state-of-the-art denoising algorithms in both speed and performance.

The Waziyu metamorphic core complex is situated at the eastern end of the Yanshan tectonic belt. The NNE-striking detachment ductile shear zone in the core complex lies between the Archean metamorphic basement and Fuxin-Yixian rift basin, dips NW gently, and shows corrugation folds. Exposure structures, microstructures, and quartz C-axis fabrics all indicate top-to-the WNW sense of shear, i.e., ca. 285°, for the shear zone. Estimates of the deformation temperatures (ca. 550−250°C) demonstrate its mid-crustal origination and progressive deformation from deep to shallow levels. The northern segment of the shear zone shows relatively weak exhumation with exposures of low-temperature mylonites whereas its middle and southern segments have more intense uplifting with exposures of high-temperature mylonites. Biotite and muscovite 40Ar/39Ar ages, U-Pb dating results of zircon from dikes and plutons as well as formation ages of the supra-detachment basin all suggest the formation time of 135–100 Ma for the core complex. The formation was also associated with syntectonic emplacement of the Early Cretaceous Shishan pluton. The western margin of the core complex was truncated by the Sunjiawan-Shaohuyingzi brittle normal fault when it uplifted to shallow crust levels, and finally exhumed to near-surface levels. The core complex was developed by the rolling-hinge model under WNW-ESE extension during the Early Cretaceous peak destruction of the North China Craton. Ductile flow did not appear in the lower plate, therefore not supporting the low-crust gravitational collapse.

We propose a novel light field super-resolution framework based on hybrid imaging system, which combines two different imaging mechanisms: conventional imaging and current art-of-the-state imaging — light field imaging. We take advantage of conventional imaging in spatial resolution to make up light field and reconstruct a higher quality light field. In our method, we classify the points of the 3D scene: First, for highlight and occlusion, dictionary learning based interpolation is utilized, Second, for other areas, an improved patch matching algorithm is applied. As shown in experimental results, compared with four methods, which include the art-of-the-state algorithms, our approach is effective.

Specular reflection removal is indispensable to many computer vision tasks. However, most existing methods fail or degrade in complex real scenarios for their individual drawbacks. Benefiting from the light field imaging technology, this paper proposes a novel and accurate approach to remove specularity and improve image quality. We first capture images with specularity by the light field camera (Lytro ILLUM). After accurately estimating the image depth, a simple and concise threshold strategy is adopted to cluster the specular pixels into "unsaturated" and "saturated" category. Finally, a color variance analysis of multiple views and a local color refinement are individually conducted on these two categories to recover diffuse color information. Experimental evaluation by comparison with existed methods verifies the effectiveness of our proposed algorithm.

Specular reflection exists widely in photography and causes the recorded color deviating from its true value, so fast and high quality highlight removal from a single nature image is of great importance. In spite of the progress in the past decades in highlight removal, achieving wide applicability to the large diversity of nature scenes is quite challenging. To handle this problem, we propose an analytic solution to highlight removal based on an L2 chromaticity definition and corresponding dichromatic model. Specifically, this paper derives a normalized dichromatic model for the pixels with identical diffuse color: a unit circle equation of projection coefficients in two subspaces that are orthogonal to and parallel with the illumination, respectively. In the former illumination orthogonal subspace, which is specular-free, we can conduct robust clustering with an explicit criterion to determine the cluster number adaptively. In the latter illumination parallel subspace, a property called pure diffuse pixels distribution rule (PDDR) helps map each specular-influenced pixel to its diffuse component. In terms of efficiency, the proposed approach involves few complex calculation, and thus can remove highlight from high resolution images fast. Experiments show that this method is of superior performance in various challenging cases., 11 pages, 10 figures, submitted to IEEE TIP

We propose a fast single image super-resolution algorithm based on adaptive local nonparametric regression. Making use of dictionary learning and regression, we learn multiple projection matrices mapping low-resolution features to their corresponding high-resolution ones directly. Different from previous linear regression that needs some constant parameters, our method would not use extra parameters for regression. We use the mutual coherence between dictionary atom and low-resolution feature as a label to reconstruct more sophisticated high-resolution feature. As we use the same form of mutual coherence as labels in both training and testing phases, our method would lead to an adaptive local linear regression model. Moreover, we investigate the statistical property of the dictionary atoms from the training features. Utilizing the learned statistical priors, our method would not only obtain more useful dictionary atoms, but also further decrease the computational time. As shown in our experimental results, the proposed method yields high-quality super-resolution images quantitatively and visually against state-of-the-art methods.

Image denoising is the basic problem in image processing and computer vision. Pictures are always contaminated with Gaussian noise in the capture and transmission process. The adaptive dictionary learning algorithms can remove Gaussian noise very well, but it takes a lot of time in training dictionary so that these methods cannot be applied in the actual scene. So we would like to solve the problem using massive image database. Firstly in the offline stage, hash coefficients are calculated and a dictionary is trained for every image in the database. Second, in the online stage, several reference images are searched by comparing the hash coefficients. For each given noisy block the best dictionary is selected according to the best sparse operator. Finally, all blocks are recovered using these selected dictionaries thus the denoised image is obtained by weighted averaging. Experiments show that the proposed method can remove Gaussian noise very well and preserve the details and the computation complexity is significantly reduced compared with other dictionary learning algorithms.

In this paper, we study how to perform edit propagation using \( L_0 \) gradient minimization. Existing propagation methods only take simple constraints into consideration and neglects image structure information. We propose a new optimization framework making use of \( L_0 \) gradient minimization, which can globally satisfy user-specified edits as well as tackle counts of non-zero gradients. In this process, a modified affinity matrix approximation method which efficiently reduces randomness is raised. We introduce a self-adaptive re-parameterization way to control the counts based on both original image and user inputs. Our approach is demonstrated by image recoloring and tonal values adjustments. Numerous experiments show that our method can significantly improve edit propagation via \( L_0 \) gradient minimization.

We propose a new model called iterative collaborative representation ICR for image super-resolution SR. Most of popular SR approaches extract low-resolution LR features from the given LR image directly to recover its corresponding high-resolution HR features. However, they neglect to utilize the reconstructed HR image for further image SR enhancement. Based on this observation, we extract features from the reconstructed HR image to progressively upscale LR image in an iterative way. In the learning phase, we use the reconstructed and the original HR images as inputs to train the mapping models. These mapping models are then used to upscale the original LR images. In the reconstruction phase, mapping models and LR features extracted from the LR and reconstructed image are then used to conduct image SR in each iteration. Experimental results on standard images demonstrate that our ICR obtains state-of-the-art SR performance quantitatively and visually, surpassing recently published leading SR methods.

The problem of state estimation for discrete-time stochastic descriptor systems with delayed measurements is addressed. Using the method of measurements reorganisation, the recursive state estimation and corresponding filtering error covariance are derived. The calculation of the estimator includes two steps. One is for the steady estimator and the other is the ‘non-steady-state’ estimator. The former is obtained by the autoregressive moving average (ARMA) innovation model and projection formula. The latter is derived by utilising the singular value decomposition and standard Kalman filtering. The authors' concept is the measurements reorganisation which converts the time-delay system into a delay-free but with-different-measurement system, and leads to the computation saving of the estimator. A numerical example is presented to illustrate the given algorithm.