Residual Feature Attentional Fusion Network for Lightweight Chest CT Image Super-Resolution

The diagnosis of COVID-19 requires chest computed tomography(CT).High-resolution CT images can provide more diagnostic information to help doctors better diagnose the disease,so it is of clinical importance to study super-resolution(SR)algorithms applied to CT images to improve the reso-lution of CT images.However,most of the existing SR algorithms are studied based on natural images,which are not suitable for medical images;and most of these algorithms improve the reconstruction quality by increasing the network depth,which is not suitable for machines with limited resources.To alleviate these issues,we propose a residual feature attentional fusion network for lightweight chest CT image super-resolution(RFAFN).Specifically,we design a contextual feature extraction block(CFEB)that can extract CT image features more efficiently and accurately than ordinary residual blocks.In addition,we propose a feature-weighted cascading strategy(FWCS)based on attentional feature fusion blocks(AFFB)to utilize the high-frequency detail information extracted by CFEB as much as possible via selectively fusing adjacent level feature information.Finally,we suggest a global hierarchical feature fusion strategy(GHFFS),which can utilize the hierarchical features more effectively than dense concatenation by progressively aggregating the feature information at various levels.Numerous experiments show that our method performs better than most of the state-of-the-art(SOTA)methods on the COVID-19 chest CT dataset.In detail,the peak signal-to-noise ratio(PSNR)is 0.11 dB and 0.47 dB higher on CTtest1 and CTtest2 at×3 SR compared to the suboptimal method,but the number of parameters and multi-adds are reduced by 22K and 0.43G,respectively.Our method can better recover chest CT image quality with fewer computational resources and effectively assist in COVID-19.

Multi-channel fast super-resolution image reconstruction based on matrix observation model

A multi-channel fast super-resolution image reconstruction algorithm based on matrix observation model is proposed in the paper,which consists of three steps to avoid the computational complexity: a single image SR reconstruction step,a registration step and a wavelet-based image fusion. This algorithm decomposes two large matrixes to the tensor product of two little matrixes and uses the natural isomorphism between matrix space and vector space to transform cost function based on matrix-vector products model to matrix form. Furthermore,we prove that the regularization part can be transformed to the matrix formed. The conjugate-gradient method is used to solve this new model. Finally,the wavelet fusion is used to integrate all the registered highresolution images obtained from the single image SR reconstruction step. The proposed algorithm reduces the storage requirement and the calculating complexity,and can be applied to large-dimension low-resolution images.

基于RNA-based NGS检测非小细胞肺癌融合基因临床实践中国专家共识

基于RNA水平的二代测序(RNA-based next-generation sequencing, RNA-based NGS)技术已被非小细胞肺癌(non-small cell lung cancer, NSCLC)临床实践指南和专家共识推荐为融合基因的检测方法之一。NSCLC可用药靶点主要包括基因突变和融合,用于评估靶向治疗可行性的基因突变和融合基因检测均不可或缺。目前,基于DNA水平的NGS(DNA-based NGS)结合RNA-based NGS一次性同步检测基因突变和融合的技术已部分应用于临床实践。然而,RNA-based NGS检测融合基因的应用时机、应用场景和质控方面在我国仍缺乏规范和标准。本共识将进一步明确RNA-based NGS在融合基因检测中的应用时机、应用场景和质控,并给予指导性建议,推动RNA-based NGS在NSCLC临床诊疗中的应用,使患者能够最大程度地从融合基因检测中获益。

Hyperspectral Image Super-Resolution Meets Deep Learning:A Survey and Perspective

Hyperspectral image super-resolution,which refers to reconstructing the high-resolution hyperspectral image from the input low-resolution observation,aims to improve the spatial resolution of the hyperspectral image,which is beneficial for subsequent applications.The development of deep learning has promoted significant progress in hyperspectral image super-resolution,and the powerful expression capabilities of deep neural networks make the predicted results more reliable.Recently,several latest deep learning technologies have made the hyperspectral image super-resolution method explode.However,a comprehensive review and analysis of the latest deep learning methods from the hyperspectral image super-resolution perspective is absent.To this end,in this survey,we first introduce the concept of hyperspectral image super-resolution and classify the methods from the perspectives with or without auxiliary information.Then,we review the learning-based methods in three categories,including single hyperspectral image super-resolution,panchromatic-based hyperspectral image super-resolution,and multispectral-based hyperspectral image super-resolution.Subsequently,we summarize the commonly used hyperspectral dataset,and the evaluations for some representative methods in three categories are performed qualitatively and quantitatively.Moreover,we briefly introduce several typical applications of hyperspectral image super-resolution,including ground object classification,urban change detection,and ecosystem monitoring.Finally,we provide the conclusion and challenges in existing learning-based methods,looking forward to potential future research directions.

SRResNet Performance Enhancement Using Patch Inputs and Partial Convolution-Based Padding

Due to highly underdetermined nature of Single Image Super-Resolution(SISR)problem,deep learning neural networks are required to be more deeper to solve the problem effectively.One of deep neural networks successful in the Super-Resolution(SR)problem is ResNet which can render the capability of deeper networks with the help of skip connections.However,zero padding(ZP)scheme in the network restricts benefits of skip connections in SRResNet and its performance as the ratio of the number of pure input data to that of zero padded data increases.In this paper.we consider the ResNet with Partial Convolution based Padding(PCP)instead of ZP to solve SR problem.Since training of deep neural networks using patch images is advantageous in many aspects such as the number of training image data and network complexities,patch image based SR performance is compared with single full image based one.The experimental results show that patch based SRResNet SR results are better than single full image based ones and the performance of deep SRResNet with PCP is better than the one with ZP.

Half space object classification via incident angle based fusion of radar and infrared sensors

In this paper,we introduce an incident angle based fusion method for radar and infrared sensors to improve the recognition rate of complex targets under half space scenarios,e.g.,vehicles on the ground in this paper.For radar sensors,convolutional operation is introduced into the autoencoder,a“winner-take-all(WTA)”convolutional autoencoder(CAE)is used to improve the recognition rate of the radar high resolution range profile(HRRP).Moreover,different from the free space,the HRRP in half space is more complex.In order to get closer to the real situation,the half space HRRP is simulated as the dataset.The recognition rate has a growth more than 7%com-pared with the traditional CAE or denoised sparse autoencoder(DSAE).For infrared sensor,a convolutional neural network(CNN)is used for infrared image recognition.Finally,we com-bine the two results with the Dempster-Shafer(D-S)evidence theory,and the discounting operation is introduced in the fusion to improve the recognition rate.The recognition rate after fusion has a growth more than 7%compared with a single sensor.After the discounting operation,the accuracy rate has been improved by 1.5%,which validates the effectiveness of the proposed method.

Hilbert空间中的fusion-Besselian框架与拟fusion-Riesz基

fusion框架作为Hilbert空间中g-框架的特例,与g-框架有许多类似的性质.该文在已有文献的基础上,借助算子理论知识,举反例说明去掉有限维空间的条件下结论不成立,进一步给出fusion-Besselian框架的算子刻画.结合fusion-Besselian框架的算子刻画和反例1,阐明在探讨该类框架性质时,应关注其适用条件和范围.随后讨论拟fusion-Riesz基与拟Riesz基、fusion-Besselian框架之间的关系.最后讨论fusion-Besselian框架和拟fusion-Riesz基的算子扰动,所得结论补充了算子扰动方面的研究.

MRI and PET images fusion based on human retina model

The diagnostic potential of brain positron emission tomography (PET) imaging is limited by low spatial resolution. For solving this problem we propose a technique for the fusion of PET and MRI images. This fusion is a trade-off between the spectral information extracted from PET images and the spatial information extracted from high spatial resolution MRI. The proposed method can control this trade-off. To achieve this goal, it is necessary to build a multiscale fusion model, based on the retinal cell photoreceptors model. This paper introduces general prospects of this model, and its application in multispectral medical image fusion. Results showed that the proposed method preserves more spectral features with less spatial distortion. Comparing with hue-intensity-saturation (HIS), discrete wavelet transform (DWT), wavelet-based sharpening and wavelet-à trous transform methods, the best spectral and spatial quality is only achieved simultaneously with the proposed feature-based data fusion method. This method does not require resampling images, which is an advantage over the other methods, and can perform in any aspect ratio between the pixels of MRI and PET images.

Video Concept Detection Based on Multiple Features and Classifiers Fusion

The rapid growth of multimedia content necessitates powerful technologies to filter, classify, index and retrieve video documents more efficiently. However, the essential bottleneck of image and video analysis is the problem of semantic gap that low level features extracted by computers always fail to coincide with high-level concepts interpreted by humans. In this paper, we present a generic scheme for the detection video semantic concepts based on multiple visual features machine learning. Various global and local low-level visual features are systematically investigated, and kernel-based learning method equips the concept detection system to explore the potential of these features. Then we combine the different features and sub-systems on both classifier-level and kernel-level fusion that contribute to a more robust system. Our proposed system is tested on the TRECVID dataset. The resulted Mean Average Precision (MAP) score is much better than the benchmark performance, which proves that our concepts detection engine develops a generic model and performs well on both object and scene type concepts.

Fusion of multi-spectral image and panchromatic image based on support vector regression

In our study, support vector value contourlet transform is constructed by using support vector regression model and directional filter banks. The transform is then used to decompose source images at multi-scale, multi-direction and multi-resolution. After that, the super-resolved multi-spectral image is reconstructed by utilizing the strong learning ability of support vector regression and the correlation between multi-spectral image and panchromatic image. Finally, the super-resolved multi- spectral image and the panchromatic image are fused based on regions at different levels. Our experi- ments show that, the learning method based on support vector regression can improve the effect of super-resolution of multi-spectral image. The fused image preserves both high space resolution and spectrum information of multi-spectral image.

A Stable Switch Method Based on Fusion in Uncalibrated Visual Servoing

Stable switch control between multiple cameras for uncalibrated visual servoing was studied. Switch images based on fusion were presented to get the continuous dynamic image Jacobian matrix among robots and distributed visual sensors. The designed fusion algorithm is suitable to have dynamically adjustable fusion weights, and the fusion structure was analyzed. Simulations and experiments without any knowledge of mobile robots and uncalibrated visual sensors show that the method has higher adaptability than the traditional instant switch control method. The method can enhance the system stability at the switching process.

Fast Image Retrieval of Textile Industrial Accessory Based on Multi-Feature Fusion

A hierarchical retrieval scheme of the accessory image database is proposed based on textile industrial accessory contour feature and region feature. At first smallest enclosed rectangle[1] feature (degree of accessory coordination) is used to filter the image database to decouple the image search scope. After the accessory contour information and region information are extracted, the fusion multi-feature of the centroid distance Fourier descriptor and distance distribution histogram is adopted to finish image retrieval accurately. All the features above are invariable under translation, scaling and rotation. Results from the test on the image database including 1,000 accessory images demonstrate that the method is effective and practical with high accuracy and fast speed.

Satellite-based detection of 16.76 MeV γ-ray from H-bomb D-T fusion

Based on the high energy γ-ray yield from the H-bomb D-T fusion reaction, it brings forward the idea applying the 16.76 MeV γ-ray to judge whether the H-bomb happens or not, and to deduce the explosion TNT equivalent accurately. The Monte Carlo N-Particle was applied to simulate the high energy γ-ray radiation characteristics reaching the geosynchronous orbit satellite, and the CVD diamond detector suit for the requirements was researched. A series of experiments were carried out to testify the capabilities of the diamond detector. It provides a brand-new approach to satellite-based nuclear explosion detection.

Fusion network for small target detection based on YOLO and attention mechanism

Target detection is an important task in computer vision research, and such an anomaly detection and the topic of small target detection task is more concerned. However, there are still some problems in this kind of researches, such as small target detection in complex environments is susceptible to background interference and poor detection results. To solve these issues, this study proposes a method which introduces the attention mechanism into the you only look once(YOLO) network. In addition, the amateur-produced mask dataset was created and experiments were conducted. The results showed that the detection effect of the proposed mothed is much better.

A MAP Approach for Vision-based Self-localization of Mobile Robot

一当场，自我本地化系统为在有深入的 3D 里程碑的 3D 环境起作用的活动机器人被开发。机器人通过合并从 odometry 和单向性的照相机收集的信息的一个地图评估者递归地估计它的姿势。我们为这二个传感器造非线性的模型并且坚持说机器人运动和不精密的传感器大小的无常操作应该全部被嵌入并且追踪我们的系统。我们在一个概率的几何学观点和使用 unscented 变换描述无常框架宣传无常，它经历给定的非线性的功能。就我们的机器人的处理力量而言，图象特征在相应投射特征的附近被提取。另外，数据协会被统计距离评估。最后，一系列系统的实验被进行证明我们的系统的可靠、精确的性能。

On-line detection system of weld-based rapid prototyping process based on structured light

In order to realize the closed-loop control for rapid prototyping process based on gas metal arc welding, the geometric parameters of weld beads should be detected. In this study, a vision sensor system consisting of a linear laser projector and a CCD camera was designed to collect images of weld beads. Then, an image processing approach which combines with a Gaassian filter and an improved gravity method was used to extract the centerline of a light stripe based on VC ＋＋. Feature points of the centerline were identical directly by means of an image fusion algorithm. Experimental results show that image fusion is an effective approach to measure the width and height of the weld bead with high accuracy. This method can identify beads effectively in multi-pass welding and avoid designing different modes to suit all kinds of shapes.

Microstructures and properties of Co-based alloy coatings prepared on surface of H13 steel

The Co-based alloy coatings had been prepared by laser cladding and vacuum fusion sintering. Microstructures of the coatings were investigated and the performance of thermal cycling was also tested using scanning electron microscopy （ SEM） and X-ray diffraction （ XRD ）. The results show that the coatings and substrates combine well. The main phase compositions of laser cladding coating are T-Co, Cr23 C6 and Ni2 9 Cro. 7 Feo. 36, while vacuum fusion sintering coating consists of Co, Cr7 C3, and Ni2.9 Cro. 7 Feo. 36. After thermal cycling, the minimum hot cracking width of laser cladding coating is 14 μm; moreover, laser cladding coating maintains high hardness and hot-cracking susceptibility. Those are beneficial to high temperature wear resistance of hot work dies.

Semantic-Based Video Retrieval Survey

There is a tremendous growth of digital data due to the stunning progress of digital devices which facilitates capturing them. Digital data include image, text, and video. Video represents a rich source of information. Thus, there is an urgent need to retrieve, organize, and automate videos. Video retrieval is a vital process in multimedia applications such as video search engines, digital museums, and video-on-demand broadcasting. In this paper, the different approaches of video retrieval are outlined and briefly categorized. Moreover, the different methods that bridge the semantic gap in video retrieval are discussed in more details.

近地小行星天地基协同监测和轨道确定试验

基于中国科学院国家天文台空间碎片试验望远镜,联合“仰望一号”“吉林一号”卫星,设计并开展了近地小行星天地基协同监测试验,利用图像处理和定轨方法,实现了天地基观测图像的目标检测和天文定位,完成了近地小行星轨道的精确确定。经分析,基于现有的天地基设备可实现对近地小行星的协同监测和轨道编目,为研判近地小行星撞击风险提供轨道数据支持。

基于动态视觉传感器的无人机目标检测与避障

针对无人机在动态环境中感知动态目标与躲避高速动态障碍物,提出了基于动态视觉传感器的目标检测与避障算法。设计了滤波方法和运动补偿算法,滤除事件流中背景噪声、热点噪声及由相机自身运动产生的冗余事件;设计了一种融合事件图像和RGB图像的动态目标融合检测算法,保证检测的可靠性。根据检测结果对目标运动轨迹进行估计,结合障碍物运动特点和无人机动力学约束改进速度障碍法躲避动态障碍物。大量仿真试验、手持试验及飞行试验验证了所提算法的可行性。