New multi-DSP parallel computing architecture for real-time image processing

The flexibility of traditional image processing system is limited because those system are designed for specific applications. In this paper, a new TMS320C64x-based multi-DSP parallel computing architecture is presented. It has many promising characteristics such as powerful computing capability, broad I/O bandwidth, topology flexibility, and expansibility. The parallel system performance is evaluated by practical experiment.

Effective distributed convolutional neural network architecture for remote sensing images target classification with a pre-training approach

How to recognize targets with similar appearances from remote sensing images(RSIs) effectively and efficiently has become a big challenge. Recently, convolutional neural network(CNN) is preferred in the target classification due to the powerful feature representation ability and better performance. However,the training and testing of CNN mainly rely on single machine.Single machine has its natural limitation and bottleneck in processing RSIs due to limited hardware resources and huge time consuming. Besides, overfitting is a challenge for the CNN model due to the unbalance between RSIs data and the model structure.When a model is complex or the training data is relatively small,overfitting occurs and leads to a poor predictive performance. To address these problems, a distributed CNN architecture for RSIs target classification is proposed, which dramatically increases the training speed of CNN and system scalability. It improves the storage ability and processing efficiency of RSIs. Furthermore,Bayesian regularization approach is utilized in order to initialize the weights of the CNN extractor, which increases the robustness and flexibility of the CNN model. It helps prevent the overfitting and avoid the local optima caused by limited RSI training images or the inappropriate CNN structure. In addition, considering the efficiency of the Na¨?ve Bayes classifier, a distributed Na¨?ve Bayes classifier is designed to reduce the training cost. Compared with other algorithms, the proposed system and method perform the best and increase the recognition accuracy. The results show that the distributed system framework and the proposed algorithms are suitable for RSIs target classification tasks.

Ultra-lightweight CNN design based on neural architecture search and knowledge distillation: A novel method to build the automatic recognition model of space target ISAR images

In this paper,a novel method of ultra-lightweight convolution neural network(CNN)design based on neural architecture search(NAS)and knowledge distillation(KD)is proposed.It can realize the automatic construction of the space target inverse synthetic aperture radar(ISAR)image recognition model with ultra-lightweight and high accuracy.This method introduces the NAS method into the radar image recognition for the first time,which solves the time-consuming and labor-consuming problems in the artificial design of the space target ISAR image automatic recognition model(STIIARM).On this basis,the NAS model’s knowledge is transferred to the student model with lower computational complexity by the flow of the solution procedure(FSP)distillation method.Thus,the decline of recognition accuracy caused by the direct compression of model structural parameters can be effectively avoided,and the ultralightweight STIIARM can be obtained.In the method,the Inverted Linear Bottleneck(ILB)and Inverted Residual Block(IRB)are firstly taken as each block’s basic structure in CNN.And the expansion ratio,output filter size,number of IRBs,and convolution kernel size are set as the search parameters to construct a hierarchical decomposition search space.Then,the recognition accuracy and computational complexity are taken as the objective function and constraint conditions,respectively,and the global optimization model of the CNN architecture search is established.Next,the simulated annealing(SA)algorithm is used as the search strategy to search out the lightweight and high accuracy STIIARM directly.After that,based on the three principles of similar block structure,the same corresponding channel number,and the minimum computational complexity,the more lightweight student model is designed,and the FSP matrix pairing between the NAS model and student model is completed.Finally,by minimizing the loss between the FSP matrix pairs of the NAS model and student model,the student model’s weight adjustment is completed.Thus the ultra-lightweight and high accuracy STIIARM is obtained.The proposed method’s effectiveness is verified by the simulation experiments on the ISAR image dataset of five types of space targets.

A Content-Based Parallel Image Retrieval System on Cluster Architectures

We propose a content-based parallel image retrieval system to achieve high responding ability. Our system is developed on cluster architectures. It has several retrieval. servers to supply the service of content-based image retrieval. It adopts the Browser/Server (B/S) mode. The users could visit our system though web pages. It uses the symmetrical color-spatial features (SCSF) to represent the content of an image. The SCSF is effective and efficient for image matching because it is independent of image distortion such as rotation and flip as well as it increases the matching accuracy. The SCSF was organized by M-tree, which could speedup the searching procedure. Our experiments show that the image matching is quickly and efficiently with the use of SCSF. And with the support of several retrieval servers, the system could respond to many users at mean time. Key words content-based image retrieval - cluster architecture - color-spatial feature - B/S mode - task parallel - WWW - Internet CLC number TP391 Foundation item: Supported by the National Natural Science Foundation of China (60173058)Biography: ZHOU Bing (1975-), male, Ph. D candidate, reseach direction: data mining, content-based image retrieval.

Non-Orthogonal Transmission for User and Control Plane Split Architecture in 5G Systems and Beyond: Performance Analysis and Design Insights

In next generation networks,mobility management will be a critical issue due to dense base station(BS)deployment,for which user and control plane split architecture provides a promising solution.Jointly designing such architecture with nonorthogonal transmission brings in more flexibility to further improve system efficiency.This paper proposes a non-orthogonal transmission design for user and control plane split architecture.In this design,user equipments(UEs)will select the BS providing the strongest received signal to associate its data channel,but constantly connect its control channel to the nearest macro-cell BS(MBS).Upon non-orthogonal transmission,an MBS can multiplex data traffics and control signals on the same resource.Stochastic geometry based analysis is carried out to investigate outage probability,which extends its regular definition by jointly considering data and control channels,and then mobility-aware outage rate.Numerical results show that:1)The proposed split architecture alleviates the increase in handover rate for ultra dense networking,compared with conventional architecture.2)Non-orthogonal transmission outperforms traditional orthogonal transmission in the split architecture,because it is capable of accommodating more control channels.3)By carefully adjusting power levels,minimum outage probabilities can be reached for macrocell UEs in the proposed design.

The ARQUIGRAFIA project: A Web Collaborative for Review Only Environment for Architecture and Urban Heritage Image

Purpose:This paper presents the ARQUIGRAFIA project,an open,public and nonprofit,continuous growth web collaborative environment dedicated to Brazilian architectural photographic images.Design/methodology/approach:The ARQUIGRAFIA project promotes the active and collaborative participation among its institutional users(GLAMs,NGOs,laboratories and research groups)and private users(students,professionals,professors,researchers),both can create an account and share their digitized iconographic collections in the same Web environment by uploading their files,indexing,georeferencing and assigning a Creative Commons license.Findings:The development of users interactions by means of semantic differentials impressions recording on visible plastic-spatial aspects of the architectures in synthetic infographics,as well as by the retrieval of images through an advanced system search based on those impressions parameters.By gamification means,the system often invites users to review images’in order to improve images’data accuracy.The pilot project named Open Air Museum that allows users to add audio descriptions to images in situ.An interface for users’digital curatorship will be soon available.Research limitations:The ARQUIGRAFIA’s multidisciplinary team gathering professorsresearchers,graduate and undergraduate students from the Architecture and Urbanism,Design,Information Science,Computer Science faculties of the University of S?o Paulo,demands continuous financial resources for grants,for contracting third party services,for the participation in scientific events in Brazil and abroad,and for equipment.Since 2016,significant budget cuts in the University of S?o Paulo own research funds and in Brazilian federal scientific agencies can compromise the continuity of this project.Practical implications:The open source template called+GRAFIA that can freely help other areas of knowledge to build their own visual Web collaborative environments.Originality/value:The collaborative nature of the ARQUIGRAFIA distinguishes it from institutional image databases on the internet,precisely because it involves a heterogeneous network of collaborators.

Modified Visual Geometric Group Architecture for MRI Brain Image Classification

The advancement of automated medical diagnosis in biomedical engineering has become an important area of research.Image classification is one of the diagnostic approaches that do not require segmentation which can draw quicker inferences.The proposed non-invasive diagnostic support system in this study is considered as an image classification system where the given brain image is classified as normal or abnormal.The ability of deep learning allows a single model for feature extraction as well as classification whereas the rational models require separate models.One of the best models for image localization and classification is the Visual Geometric Group(VGG)model.In this study,an efficient modified VGG architecture for brain image classification is developed using transfer learning.The pooling layer is modified to enhance the classification capability of VGG architecture.Results show that the modified VGG architecture outperforms the conventional VGG architecture with a 5%improvement in classification accuracy using 16 layers on MRI images of the REpository of Molecular BRAin Neoplasia DaTa(REMBRANDT)database.

Designing a High-Performance Deep Learning Theoretical Model for Biomedical Image Segmentation by Using Key Elements of the Latest U-Net-Based Architectures

Deep learning (DL) has experienced an exponential development in recent years, with major impact in many medical fields, especially in the field of medical image and, respectively, as a specific task, in the segmentation of the medical image. We aim to create a computer assisted diagnostic method, optimized by the use of deep learning (DL) and validated by a randomized controlled clinical trial, is a highly automated tool for diagnosing and staging precancerous and cervical cancer and thyroid cancers. We aim to design a high-performance deep learning model, combined from convolutional neural network (U-Net)-based architectures, for segmentation of the medical image that is independent of the type of organs/tissues, dimensions or type of image (2D/3D) and to validate the DL model in a randomized, controlled clinical trial. We used as a methodology primarily the analysis of U-Net-based architectures to identify the key elements that we considered important in the design and optimization of the combined DL model, from the U-Net-based architectures, imagined by us. Secondly, we will validate the performance of the DL model through a randomized controlled clinical trial. The DL model designed by us will be a highly automated tool for diagnosing and staging precancers and cervical cancer and thyroid cancers. The combined model we designed takes into account the key features of each of the architectures Overcomplete Convolutional Network Kite-Net (Kite-Net), Attention gate mechanism is an improvement added on convolutional network architecture for fast and precise segmentation of images (Attention U-Net), Harmony Densely Connected Network-Medical image Segmentation (HarDNet-MSEG). In this regard, we will create a comprehensive computer assisted diagnostic methodology validated by a randomized controlled clinical trial. The model will be a highly automated tool for diagnosing and staging precancers and cervical cancer and thyroid cancers. This would help drastically minimize the time and effort that specialists put into analyzing medical images, help to achieve a better therapeutic plan, and can provide a “second opinion” of computer assisted diagnosis.

Seismic Imaging and 3D Architecture of Yongle Atoll of the Xisha Archipelago, South China Sea

Yongle atoll in the Xisha(Paracel) Archipelago is an isolated carbonate platform developed on Precambrian metamorphic and Mesozoic volcanic rocks since the early Miocene. To identify the 3D stratigraphic architecture and evolution of this platform, 13 high-resolution seismic profiles and shallow-to-deep water multi-beam data were processed and analyzed to reveal seismic facies, sequence boundary reflectors, seismic units, and platform architecture. Nine types of seismic facies were recognized based on their geometry, which included seismic amplitude, continuity, and termination patterns;additionally, six reflections, i.e., Tg, T60, T50, T40, T30, and T20, were identified in the Cenozoic strata. Five seismic units, SQ1(lower Miocene), SQ2(middle Miocene), SQ3(upper Miocene), SQ4(Pliocene), and SQ5(Quaternary), were identified from bottom to top across the platform. The platform grew rapidly in the middle Miocene and backstepped in the late Miocene–Pliocene. Here, we discuss the developmental characteristics and evolution of the Yongle Atoll, in combination with drilling wells, which can be divided into four stages: the initiation stage in the early Miocene, the flourishing stage in the middle Miocene, the partial-drowning stage in the late Miocene–Pliocene, and modern atoll in the Quaternary.

Computerized Scheme for Histological Classification of Masses with Architectural Distortions in Ultrasonographic Images

Architectural distortion is an important ultrasonographic indicator of breast cancer. However, it is difficult for clinicians to determine whether a given lesion is malignant because such distortions can be subtle in ultrasonographic images. In this paper, we report on a study to develop a computerized scheme for the histological classification of masses with architectural distortions as a differential diagnosis aid. Our database consisted of 72 ultrasonographic images obtained from 47 patients whose masses had architectural distortions. This included 51 malignant (35 invasive and 16 non-invasive carcinomas) and 21 benign masses. In the proposed method, the location of the masses and the area occupied by them were first determined by an experienced clinician. Fourteen objective features concerning masses with architectural distortions were then extracted automatically by taking into account subjective features commonly used by experienced clinicians to describe such masses. The k-nearest neighbors (k-NN) rule was finally used to distinguish three histological classifications. The proposed method yielded classification accuracy values of 91.4% (32/35) for invasive carcinoma, 75.0% (12/16) for noninvasive carcinoma, and 85.7% (18/21) for benign mass, respectively. The sensitivity and specificity values were 92.2% (47/51) and 85.7% (18/21), respectively. The positive predictive values (PPV) were 88.9% (32/36) for invasive carcinoma and 85.7% (12/14) for noninvasive carcinoma whereas the negative predictive values (NPV) were 81.8% (18/22) for benign mass. Thus, the proposed method can help the differential diagnosis of masses with architectural distortions in ultrasonographic images.

Beautification of Chinese Architectural Images in the New Media Age

From the perspective of communication science,the communication of architectural images in the new media age has an obvious beautifying trend.Due to the differences in politics,economics,and cultural environment between China and western countries,the beautification of architectural images in China is a unique phenomenon.This study classifies the beautification of Chinese architectural images into different types in terms of image communication:audience orientation,time orientation,space orientation,and cultural orientation.By investigating and analyzing relevant cases,this study explores the beautification of Chinese architectural images in the new media age and puts forward thoughts and evaluation,aiming to better comprehend the relationship between beautification and architectural communication.

基于ImageJ的数字图像处理课程实验教学案例

针对计算机科学与技术专业中的数字图像处理实验教学问题,提出一种基于Java语言和Image J平台的数字图像处理实验教学方案。数字图像处理课程实验教学的目的是通过给学生布置难易适中的实验,让学生在实验课程中独立完成数字图像处理问题的解决。虽然学生已经学过Java语言,考虑到并不是所有学生都熟悉Image J软件,实验任务的起点通常是首先让学生理解并且测试已有的Image J插件的代码模板。其次教师要求学生在已有数字图像处理代码的基础上根据实验要求逐步对现有的插件进行扩充。由于Image J软件是开源的,并且本身是开放式的插件架构体系,使得这种构造性的实验教学方法成为可能。

Grayscale image encryption algorithm based on chaotic maps

A new image encryption/decryption algorithm has been designed using discrete chaotic systems as aSP (Substitution and Permutation) network architecture often used in cryptosystems. It is composed of two mainmodules: substitution module and permutation module. Both analyses and numerical results imply that the algo-rithm has the desirable security and efficiency.

Parallel Extraction of Marine Targets Applying OIDA Architecture

Computing resources are one of the key factors restricting the extraction of marine targets by using deep learning.In order to increase computing speed and shorten the computing time,parallel distributed architecture is adopted to extract marine targets.The advantages of two distributed architectures,Parameter Server and Ring-allreduce architecture,are combined to design a parallel distributed architecture suitable for deep learning–Optimal Interleaved Distributed Architecture(OIDA).Three marine target extraction methods including OTD_StErf,OTD_Loglogistic and OTD_Sgmloglog are used to test OIDA,and a total of 18 experiments in 3categories are carried out.The results show that OIDA architecture can meet the timeliness requirements of marine target extraction.The average speed of target parallel extraction with single-machine 8-core CPU is 5.75 times faster than that of single-machine single-core CPU,and the average speed with 5-machine 40-core CPU is 20.75 times faster.

融合全局聚合与局部挖掘的建筑图像检索

针对建筑图像易受到尺度变化和局部遮挡干扰而导致检索准确率低的问题,本文提出了一种融合全局聚合与局部挖掘的建筑图像检索网络。以ResNet50为骨干网络并在其后引入多尺度特征聚合的全局分支和注意力引导特征挖掘的局部分支,再通过正交融合策略高效整合双分支互补特征。其中,多尺度特征聚合模块结合混合空洞卷积和通道注意力对全局不同尺度的目标进行自适应加权聚合,增强网络对建筑多尺度显著特征的提取;注意力引导特征挖掘模块通过信息互补注意力对最显著特征标记擦除,实现对局部区域中潜在的细节信息的挖掘。所提方法在主流建筑数据集ROxf和RPar上的平均精度均值(mAP)指标分别达到了81.54%(M)、62.43%(H)和90.28%(M)、78.35%(H)。实验结果表明,该方法有效克服了尺度变化和局部遮挡的干扰,显著提升了建筑图像检索的准确率。

模块类网络安全产品的智能测试系统设计

为提高模块类网络安全产品的批量测试效率,降低人力成本及换线时间,开展了模块类网络安全产品的智能测试系统设计;通过对模块类网络安全产品的智能测试需求进行分析,采用将搭载图像传感器的工业机械手与基于PXI架构的测试系统相结合的技术路线,并对机械手控制技术和PXI总线测试平台设计技术进行了研究;创新设计了模块类网络安全产品的智能测试系统,对测试工艺流程进行优化与再造,实现了产品上下料、型号识别与柔性换线、智能装夹、批量测试、合格品与不合格品分拣等流程的一站式无人值守运转;经实际应用验证,智能测试系统的单日测试产能提升超过180%,单日人工工时减少93%以上,换线时间从平均70分钟减少到10分钟以内。

结合沙漏注意力与渐进式混合Transformer的图像分类方法

Transformer在图像分类任务中具有广泛应用,但在小数据集分类任务中,Transformer受到数据量较少、模型参数量过大等因素的影响,导致分类精度低、收敛速度缓慢。本文提出了一种融合沙漏注意力的渐进式混合Transformer模型。首先,通过下-上采样的沙漏自注意力建模全局特征关系,利用上采样补充下采样操作丢失的信息,同时采用可学习温度参数和负对角掩码锐化注意力的分数分布,避免因层数过多产生过度平滑的现象;其次,设计渐进式下采样模块获得细粒度多尺度特征图,有效捕获低维特征信息;最后,使用混合架构,在顶层阶段使用设计的沙漏注意力,底层阶段使用池化层替代注意力模块,并引入带有深度卷积的层归一化,增加网络局部性。所提方法在T-ImageNet、CIFAR10、CIFAR100、SVHN数据集上进行实验,分类精度可以达到97.42%,计算量和参数量分别为3.41G和25M。实验结果表明,与对比算法相比,该方法的分类精度有明显提升,计算量和参数量有明显降低,提高了Transformer模型在小数据集上的性能表现。

建筑类非遗影像创作策略探究

在多年的拍摄实践中,我国的建筑类非遗影像创作者探索出了一系列创作策略,主要包括:通过表现建筑营造技艺和建筑本身给观众以知识和审美享受,通过讲述建筑背后的故事吸引观众,通过挖掘建筑文化及与建筑相关的民俗文化启迪观众,通过表现建筑与人的情感联系打动观众。将这些创作策略运用于建筑类非遗影像创作中,将有助于提升作品质量,提高作品对观众的吸引力。

基于街景分析技术的滨江路空间品质研究——以重庆三代滨江路为例

滨江路是滨江城市中极具特征性的线性空间,但针对其特殊的地理环境条件,在特征性指标体系建构及量化评价等方面还有待拓展。结合滨江路环境要素分析构建特征性指标体系,运用基于语义分割的街景分析技术,对重庆市不同时期的3代滨江路展开特征总结与问题剖析。结果表明:1)在指标体系研究方面,应依据不同研究对象和研究目的,有针对性地选取指标;2)在个案层面,由于道路功能的综合提升及城市地形限制的减弱,使新一代滨江路空间品质显著提高;3)跨江立体交通节点对滨江路空间品质影响较大。利用街景技术建立滨江路空间品质评价指标体系,所得结论可为其品质提升提供科学依据。

基于改进MMAL的细粒度图像分类研究

针对细粒度图像分类中目标区域难以精准定位及其内部细粒度特征难以识别的问题,提出了一种基于改进MMAL的细粒度图像分类方法。首先,利用形变卷积的感知区域可变性原理,动态地感知样本图像中不同尺度和形状的目标区域特征,从而增强网络对目标区域位置的感知能力。随后,采用GradCAM梯度回流的方法生成网络注意力热图,以减小特征背景噪声的干扰,实现对图像目标区域的精准定位。最后,提出位置感知空间注意力模块,通过融合坐标位置和双尺度空间信息,显著提升了网络对目标区域细粒度特征的提取能力。实验结果表明,与基线算法相比,该方法在CUB-200-2011、Stanford Car和FGVC-Aircraft三个公共数据集上分类精度分别提升了1.4%、1.5%、1.9%,该结果验证了所提方法的有效性。