Saliency motivated pulse coupled neural network for underwater laser image segmentation

The detection range of underwater laser imaging technology achieves 4—6 times of detection range of conventional camera in intervening water medium, which makes it very promising in oceanic research, deep sea exploration and robotic works. However, the special features in underwater laser images, such as speckle noise and non-uniform illumination, bring great difficulty for image segmentation. In this paper, a novel saliency motivated pulse coupled neural network(SM-PCNN) is proposed for underwater laser image segmentation. The pixel saliency is used as external stimulus of neurons. For improvement of convergence speed to optimal segmentation, a gradient descent method based on maximum two-dimensional Renyi entropy criterion is utilized to determine the dynamic threshold. On the basis of region contrast in each iteration step, the real object regions are effectively distinguished,and the robustness against speckle noise and non-uniform illumination is improved by region selection. The proposed method is compared with four other state-of-the-art methods which are watershed, fuzzy C-means, meanshift and normalized cut methods. Experimental results demonstrate the superiority of our proposed method to allow more accurate segmentation and higher robustness.     

基于高分辨率遥感影像的道路提取方法

分别采用面向对象、最大似然、支持向量机3种分类方法进行实验区高分辨率遥感影像的道路提取,并作分类精度评定。结果表明,面向对象法能够综合利用高分辨率遥感影像的波谱、几何、纹理等多种特征信息,可有效克服影响因素的干扰。本案面向对象法道路提取的总分类精度较高,获得的分类结果较好。     

Research of underwater self-reconfigurable system

This article describes a novel underwater robot, which is called underwater self-reconfigurable system （USS）. USS is studied and developed by Shanghai Jiao Tong University. It is formed by a quantity of the same modules and works in group. Based on that, USS can change its form while working in the complicated underwater structures according to different tasks. New multi-jaw underwater docking system, piston style regulator and clamshell-like sampling module are designed for USS, these developments address the problems of underwater docking, floating condition adjusting and sampling. Besides, the gaits like wriggle, creeping, swimming and fourlegged moving have been successfully implemented in tank and lake test, which verifies the validity of the concept.     

Research of the image processing in dynamic flatness detection based on improved laser triangular method

As a commonly used non-contact flatness detection method, laser triangular detection method is designed with low cost, but it cannot avoid measurement errors caused by strip steel vibration effectively. This paper puts forward a dynamic flatness image processing method based on improved laser triangular detection method. According to the practical application of strip steel straightening, it completes the image pre-processing, image feature curve extraction and calculation of flatness elongation using digital image processing technology. Finally it eliminates elongation measurement errors caused by the vibration.     

改进SSD模型在高分二号遥感影像中高速公路收费站位提取的应用

以高分二号遥感影像中的高速公路收费站为研究对象,选取了北京、山西、河南、广东、福建5个省市2019年的高速公路收费站点位和0.8 m遥感影像,通过图像预处理、样本标注、裁切、数据增强、样本集划分的步骤制作训练样本集;引入“多尺度特征融合”的方法对SSD目标检测模型进行改进,通过增加“转置卷积”和“拼接”操作,将高层次特征图像的语义特征赋予低层次特征图像,以增强上采样质量与特征融合能力,从而提升了模型对小目标收费站的检测效果;将改进SSD模型用于2019年福建省高分二号影像中的收费站点位提取,沿福建省高速公路路网矢量对影像进行自动切片,将切片输入模型中进行目标检测;保留有收费站的切片,使用非极大值抑制去除多余的检测框,将剩余的检测框的坐标变换为中心点的坐标,可以直接输出得到高速公路收费站的中心点矢量,从而实现对于收费站点位的端到端自动化提取。研究结果表明：改进SSD模型的精度、召回率及二者的调和平均数分别为0.86、0.88和0.87,均优于传统的SSD, VGG, Faster R-CNN和特征金字塔网络模型。可见,对收费站点位的自动提取可以大大提高公路管理者的工作效率,有效满足公路管理者的实际工作需求。     

Formation control of underwater mobile sensing networks

Formation control is essential for an underwater mobile sensing network (UMSN), and an ad hoc network which wirelessly connects underwater vehicles of sensing and/or observing types via acoustic communications, to fulfill mobile sensing tasks. The problem of formation control for a UMSN with varying topology is studied in this paper. The methodology of synthesizing distributed formation controller which stabilizes a UMSN with varying topology is proposed on the basis of the stability analysis of linear time-varying systems.     

水下涡轮发动机动力推进系统数学模型

为了维持水下开式循环涡轮发动机系统的高性能和经济性, 针对系统对于工况变化极为敏感的工作特点, 计算了喷管在各种工况下的出口速度, 分析了变转速、变压力条件下的涡轮输出转矩, 建立了适用于系统优化和新型控制器设计的系统数学模型。以设计压力比为关键参数, 取该模型的稳态特性, 可得到由系统转速、航速、燃烧室压力、航深、推进剂耗量以及设计压力比构成的封闭代数方程, 以多速制组合航程最大为寻优目标, 可完成系统所有设计参数的优化。实践证明采用该模型可使系统经济性提高8%以上。     

水下不分散混凝土配筋梁抗弯性能试验

为了研究水下不分散混凝土结构的性能,进行了水下不分散混凝土配筋梁与普通钢筋混凝土配筋梁的抗弯性能对比试验.试验结果表明,水下不分散混凝土配筋梁的抗弯性能与普通钢筋混凝土配筋梁的抗弯性能接近.因此,可以用计算普通钢筋混凝土梁受弯的理论计算水下不分散混凝土配筋梁的受弯.     

一种基手双目线结构光视觉系统的简单标定方法

提出了一种基于双目线结构光主动三角法的视觉系统的简单标定方法．该视觉系统采用线结构光原理,使用双目摄像机避免了常用的单摄像机系统中容易出现的遮挡等问题．该系统的标定综合了线结构光传感器模型以及双目立体视觉模型,推导了各个标定坐标系之间的联系,依据蔡氏两步法,使用方格靶标和最小二乘法求得各种标定参数．实验结果表明,该标定方法操作简单、实用可行．     

A low-cost testbed of underwater mobile sensing network

Comprised by a swarm of acoustically linked and cooperative autonomous underwater vehicles (AUVs) with onboard sensors, an underwater mobile sensing network (UMSN) will be a complementary means to fixed observatory networks, e.g. seafloor observatory networks and moored buoy arrays. It has obvious advantages over a single large AUV in higher efficiency due to parallel observation, stronger robustness to vehicle failures and lower cost. Although an UMSN can be viewed as a counterpart of wireless mobile sensing networks for air and terrestrial applications, it is much more challenging due to poor performance of underwater acoustic communication, poor performance of underwater positioning and high degree of uncertainty in vehicle dynamics and underwater environment. In order to verify key technologies involved in an UMSN, e.g. cooperation of multi-AUVs based on acoustic communication, a low cost testbed has been developed for experimental study. The design of both hardware and software is introduced. Also the results of a functional test for verification of the effectiveness of the testbed are presented.     

强机动超空泡水下航行器空间运动控制面布局

针对超空泡航行器航向舵后置布局在强机动航行时舵效不足的问题, 建立了一种全新的兼顾直航与机动运动的控制面布局模式, 首次提出了非对称分布艏舵的超空泡航行器控制面总体方案, 采用双自由度空化器作为控制面, 水平偏转控制航向, 上下偏转控制俯仰和深度, 且在空化器下部复合抗横滚鳍片。针对某超空泡水下航行器进行了系统动态特性仿真分析, 横滚剩余力矩约从0.1s衰减至0, 极限横滚角小于0.1°。分析结果表明: 以空化器作为艏部航行舵替代常规布局的尾部航向舵, 可以显著提高舵效。     

海洋拖曳系统对船舶操纵性能的影响

将拖曳母船、拖缆和拖曳体视为一个相互作用的整体,利用耦合边界条件,将拖缆顶端和底端的张力与其产生的力矩,分别计入船舶操纵性运动方程和拖曳体六自由度运动方程,结合拖缆的有限差分方程,建立了船/缆/体耦合运动模型,采用数值计算方法,对比分析了海洋拖曳系统对船舶操纵性产生的影响.计算结果表明:当计入拖缆和拖曳体耦合影响后,船舶稳态运动时的速度会降低,改变量为3％～5％;船舶回转机动时,速度、回转半径与横摇角会降低,改变量分别为2％～3％、2％～4％和11％～21％.采用船/缆/体耦合运动模型计算得到的船舶操纵性能符合实际,可为预报海洋拖曳系统的运动信息提供理论依据.     

如何控制水下灌注混凝土施工质量

从几个方面介绍对水下灌注混凝土施工质量的控制。     

钻孔桩水下混凝土质量控制的探讨

结合钻孔灌注桩的施工实践,介绍水下混凝土施工质量控制的要点,简析造成断桩的原因,提出预控断桩的措施。     

基于小波的OFDM调制解调

用FFT实现OFDM调制解调，有效地提高了OFDM调制解调速度，但是各子信道之间存在着较大的频谱重叠，导致了子信道之间的功率泄漏，从而引起了严重的串扰．利用小波变换实现OFDM的调制解调可以较好地解决这一问题．     

水下不分散混凝土柱受压性能试验研究

基于跨江、跨海、深水港工程的环境特点,工程中采用大量水下不分散混凝土材料,进行了水下不分散混凝土短柱的轴心和偏心受压实验,分析了水下不分散混凝土的破坏形态和承载能力.试验结果表明,水下不分散混凝土的抗压性能与普通钢筋混凝土的抗压性能具有相似的特征.     

中国航空网络的多分辨率小波分解研究

为解决航空网络中节点与边数量过多导致整体特征分析计算量大的困难,提出了基于多分辨率小波分解理论的复杂网络数据压缩方法,论证了选用Haar小波基进行航空网络小波分解的适用性及分解形式,提出了确定网络分解层数和分解后还原参数的方法.对2011年5月我国163座通航城市和2 198条航线构成的复杂航空网络,选用Haar小波基对该网络的邻接矩阵进行4层小波分解,得到的网络最低频子带10×10阶矩阵,包含了原网络的大部分信息.实证研究结果表明:利用分解后的最低频子带可以还原出原网络节点城市的平均度、平均最短路径长度和聚类系数.      

关于不规则小波系统的框架研究

对于任意给定的g∈L2(Rn),考虑了具有这种形式{λn/2g(λjx-kb)}j∈,k∈xn(这里λj＞0,b＞0)的不规则的小波系统.对于这类小波,给出了它构成L2(Rn)空间中的框架的充分条件.同时,对于一类具有某种性质的函数g∈L2(Rn),证明了当对{λj}j∈x加一定的限制后,具有上述形式的不规则小波也能构成L2(Rn)空间中的框架.     

双正交小波包提升格式算法

提升格式是一种新的小波构造方法,它是因构造紧支的第二代小波之需而产生的从双正交小波包出发,研究提升格式在双正交小波包理论中的应用,给出提升格式的一种快速小波包变换实现方法——分解和重构公式,并给出实例.