Shenyang Institute of Automation,Chinese Academy Of Sciences
Not a member yet
23582 research outputs found
Sort by
Large dynamic range and wideband mid-infrared upconversion detection with BaGa4Se7 crystal
No full textDirect detection of mid-infrared (MIR) at room temperature commonly suffers from the high background noise and the lack of ideal sensor material. An indirect upconversion method based on nonlinear optics bridges the desired MIR region to the well-developed near-infrared devices. In this paper, we report an experimental study of 3-8 mu m MIR upconversion detection with a 1 mu m laser-pumped BaGa4Se7 crystal. This detection scheme is integrated with an optical parametric oscillator and an optical receiver to forma coherent source and detection system that offers high temporal (nanosecond time scale) and spectral resolution (1.4 octave) and large dynamic range (11.5 orders) of this detection scheme benefits from the superior optical properties of this non-oxide crystal, and a minimum detectable energy of 1.02 fJ/pulse is observed with an In GaAs p-i-n photodiode. The upconversion technique outperforms commercial HgCdTe detectors in terms of both sensitivity (2-4.5 orders better) and wavelength response flatness. The aforementioned characteristics of this integrated MIR source and detection system make it of significant interest for applications in high-resolution gas analysis and spectral imaging. (C) 2022 Optical Society of America under the terms of the OSA Open Access Publishing Agreemen
Single infrared image stripe removal via deep multi-scale dense connection convolutional neural network
No full textStripe noise removal is a crucial step for the infrared imaging system. Existing stripe removal methods are hard to balance stripe removal and image details preservation. In this paper, a deep multi-scale dense connection convolutional neural network (DMD-CNN) is proposed to address this problem. In DMD-CNN, a multi-scale feature representation unit (FR-Unit) is designed to decompose raw image into different scales which can extract diverse fine and coarse features. Dense connection is introduced into the network, which makes full use of the multi-scale information obtained by FR-Unit and avoids performance degradation. Moreover, the regularization term Lh is defined to depict the vertical direction smoothness property of stripe. Experiment results show that DMD-CNN performs more stable stripe removal effects in different scenes and diverse stripe intensity. Meanwhile, DMD-CNN outperforms seven state-of-the-art stripe removal methods on qualitative and quantitative evaluation
一种包络形深海软体手指及深海软体抓手
No full text本发明涉及软体机器人的软体手指，具体地说是一种包络形深海软体手指及深海软体抓手。软体手指设有手指指尖、包络结构板、小压力囊、大压力囊、密封夹紧片、压力管道等结构，软体手指内部设有两个压力通道，当向压力通道充入液体时，充入的液体通过压力通道导入到小压力囊和大压力囊中，压力囊充液膨胀使得手指实现弯曲。本发明通过手指、夹紧件、直角宝塔接头与三指手掌组装，能够得到三指深海软体抓手。本发明设置的大小压力囊能够在加压时让软体手指不同位置均达到合适的弯曲角度，包络结构板相互配合能够牢牢地包裹住被抓物体，压力管道与直角宝塔接头相互配合能够实现水下密封
一种用于增材制造的陶瓷先驱体浆料和增材制造工艺
No full text本发明公开了一种用于增材制造的陶瓷先驱体浆料和增材制造工艺，属于陶瓷材料的增材制造技术领域。该工艺以硅元素含量较高的双官能团丙烯酸酯硅树脂为主要原料制备先驱体溶液，同时将不同质量分数的惰性填料纳米SiC粉末加入到先驱体液相溶液中，利用光固化和原位烧结陶瓷化技术实现陶瓷类材料的直接成型，先平铺一层先驱体材料，采用光固化技术实现材料固化，再采用红外光烧结的方式将材料陶瓷化，如此依次逐层固化、逐层烧结最终实现复杂陶瓷结构的直接一体化成型。将惰性填料加入到先驱体溶液中，以减少先驱体陶瓷在烧结过程中体积收缩，同时减少裂纹和孔隙的产生，兼顾效率与产率的最大化
一种基于多视图约束的导管中心线匹配方法
No full text本发明涉及一种基于多视图约束的导管中心线匹配方法。其步骤为:将多目视觉进行两两标定，确定任意两个相机之间的内参数和外参变换矩阵，同步采集图像，对图像进行中心线提取，得到中心线离散坐标点，利用双目三角测量原理得到中心线离散点的三维空间点云，借助于多视图约束，剔除三维重建过程中相邻视图之间的的误匹配点，得到正确的匹配点，拟合中心线曲线，扫掠得到导管三维模型，本发明方法可以实现任意异形导管的中心线三维重建，精度高，效率高，应用范围广泛
多线程交替实现增减材的设备控制系统
No full text本发明涉及多线程交替实现增减材的设备控制系统，属于金属增减材制造技术领域。本发明的增材方式为铺粉式，减材方式为三轴数控铣削。包括：确定需要监测的状态量，监测线程作为守护线程；将增材控制过程封装进增材线程；将减材控制过程封装进减材线程；守护线程负责协调增材线程与减材线程的交替。本发明的系统可以实现对增减材设备的快速高效控制，可以避免软件在空闲时占用CPU资源，提高了计算效率，并且避免了因CPU负载过重导致软件界面卡死，从而导致设备失去控制的危险情况
Quantitative analysis of iron slurry based on laser induced breakdown spectroscopy combined with mutual information feature selection partial least squares method
No full text目前检测矿浆品位相对准确的方法是传统化学分析,但周期长、有滞后性,无法实现在线检测。实验利用激光诱导击穿光谱（Laser induced breakdown spectroscopy, LIBS）在线、原位、快速等优点,分析了铁矿选矿过程尾矿浆中铁元素的品位值。由于LIBS采集到的光谱数据中存在大量对成分分析无用的冗余信息,进而增加了建模复杂程度,导致建立的模型精确度不够、泛化能力不强。因此,在偏最小二乘（PLS）模型基础上,提出了基于互信息特征筛选的偏最小二乘模型。实验结果表明,与传统的PLS模型相比,基于互信息特征筛选的偏最小二乘模型在分析精度上得到了明显改善,测试样品的决定系数R2从0.52提高到0.90,测试样本的平均绝对误差（MAEP）从2.87%下降到1.38%,总样本的平均绝对误差（MAE）从1.0%下降到0.60%。</p
Analysis of Analytical Model for Space Deformation of Inspection Robot Manipulator
No full text根据高压输电线路检测环境,提出一种高压输电线路巡检机器人。以该输电线路巡检机器人为研究对象,采用Denavit-Hartenberg法建立巡检机器人的连杆坐标系,推导巡检机器人的变换矩阵。并应用微分变换法推导巡检机器人的雅可比矩阵,得到操作空间外力(速度)与关节空间内力(速度)的线性映射关系。机器人操作臂的刚度影响整个机器人的动态特性和定位精度,建立巡检机器人的操作臂刚度矩阵,推导了机器人的末端柔度矩阵。通过关节传递运动和动力,得到末端执行器因各个关节力作用而产生的变形方程,结合变形模型计算末端执行器的最大变形量,并在ANSYS中进行验证末端的最大变形量。最后利用机器人在单位力作用下末端执行器的最大变形量选择最佳的反解路径
Multi Stereo Vision SLAM for Local Weak Texture Environment
No full text在局部弱纹理环境下，使用特征法视觉SLAM的移动机器人存在定位丢失的问题。为此，提出了一种基于多个双目相机紧耦合的视觉SLAM算法。该视觉SLAM算法通过将多个双目相机的测量信息融合在一个统一的最小二乘方程中，完成对周围环境的建模和对自身的定位。在该多双目视觉SLAM中，即使一个双目相机观测到弱纹理场景，依靠其他相机的观测信息，依然可以鲁棒地完成定位和建图的工作。实验结果表明，多双目视觉SLAM有效地解决了局部弱纹理环境下的移动机器人定位丢失问题。</p
Infrared small target detection based on multiscale local contrast learning networks
No full textRecently, model-driven deep networks have achieved excellent detection performance on infrared small targets in cluttered environments. However, its detection performance is sensitive to the hyperparameters in the embedded model-driven module. Therefore, we propose a novel multiscale local contrast learning network (MLCL-Net), which is an end-to-end fully convolutional infrared small target detection network. By constructing a local contrast learning (LCL) structure, it can learn to generate local contrast feature maps during training. Considering the difference in target size, we further build a multiscale local contrast learning (MLCL) module based on LCL. By extracting and fusing local contrast information of different scales from feature maps of the same level, the feature information of targets is fully excavated. At the same time, due to the small size of the target, a slight pixel shift will cause a severe loss of accuracy. We propose a bilinear feature pyramid network (BFPN) based on the feature pyramid network (FPN). Compared to state-of-the-art methods, the proposed MLCL-Net achieves superior performance with an intersection-over-union (IoU) of 0.772 and normalized IoU (nIoU) of 0.755 on the public SIRST dataset.</p