IVS201SH视野检查系统控制软件实现

VS201SH视野检查系统的软件控制基于Windows平台和VC6．0开发环境,使用了对象建模方法和消息、多线程机制。系统通过人性化的操作界面,实NT高可靠性,强扩展性的医疗视野控制检查功能。该系统实NT盲点、青光眼、黄斑、低视力、区域、阈值等视野检查功能。     

粒子测速系统中刚性平移运动假设条件

在对流场进行的粒子示踪测量中,针对空间运动粒子透视成像后易出现的误匹配问题,研究了粒子图像测速技术中近邻粒子平移运动假设,提出了三点式结构约束.该约束条件基于粒子图像测速技术的近邻性匹配准则和机器视觉映射原理,利用视平面内相同运动的特征平面共线的性质,对前后2帧图像中的粒子进行运动检测.最后通过仿真实验证明了该约束条件的有效性.     

基于最优形态学滤波的面料缝纫加工检测

为了对服装加工质量进行自动检测,提出了一种对针织面料缝纫加工的疵点进行自动检验的方法。该方法首先基于遗传算法进行结构元素的寻优,然后通过图像的数学形态学滤波处理来消除面料组织的正常纹理对图像分割的干扰;最后利用灰度阈值实现图像分割来得到表现缝线和加工疵点特征的二值图像。实验结果表明,该方法是有效的。     

曲面工件超声检测中碰撞干涉的研究

在超声检测自动探伤过程中,探头可能和工件发生碰撞干涉。根据曲面工件和圆柱形探头碰撞干涉的特点,给出了一种分级碰撞干涉检测方法。首先利用包围盒算法和几何求交算法剔除大量被检对象分离的情况,然后将可能干涉的3维对象正投影到2维平面中,空间碰撞问题就转化为平面碰撞问题,只需计算投影图是否有重叠就可精确判断是否干涉。工程应用表明,该方法能满足超声检测机器人碰撞干涉检测的高效性和精确性要求,该方法也适用于装配和数控加工中复杂曲面和圆柱体之间的碰撞干涉快速检查。     

基于 FEF-DeepLabV3+的电力金具锈蚀分割方法

金具锈蚀在输电线路航拍图像中细节丰富且分布不规律,为克服分割检测过程中局部信息丢失、精度低和速度慢等问题,提出基于DeepLabV3+的金具锈蚀语义分割模型。替换其主干网络为轻量化改进MobileNetV3网络加快运算速度,并提出自适应特征金字塔(adaptive feature pyramid, AFP)结构融合多尺度。结合FRN层提出特征融合空洞空间金字塔池化(feature fusion atrous spatial pyramid pooling, FEF-ASPP)结构,能够在加强像素间联系的同时不降低分辨率。最后优化损失函数,提高算子的有效性。实验表明,mIoU和mPA分别达到了87.15%、96.64%,相较于原模型提高了3.09%、4.29%。参数量仅为原模型的48%,推理时间仅为15.94 ms,降低了对设备算力的要求,实现高效高精度、轻量化的输电设备锈蚀缺陷分割检测。     

基于改进YOLOv5的输电线路鸟巢检测方法研究

输电线路上的鸟巢会对电力设备的安全运行构成威胁,甚至影响整个电力系统的稳定性。针对复杂场景下输电线路鸟巢检测方法适用性较差的问题,提出一种基于改进YOLOv5的输电线路鸟巢检测方法。该方法结合通道注意机制和空间注意机制设计特征平衡网络,以通道权值和空间权值作为引导,实现检测网络不同层次特征之间语义信息和空间信息的平衡。同时,为了避免因网络层数增加导致特征信息不断被弱化的问题,设计特征增强模块以捕获与鸟巢相关的通道关系和位置信息。最后,利用输电线路无人机巡检图像建立鸟巢数据集进行训练和测试。实验结果表明,所提出的输电线路鸟巢检测方法具有较强的泛化能力和适用性,同时也为电力图像缺陷检测提供技术参考。     

AR-based deep learning for real-time inspection of cable brackets in aircraft

In the process of aircraft assembly, there exist numerous and ubiquitous cable brackets that shall be installed on frames and subsequently need to be manually verified with CAD models. Such a task is usually performed by special operators, hence is time-consuming, labor-intensive, and error-prone. In order to save the inspection time and increase the reliability of results, many researchers attempt to develop intelligent inspection systems using robotic, AR, or AI technologies. However, there is no comprehensive method to achieve enough portability, intelligence, efficiency, and accuracy while providing intuitive task assistance for inspectors in real time. In this paper, a combined AR+AI system is introduced to assist brackets inspection in a more intelligent yet efficient manner. Especially, AR-based Mask R-CNN is proposed by skillfully integrating markerless AR into deep learning-based instance segmentation to generate more accurate and fewer region proposals, and thus alleviates the computation load of the deep learning program. Based on this, brackets segmentation can be performed robustly and efficiently on mobile devices such as smartphones or tablets. By using the proposed system, CAD model checking can be automatically performed between the segmented physical brackets and the corresponding virtual brackets rendered by AR in real time. Furthermore, the inspection results can be directly projected on the corresponding physical brackets for the convenience of maintenance. To verify the feasibility of the proposed method, experiments are carried out on a full-scale mock-up of C919 aircraft main landing gear cabin. The experimental results indicate that the inspection accuracy is up to 97.1%. Finally, the system has been deployed in the real C919 aircraft final-assembly workshop. The preliminary evaluation reveals that the proposed real-time AR-assisted intelligent inspection approach is effective and promising for large-scale industrial applications.     

Data-enabled learning and control algorithms for intelligent glucose management: The state of the art

External insulin administration is an effective way for patients with diabetes mellitus to regulate their blood glucose. Multiple daily injections (MDIs), sensor-augmented pump (SAP) and artificial pancreas (AP) are widely adopted approaches in insulin therapy. With the increasing popularity of continuous glucose monitoring (CGM) sensors, a large number of data-enabled learning and control algorithms have been developed for MDI, SAP and AP. In this paper, we perform a systemic review concerning the state-of-the-art methodologies that are developed for MDI, SAP and AP with feedback from CGM data or other available data, from a systems and control perspective. The review characterizes the traditional learning and control methods developed for the MDI, SAP and AP, including run-to-run control, proportional–integral–derivative control, fuzzy logic control and model predictive control, as well as the discussions on the roles of machine learning technologies in MDI, SAP and AP. Finally, potential future directions on the algorithm architecture design, a unified control framework for MDI, SAP and AP algorithm design and practical usage of the MDI, SAP and AP are discussed.     

基于改进混合粒子群算法和匹配理论的无人机电力巡检卸载策略

无人机搭载深度神经网络进行自主电力巡检时由于受到设备本身计算能力、电池容量、深度神经网络计算负载的限制,无法独立处理巡检任务中产生的海量图像数据。为解决该问题,提出了一种基于改进混合粒子群算法和匹配理论的无人机电力巡检卸载策略,该策略将系统成本最小化问题分解为深度神经网络计算任务协同分割和边缘服务器选择两个子问题。针对协同分割子问题,基于深度神经网络计算任务的执行流程提出了一种错时传输方法,通过改进混合粒子群算法求解多无人机任务协同分割层。针对边缘服务器选择子问题,定义无人机与边缘服务器各自偏好函数,根据偏好函数通过匹配理论建立两者间的稳定匹配,得到边缘服务器选择策略。仿真结果表明,与其他卸载策略相比,所提策略能有效降低无人机能耗和计算任务处理时延,促进边缘服务器负载均衡。     

融合改进 Padim 建模和 ResNet 网络的 喷涂质量检测算法

为了满足喷涂机器人对于喷涂质量检测的需求,采用迁移学习对改进 Padim 建模和 ResNet 网络进行融合,构建自主喷 涂机器人喷涂质量检测一体化模型。 该模型提取一次图像特征可同时用于缺陷定位和分类。 在缺陷定位端,通过改进 Padim 模型以减少特征冗余所造成网络的计算消耗,首先将 ResNet-18 网络获取的 patch 嵌入向量语义层由原先前 3 层改为单 2 层,然 后特征表达由 100 维降维至 20 维,最后训练正样本得到正态分布模型与测试图像进行缺陷定位。 在缺陷分类端,对预训练 ResNet-18 网络进行负样本二次训练,得到 ResNet-18 分类模型对测试图像进行缺陷分类。 经过实验,将一体化模型移植在 jetson nano 移动端中,参数量仅为 11. 69 M,定位精度 94. 5%,分类准确率高达 99. 6%,在机器人位移速度 0. 02 m/ s 下检测时间 为 0. 730 s,不会出现缺帧漏检情况,满足实时检测的要求。