高速重轨平直度在线检测研究

重轨的平直度检测是重轨生产过程中重要的一个环节,该项检测的速度、准确性、稳定性等因素直接影响重轨总体质量。 本文以中国某大型钢铁集团轨梁厂的重轨平直度检测现状为背景,利用激光传感、数据采集、计算技术等现代技术,设计了一套基于并行计算的重轨平直度在线检测系统。系统采用14个激光位移传感器和2个旋转编码器分别对于重轨的7个水平方向位移测点、7个竖直方向位移测点、和重轨传输方向的位移进行实时测量,得到的信号转换为数字信号后,在PC机内通过GPU并行计算进行同步配准、块内数据分析、块间数据分析、帧间数据分析等处理,除去粗糙度、氧化铁皮、高频振动、低频振动等干扰,最后得到并输出平直度信息。实验及现场应用表明,该方法检测速度可达到3m/s、精确度可达到0.1mm.     

基于注意力机制的双向长短时记忆网络模型突发事件演化关系抽取

针对现有突发事件关系抽取研究多集中于因果关系抽取而忽略了其他演化关系的问题,为了提高应急决策中信息抽取的完备性,应用一种基于注意力机制的双向长短时记忆(LSTM)网络模型进行突发事件演化关系抽取。首先,结合突发事件演化关系的概念,构建演化关系模型并进行形式化定义,依据模型对突发事件语料进行标注;其次,搭建双向LSTM网络结构,并引入注意力机制计算注意力概率以突出关键词汇在文本中的重要程度;最终,使用搭建的网络模型进行演化关系抽取得到结果。在演化关系抽取实验中,相对于现有因果关系抽取方法,所提方法不仅抽取出更加充分的演化关系,为突发事件应急决策提供了更完善的信息;同时,在正确率、召回率和F1分数上分别平均提升了7.3%、6.7%和7.0%,有效提高了突发事件演化关系抽取的准确性。     

无锚点的遥感图像任意角度密集目标检测方法

针对基于深度学习的遥感图像目标检测方法密集目标漏检率高、分类不准确的问题,建立了一种基于深度学习的无锚点的遥感图像任意角度的密集目标检测方法。首先采用CenterNet作为基线模型,经过主干网络提取特征,并改造原有检测器结构,即加入角度回归分支进行目标角度回归;然后提出一种基于非对称卷积的特征增强模块,并将主干网络提取到的特征图输入特征增强模块,从而增强目标的旋转不变性特征,消除由于目标的旋转、翻转带来的影响,进一步提升目标中心点、尺寸信息的回归精度。采用HourGlass-101作为主干网络时,该方法在DOTA数据集上的平均精度均值（mAP）比旋转区域候选网络（RRPN）提升了7.80个百分点,每秒处理帧数（FPS）提升了7.5;在自建数据集Ship3上,该方法的mAP比RRPN提升了8.68个百分点,FPS提升了6.5。结果表明,所提方法能获得检测精度和速度的平衡。     

Enhanced 3D residual network for video event recognition in shipping monitoring

The three-dimensional convolutional neural network is widely used in video recognition, action recognition and other tasks because it can directly extract temporal and spatial features. Due to the large number of parameters, many computing resources, and difficulty in training, the structure of three-dimensional convolutional neural network is generally shallow. For example, the traditional C3D [17] method uses only the 11-layer VGGNet structure, and the traditional Res3D [18] method adopts a residual network of 18 and 34 layers. Some experience of two-dimensional convolutional neural network shows that the deeper the network structure is, the higher the recognition accuracy will be. Therefore, this paper proposes a new method 3D ResNet-66, which combines a 50-layer 3D residual network and four-layer residual blocks, effectively reducing the number of parameters while increasing the depth of the network, and we finally obtain a better video recognition model through experiments. We evaluate our method on shipping event datasets. Compared to the traditional C3D and Res3D method, our method has improved the accuracy from 91.48% to 96.33%, the model size has been reduced from 561 MB to 135 MB, and the average processing time has become half of the original.

     

Geometric calibration based on B-spline with multi-parameter and color correction based on transition template and decay function

Multimedia Tools and Applications - To achieve low-cost and fast multi-channel surface projection geometric correction, the quadratic quasi-uniform B-spline surface is used to reconstruct the...     

A learning-based approach for leaf detection in traffic surveillance video

Multidimensional Systems and Signal Processing - Traffic surveillance video is recorded in uncontrolled outdoor scenarios. If the camera view gets obstructed by the leaves, the video will fail to...     

基于稳定匹配的实时ETL弹性调度机制

在数据生产速度波动较大的场景,为了实时ETL资源利用更合理,提出基于稳定匹配的ETL弹性调度机制.预测数据源的数据生产速度,并计算满足预测值的消费数据速度;使用贪婪负载均衡算法,调整ETL服务个数使节点负载均衡;确定ETL操作匹配关系,使消费数据速度最大且代价最小.该调度机制将匹配问题转化为最小费用最大流问题,并提出基...