基于改进模糊算子和形态学的舌体分割方法

舌诊是中医望诊中的重要内容,为了自动分析舌体的各项指标,必须将舌体从背景区域中分割出来.由于舌体与脸部等背景区域颜色较为接近,提出一种基于改进模糊算子和形态学的舌体分割方法,该方法根据由大津法确定的阈值定义隶属函数,取得了令人满意的分割效果.     

遗传算法在图像分割中的应用

图像分割是图像分析的预处理阶段,被认为是计算机视觉中的一个瓶颈.基于扩展的Otsu最优阈值图像分割方法,提出用一种改进遗传算法进行图像分割的方法,并给出了遗传算法中参数的设定.仿真结果表明,改进算法的计算速度不仅明显优于传统的Otsu方法,而且算法的分割效果也很好.     

基于彩色图像车牌分割研究

智能运输系统中车牌识别技术得到了广泛应用,车牌分割是车牌识别的重要部分。基于彩色图像车牌分割与采用灰度图像车牌分割相比,可以有效消除阴影影响,同时车牌颜色也是车牌识别的一个参数。颜色分类处理使用特征函数,可以减少颜色坐标转换运算,提高颜色分类速度。文中详细讨论中国车牌特征,给出车牌分割详细步骤。车牌区域判别采用信息融合技术。车牌倾斜矫正结合车牌倾斜特点,提出快速算法。     

基于视觉的地铁列车前向目标识别系统研究

为解决地铁列车前向运行环境中障碍物判断问题,提出基于视觉的地铁列车前向目标识别系统。该系统利用语义分割算法提取行驶轨道区域,结合轨道区域识别,基于SSD算法提取前向运行环境中轨道限界范围内的列车、行人等目标。针对上述场景,建立列车前向运行环境样本库,并基于该样本库完成模型训练。实验结果证明所得模型可有效识别列车前向行驶多目标。     

基于车载LiDAR数据的公路竖向净空自动化评估

现役道路基础设施管理过程中缺乏大范围区域内不同路段的现状或实时的竖向净空数字化资料，导致部分过高车辆撞击跨线桥或其他上空构造物的事故时有发生，造成了重大财产损失与人员伤亡。针对该问题，基于车载LiDAR数据构建公路竖向净空自动化评估方法框架。通过数据重构方法将复杂道路线形的车载LiDAR点云转化为简单的直线形式，利用基于线性索引的点云数据分块方法实现重构场景下车载LiDAR数据的条形、柱形与体素单元的快速分块，建立柱形单元非平面点初步滤波、基于K-Means与体素聚类的复杂LiDAR点云环境中路面优化分割流程。在基于条形单元划分提取道路边界后，利用体素聚类将路面上方点云进行划分。以提取的路面点云作为二维插值基准面，完成不同物体的竖向净空计算，并利用江苏南京市内的2条公路LiDAR数据的对算法框架进行测试。研究结果表明：所提方法在噪音存在的复杂LiDAR环境中可以有效分割出道路上方物体并完成竖向净空的计算；通过部分算法提取与人工标注结果的对比，显示公路1与公路2的竖向净空平均绝对误差分别为0.94、1.57 cm，具有较好的可靠性；在32 GB内存、Intel® Xeon® E5-1650 v4@3.6 GHz六核处理器的计算机上完成公路1与公路2竖向净空评估的平均时间分别为6.62、7.83 s·km^-1，算法效率可满足大尺度场景下的公路竖向净空自动化计算；相比于已有研究方法，所提方法框架考虑了车载LiDAR点云环境内的路面上测量噪音的存在，对变宽度路面条件复杂场景下的公路竖向净空评估具有更好的适用性。     

边缘提取在涂胶缺陷检测中的方法研究

船舶发动机盖外板与内板之间常涂覆一层防震密封胶,但具体涂胶形态难以观测,常用机器视觉手段获取产品涂胶图片。针对涂胶区域分割和检测精度低,效率慢的问题,提出一种新的边缘提取算法获取涂胶区域。为了避免传统Canny边缘检测方法易受噪声干扰的缺点,采用具有保边特性的加权最小二乘法滤波器对图像滤波去燥,增加45°和135°两个方向计算图像梯度,利用改进后的Canny算法获取边缘,对获取到的边缘点用区域生长方式将涂胶区域提取出来,最后将提取到的涂胶区域与标准涂胶区域进行配准和减影,得到减影结果即为涂胶缺陷。试验仿真表明：本算法能够很好的提取到胶水区域,并且获取胶水缺陷。     

基于多波束点云的水下柱桩及桩间距高精度确定方法

受海水介质环境复杂、桩体倾斜以及测量误差大的影响,水下圆柱桩位置的高精度、低成本、高效率确定一直是港口工程建设中的一个难点。为此,提出了一种基于多波束点云的水下圆柱桩及桩间距高精度确定方法。该方法首先基于多波束测量实现桩体及海床点云数据低成本、高效获取。考虑桩体属性信息、点云法向量和高程等几何信息,利用一种综合阈值分割方法实现了桩体点云的有效提取。还提出了顾及桩体形态、测量误差的桩体中心位置综合确定方法,实现桩体相对位置和桩间距的高精度确定。该方法得到了实际工程验证,取得了优于5 cm的桩体相对位置和桩间距确定精度。     

Predictive usage mining for life cycle assessment

The usage modeling in life cycle assessment (LCA) is rarely discussed despite the magnitude of environmental impact from the usage stage. In this paper, the usage modeling technique, predictive usage mining for life cycle assessment (PUMLCA) algorithm, is proposed as an alternative of the conventional constant rate method. By modeling usage patterns as trend, seasonality, and level from a time series of usage information, predictive LCA can be conducted in a real time horizon, which can provide more accurate estimation of environmental impact. Large-scale sensor data of product operation is suggested as a source of data for the proposed method to mine usage patterns and build a usage model for LCA. The PUMLCA algorithm can provide a similar level of prediction accuracy to the constant rate method when data is constant, and the higher prediction accuracy when data has complex patterns. In order to mine important usage patterns more effectively, a new automatic segmentation algorithm is developed based on change point analysis. The PUMLCA algorithm can also handle missing and abnormal values from large-scale sensor data, identify seasonality, and formulate predictive LCA equations for current and new machines. Finally, the LCA of agricultural machinery demonstrates the proposed approach and highlights its benefits and limitations.     

The design of electric vehicle charging network

The promotion of Electric Vehicles (EVs) has become a key measure of the governments in their attempt to reduce greenhouse gas emissions. However, range anxiety is a big barrier for drivers to choose EVs over traditional vehicles. Installing more charging stations in appropriate locations can relieve EV drivers’ range anxiety. To determine the locations of public charging stations, we propose two optimization models for two different charging modes - fast and slow charging, which aim at minimizing the total cost while satisfying certain coverage goal. Instead of using discrete points, we use geometric objects to represent charging demands. Importantly, to resolve the partial coverage problem (PCP) for networks, we extend the polygon overlay method to split the demands on the road network. After applying the models to Greater Toronto and Hamilton Area (GTHA) and to Downtown Toronto, we show that the proposed models are practical and effective in determining the locations of charging stations. Moreover, they can eliminate PCP and provide much more accurate results than the complementary partial coverage method (CP).     

How might potential future plug-in electric vehicle buyers differ from current “Pioneer” owners?

The entry of various plug-in electric vehicles (PEVs) into the passenger vehicle sector provides novel opportunities to learn about the potential for future PEV markets. However, early PEV buyers (or “Pioneers”) can substantially differ from present conventional vehicle owners that have interest in purchasing PEVs in the future (or the “Potential Early Mainstream buyers”). To compare the characteristics, preferences, and motivations of Pioneers and Potential Early Mainstream buyers, we draw data from the Canadian Plug-in Electric Vehicle Study, a three-part mixed-mode survey with samples of PEV owners (n = 94) and conventional new vehicle buyers (n = 1754). We identify several significant differences in household characteristics, including income, education, and recharge access. In terms of preferences, Pioneers express extremely high valuation of PEVs and prefer pure battery electric vehicle (BEV) designs over plug-in hybrid electric (PHEV) designs. In contrast, Potential Early Mainstream respondents prefer PHEVs. Both Pioneer and Potential Early Mainstream respondents are similarly cautious about controlled charging programs, but Pioneers place five times as much value on using electricity generated from renewable sources than the Potential Early Mainstream. Pioneers also tend to have different motivations, including significantly higher levels of environmental concern, and higher engagement in environment- and technology-oriented lifestyles. Policymakers, automakers, and electric utilities that anticipate a transition to electric mobility ought to consider how potential future PEV buyers may differ in their vehicle preferences, usage and motivations relative to current PEV owners.