ID 原文 译文
18005 否则，采取上述准则更新用户支撑集，提高支撑集中活跃用户数估计精度。 Otherwise, the above criteria are usedto update the user support set, and the estimation accuracy of the active users in the support set is improved.
18006 在迭代过程中，根据最近两次残差能量之比，选取不同的迭代步长，以降低检测迭代次数。 In the iteration process, different iteration steps are selected according to the ratio of the last two residualenergies, so as to reduce the number of detection iterations.
18007 仿真结果表明，所提算法与传统基于CS的MUD算法相比，误码率降低约9%，迭代次数减少约10%。 The simulation results show that, compared withthe traditional CS-based MUD algorithm, the proposed algorithm reduces the bit error rate by about 9% andthe number of iterations by about 10%.
18008 色环电阻是印刷电路板(PCB)中最常用的电子元器件之一，主要依靠色环的排列顺序和颜色等视觉信息进行区分，易发生装配错误。 The color-ring resistor is one of the most commonly used electronic components in Printed CircuitBoard (PCB). It is featured by sequential color rings, which often brings assembling errors, however.
18009 但是色环电阻装配质量的人工检测方法效率低、误检率高，而传统的基于图像处理技术的自动检测方法鲁棒性较差，难以解决不同拍摄角度、物距及光照条件下的PCB板色环电阻检测问题。 Manualdetection of color-ring resistors has low efficiency and high false detection rate. Traditional image-basedautomatic detection methods have difficulties in dealing with PCB images under various illuminations, imagingdistance and views.
18010 针对这一问题，该文提出一种基于卷积神经网络(CNN)的PCB板色环电阻自动检测与定位方法， To solve this problem, an automatic detection and localization method for PCB color-ringresistor is proposed based on Convolution Neural Network (CNN).
18011 首先采用编码器-解码器结构的卷积神经网络模型及带有权重的交叉熵损失函数的网络训练方法，较好地解决了复杂光照及场景下PCB板色环电阻的图像分割问题； Firstly, the encoder-decoder CNN model isestablished and trained using weighted cross-entropy loss function. With CNN, color-ring resistors aresegmented from PCB images with complex illumination and scenes.
18012 然后采用最小面积外接矩形方法定位单个色环电阻，并通过仿射变换对色环电阻位置进行垂直校正； Secondly, each color-ring resistor islocalized using minimum area bounding rectangle, and its position is adjusted to the vertical direction by affinetransformation.
18013 最后通过高斯模板匹配方法实现了色环电阻的色环定位。 Finally, the localization of color rings on the resistor is achieved by Gaussian templatematching.
18014 采用1270幅PCB图像对该文方法进行了实验和验证，并与传统的基于形态学和基于模板匹配的色环电阻检测方法进行了对比， The proposed method is tested and verified by 1270 PCB images, and the result is compared withthat of the traditional method (method based on geometric contour, and method based on template matching).