ID 原文 译文
41396 采用K均值聚类算法训练轻量化模型以优化预选框参数，提高算法对特定目标的检测精度。 The K-means clustering algorithm was used to train the light-weighted model to improve the parameters of the pre-selection box and improve the detection accuracy of the algorithm for specific targets.
41397 测试表明，改进后的轻量化算法对特定目标的预测精度为98%，召回率为92%，均值平均精度达90.70%，单张图片检测时间为7.8ms，对视频的处理速度可达125.6f/s，满足微型无人飞行器的实时应用要求。 Test results show that the proposed algorithm can reach a prediction accuracy of98%for a specific target, a recall rate of 92%, a mean average accuracy of 90.70%, a single image detection time of 7.8 ms, and a video processing speed of 125.6 f/s, which can meet the real-time application requirements of micro unmanned aerial vehicles.
41398 数字同轴全息对刀技术中微细铣刀全息再现像的自动聚焦是实现高精度自动对刀的关键技术，其中聚焦评价函数是判别图像质量的依据。 The automatic focusing of the holographic reconstructed image of the micromilling tool in the digital in-line holographic tool setting technology is the key technology for realizing high-precision automatic tool setting, and the focus evaluation function is the basis for discriminating image quality.
41399 通过比较几种常用的聚焦评价函数的评价性能，探讨了数字全息自动对焦过程中刀具的成像特点以及适用的聚焦评价方法，发现小波变换聚焦评价函数能适应高精度聚焦的需要。 In this paper, by comparing the performance of several commonly used focusing evaluation functions, the imaging characteristics of the tools in the digital holographic autofocus process and the applicable focusing evaluation method are discussed.
41400 针对自动调焦问题，提出一种分段递进搜索方法，将搜索过程分为两个过程:大步距粗调和小步距精调，分段搜索最优解。 Aiming at the problem of automatic focusing, apiecewise progressive search method is proposed, which divides the search process into two processes:coarse adjustment of large step and fine adjustment of small step, and piecewise search for the optimal solution.
41401 然后比较分析了所提搜索方法与经典的模拟退火算法和Levenberg-Marquardt算法的搜索性能。 Then the performance of the proposed search method is compared with that of the classical simulated annealing algorithm and Levenberg-Marquardt algorithm.
41402 实验结果表明，所提分段递进搜索方法适用性更强，并通过计算机模拟实验进一步验证了其有效性。 The experimental results show that the proposed piecewise progressive search method has better applicability, and the computer simulation experiment further verifies the effectiveness of the proposed method.
41403 针对铂等常用金属热敏材料电阻温度系数(TCR)不高，导致热式MEMS流速传感器宽量程测量时功耗高的问题，设计了一种基于非晶锗(a-Ge)薄膜热电阻的低功耗、宽量程柔性MEMS流速传感器。 In order to solve the problem of high power consumption in wide range measurement of flow velocity caused by low temperature coefficient of resistance(TCR)of commonly used metal thermistors materials such as platinum, a flexible MEMS thermal flow sensor with low power consumption and wide range based on amorphous germanium(a-Ge)film thermistors was designed.
41404 非晶锗热电阻材料具有较高的TCR系数(约为-0.02/K)和室温电阻率(5Ω·m)，传感器在较低的工作温差和功耗下可获得宽量程的流速测量。 The a-Ge thermistor material has high TCR coefficient(about-0.02/K)and room temperature resistivity(5Ω·m), which makes the sensor can achieve a wide range of flow velocity measurement under low working temperature difference and power consumption.
41405 阐述了该柔性MEMS流速传感器的设计结构、工作原理、3D有限元建模和热-流场仿真结果。 The design structure, working principle, 3 Dfinite element modeling and thermalflow field simulation results of the flexible MEMS flow sensor are described.