ID 原文 译文
2453 结果表明，APFCN 在像素精度和交并比等方面均优于 FCN(输入为 RGB图像或红外图像)，适用于背景一致下地面目标的语义分割任务。 The results show that APFCN is better than FCN(input RGB image or infrared image)in PA (Pixel Accuracy)and MIoU (Mean Intersection over Union). APFCN is suit-able for the semantic segmentation task of ground targets with consistent background colors.
2454 为了驱动后级开关功放，全数字发信机需要利用数字射频调制器将输入信号转换为相应的脉冲序列信号。 In order to drive the switched mode power amplifier (SMPA)for full-digital transmitter, the input RadioFrequency (RF)signal of full-digital transmitter first needs to be transformed into pulse sequences by using the digital RFmodulator.
2455 针对现有脉宽调制(PWM)和 Delta-Sigma 调制(DSM)策略在调制性能及系统实现方面存在的不足，本文利用面积等效原理提出一种新的数字射频调制策略。 Considering the disadvantages in modulation performance and system realization of existing digital RF modulation strategies, especially the Pulse Width Modulation (PWM )and Delta-Sigma Modulation (DSM ), a digital RF modulation strategy was proposed and analyzed in this paper by using the area equivalent theory.
2456 理论分析和仿真结果表明，本文策略相比 PWM 和 DSM，不仅简化了硬件实现难度，而且能够获得更优的带内 SNR 和编码效率性能。 The theoretical analysis and simulation results show that, compared to the PWM and DSM, the proposed modulation strategy not only reduces the hardware imple-mentation difficulty, but also achieves better performance of the SNR and coding efficiency.
2457 传感器配准是多传感器数据融合系统获得性能优势的关键前提。 Sensor registration is the key precondition of the performance advantages of the multisensor data fusion system.
2458 受随机噪声、系统误差、虚警、漏报等因素的干扰，传感器配准常常工作在非理想关联环境中，依赖于理想关联假设的传统配准方法性能衰退严重。 In the presence of random errors, sensor biases, false alarms and missed detections, sensor registration usually worksin a nonideal association envrionment. Traditional registration approches relying on the ideal association condition degrade seriously.
2459 另一方面，传统传感器配准方法对目标分布场景敏感，当目标密集分布时，配准问题呈现病态性，估计结果数值不稳定。 On the other hand, traditional registration methods are sensitive to the target distribution. When targets are densely distributed, the registration problem is ill-conditioned and the estimation encounters the numerical instability phenomena.
2460 本文重点研究非理想关联及场景病态性共存时的传感器稳健配准问题，提出了系统误差的岭最小截平方(Ridge LeastTrimmed Squares，RLTS)估计方法。 Fo-cusing on sensor registration in the context of nonideal association and ill-condition, this paper presents the robust registration approach based on ridge least trimmed squares (RLTS).
2461 该方法结合了岭回归(Ridge Regression，RR)与最小截平方(Least Trimmed Squares，LTS)估计的优点，能够有效应对错误关联及病态性的不良影响。 The proposed approach combines the advantages of the ridge re-gression (RR)and the least trimmed squares (LTS)estimation. The RLTS can deal with nonideal association and ill-condi-tion simultaneously.
2462 仿真实验证实了所提方法的稳健性能。 Simulation results verify the robust performance of the RLTS method.