ID 原文 译文
5994 提出快速加权核范数最小化(fast weighted nuclear norm minimization,FWNNM)的合成孔径雷达(synthetic aperture radar,SAR)图像去噪算法。 Quick weighted nuclear norm minimization (fast weighted nuclear norm minimization, FWNNM) of synthetic aperture radar (synthetic aperture radar, SAR) image denoising algorithm.
5995 首先采用对数变换将SAR图像的乘性噪声变换为加性噪声; First USES the logarithmic transformation of the SAR image as the additive noise, multiplicative noise transformation and then transform the image to make use of the local similarity matching;
5996 然后利用非局部相似性对变换后的图像进行块匹配,随后根据低秩模型框架,用随机奇异值分解替换加权核范数最小化(weighted nuclear norm minimization,WNNM)算法中的奇异值分解进行低秩矩阵逼近; then according to the low rank model framework, replace with singular value decomposition (SVD) random weighted nuclear norm minimization (weighted nuclear norm minimization, WNNM) algorithm of matrix singular value decomposition of low rank approximation;
5997 再采用梯度直方图保存的方法对图像进行纹理增强; then adopt the method of gradient histogram save for texture image enhancement;
5998 最终实现了对SAR图像快速去噪。 finally achieved the SAR image denoising quickly.
5999 在MSTAR数据库上的实验结果表明,与已有方法相比,所提方法在SAR图像去噪和边缘保持方面是有效的,并且比WNNM去噪速度快3倍。 In MSTAR database on the experimental results show that, compared with existing methods, the proposed method in SAR image denoising and edge is valid, and three times faster than WNNM denoising speed.
6000 针对现有路径规划方案忽略侦察区域优先级以及缺乏对侦察区域中新发生事件的跟踪,而导致规划路径不能适应动态环境和无法根据侦察区域重要性来执行优先侦察等问题。 In view of the existing path planning ignore reconnaissance area and the lack of priority of reconnaissance area new event tracking, causes the planning path can not adapt to the dynamic environment and not according to the importance to carry out reconnaissance area first reconnaissance and other issues.
6001 首先,提出将侦察区域重要性作为必要优化指标,与无人机能耗和飞行风险值等7个优化指标加权联合,构成路径优化过程中评估生成路径的多目标效用函数,从而使得规划路径可以反映侦察区域优先级特性。 First, importance as necessary to optimize the reconnaissance area index is put forward, and the energy consumption of unmanned aerial vehicle (uav) and flight seven optimization indexes such as risk value weighted joint, a route optimization evaluation path is generated in the process of multi-objective utility function, so as to make the planning path can reflect priority features reconnaissance area.
6002 然后,提出了基于事件检测的侦察区域重要性值更新机制以提升路径规划方案对动态环境的适应性。 Then, event detection is proposed based on importance of reconnaissance area value update mechanism to improve the environment of path planning scheme for dynamic adaptability.
6003 最后,采用粒子群算法求解最优路径。 Finally, using the particle swarm algorithm to solve the optimal path.