共查询到17条相似文献,搜索用时 62 毫秒
1.
2.
3.
采用基于支持向量机的模型预测控制法来实现非线性模型预测控制.控制器设计采用改进MOUSE(Modifled Univariate Search)方法来解决非线性约束优化问题.其具体实现通过计算种群适应值函数、检查每个解是否满足约束条件、选取一定数量产生子代的种群计算相应适应值,更新设计变量等步骤完成. 相似文献
4.
5.
6.
7.
8.
9.
10.
支持向量机的手写体数字识别中,采用美国邮政服务数据库.并取多个2层神经网络中的最好者得出2层神经网络结果,专门设计5层卷积神经网络Lenetl.所有的结果均直接采用点阵输入,将像素值归正到相应区域间,且不施加任何预处理.该方法与人工分类、神经网络、决策树等方法比较,其测试误差低,测试速度高. 相似文献
11.
12.
13.
14.
基于最小二乘支持向量机时域背景预测的红外弱小目标检测 总被引:2,自引:1,他引:2
针对信噪比较低时,如何有效地抑制自然背景对目标检测的影响,提出了一种基于最小二乘支持向量机(LS-SVM)时域背景预测的红外弱小目标检测方法。首先针对前几帧图像中对应同一位置像素点的灰度值序列,利用参数经粒子群优化的最小二乘支持向量机进行函数拟合,并据此预测下一帧图像在该位置处像素点的灰度值;然后将原始图像与预测图像相减得到预测残差图像,利用基于二维Tsallis-Havrda-Charvat熵的阈值选取快速算法进行分割,并根据小目标运动的连续性和轨迹的一致性进一步分离噪声和小目标。文中给出了实验结果及分析,并与现有的检测红外小目标的空域和时域背景预测算法进行了比较。结果表明所提出的算法具有更高的检测概率,明显优于已有的基于背景预测的红外小目标检测算法。 相似文献
15.
SVM handles classification problem only considering samples themselves and the classification effect depends on the characteristics of the training samples but not the current information of classified problem.From the phenomena of data crossing in systems,this paper improves the classification effect of SVM by adding the prior probability item reflecting the classified problem information into the decision function,which fuses the Bayesian criterion into SVM.The detailed deducing process and realizing steps of the algorithm are put forward.It is verified through two instances.The results showthat the new algorithm has better effect than the traditional SVM algorithm,and its robustness and sensitivity are all improved. 相似文献
16.
针对传统建模方法无法解决“小样本、高维度、非线性”类大型复杂装备系统装备的费用估算问题,提
出基于支持向量机(support vector machine,SVM)的方法。针对SVM 方法固有的缺陷进行优化改进。结果表明,该
方法为装备全寿命费用估算做了有益探索。 相似文献
17.
Threat Assessment of Targets Based on Support Vector Machine 总被引:1,自引:0,他引:1
In the context of cooperative engagement of armored vehicles, the threat factors of offensive targets are analyzed, and a threat assessment (TA) model is built based on a support vector machine (SVM) method. The SVM-based model has some advantages over the traditional method-based models: the complex factors of threat are considered in the cooperative engagement; the shortcomings of neural networks, such as local minimum and "over fitting", are overcome to improve the generalization ability; its operation speed is high and meets the needs of real time C2 of cooperative engagement; the assessment results could be more reasonable because of its self-learning capability. The analysis and simulation indicate that the SVM method is an effective method to resolve the TA problems. 相似文献