期刊文献+
共找到3篇文章
< 1 >
每页显示 20 50 100
基于GA-SVR模型的无参考立体图像质量评价 被引量:8
1
作者 马允 王晓东 +2 位作者 富显祖 娄达平 章联军 《计算机工程》 CAS CSCD 北大核心 2017年第5期234-239,247,共7页
针对支持向量回归(SVR)中惩罚因子和径向基函数选取具有较大不确定性和随机性的问题,结合单双目信息与基于遗传算法(GA)的SVR优化模型,提出无参考立体图像质量评价方法。提取左右失真图像的单双目特征,将梯度幅值和拉普拉斯特征作为单... 针对支持向量回归(SVR)中惩罚因子和径向基函数选取具有较大不确定性和随机性的问题,结合单双目信息与基于遗传算法(GA)的SVR优化模型,提出无参考立体图像质量评价方法。提取左右失真图像的单双目特征,将梯度幅值和拉普拉斯特征作为单目视觉特征。为更好地结合人类双目视觉特性,使左右图像融合成一幅独眼图,对独眼图提取空域自然场景统计特征。利用GA选择、交叉和变异等操作优化SVR参数组合,选出最优的参数组合,引入到SVR中预估左右图像质量。考虑到人眼对于左右失真图像的响应不同,通过增益控制模型融合左右图像质量,从而得到最终的质量评价值。应用该评价方法对宁波大学建立的立体数据测试库进行评价,结果表明其Pearson线性相关系数在0.95以上,Spearman等级相关系数值在0.94以上,与人类主观感知具有高度一致性。 展开更多
关键词 遗传算法 支持向量回归 立体图像质量评价 单双目视觉特性 增益控制模型
下载PDF
Using Adaptive Gain Scheduling LQR Method Control of Arm Driven Inverted Pendulum System Based on PIC18F4431 被引量:1
2
作者 Huu Chan Thanh Nguyen An-Wen Shen 《Journal of Harbin Institute of Technology(New Series)》 EI CAS 2013年第4期85-92,共8页
The arm driven inverted pendulum system is a highly nonlinear model, muhivariable and absolutely unstable dynamic system so it is very difficult to obtain exact mathematical model and balance the inverted pendulum wit... The arm driven inverted pendulum system is a highly nonlinear model, muhivariable and absolutely unstable dynamic system so it is very difficult to obtain exact mathematical model and balance the inverted pendulum with variable position of the ann. To solve this problem, this paper presents a mathematical model for arm driven inverted pendulum in mid-position configuration and an adaptive gain scheduling linear quadratic regulator control method for the stabilizing the inverted pendulum. The proposed controllers for arm driven inverted pendulum are simulated using MATLAB-SIMULINK and implemented on an experiment system using PIC 18F4431 mieroeontroller. The result of experiment system shows the control performance to be very good in a wide range stabilization of the arm position. 展开更多
关键词 Arm Driven Inverted Pendulum (ADIP) adaptive gain scheduling LQR control LQR control swing up pendlum
下载PDF
Gain-scheduling model predictive control for unmanned airship with LPV system description 被引量:1
3
作者 Liu Cui Li Chen Dengping Duan 《Journal of Systems Science & Complexity》 SCIE EI CSCD 2015年第5期1043-1051,共9页
This paper presents a gain-scheduling model predictive control(MPC) for linear parameter varying(LPV) systems subject to actuator saturation. The proposed gain-scheduling MPC algorithm is then applied to the lateral c... This paper presents a gain-scheduling model predictive control(MPC) for linear parameter varying(LPV) systems subject to actuator saturation. The proposed gain-scheduling MPC algorithm is then applied to the lateral control of unmanned airship.The unmanned airship is modeled by an LPV-type system and transformed into a polytopic uncertain description with actuator saturation. By introducing a parameter-dependent state feedback law, the set invariance condition of the polytopic uncertain system is identified. Based on the invariant set, the gain-scheduling MPC controller is presented by solving a linear matrix inequality(LMI) optimization problem. The proposed gain-scheduling MPC algorithm is demonstrated by simulating on the unmanned airship system. 展开更多
关键词 unmanned airship gain-scheduling model predictivecontrol polytopic uncertain actuator saturated systems
原文传递
上一页 1 下一页 到第
使用帮助 返回顶部