In order to improve the accuracy of wireless network positioning,the triangulation method of wireless network positioning technology is proposed,which is based on the linear least square fitting method.It makes the ob...In order to improve the accuracy of wireless network positioning,the triangulation method of wireless network positioning technology is proposed,which is based on the linear least square fitting method.It makes the observed value and the fitting value very close,effectively solves the problem of significant contradiction between the fitting result and the observed value in the principle of least square method,and can realize the accurate measurement of geographic information by wireless network positioning technology.展开更多
为解决指针式仪表示数读取中识别精度低和算法读取速度慢的问题,提出一种基于戴明回归和感兴趣区域(region of interest, ROI)细化的指针式仪表读数技术.给出了仪表示数读取的算法流程:首先选择ROI,采用基于颜色通道的剪影法和二值化形...为解决指针式仪表示数读取中识别精度低和算法读取速度慢的问题,提出一种基于戴明回归和感兴趣区域(region of interest, ROI)细化的指针式仪表读数技术.给出了仪表示数读取的算法流程:首先选择ROI,采用基于颜色通道的剪影法和二值化形态学操作进行图像预处理;接着运用图像帧差法消除指针的抖动;然后利用ROI细化算法对待识别仪表的指针进行细化;再使用戴明回归法拟合出仪表指针所在直线的方程和斜率;最后根据指针斜率利用角度法计算仪表的实时示数.通过3组试验,测试了该方法的可行性和防抖动能力,比较了戴明回归拟合直线与霍夫直线检测拟合直线的检测精度,还比较了ROI细化算法与全局细化算法的计算速度.结果表明该方法检测的平均误差比霍夫直线检测减小了37.85%,每张图像的平均计算时间比全局细化算法减少了192.717 s,同时具有防抖动能力.展开更多
文摘In order to improve the accuracy of wireless network positioning,the triangulation method of wireless network positioning technology is proposed,which is based on the linear least square fitting method.It makes the observed value and the fitting value very close,effectively solves the problem of significant contradiction between the fitting result and the observed value in the principle of least square method,and can realize the accurate measurement of geographic information by wireless network positioning technology.
文摘为解决指针式仪表示数读取中识别精度低和算法读取速度慢的问题,提出一种基于戴明回归和感兴趣区域(region of interest, ROI)细化的指针式仪表读数技术.给出了仪表示数读取的算法流程:首先选择ROI,采用基于颜色通道的剪影法和二值化形态学操作进行图像预处理;接着运用图像帧差法消除指针的抖动;然后利用ROI细化算法对待识别仪表的指针进行细化;再使用戴明回归法拟合出仪表指针所在直线的方程和斜率;最后根据指针斜率利用角度法计算仪表的实时示数.通过3组试验,测试了该方法的可行性和防抖动能力,比较了戴明回归拟合直线与霍夫直线检测拟合直线的检测精度,还比较了ROI细化算法与全局细化算法的计算速度.结果表明该方法检测的平均误差比霍夫直线检测减小了37.85%,每张图像的平均计算时间比全局细化算法减少了192.717 s,同时具有防抖动能力.