摘要
提出了一种能够在存在干扰光源的情况下正确判别指针、去除干扰阴影的算法。通过使用Hough概率变换对二值图像处理,通过任意两个不同位置的指针来确定指针的回转中心,根据回转中心坐标固定不变的特点,当图像中出现干扰时,通过图中所有检测到的直线与仪表零刻度指针所在直线的交点与回转中心坐标的大小关系,确定正确代表指针的直线。所提算法提高了仪表检测的稳定性,通过对电流表的判别实验证明了方法的有效性与正确性。
Put forward an algorithm to distinguish pointers,remove interference of shadow,in the presence of interference. By processing binary image,using the probabilistic Hough transform,center of rotation of a pointer is determined by any pointer of two different positions,according to fixed features of coordinates of rotation center,when there is interference in the image,through the relationship between all lines dected and the intersection point of the zero scale pointer of the instrument and the size of the coordinate of rotation center,the straight line which represents pointer correctly is determined. The stability of instrument detection is improved,and the validity and correctness of the method are proved by the experiment of the ammeter.
引文
[1]赵艳琴,杨耀权,田沛.基于计算机视觉技术的指针式仪表示值的自动判读方法研究[J].电力情报,2001(3):39-42.
[2]戴亚文,王三武,李金龙.基于图像处理技术复杂仪表的自动识别[J].计量技术,2003(3):33-35.
[3]陈彬,金连文.一种仪表指针位置检测的中心投影法[J].计算机应用研究,2005,22(1):246-248.
[4]Alegria E C,Serra A C.Automatic calibration of analog and digital measuring instruments using computer vision[J].IEEE Trans on Instrumentation and Measurement,2000,49(1):94-99.
[5]Matas J,Galambos C,Kittler J.Robust detection of lines using the progressive probabilistic hough transform[J].Comput Vision Image Underst,2000,78(1):119-137.