视觉测量红外光点图像中心精确提取

提出了一种视觉测量中的红外光点图像中心提取方法。首先,使用一组阈值平面与红外光点灰度能量分布相交确定光点能量等高线,得到各等高轮廓点和阈值平面截得的能量包络。然后,利用最小二乘椭圆拟合等高线上轮廓点的椭圆中心,并计算阈值平面所截得的能量包络的质心,以椭圆中心和包络质心的黄金分割点作为等高层面上的光点中心。最后,利用各等高层面光点的中心获得红外光点图像中心的精确位置。试验验证结果表明,算法对红外光点图像中心提取可取得较高精度。目前,方法已在面阵静态红外地球敏感器标定中得到应用。

High precision extraction of infrared light spot center in vision measurement

An extraction method for center location of infrared light spot image in vision measurement was proposed. Firstly,the contour line of spot gray energy is segmented by the intersection of a group of threshold planes and infrared spot gray level energy distribution. The contour points and energy envelopes of each contour are obtained. Then,by using least square fitting method,the ellipse center of contour points is calculated. The centroid of spot energy envelope which is generated by threshold plane is gotten. The golden section point of ellipse center and envelope centroid is taken as the center of each contour. Finally,infrared light spot center location is extracted by the centers of each contour. Experimental results show that the proposed method has high accuracy for center extraction of infrared light spot image. The method has been used for the calibration of array static infrared earth sensor.