基于openCV的互相关图像测速方法研究

通过对比互相关技术和多普勒技术的差别,从一个新的角度把互相关技术应用于微小气泡测试的研究中.利用高速CCD相机拍摄微小气泡运动图像,利用快速傅里叶变换实现互相关算法,为提高计算准确度采用高斯曲线拟合互相关函数的峰值并给出拟合公式,得到了亚像素级的运动速度.为标定互相关算法的准确度,采用图像平移的方法验证算法的可行性和可靠性.该算法具有潜在的广泛应用价值.

Researches on the Measurement of Cross-correlation images Based on openCV

After comparing the differences between Cross-Correlation and Doppler method, the Cross-Correlation method has been applied to the measurement of the velocity of mini-bubbes in to a new aspect. The high-speed CCD camera is used to acquire the images of moving mini-bubbles, the velocity of mini-bubbles is computed from image post-processing. Through improving cross-correlation algorithm commonly used for analysis of flow field, the Fast Fourier Transform(FFT) has been used to compute the cross-correlation function fastly. In order to enhance the computing accuracy, Gauss curve fitting is used to modify the location of the max value point, and the fitting formula is listed, so the bubble velocity with subpixel accuracy is obtained. In order to calibrate the algorithm, the image transition is used to proceed its feasibility and dependability.