多直线目标的影像畸变计算

Image Distortion Calculation Based on Multi-line Targets Condition

对于未知来源或未知摄影机摄取的近景影像,通过提取影像上多个直线目标,恢复其曲线构像为直线的过程来计算影像畸变。首先分析影像畸变类型,确定影像畸变模型,然后采用二分法进行多个直线目标影像畸变的计算方法。经过室内、室外场地的实例计算,该方法计算得到的影像畸变参数与三维鉴定场计算的结果非常接近。试验表明该方法可行,能满足小像幅图像解算的精度要求,适用于历史影像及未知来源的影像解算和纠正。

For the unknown sources or unknown camera close range images, the distortion is calculated by extracting the multiple linear targets in the image, and recovering the curve conformation as a straight line. Firstly, analyze the types of image distortion, determine the image distortion model, and then calculate the distortion of multiple linear objects by applying dichotomy method. Through examples calculating of indoor and outdoor test sites, the image distortion parameters calculated by this method are almost same to the results calculated by the three-dimensional calibration field. The experiments results show the method is feasible and can satisfy small image solution precision requirements, suitable for historical images and unknown source images calculation and correction.