Image Geometric Correction Based on Android Phone Sensors

For the purpose of image distortion caused by the oblique photography of Android CMOS camera, a auto-rectification approach was proposed in this thesis. According to the camera stance information took into geometric correction model and provided by the phone built-in acceleration sensor and geomagnetism sensor, the relationship of pixel coordinates between distortion image and standard image was obtained. The pixel coordinates was resampled to obtain orthorectified images without the squint distortion by OpenCV4Android perspective projection transformation. The experiment results show that the approach is efficient, real-time and has great significance for implementing precise position of the target.