Agricultural Unmanned Aerial Vehicle (AUAV) is widely used in the fields of aviation plant protection, remote sensing, yield estimation and so on.Acquiring the heading angle changing is important to flight path planning and control for the AUAV.Founding a kind of attitude solution algorithm which can navigate out heading angle changing and the posture information relative to the ground of AUAV, and then improve the navigation accuracy has becoming the focus of attention.In the past, the attitude measurement methods for small unmanned aerial vehicle are to use the gyroscope and the positioning system in the heading angle measurement.But this kind of heading angle measurement method is vulnerable to electromagnetic interference, which will affect the accuracy of heading angle for the unmanned aerial vehicle.In this paper, the Industrial Personal Computer (IPC) has been chosen as the hardware development platform, and the program of image processing has been designed based on MATLAB.According to the characteristics of visualized environment for AUAV working, a method based on the dynamic change of farmland environment feature points to estimate the changes of AUAV heading angle has been propose.It avoids the limitation of traditional machine vision algorithm which based on crop rows for acquiring the changes of heading angle.A dynamic variable interested area has been established in two consecutive frames, respectively.Then Speeded Up Robust Features (SURF) algorithm has been used to detect the key features points of farmland environment, to realize precise positioning of key feature points.The key feature points are described, and then the key feature point's descriptors have been constructed.Then local feature points have been extracted from farmland environment in the image.In two continuous images, the image of the previous frame is used as a reference frame, and then the next frame image has been matched with the image of the previous frame.Thus,the local feature points get from two consecutive frames have been matched.The relative motion parameter of the two consecutive frames can be obtained.According to the relative motion characteristics between AUAV and farmland, the image matching technology has been used to acquire its heading angle changes.Field experiment shows that the algorithm can extract heading angle changes of AUAV in irregular farmland environment.