Image processing method, and photographing device

A photographing device, comprising an inertia measurement unit, a memory, a processor and a computer program stored in the memory and executable by the processor, wherein the processor is configured to, when executing the computer program, implement an image processing method, comprising:

• obtaining first estimated data, wherein the first estimated data is posture data of the inertial measurement unit corresponding to a previous moment of an estimated specified moment;
• obtaining second estimated data according to the first estimated data, wherein the second estimated data is estimated posture data of the inertial measurement unit corresponding to the estimated specified moment;
• determining a target scaling parameter matrix and a target covariance matrix, wherein the target scaling parameter matrix is a scaling parameter matrix corresponding to a temperature value of an environment where the inertial measurement unit at the specified moment is located, and the target covariance matrix is a covariance matrix of measured noises of the inertial measurement unit, which corresponds to the temperature value of the environment where the inertial measurement unit at the specified moment is located;
• determining a Kalman filtering parameter corresponding to the specified moment according to the target scaling parameter matrix and the target covariance matrix;
• obtaining target posture data according to the Kalman filtering parameter corresponding to the specified moment, the second estimated data and measurement data of the inertial measurement unit, wherein the target posture data represents a posterior state quantity corresponding a posture of the inertial measurement unit at the specified moment;
• and
• performing a rotation processing on the image photographed by the photographing device according to a rotation matrix corresponding to the target posture data.