Abstract: The present application provides an image stitching method, apparatus and device based on reinforcement learning and a storage medium. The method includes: acquiring initial calibration parameters, collecting a sample image and position information of a motion platform; setting a negative reward function; acquiring a state set and a negative reward value set according to a randomly generated action set, the initial calibration parameters, the position information of the motion platform and the negative reward function to construct a probability kinematics model; constructing a state value function based on an occurrence probability of the state, and acquiring an optimal action by optimizing the state value function; and acquiring optimized calibration parameters through the optimal action and the initial calibration parameters, and carrying out image stitching on corresponding sample images through the optimized calibration parameters.

Abstract: The present application provides an image stitching method, apparatus and device based on reinforcement learning and a storage medium. The method includes: acquiring initial calibration parameters, collecting a sample image and position information of a motion platform; setting a negative reward function; acquiring a state set and a negative reward value set according to a randomly generated action set, the initial calibration parameters, the position information of the motion platform and the negative reward function to construct a probability kinematics model; constructing a state value function based on an occurrence probability of the state, and acquiring an optimal action by optimizing the state value function; and acquiring optimized calibration parameters through the optimal action and the initial calibration parameters, and carrying out image stitching on corresponding sample images through the optimized calibration parameters.

Abstract: Provided are a neutralizing antibody, which can specifically bind to NGF, or an antigen-binding fragment thereof, and a preparation method for producing the antibody. Further provided is an application of the antibody in the treatment of NGF-related diseases or disorders, especially in relieving or improving individual pain.

Abstract: An XY macro-micro motion stage and a control method thereof based on end feedback. The XY macro-micro motion stage includes a fine-adjusting assembly, a macro-driving assembly and an XY position detecting device. The macro-driving assembly includes an X-direction macro-driving unit, a Y-direction macro-driving unit, an X-direction mounting plate and a Y-direction mounting plate. The XY position detecting device is configured to obtain its position on the X axis and Y axis.