| High-speed moving object tracking is a widely used technology in the fields of military,transportation,and industry.Most of high-speed object tracking technology obtain the spatial position of the target object by methods based on image information.These methods need to collect image sequences,which has a large data size,for analysis.The large size of redundant information of images,image acquisition,data transmission,data storage and so on,brings great pressure to the limited bandwidth.And the high complexity of positioning algorithm generally makes it difficult to real-time tracking and track high-speed moving objects for a long time.Structured light active illumination detection technology is a novel signal detection technology.It uses spatial light modulator to produce spatial modulation of structure light field,which is used to illuminate target object.And it encodes the spatial information of the target by spatial light which is produced by a spatial light modulator.Then it uses a single-pixel detector as a signal acquisition device to obtain the light signal which is modulated by the target.Finally,the target will be restored by the reconstruction algorithm.For reducing the sample redundant data of target scene,this thesis proposes a high-speed moving object tracking technology with spatial light modulation based on the Fourier single pixel imaging principle.By means of structured light active illumination detection,high-speed moving object tracking technology based on Fo urier slice theorem and high-speed moving object tracking technology based on Fourier shift theorem are proposed respectively.The method based on the Fourier slice theorem,whose scene is illuminated by the structure light of the Fourier pattern.The information of the scene(including the information of the moving object)is encoded into the Fourier domain,and the signal is collected by the single-pixel detector.This method collects the slices which located in k_x axis and k_y axis of the Fourier transform spectrum of the scene.And then,this method obtains its projection in the x and y directions by inverse Fourier transform,thereby calculating the moving object's spatial coordinates.In order to verify the feasibility of the method,numerical simulation and physical experiment are carried out.The frame rate of these method can reach 556fps when the projection frequency of structed light is 10k Hz,and the data size of location in single frame is 1 800 bytes.The method based on the Fourier shift theorem,whose scene is also illuminated by the structure light of the Fourier pattern.The information of the scene(including the information of the moving object)is encoded into the Fourier domain,and th e signal is collected by the single-pixel detector.The difference between the above method,these method calculates the relative displacement of the moving target between the previous and subsequent frames by Fourier shift theorem.Finally,applying this method to image sequences,and obtain the trajectory of the target.In order to verify the feasibility of the method,numerical simulation and physical experiment are carried out.The frame rate of these method can reach 1667fps when the projection frequen cy of structed light is10k Hz,and the data size of location in single frame is 600 bytes. |