Research On Distortion Correction And Edge Fusion Method For Multi-projection Body Surface

In order to solve the problems of limited scope and complicated operation of traditional minimally invasive surgery,some researchers have proposed a patient's surface projection method based on augmented reality technology,that using a projector to project an image taken by a endoscope to a patient's body surface to help the doctor.Judging from the corresponding operation,this paper combines the projection image algorithm to study the image distortion correction and edge fusion in multi-projection system.In this paper,the causes of the distortion of the projected image are analyzed.The correction method of the quadric surface fitting distortion is studied.The algorithm is analyzed and the correction effect of the method on the complex projection surface is not ideal.Then the distortion correction based on the corresponding point mapping is studied.Methods,through the experimental comparison and analysis of the quadratic surface fitting method and the corresponding point mapping based correction method for the image correction effect,the correction method based on the corresponding point mapping is more accurate on the irregular complex surface.For the edge fusion problem of multi-projection images,this paper studies the fusion method based on quadratic surface fitting.The analysis results show that the method has poor fusion effect on irregular projection surfaces,and then studies an edge based on corresponding point mapping.The fusion method is then combined with the surface fitting method based on multi-point matching and the corresponding point mapping fusion method.The experimental results are analyzed.It is concluded that for most projection surfaces,the edge fusion method based on corresponding point mapping has better Good fusion effect.Aiming at the color difference in the fused area of the fusion image,this paper analyzes the specific reasons for the brightness of the fused image.The linear colorcorrection method based on brightness attenuation and the nonlinear color correction method based on control point are studied.Two color correction methods are used respectively.The experimental simulation is carried out,and the experimental results are analyzed.It is concluded that the nonlinear color correction method based on the control point effectively eliminates the luminance stripe in the image fusion area,and the color correction effect is better.Finally,this paper designs and constructs a multi-projection experimental system.On different projection surfaces,the same cavity image is simulated by surface fitting and corresponding point mapping distortion correction method;the two cavity images are based on quadratic surface.The fusion method of the fitting and the fusion method based on the corresponding point mapping are compared experimentally.At the same time,the linear color correction method based on luminance attenuation and the nonlinear color correction method based on control points are used for color correction of the projected image,and the experimental results are analyzed.The theoretical analysis and research results of this paper are verified.