Algorithm Optimization And Display System Research For Low-coherence Led Illuminated CGH

In recent years,the reliance on visualized information has been increasing for more and more people,while the traditional two-dimensional display technology is approaching a bottleneck,and the three-dimensional display technology has achieved considerable development as the successor.As one of the important fields of 3D display technology,holographic display technology has a perfect basic theory and the best display quality under ideal conditions.However,the laser diode(LD)is used as a light source in traditional holographic technology.The LD is harmful to human eyes,and it brings speckle noise into the display system.In this paper,light-emitting diode(LED)was used to avoid speckle noise in the computer-generated hologram(CGH)system.The negative impact of the LED on the reconstruction was suppressed,and the generation speed of the computed hologram was increased.A pre-dispersion method was proposed in this paper to suppress image blur of LED-illuminated CGH.The LED light source was pre-dispersed by a special device,the generation algorithm of CGH was modified according to the characteristics of the pre-dispersed light source,and the light source was used to reconstruct the corrected hologram.The results of numerical simulations and optical experiments were relatively consistent,and it could be seen that the clarity of the reconstructions were improved,which proved the effectiveness and feasibility of the proposed method.While an LED light source with a full width at half maximum(FWHM)of approximately 30nm was used as a reference light,the average equivalent FWHM of the optical reconstruction was less than 10nm.Then,an accelerating method of heterogeneous computing based on a new look-up table method for LED-illuminated CGH was proposed in this paper.Some additional calculation was performed to generate look-up tables in this method.The CGH could be generated in a new way based on the data in the look-up tables.The calculation result of this method is equivalent to the traditional method,and the calculation speed is faster.This look-up table method could also be accelerated by using heterogeneous computing.The generation of look-up table of high-complexity and low-repetition was handed over to the central processing unit(CPU)of the heterogeneous computer system.Subsequently,the hologram generation of low-complexity and high-repetition was handed over to the graphics processing unit(GPU)to improve the utilization of computing resources and speed up the generation of holograms.For a target image of 256*256 sampling points,the generation time of the computed hologram was shortened to about 20 seconds,and the calculation speed is significantly improved compared with the traditional method.