Research On Variable Ratio Spectroscopic High Dynamic Optical Imaging Technology

With the increasing complexity of optical imaging scenes and the continuous expansion of application fields,people have higher and higher requirements for highquality image information acquisition such as resolution and dynamic range.The dynamic range of real-world scenes can exceed 200 dB,but traditional imaging sensors,due to their own limitations,usually have a maximum dynamic range of less than 60 dB,which cannot cover the entire dynamic range in actual scenes.High Dynamic Range imaging technology(HDR)can effectively solve this problem,and has been widely used in mobile devices,machine vision,smart cities,medical health,scientific research and aerospace and other fields.Based on this,this paper studies the HDR optical imaging implementation scheme combining software and hardware,and expects to achieve the purpose of expanding the dynamic range of the imaging system,reducing the hardware resource requirements and simplifying the fusion algorithm to achieve real-time processing through research.The main work of the paper is as follows:1.A survey was carried out on the related technologies of variable ratio spectroscopic HDR optical imaging: firstly,the related technologies of HDR imaging were reviewed,and then the variable ratio spectrophotometric HDR optical imaging technology that could eliminate artifacts and real-time imaging was analyzed in detail.The optical path design of the imaging system and the fusion algorithm are theoretically analyzed and sorted out.2.A variable spectrophotometric HDR imaging system is designed.First,a variable spectrophotometric imaging model is constructed.Simultaneous multiexposure imaging is realized based on a spectroscopic prism,which avoids the problem of image registration.Then based on the radiance transfer calculation,a variable spectroscopic imaging analysis of the system was carried out.Based on this analysis method,combined with a typical CMOS sensor,a variable ratio spectroscopic HDR imaging simulation was carried out,and then the exposure time of the two sensors was continuously changed under the condition of a fixed spectroscopic ratio,and three HDR imaging schemes of 80 dB,100 dB and 120 dB were obtained.The exposure time selection range,and finally the 9:1 beam splitting prism is used to carry out specific calculations to achieve three HDR imaging schemes.3.A variable ratio spectroscopic image fusion method is proposed,which adds local windows to the high-brightness and low-brightness images,and divides the pixel values into three categories: complete overexposure,incomplete overexposure and good overexposure according to the pixel saturation in the highlight image window.a situation.For different exposure situations,different weight coefficients are determined for multi-exposure weighted fusion according to the saturation of neighborhood pixels in the highlight image window.The experimental results show that 1)the local weighted superposition method well preserves the dark detail information and highlight scene information in the two images,and compared with other algorithms,the peak signal-to-noise ratio and structural similarity of the fused images are improved.2)When selecting the window size,the influence of local noise on the quality of the fused image and the constraining ability of the neighboring pixels to the highlight center pixel value should be considered at the same time.3)For low signal-to-noise ratio images,the dark scene information in the low-brightness image can still be effectively recovered after fusion,and the overall imaging effect of the fusion result can be guaranteed.4.A variable ratio spectroscopic imaging system was built,and the multiexposure images with a spectroscopic ratio of 9:1 and an exposure time of 1ms were collected in the experiment,and the fusion image was obtained by algorithm processing.range,indicating that the algorithm has a better performance in improving the dynamic range.