Research On Adaptive Multi-scale 3D Ghost Imaging Method

Ghost imaging(GI),also known as correlation imaging,is a novel imaging method that images with correlation between light fields.Compared with traditional imaging technology,ghost imaging performs better in the field of anti-scattering and anti-air turbulence imaging.Also,ghost imaging can be realized without lens and recently,three-dimensional imaging technology based on ghost imaging has appeared.Of course,ghost imaging technology has some limitations.The most obvious is its rather long imaging time,including the time of data acquisition and post-processing.Furthermore,the higher the resolution of the acquired image,the larger the amount of data required,and the longer the sampling and reconstruction time.Therefore,the resolution and sampling rate of ghost imaging system are a contradiction.This contradiction also limits the development of three-dimensional ghost image technology.Aiming at this issue,we propose a multi-scale adaptive three-dimensional ghost imaging technology based on the idea of adaptive compressed sampling(ACS).The main work of this paper are as follows:(1)Based on the concept of surface normals,we explore the compressibility of threedimensional information in this paper.The local flatness degree of the target surface is calculated with surface normals.For the flat area,the three-dimensional shape can be restored with less samplings.For the non-flat area,relatively more measurements are required to ensure the reservation of details of the target’s 3D shape.(2)We propose a multi-resolution 3D ghost imaging scheme,combining photometric stereo vision technology.The imaging region is divided into two parts according to the object’s local flatness degree.This degree is evaluated by the variation of surface normal calculated by low-resolution images.In the flat area,sampling can be reduced,while more samples are performed in the unflat area to maintain the stereo feature of the object.By this way,the redundancy in sampling process can be digged out to reduce the number of measurements and thus achieve adaptive compressed sampling.This technology can alleviate the contradiction between sampling rate and image resolution,and of course,it is in favor of improving the speed of imaging system.(3)Also,benefiting from the flexible of the single-pixel configuration,our technique can be extended to infrared wavebands.In addition,we propose a portion imaging method based on Hadamard,which makes full use of the orthogonality of Hadamard matrix and takes advantage of single-pixel imaging at the same time.The imaging area of traditional imaging based on array camera is usually in a shape of square,while the imaging area of the single-pixel imaging technology is determined by the modulated light field.An image of the region of interest in any shape can be directly obtained by modulating a measurement template.This kind of regional imaging method can capture high-quality images of the target with fewer measurements.Also,compared with imaging a region by scanning point by point,the method based on Hadamard modulation is more resistant to noise.The results of simulation and experiment results show that our proposed scheme can reduce the number of samples while preserve the quality of the formed 3D image.The sampling compression ratio is no less than 25%.The error of the face model used in the experiment is increased by 0.61 mm compared with the traditional scheme,and the error of the simple test target is reduced by 0.49 mm.The error fluctuations are less than one pixel(1.25mm),which means,the accuracy of our results isn’t depressed.Also,because of the decrease of the sampling numbers for flat area,the noise introduced by measurement process reduced,and the image quality is improved to some extent.