Research On Compressed Sensing Method Of Scattered Light Signal Of Particles Based On DMD

In recent years,due to the increased pollution caused by industrial emissions and the frequent occurrence of global natural disasters,the harmful effects of particulate matter on human health,the global environment and climate,and industrial production are becoming more and more serious.And the importance of particle measurement technology is also gradually increasing.The particle measurement method based on scattered light signal has the advantages of high measurement efficiency,high measurement accuracy,wide measurement coverage,real-time online measurement and so on.Therefore,it has become one of the most important particle measurement techniques.Generally,the traditional particle measurement device based on scattered light signal receives the scattered light signals of particles at one or several specific angles,and send discrete samples of these to the back-end processor to analysis in order to obtain the relevant parameters of particles.The method of combining particle scattering and scattered light imaging can not only receive more comprehensive multi-dimensional scattered light signals of particle,but also obtain more detailed information by using image processing method.However,the traditional digital imaging system is limited by the Nyquist sampling theory.It needs to initially sample the original image at a very high sampling rate,but discard a large part of these values in the later compression stage,which is a great waste of the hardware resources of imaging.Moreover,the time cost required for sampling and compression is more and more expensive,and the number of image sensors is also increasing,which is a critical challenge for traditional digital imaging technology.The Compressed Sensing(CS)theory,which breaks through the restriction of the Nyquist sampling theory,can achieve sampling and compression simultaneously in one-step data measurement by flipping the mirrors of the Digital Micro-mirror Device(DMD)rapidly.Then the time cost of compression is removed completely and the number of image sensors is reduced successfully.And the heavy task of signal processing is transferred from the front-end sensors to the back-end data processors.In this paper,the traditional particle measurement method based on scattered light of particles and the single-pixel imaging method of scattered light signal based on the CS theory are combined to design and implement a simulation of the DMD based single-pixel imaging system for the scattered light signal of particles.Based on simulated images,the proposed compressed sensing method for imaging scattered light signal of particles is compared with several classical image compression and decompression methods in terms of compression and reconstruction.And the results show the advantages of the CS theory in the image processing of scattered light signal of particles.