Research On Imaging Technology Of High Dynamic Range Pixel-level Light Adjusting Based On Digital Micromirror Device

In the real world,there is a great variety in the brightness of scenes,ranging from direct sunlight to the deepest shadow.But when capturing scenes that include drastic lighting changes,the stronger light saturates the receiving elements and the actual radiance can not be obtained by normal digital cameras.High dynamic range images offer better visual quality by allowing a wider range of luminance and the visual experience in much closer to reality.It has been widely used in military,medical,automotive,astronomy and other fields.However,most of the high dynamic range imaging techniques have their own limitations,and sometimes the dynamic range they can reach is not enough.This dissertation presents a high dynamic range pixel-level light adjusting and imaging technology based on digital micromirror device(DMD).It uses DMD as a spatial light modulator.The intensity of each pixel on the image sensor can be modulated separately,so that the dark target and bright target of the scene can be detected at the same time.Thus the dynamic range of the imaging system can be improved.By analyzing the working principle of DMD,we have designed a binocular high dynamic range pixel-level light adjusting and imaging system.The system uses two sCMOS cameras which have the same property as the image acquisition equipments.Its theoretical dynamic range can reach 209 dB,the dynamic range of experimental demonstration is higher than 180 dB.We have designed an experiment to verify the modulation accuracy of DMD.The performance parameters of the sCMOS camera are tested according to the EMVA 1288 standard.An experimental platform is built to verify the modulation linearity of the system.We have proposed a real time binocular self-calibration method to calibrate the two cameras at the same time.This method is simple but effective,it does not require external calibration object,and the system itself can correct the distortion of the two image sensors and make their pixels of the same coordinate correspond to the same micromirror.Experiments indicate that the method can calibrate the two image sensors of the system at the same time,the average deviation of the calibrated sensor is less than 1pixel and the average deviation between the two sensors is only 0.57 pixels.We have proposed several light adjusting algorithms for different high dynamic scenes and different observation requirements.Experimental results indicate that these algorithms can effectively adjust the high dynamic scene.A high dynamic scene recovery method and an improved Tone Mapping Operator are proposed.The recoverd scene can show all the details of the scene and has a rich contrast.The image conforms to the subjective feelings of the human eyes.