Research On The Optimization Design For The Facular Image Processing Algorithm In Satellite Optical Communications

As the aerospace industry develop rapidly，CMOS image sensors is widely used insatellites, space-based observations and other fields for its excellent performance.Since CMOS image sensor has much advantage such as high integration, low powerconsumption, strong anti-radiation, it has great potential applications in the spacesector. But because of the space radiation environment and high-energy particles andrays eradiating the characteristic of CMOS image sensors, it is easy to aging and toarise dead pixels which maybe affect the spot identification, location, and cause seriousCMOS image sensor invalid.Dead pixels in the CMOS image sensor are usually caused by the following twofactors. One is the manufacturing process. The other is the space environment radiation,Dead pixels on the image display in the three forms, one case is in the image grayvalue of255on the performance of the highlights, and another case is the imageappearing as a dark gray value of0points, the other dead pixels is gray in the imagetable value between0and255, this case is the pixel to be bad but it has not completelybad result. In this paper we research on the CMOS image sensors which is used insatellite optical communication, subject to space radiation dead pixel gray scale valueof255conditions, include following aspects:Research on space radiation environment on the impact of CMOS image sensors.This part of the study includes four aspects:1. Presentation space radiationenvironment.2. Analysis of the space radiation environment of the CMOS imagesensor causes accelerated aging, this part is mainly from the two core CMOS imagesensor photodiode and MOSFET devices starting its parameters by space radiationeffects were analyzed.3. Establish dead pixel CMOS image sensors on the spot centriclocation error mathematical model.4. Introduces the space can be used for imagedemising method and its basic principles.MSG（Modifying Single Gray Value）algorithm and simulation. This part of thestudy includes two aspects:1. MSG algorithm proposed theoretical basis andflowcharts.2. Through the static experiments and simulation experiments, we have thetraditional image demising methods median filter, mean filter, weighted average filtercompares the image de-noising, and then, will be one of the best demised filtering algorithm and MSG algorithm in image demising, spot centric errors were compared.Through dynamic simulation experiment to test MSG is effective. This sectionmainly simulated space environment context appears dead pixel in CMOS imagesensor on the LABVIEW experiment platform, then, respectively, utilize medianfiltering algorithm and MSG algorithm for image demising spot after spot centric errorcorrection for contrast.This paper study on space radiation environment on radiation dead pixel CMOSimage sensor noise to verify the effectiveness of MSG algorithm, in the researchprocess,we obtain large amounts of data which will be the foundation continued tostudy the issue.