The CMOS image sensors are used for more and more remote sensing camera systems because the system with the CMOS sensor consumes less power and can be made compact. Drawbacks of the CMOS sensors from the point of performance are being overcome. TDI(Time Delayed Integration) function is implemented in the CMOS image sensors to increase the signal to noise ratio in the high resolution earth observation camera systems. Intrinsic weak point of the TDI CMOS sensors, compared to the TDI CCD sensors, is the motion blur that affects image quality when it takes images by scanning the targets. Several ways to mitigate the motion blur have been introduced. As the motion blur is not critical in the TDI CCD camera thanks to the multi-phased clocking scheme, some TDI CMOS sensors are implemented to sample the signal more than once to get better synchronization. One active area (photo diode) can be divided into two or more sections for better synchronization between scanning and sampling. Physical masks for spatial gaps between pixels can be placed to mitigate the motion blur in the scanning direction. In this case, the motion blur is minimized effectively, but the amount of the collected signal is reduced and eventually the signal to noise ratio is reduced.
Two contradictory performance parameters, signal to noise ratio and MTF can be optimized by determining the size of the pixel masks optimally after the analysis and simulation. Key design parameters for the camera system have been set in this paper. Then, it has been analyzed how the size of the pixel mask affects signal to noise ratio and MTF performance. The methods and results have been presented. The motion blur can also be reduced by adjusting the integration time in the TDI CMOS camera system. Parts of the line time, not whole of the line time, can be used for signal integration and the amount of the motion blur will be reduced accordingly. As the integration time decreases from 100% of the line scan time to 90%, 80%, and 70%, it has been shown by the analysis and simulation how these two performances are being affected respectively. As the signal to noise ratio decreases from 215 to 202, 189, and 174, the dynamic detector MTF increases from 40.6% to 44.5%, 48.2%, and 51.6%.
The results of analysis and simulation show that reduction of the motion blur with the pixel masks can also be achieved by reducing the integration time from the line scan time. Lots of efforts need to be invested to implement the pixel masks on the CMOS sensors and the masks will be permanent once it have been implemented. However, the method to control the motion blur by adjusting the integration time can be flexibly applied in the on-orbit operation phase, not in the system design phase. It is possible with the help of versatile camera control electronics to focus on one of the two performances, signal to noise ratio and MTF, depending on the amount of radiance from the target and the aim of the target imaging.