Pixel structure of high-speed global shutter and signal control method of pixel structure

Abstract

The invention discloses a pixel structure of a high-speed global shutter. According to the pixel structure, innovation is made to the electric structure of a sampling switch, and two sampling capacitors can be independently managed, so that sampling capacitance during charge transfer in the process of reading does not change; in addition, the invention further discloses a signal control method applied together with the pixel structure. The pixel structure and the signal control method thereof have the advantages that voltage amplitude of output signals of the pixel structure can be improved, dynamic range of pixels can be widened, and image gradation quality can be improved; meanwhile, indexes such as CVF, PLS and noise of a typical 8T structure are not changed.

Description

technical field [0001] The invention relates to a CMOS image sensor, in particular to a high-speed global shutter pixel structure and a signal control method thereof. Background technique [0002] When shooting a high-speed moving object, it is necessary to use a global shutter to avoid image distortion; on the existing CMOS image sensor, the typical pixel structure used to realize the global shutter function is as follows: figure 1 As shown, the pixel structure is changed based on the traditional 4T pixel structure, which consists of a photodiode, a transmission tube, a reset switch, a source follower, a current source load, two sampling switches (M5 and M6), two sampling Composed of capacitors (C1 and C2), source follower 2 and row selection switch; its working principle is: 1) Drive the transfer tube and reset switch to close to reset the photodiode, and then disconnect the transfer tube and reset switch, the image Start the exposure; 2) After the exposure starts, cl...

AI Technical Summary

Problems solved by technology

[0004] The principle of the present invention is: in the prior art, since the two sampling capacitors are only connected by one sampling switch, when the optical signal is read out, the optical signal charge is distributed in the two sampling capacitors at the same time, and finally the row selection switch outputs light The voltage amplitude of the signal will be reduced by C1·(V reset -V signal ) / (C1+C2), and after adopting the scheme of the present invention, the two sampling capacitors can be managed separately, and the size of the sampling capacitor remains unchanged during the charge conversion in the readout process, which solves the problem that the voltage amplitude of the output signal is reduced Problem; at the same time, the pixel can adopt correlated double sampling, which eliminates KTC noise; moreover, the storage environment of the reset voltage signal and the photovoltage signal is the same, and has a small PLS through correlated double sampling; compared with the typical 8T pixel structure, there is no reduced PLS , CVF, noise and other indicators

Images