Based on the principle and design of a novel CMOS image sensor 2

Each pixel of the four field-effect transistor (4T) active pixel CMOS image sensor consists of a photodiode, a reset transistor T2, a transfer transistor T1, a source follower T3, and a row gate switch T4, as shown in Figure 2.

Transfer tube T1 is used to connect the photodiode to source follower T3 and to connect it to VDD through reset tube T2. The gate of T3 is connected to the N+diffusion region between T1 and T2. Compared with APS with 3T structure, it reduces the leakage current effect related to the gate of T3. The function of the source follower T3 is to amplify and buffer the signal, and improve the noise problem of APS. T4 is used to connect signals to the column bus. The working process is as follows: first enter the "reset state", T2 opens, and the photodiode is reset; transistor Then enter the "sampling state", T2 is turned off, and light is irradiated onto the photodiode to generate photo generated carriers, which are amplified and output through the source follower T3; Finally entering the 'read state', the row gate T4 is turned on, and the signal is output through the column bus.transistor

APS has the advantages of low readout noise and high readout rate, but the pixel unit structure is complex and the fill factor is reduced, usually only 20% to 30%. In order to improve the fill factor of pixels, APS has set micro lenses above the pixels, as shown in Figure 3.transistor

The image information obtained from the APS array is converted into a digital signal by the Column ADC in Figure 1, and then undergoes a series of subsequent processing steps to obtain the output frame image data structure shown in Figure 4.transistor