CMOS image sensor (CIS) is used in various applications such as surveillance cameras, automobile cameras, mobile phones and digital single lens reflex (DSLR). The photodetectors used in the CIS are p-n junction photodiodes, pinned photodiodes, MOSFET-type photodetectors, and bipolar junction transistor-type photodetectors. A CMOS active pixel sensor (APS) with adjustable sensitivity is presented which uses MOSFET-type photodetector with a built-in transfer gate. The sensitivity of the APS using the MOSFET-type photodetector is much higher than that of the APS using the pn junction photodiode, since the MOSFET-type photodetector is composed of a floating-gate tied to an n-well and the photocurrent is amplified by the MOSFET. Although the APS using conventional MOSFET-type photodetector cannot control the current flowing through the channel, the APS using MOSFET-type photodetector with a built-in transfer gate can control the photocurrent by adjusting the pulse level of the transfer gate. Since the transfer gate controls the amount of electric charge that is transferred from the drain of the MOSFET to the integration node, the sensitivity of the APS can be adjusted by controlling the pulse level of the transfer gate. Using the high sensitivity characteristic of MOSFETtype photodetector and the function of transfer gate, the APS maintains high sensitivity under low intensity of illumination and adjusts to low sensitivity under high intensity of illumination. These results might be useful for extending the dynamic range of the APS using the MOSFET-type photodetector. The CMOS APS was designed and fabricated using 2-poly 4-metal 0.35 μm standard process and its performance was evaluated.
A CMOS image sensor with off-center circular apertures for two-dimensional (2D) and three-dimensional (3D) imaging was fabricated, and its performance was evaluated, including the results of 2D and 3D images. The pixel size, based on a four-transistor active pixel sensor with a pinned photodiode, is 2.8 μm × 2.8 μm. Disparate images as well as focused images for depth calculation can be obtained using the designed pixel pattern. The pixel pattern is composed of one white subpixel with a left-offset circular aperture, a blue pixel, a red pixel, and another white subpixel with a right-offset circular aperture. The proposed technique was verified by simulation and measurement results using a point light source. In addition, the depth image was implemented by calculating the depth information from the 2D images.