A CMOS image sensor is an optical sensor and the core component of camera modules, which exerts critical influence over light perception and image quality of cameras.

    With advancing manufacturing processes, CMOS Image Sensor (CIS) technology has achieved remarkable progress with greatly improved performance. Its performance has matched or even surpassed that of CCD sensors. Moreover, CMOS image sensors feature high‑level process compatibility with conventional CMOS chips, making them the dominant sensor type for cameras.

CMOS stands for Complementary Metal‑Oxide‑Semiconductor. It refers to a manufacturing technology for large‑scale integrated circuit chips, or chips fabricated by such technology. On computer motherboards, CMOS is a readable‑writable RAM chip. Thanks to its read‑write capability, it stores hardware configuration data after BIOS setup. This chip is dedicated solely to data storage.

Additional digital signal processing circuits can be integrated onto a CMOS image sensor chip, including AD converters, auto‑exposure control, non‑uniformity compensation, white‑balance processing, black‑level control, gamma correction and more. For high‑speed computation, programmable DSP devices can even be integrated together with CMOS components to form single‑chip digital cameras and image processing systems.

Overall Architecture of Image Sensor

Essentially a semiconductor chip, a CMOS image sensor mainly consists of these modules: photosensitive array (Bayer array / pixel array), timing control, analog signal processing and analog‑to‑digital conversion. Functions of each block are described below:

The pixel array occupies the largest area of the entire chip. Composed of individual pixels, it corresponds to each pixel in output pictures. Each pixel comprises a photosensitive region and readout circuitry (detailed in subsequent sections). Pixel signals go through analog signal processing before being converted by ADC and delivered to digital processing modules.

Working Principle

According to the functional block diagram, the operating workflow of a CMOS image sensor falls into three main steps.

Step 1: Incident light hits the pixel array and triggers the photoelectric effect to generate electric charges within pixel units.

Scenes are focused onto the image‑sensor array via imaging lenses. The array is a two‑dimensional grid of pixels, each fitted with a photodiode. Every photodiode converts incident light intensity into electrical signals.

Step 2: Row‑select and column‑select circuits target designated pixels and read out their electrical signals.

During pixel addressing, row‑selection logic supports progressive or interlaced scanning of the pixel array; the same applies for columns. Coordination between row‑selection and column‑selection logic enables image window extraction.

Step 3: Signal processing for selected pixel units.

Image signals from row pixels are transmitted via column‑shared signal buses to corresponding analog‑signal‑processing units and A/D converters for digital‑image‑signal output. Analog‑signal‑processing units mainly amplify signals and boost signal‑to‑noise ratio.

Amplified pixel signals are fed into the Correlated Double Sampling (CDS) circuit. CDS is an essential technique for high‑performance devices to eliminate interference. Its operating principle relies on dual output channels from the image sensor: one channel delivers real‑time signal while the other provides reference signal. Differential calculation between the two channels removes common‑mode or correlated interference.

　　Processed signals are then sent to analog‑to‑digital converters for digital‑signal output.

To deliver qualified practical cameras, the chip must embed various control circuits for exposure‑time adjustment, auto‑gain control and other functions.

Application Scenarios

Digital Cameras

Controlled by electronic shutters, color CMOS cameras capture photos and store them in DRAM before transferring data to ROM for permanent storage. CMOS chips also handle analog‑to‑digital conversion, load signal processing, white‑balance adjustment and camera control. Almost all entry‑level digital cameras adopt CMOS image sensors today.

Automotive Industry

Covers Rear‑View Camera (RVC), Surround‑View System (SVS), Camera Monitoring System (CMS), FV/MV, DMS/IMS systems and others.

Satellite Remote Sensing

For nano‑satellites weighing under 10 kg, optical imaging (primarily visible‑light band) serves as the primary approach for earth observation missions. Solid‑state imaging devices such as CCD are widely deployed within optoelectronic systems for remote‑sensing imaging.

Benefiting from inherent merits, CMOS imaging devices hold promising application prospects for micro‑nano satellites, including CMOS cameras, satellite sensors, MEMS micro‑gyroscopes, micro‑accelerometers and other components.

Looking ahead, the CMOS image sensor market will keep expanding. Continuous technological innovation unlocks huge growth potential across smartphones, automotive electronics, security surveillance and many other fields.