17.1 CCD and CMOS image sensors

Image sensors are the heart of digital cameras, converting light into electrical signals. CCD and CMOS are the two main types, each with unique strengths. CCDs excel in image quality, while CMOS sensors offer faster processing and lower power consumption.

These sensors use photodiodes to capture light, organizing them into pixels for image formation. Key performance factors include quantum efficiency, dark current, signal-to-noise ratio, and dynamic range. Understanding these helps in choosing the right sensor for specific applications.

Image Sensor Types

Charge-Coupled Device (CCD) and Complementary Metal-Oxide-Semiconductor (CMOS) Sensors

• Charge-Coupled Device (CCD) uses an array of light-sensitive capacitors to capture and store charge generated by incoming photons
• CCD transfers the stored charge across the chip and reads it at one corner of the array
• Complementary Metal-Oxide-Semiconductor (CMOS) uses an array of photodiodes and transistors to capture and process light
• CMOS incorporates amplifiers, noise-correction, and digitization circuits next to each photodiode, enabling on-chip image processing
• CCD sensors generally have lower noise and higher sensitivity compared to CMOS sensors
• CMOS sensors consume less power, have lower manufacturing costs, and enable faster readout speeds than CCD sensors

Photodiodes and Pixels

• Photodiode is a semiconductor device that converts light into an electrical current
• Photodiodes are the fundamental light-sensing elements in both CCD and CMOS image sensors
• When a photon strikes a photodiode, it generates an electron-hole pair, which contributes to the electrical signal
• Pixel (picture element) refers to the smallest addressable element in an image sensor
• Each pixel typically contains a photodiode and associated circuitry for readout and processing
• The number of pixels in an image sensor determines its spatial resolution (megapixels)

Image Sensor Performance

Quantum Efficiency and Dark Current

• Quantum efficiency is the ratio of the number of collected electrons to the number of incident photons on the sensor
• Higher quantum efficiency indicates better light sensitivity and improved low-light performance
• Quantum efficiency varies with wavelength, and image sensors are designed to optimize sensitivity for specific spectral ranges
• Dark current refers to the small electrical current that flows through the image sensor even in the absence of light
• Dark current arises from thermal generation of electron-hole pairs and is a source of noise in the image
• Cooling the image sensor can reduce dark current and improve the signal-to-noise ratio

Signal-to-Noise Ratio (SNR) and Dynamic Range

• Signal-to-Noise Ratio (SNR) compares the level of the desired signal to the level of background noise in an image
• Higher SNR indicates better image quality, with clearer details and less visible noise
• SNR can be improved by increasing the light exposure, using a lower ISO setting, or applying noise reduction techniques
• Dynamic range is the ratio between the maximum and minimum measurable light intensities of an image sensor
• Wide dynamic range allows an image sensor to capture details in both bright and dark areas of a scene without saturation or loss of information
• High-end image sensors often employ techniques like multiple exposures or logarithmic response to extend the dynamic range

Shutter Mechanisms

Rolling Shutter and Global Shutter

• Rolling shutter is a capture method where the image sensor scans the scene row by row, with a slight time delay between each row
• Rolling shutter can cause image distortion for fast-moving objects or when the camera is in motion (skew, wobble, or partial exposure)
• Global shutter captures the entire image frame at the same time, eliminating the distortion issues associated with rolling shutter
• Global shutter is preferred for applications involving fast motion or precise timing, such as machine vision or scientific imaging
• CMOS sensors can be designed with either rolling or global shutter, while CCD sensors typically use global shutter

Color Capture

Color Filter Arrays and Bayer Pattern

• Color filter array (CFA) is a mosaic of tiny color filters placed over the image sensor to capture color information
• Each pixel in the image sensor is covered by a single color filter, typically red, green, or blue
• The Bayer pattern is the most common CFA arrangement, consisting of a repeating 2x2 grid of red, green, and blue filters
• In the Bayer pattern, there are twice as many green filters as red or blue, mimicking the human eye's higher sensitivity to green light
• Demosaicing algorithms interpolate the missing color values at each pixel location based on the surrounding pixels to reconstruct a full-color image
• Alternative CFA patterns (X-Trans, RGBW) and multi-layer sensors have been developed to improve color accuracy and low-light performance