The heart of the camera: the image sensor (CMOS and CCD)
The image sensor is the cornerstone of all camera modules. It converts light into electrical signals and there are two main types on the market today:
- CMOS sensors: 90% of smartphones are equipped with these sensors, which are very energy efficient and support advanced features such as HDR video. Their pixel size (e.g. 1.12µm to 2.4µm) has a direct impact on low-light performance.
- CCD Sensors: Used in medical imaging and telescopes, they offer a higher dynamic range but consume more power.
Practical example: Sony's IMX800 series sensors combine a 1/1.49-inch sensor size with Quad Bayer technology to take 50-megapixel photos.
Lens: More than just glass
A camera's lens assembly is made up of a bunch of curved glass or plastic elements that focus light onto the sensor. Key components include：
- Aperture (f-stop value): A lower aperture value (e.g. f/1.8) lets in more light, which is critical for nighttime photography.
- Field of View (FoV): Wide angle lenses (120° FoV) vs. telephoto lenses (30° FoV).
- Material: High-end lenses use glass to improve sharpness, while budget modules opt for polycarbonate.
Autofocus system
Modern camera modules rely on advanced autofocus mechanisms, commonly available:
- Voice Coil Motor (VCM): moves the lens electromagnetically for fast focusing (common in smartphones).
- PDAF (Phase Detection Autofocus): Calculates the focus distance using dual pixels on the sensor.
- Laser AF: emits infrared pulses to map the subject in low light (e.g. Samsung Galaxy S24 Ultra).
- Data Insight: High-end VCMs can adjust focus in 0.3 seconds, faster than the blink of a human eye!
Image Signal Processor (ISP)
The ISP is the equivalent of the brain of image quality processing. It effectively performs operations such as noise reduction and color correction:
- Noise Reduction: Use multi-frame stacking in night mode.
- Color Correction: Adjust white balance for natural skin tones.
- Artificial Intelligence Enhancement: Google's Tensor G3 chip uses machine learning to sharpen faces and skies separately.
Isps play an important role in the camera imaging process, which can be detailed in the previous article on the camera module imaging process.