When shooting against the light, the sky is overexposed and lacks detail, while the foreground subjects are reduced to black “silhouettes” (see Figure 1); when shooting streetlights at night, the light sources are glaringly overexposed, while pedestrians and road signs on the roadside fade into the shadows and are hard to distinguish (see Figure 2); under mixed indoor and outdoor lighting, the strong light by the window is overexposed, while the corners of the room are blurred and unclear (see Figure 3) — these common “regrets” in photography are essentially caused by insufficient camera dynamic range. Dynamic range is the “boundary of a camera's ability to capture light and dark details.” Next, this article will delve into the principles behind dynamic range.

Figure 1	Figure 2	Figure 3

I. Concept of Dynamic Range
Dynamic Range (DR) in photography refers to the luminance range between the brightest and darkest areas of a real scene where “distinguishable details” can be simultaneously recorded by an imaging device.

Lower DR	Higher DR

Its commonly used units are as follows:
1. Decibels (dB): The preferred unit for professional equipment specifications. The formula is: $DR_{dB}=20log_{10}(\frac{L_{max}}{L_{min}})$, where $L_{max}$ is the maximum distinguishable luminance and $L_{min}$ is the minimum distinguishable luminance.
2. Exposure Value (EV) or f-Stops: Commonly used in photographic practice to vividly describe the magnitude of dynamic range. 1 EV represents a doubling or halving of luminance.

Note:
1. Dynamic Range ≠ Contrast: Dynamic range is the “luminance span” (i.e., the range from dark to bright), while contrast is the “intensity of light-dark difference” (i.e., the ratio between bright and dark areas). High dynamic range is the foundation for “preserving both bright and dark details while maintaining contrast,” rather than being directly equivalent to high contrast.
2. Higher Value ≠ Better Image Quality: It must be combined with “detail retention capability.” If the dynamic range is high but the image is accompanied by severe noise or color distortion, the actual image quality will be poor.
3. “Distinguishable Details” is the Premise: If bright areas are overexposed (loss of detail) or dark areas are underexposed (no texture), no matter how large the calculated “luminance ratio” is, it does not constitute effective dynamic range.

II. Influencing Factors:
(1) Hardware: The “Physical Upper Limit” of Dynamic Range
Sensor:
Affected by two factors: Full Well Capacity and Read Noise.
(Image source: https://commons.wikimedia.org/wiki/File:Pier2.3_OV7120_CMOS_image_sensor.jpg)

• Full Well Capacity: The maximum number of photo-generated electrons that a single pixel can hold, which determines the upper limit of luminance. Full well capacity is like a water bucket; the larger the bucket, the more water (light signal) it can hold, making it less likely to “overflow” (overexpose), thereby preserving more highlight details.
  

• Read Noise: The background noise introduced by the sensor during the process of reading the signal from each pixel, measured in electrons. This noise determines the lower limit of the signal that the sensor can detect. The lower the read noise, the more sensitive the sensor is to dark or low-light signals, resulting in richer details in dark areas.
  

Lens: Lens glare directly affects dynamic range.

Glare is mainly caused by unintended reflection, scattering, or diffraction of light within the imaging system, which interferes with the effective dynamic range. To address this, high-quality lenses typically use multi-layer coatings and low-reflection lens elements to minimize glare interference and ensure that relevant details are not obscured.

(2) Algorithms: The “Efficiency Amplifier” of Dynamic Range
Hardware determines the upper limit, while algorithms determine “whether the hardware's potential can be fully utilized.”

1. HDR (High Dynamic Range) Algorithms: Solves the problem that “a single frame cannot cover a wide dynamic range.”
① Multi-frame HDR: Captures photos with different exposures (underexposed to capture bright areas, overexposed to capture dark areas, and normally exposed to capture mid-tones), and fuses them through algorithms into a single high dynamic range image containing details in both dark and bright areas;
② Single-frame HDR: Achieves exposure differences in different regions within a single frame through “sub-pixel partitioned exposure” of the sensor, adapting to fast-paced scenarios such as snap shots;

2. Tone Mapping: Solves the problem that “high dynamic range data cannot be adapted to display devices.” The luminance span of high DR images far exceeds the display range of screens or prints. Tone mapping compresses the luminance range under the premise of “preserving details,” ensuring that bright areas are not glaring and dark areas are not pitch black.

HDR Image	Locally Tone-Mapped Image

(Image source: https://commons.wikimedia.org/w/index.php?curid=77755416)