What is Dynamic Range? Dynamic Range is one of the key parameters of an image sensor. It defines the range of light intensity distribution that the image sensor can capture from shadow areas to highlight areas, and determines the details, layers and features of the captured images. Traditional Film Imaging Based on photochemical theory, when a camera takes a photo, light passes through the lens and reaches the light-sensitive silver halide crystals on the film, causing changes in the optical density of the film. A higher exposure level results in lower optical density, presenting a non-linear relationship. When light irradiates silver halide, it is converted into black silver, which is fixed on the film base through the developing process to form black-and-white film. Color film is coated with three layers of silver halide to reproduce the three primary colors. Digital Image Imaging An image sensor (CCD, CMOS) converts received optical signals into tens of thousands of discrete pixel points proportional to exposure via photosensitive units on the sensor, which are then transformed into analog voltage signals. After A/D conversion processing, the signals become digital data. Finally, microprocessors perform non-linear calculations to convert the data into standard image storage formats such as BMP, JPEG and TIFF, which are saved onto physical storage media.

Definition of Dynamic Range Dynamic range is defined as the ratio of the maximum detectable light intensity to the minimum detectable light intensity. It can also be defined as the ratio of the full-well capacity at saturation to the pixel noise electron count under no-light conditions. It is expressed in decibels (dB), with the formula shown below:

In natural scenes, the dynamic range from a moonless night to direct sunlight is approximately 180 dB. However, due to the limitation of full-well capacity, the dynamic range of ordinary image sensors is typically around 70 dB. The dynamic range within a single real-world scene can vary drastically; this is referred to as High Dynamic Range (HDR). In contrast, the dynamic range of standard images is defined as Low Dynamic Range (LDR). The imaging process of a camera is essentially a **nonlinear mapping** from the high dynamic range of the real world to the low dynamic range of a photograph. Factors Affecting Dynamic Range Full‑Well Capacity Conventional image sensors have a limited full‑well capacity in their photoreceptors, restricting their ability to store photogenerated charges. When light intensity exceeds a certain threshold, the stored charges reach saturation and can no longer accumulate further photons. Once pixels become fully saturated, charge overflow occurs, resulting in irreversible image information loss. Taking red color as an example: highlight saturation overflow forces adjacent pixel values near saturated red pixels to clip at 255, even though their actual brightness levels are far lower than this value. In other words, fine details in bright areas are permanently lost. If the exposure time is reduced to prevent highlight clipping, pixels capturing dark regions fail to collect sufficient photons within the shortened exposure, causing their values to drop to 0 and leading to missing details in shadow areas. Image Bit Depth Due to hardware and software limitations of common output and display devices, most digital image processing operates at 8‑bit depth, which only supports 256 grayscale levels. This creates a large gap between the brightness gradients displayed digitally and those existing in real physical scenes. Therefore, no matter how wide the original brightness contrast captured by an image sensor is, once converted into an 8‑bit digital image, it is confined to only 256 grayscale steps. Sensors with different native dynamic ranges map different real‑world brightness spans onto these 256 levels. A higher dynamic range enables the image to reproduce natural brightness variations far more accurately. Methods to Improve Dynamic Range Full‑Well Capacity Optimization Increasing the pixel full‑well capacity or reducing readout noise are direct approaches to boost sensor dynamic range. Nevertheless, modern pixel sizes continue shrinking, and a given pixel architecture features a fixed fill factor. For this reason, expanding dynamic range simply by raising full‑well capacity has become extremely challenging. Bit Depth Adjustment Raising the image bit depth is a widely adopted solution. More grayscale levels are rendered naturally, delivering richer tonal layers and an extended effective dynamic range. Logarithmic Response Technology This method modifies the pixel signal output curve to delay charge saturation time. It effectively extends the equivalent charge storage capability of pixels and broadens the overall dynamic range of the sensor. Multi‑Exposure Fusion Technology Without modifying the design of pixel photodiodes, traditional low‑dynamic‑range image sensors can synthesize multiple images captured at different exposure settings to reconstruct a final high‑dynamic‑range result.