HDR (High Dynamic Range)

High dynamic range (HDR) imaging is a feature for capturing images that have a wide contrast range between the lightest and darkest areas. It helps make such images more natural-looking by stretching the dynamic range. Toshiba has developed HDR function for image sensors that can correct high-contrast scenes to produce natural-looking images for both still and video capture. Toshiba's HDR function delivers a smooth and crisp video by reducing blur and false colors that are inherently produced as a result of HDR composition.

There are single- and multi-frame HDR methods.
Toshiba's image sensors use a single-frame HDR implementation.

• Single-frame HDR … Captures lines of an image with different exposure times simultaneously and composes them into a single HDR image.

Toshiba's HDR: example of single-frame HDR

(Captured with different exposure times every two lines alternately)

• Multi-frame HDR … Composes multiple frames captured with different exposure times into a single HDR image.

To compose a single HDR image, multi-frame HDR must capture multiple frames that are short- and long-exposure versions of a scene while single-frame HDR requires only one frame.
Thus single-frame HDR allows video capturing at faster frame rates than multi-frame HDR. Toshiba offers image sensors that support a fast frame rate of up to 60 fps at full-HD resolution.

Long exposures can blur moving objects. To address this problem, Toshiba's HDR composes lines capturing with short and long exposure times into a single frame.
This approach helps reduce motion blur that is typically produced as a result of HDR composition and provide a smooth video.

Capturing moving objects with multi-frame HDR causes motion blur between consecutive frames since their images differ in a certain area of the frames. Composing these frames together into a single HDR image can lead to false colors. Since single-frame HDR combines short- and long-exposure lines captured into a single HDR image, it causes less motion blur.
Consequently, single-frame HDR delivers images with crisper colors and less false colors.

Since single-frame HDR uses different exposure times for different lines of a frame (ref. The figure of Single-frame method), it generally has a lower resolution in the vertical direction. To improve vertical resolution, Toshiba's single-frame HDR incorporates various innovations according to target applications and resolution requirements.
Toshiba's image sensors designed for smartphone and tablet applications provide twice the number of pixels that is actually necessary in the vertical direction while those for surveillance and automotive cameras applications implement a unique capturing method.

The pixel data of composed HDR images are more than 10 bits data.
To perform other types of image processing, HDR pixel data must be compressed to use fewer bits.
Generally, there are two compression methods to do this: global tone mapping that applies the same tone curve to the entire images and local tone mapping that only applies tone curves locally.
Toshiba's HDR function is implemented a unique compression method that was originally developed by Toshiba based on a local tone mapping technique. It helps reduce false colors and contrast degradation and maintain natural-looking tones.