A good photo of HDR Core technology Introduction

Alignment correction of a multiple exposure picture group

In the process of photographing the HDR multiple exposure picture Group, it is best to use a tripod to ensure that the multiple exposure picture group that is photographed is fully one by one corresponding to each pixel coordinate to obtain a clear HDR image. But the process of using a tripod is cumbersome, and in most cases we prefer to shoot multiple exposure groups using a hand-held camera. As shown in figure (i) three images, due to the inevitable slight shaking and other unstable factors, more or less will result in the image group of various images in the horizontal/vertical direction, rotation offset, deformation offset and so on the misaligned factors.

If you use a group of pictures directly to synthesize an HDR image, you will create blurry, even overlapping pictures, such as figure (e). Therefore, before you synthesize an HDR image, you need to align the image groups of different exposures so that their pixel coordinates are exactly one by one corresponding. Good photos of the international leading HDR calibration Technology program can not only correct the picture group at the level, offset vertically and in the direction of rotation can also be used to correct the displacement caused by the deformation between photos in the plane range, and to a great extent ensure the quality of the HDR image shooting and making by means of a handheld camera, as shown in (f).

Misaligned Photo Alignment Technology

Second, the multiple exposure picture group "Ghost Shadow" removal

Scene movement is one of the biggest obstacles to a high quality HDR image, and the photographer is in the process of photographing multiple images of the same scene with different exposures, even though we ensure that the camera is absolutely stationary in the course of the filming, but if there is a scene moving (as the red flag in the bottom image), The resulting banner will have different positions in each picture, so in the final HDR image, the moving object will appear in the photo multiple times, creating a "ghost", as shown in (g).

For this good photo team has developed a powerful removal of "phantom" technology to help HDR photography enthusiasts solve such problems encountered in the shooting, such as figure (h). This process is all automatic and does not require manual intervention, which means that we can spend more time doing more work related to image creation.

Original

Did not go to "ghost shadow" effect

To the "ghost" effect.

Three, 32-bit spectral graph (digital film) generation technology

After the alignment correction for the multiple exposure picture group, "Ghost Shadow" after removal, good photos of the 32-bit floating point number format (digital film) generation technology through the multiple exposure image group to restore the camera as a result of a low dynamic range of the loss of the signal, the original scene to reproduce the actual light intensity signal, and recorded as 32-bit floating-point format (digital negative) (. HDR,. exr).

Compared to traditional negatives, 32-bit floating-point format (digital film) records a higher dynamic range, more image details. 32-bit floating-point spectra (digital negatives) can produce final HDR images using our Digital film stamping technology (hue mapping), and can be retained for later use, as in traditional photographic negatives.

Four, tonal mapping technology

Although the spectral map accurately records all the information of the original scene, there is a higher dynamic range and more realistic image details than the direct output of a camera, but the spectral map cannot be visualized directly on an existing 8-bit display or print device that can only render 102~103 of an order of magnitude dynamic range. So no matter how accurately the spectral chart records the information of the original scene, if the information cannot be visualized, then these numbers will be a bunch of numbers that don't make much sense to the user.

In response to this challenge, tonal mapping technology (Tone Mapping) came into being. The so-called tonal mapping technique is how to convert a spectrum of a higher dynamic range number format to a lower dynamic range integer format acceptable to the usual display device, and make it appear more accurate, more realistic reproduction of the original scene, and the original scene appears to have a consistent degree of light and shade. This is also the most critical and important step in high dynamic range photography (HDR) technology to produce high quality images.

Because the tonal mapping technology makes the spectrum visualization, compared with the traditional photographic film stamping process, we can also call the hue mapping technology as the development technology.

In order to obtain high quality picture, produce the actual visual effect that the human eye sees, the tonal mapping technique of good photograph designs based on the characteristic of the human eye Vision system, and provides two kinds of local tone mapping algorithm, a global tonal mapping algorithm, which ensures the photographer to obtain high-quality HDR images. The tonal mapping technology of good photos has been recognized by the depth of photography enthusiasts in international scope.

Local tone mapping algorithm-rich color

Local tone mapping algorithm-natural restore

Global tonal mapping algorithm-high Contrast

High Dynamic range image Technology (HDR) allows professional and amateur photographers to use image processing techniques that have been used for centuries to enter a new digital arena, with high dynamic range image technology (HDR) allowing photographic enthusiasts and artists to create images that are the same as the real world.

High Dynamic range image Technology (HDR) is the final step in digital transcendence simulation, and the extension of digital photography to the simulation of the darkroom is far from reaching the realm. High dynamic range image technology represents a leap in technology, just as from black-and-white images to color images of the leap.

Note : More wonderful tutorials Please pay attention to the triple computer Tutorials Section