This project comes from EECS 331: Introduction to Computational Photography.

 

HDR synthesis technology has been widely used in mobile phone photography software. I think mastering HDR synthesis is a necessary skill for image processing, so I made a HDR synthesis program by using Matlab.

 

It can generate photos with excessive light smoothness. Whether it is taking high-brightness objects like the sun or taking low-brightness night scenes, it can achieve good results.

 

At the same time, it can ensure that there will be no over-exposure and under-exposure parts in the photos. The final treatment effect was very successful. The improved technology is expected to be applied in the field of automobile automatic driving in the future.

​

The next four pictures are used for HDR synthesis. They are taken at different exposure times, with varying degrees of over-exposure and under-exposure.

​

​

​

​

​

​

​

When we synthesize HDR, we can get the light energy value of each pixel and use it to generate the HDR synthesis result with soft light distribution. 

​

​

​

​

​

​

Here is another example: 5 pictures were used for HDR Synthesis

​

​

​

​

​

​

​

​

​

​

​

​

​

​

The result can be seen below: