Imaging technology has come a long way since the beginning of photography in the mid-19th century. Now, many state-of-the-art cameras for demanding applications rely on significantly different mechanisms than consumer devices. One of these cameras uses what is called “single photon imaging”, which can produce significantly better results in dark conditions and fast dynamic scenes. But how does single-photon imaging differ from conventional imaging?
When you take a photo with a regular CMOS camera, like those in smartphones, the camera’s sensor is open to a large influx of photons for a preset exposure time. Each pixel of the sensor grid delivers an analog value which depends on the number of photons that reach that pixel during exposure. However, this type of imaging has little ability to process moving objects; the movement of the object should be much slower than the exposure time to avoid blurring. In contrast, single-photon cameras capture a rapid burst of consecutive images with very short individual exposure times. These images are binary – a grid of 1s and 0s that respectively indicate whether a photon has arrived at each pixel or not during exposure. To reconstruct an actual image from these binary frames (or bit planes), many of them must be processed into a single non-binary image. This can be done by assigning different brightness levels to all the pixels in the grid, depending on the number of bit planes having a “1” for each pixel.
Besides its higher speed, the fully digital nature of single-photon imaging allows for the design of intelligent image reconstruction algorithms that can compensate for technical limitations or difficult scenarios. At Tokyo University of Science, Japan, Professor Takayuki Hamamoto led a research team focused on improving the capabilities of single-photon imaging. In the latest study by Professor Hamamoto and his team, published in IEEE Access, they developed a very efficient algorithm to correct motion blur in imaged objects, as well as common blur of the entire image like that caused by camera shake.
Their approach addresses many limitations of existing blur removal techniques for single-photon imaging, which produce low-quality images when multiple objects in the scene move at different speeds and dynamically overlap. Instead of adjusting the entire image based on the estimated movement of a single object or on the basis of spatial regions where the object is considered to be in motion, the proposed method employs a more versatile strategy.
First, a motion estimation algorithm tracks the motion of individual pixels through statistical evaluations of how bit values change over time (on different bit planes). In this way, as experimentally demonstrated by the researchers, the movement of individual objects can be accurately estimated. “Our tests show that the proposed motion estimation technique produced results with errors of less than one pixel, even in dark conditions with few incident photons,” notes Professor Hamamoto.
The team then developed a blur removal algorithm that uses the results of the motion estimation step. This second algorithm groups pixels with a similar movement, thus identifying in each binary plane separate objects moving at different speeds. This makes it possible to blur each region of the image independently according to the movements of objects passing through it. Using simulations, the researchers showed that their strategy produced very sharp, high-quality images, even in dynamic, low-light scenes crowded with objects moving at disparate speeds.
Overall, the results of this study rightly show how much single-photon imaging can be improved if one sets out to develop efficient image processing techniques. “Methods for obtaining sharp images in limited photon situations would be useful in several fields, including medicine, security and science. We hope that our approach will lead to new technology for high quality imaging in high-quality environments. dark environments, like outer space, and slow recording that will far exceed the capabilities of today’s fastest cameras, “says Professor Hamamoto. He also says that even consumer cameras could benefit from this. timely advancements in single-photon imaging.
Source of the story:
Material provided by Tokyo University of Science. Note: Content can be changed for style and length.