Shutterbug

The HS-MAC high-speed camera, inspired by insect eyes, captures 9,000+ FPS with minimal distortion, even in low-light conditions.

Nick Bild
27 days agoSensors
HS-MAC is a high-speed camera that works well in low-light conditions (📷: H. Kim et al.)

Superman isn’t the only thing faster than a speeding bullet. And when we need to understand one of these speeding objects, high-speed cameras are one of the best tools available. These cameras provide invaluable insights into events by recording at thousands or even millions of frames per second, enabling detailed analysis of split-second occurrences. This comes in very handy in the world of scientific research, where they are used to study phenomena like shock waves, fluid dynamics, and the mechanics of microscopic cellular processes. Industry finds many uses for the technology as well, such as diagnosing equipment malfunctions and identifying defects in assembly lines.

However, traditional high-speed cameras have some significant limitations. For starters, they are commonly very bulky and complex. This makes them difficult to integrate into compact systems or mobile applications. Furthermore, many high-speed cameras suffer from rolling shutter distortions, where fast-moving objects appear skewed due to the sequential nature of image capture. These issues become even more problematic when capturing images in low-light conditions. Small apertures and limited light sensitivity make it difficult for these cameras to gather enough photons for clear, high-quality imaging, particularly in dim environments.

A team of researchers at the Korea Advanced Institute of Science and Technology has drawn inspiration from insects that are active in twilight or nighttime conditions to develop a new type of imaging technology called the High-Speed and High-Sensitive Microlens Array Camera (HS-MAC). This camera can capture over 9,000 frames per second, but thanks to some clever design features, it can do so with minimal distortion, even in low-light conditions.

Insects, such as moths and dragonflies, have eyes that are composed of multiple optical units called ommatidia, which work in parallel to process visual information. This parallel processing allows insects to detect rapid motion, while a process called temporal summation enhances their sensitivity to low light by integrating visual signals over time. This insight was the key to the development of HS-MAC.

HS-MAC features offset microlens arrays combined with a rolling shutter CMOS image sensor. This design enables a process known as channel fragmentation, where images are divided into discrete sections across multiple channels. By capturing fragmented visual data at exceptionally high frame rates, the camera can record events occurring far too quickly for conventional cameras.

To address the challenges associated with low-light imaging, HS-MAC employs an artificial temporal summation process. This technique accumulates light over time from fragmented images, boosting sensitivity and enabling the camera to perform well in dim conditions. The captured fragments are then reconstructed into clear, cohesive frames using a process called compressive frame reconstruction. This process not only stitches the images together but also reduces motion blur by resolving overlapping temporal information.

The performance of HS-MAC was evaluated through a series of experiments focusing on its frame rate, distortion control, light sensitivity, and imaging capabilities under challenging conditions. In one test, HS-MAC and a conventional MAC were compared by capturing a free-falling ball. HS-MAC demonstrated significantly reduced rolling shutter distortion, with the ball appearing nearly spherical. The distortion, measured by the ratio of the ball’s major to minor axis, showed a 50% improvement in accuracy compared to the conventional MAC.

To assess its speed and accuracy, HS-MAC recorded a rotating disk at 9,120 frames per second. The disk, spinning at 1,950 revolutions per minute, was captured without rolling shutter distortions, and the rotational speed was measured with high precision. The camera’s ability to capture rapid events was further validated by successfully reconstructing sharp, sequential images of the disk.

Whether improving crash simulations, advancing neuroscience research, or enabling new capabilities in machine vision, HS-MAC offers a glimpse into the future of high-speed imaging. By combining the ingenuity of nature with cutting-edge engineering, this bio-inspired innovation is on track to set a new standard for what high-speed cameras can achieve.

Nick Bild
R&D, creativity, and building the next big thing you never knew you wanted are my specialties.
Latest articles
Sponsored articles
Related articles
Get our weekly newsletter when you join Hackster.
Latest articles
Read more
Related articles