A method for restoring distorted images without expensive equipment has been developed, and is expected to be applied to high-quality images for autonomous driving and observation of the inside of living animals.

UNIST (President Yong-Hoon Lee) announced on the 4th that a research team led by Professor Jeong-Hoon Park of the Department of Biomedical Engineering recently developed a method to restore high-resolution images using information hidden in distorted images, and published it in the international academic journal 'Laser & Photonics Reviews' on the 1st.

'Adaptive optics technology' that corrects image distortion is already being used in the field of astronomy and space science. It corrects starlight distorted by the atmosphere to observe the universe clearly.

On the other hand, this technology requires expensive specialized equipment such as a wavefront meter or wavefront controller, making it difficult to use in everyday life to overcome image distortion.

Meanwhile, the method developed by Professor Park Jeong-hoon's team can restore distorted images without expensive specialized equipment.

First, we divided the distorted image into the ‘resolution-lowering component’ and the ‘component that only changes position.’ Then, we removed the effect of the position change using a computer. All positional components are moved and placed in place based on one image. In this state, only the components that reduce the resolution can be collected and the average value can be obtained. This is a concept of finding the average of random resolution-degrading cause factors and removing them, and only the information of the actual object is extracted to restore the high-resolution image.

Phenomena obscured by fog, smoke, or haze are captured using videos shot over time. “Videos are created by combining multiple images over time,” said the first author, Dr. Byungjae Hwang of UNIST’s Department of Biomedical Engineering. “Even though there are different distortions in each scene, the necessary information is hidden, so we can extract it to obtain a clear image.”

The developed method can also be applied to objects with much higher light scattering, such as biological tissues. In this case, the degree of distortion can be used to arbitrarily divide the 'space' of an image to obtain necessary information. It is to collect image information that is fragmented into each space, calculate the average value, and process it as a function.

Professor Park Jeong-hoon said, “Image distortion caused by the atmosphere or biological tissue changes randomly over time and space, and this phenomenon is closely related to our daily lives.” He added, “The technology developed this time can be applied to stable autonomous driving even in bad weather, as well as remote surveillance and astronomy.”

He continued, “Furthermore, we have presented a method to observe the interior of living animals in high resolution and detail,” adding, “It will help enable non-invasive observation of life phenomena.”