Shulou(Shulou.com)11/24 Report--

Two bioengineers and a former postdoctoral scholar from the University of California, Los Angeles (UCLA) have jointly developed a new bionic 3D camera system that can mimic the multi-angle vision of flies and the natural sonar induction of bats to achieve multi-dimensional imaging of depth-scannable blind spots.

With the support of computational image processing technology, the camera can recognize the size and shape of objects hidden in corners or behind other obscured objects. The sensing capability of this technology far exceeds that of today's cutting-edge technology and is expected to be used in the field of self-driving or medical imaging, and the research has been published in the journal Nature Communications.

Screenshot of ▲ paper

In the dark, bats can visualize the vivid images around them by using echolocation (sonar). They constantly emit high-frequency calls, which are reflected by their surroundings and are picked up by their ears. Through small differences in the time required to receive the echo and the intensity of the sound, bats can determine in real time the location of objects in their path and their distance from potential prey.

Many insects have geometric compound eyes, each of which consists of hundreds to tens of thousands of independent visual units, which allows them to see the same object from multiple perspectives. For example, flies' spherical compound eyes, although with a fixed focal length, make it difficult for them to see distant objects-such as a high fly swatter-but also give them a nearly 360-degree field of vision.

Compound eyes of ▲ flies | Wikimedia Commons, Thomas Sn / CC BY 2.0 (https://creativecommons.org/ licenses / by / 2.0)

Inspired by these two animals, a team led by the University of California, Los Angeles (UCLA) began to design a high-performance 3D camera system with advanced features. Gao Liang, associate professor of bioengineering at the UCLA Samueri School of Engineering (Samueli School of Engineering), said: "although the idea has been tried before, the main obstacle has always been the identification of long distances and occlusion. In order to solve this problem, we have developed a new computational imaging framework, which for the first time uses simple optical devices and a small sensor array to obtain panoramic images with both width and depth. "

This framework, known as compact light field imaging, or CLIP (Compact Light-field Photography), allows the camera system to observe objects at greater depths or around them. In the experiment, the researchers proved that their system could "see" hidden objects that traditional 3D cameras could not.

Principle of ▲ Compact Light Field Imaging (CLIP) | reference [1]

The researchers also used a "light detection and ranging" (Light Detection And Ranging, LiDAR) technology that uses lasers to scan the surrounding environment to create a three-dimensional map of the area. Without CLIP, traditional lidar would take high-resolution snapshots of the scene, but ignore hidden objects like human eyes.

The researchers used seven LiDAR camera arrays with CLIP to take a low-resolution image of a scene, processed the results of each camera, and then reconstructed and synthesized the scene in high-resolution 3D imaging. The results show that the camera system can image complex 3D scenes with multiple objects with different distances.

Lidar snapshot of ▲ depth expansion | reference [1]

"if you cover one eye and look at your laptop, a coffee cup is only slightly hidden behind the computer, you may not see it because the laptop is blocking the view." "but if you look with two eyes at the same time, you'll get a better view of the object," explained Gao Liang, who is also a member of the California Nanosystems. it's kind of like what's happening here, but now imagine looking at the cup with the compound eyes of insects. Now, multi-perspective observation of objects is possible. "

According to the highlight, CLIP can help camera arrays observe hidden objects in a similar way. Combined with LiDAR, the system can achieve echolocation in bats, so people can detect hidden objects by the time it takes for light to be reflected to the camera.

reference

[1] Feng X, Ma Y, Gao L. Compact light field photography towards versatile three-dimensional vision [J]. Nature Communications, 2022, 13 (1): 1-10.

[2] https://phys.org/news/2022-08-bug-eyes-sonar-bioengineers-animal.html

Source: Intelligent Optics Laboratory, Liang Gao / UCLA

Research team

Communications author Gao Liang: associate Professor, Henry Samueli School of Engineering, University of California, Los Angeles, majoring in biomedical optics, ultrafast optical imaging, computational optical imaging, etc.

Paper information

Publish the journal Nature Communications Nature Communications

Release date: June 9, 2022

Paper title Compact light field photography towards versatile three-dimensional vision

(DOI: https://doi.org/10.1038/s41467-022-31087-9)

Optical imaging in the field of article

This article comes from the official account of Wechat: I am a scientist iScientist (ID:IamaScientist), compiled by Matrix Star Editor: Jin Xiaoming typesetting: Yin Ningliu

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.