For even the most sharp-eyed ornithologists (鸟类学家), a small bird can look much like another. But luckily, researchers have built an artificial intelligence (AD) model. “Our study provides the means of overcoming one of the greatest limitations on the study of wild birds—reliably recognizing individuals.” said Andre Ferreira, lead author of the study.

Ferreira began exploring AI’s potential while in South Africa, where he studied the sociable weaver, a bird which works with others to build the world’s largest nest. Ferreira wanted to understand each individual’s contribution to building the nest, but it was hard through direct observation because they were often hiding in trees or building parts of the nest out of sight. This experience contributed to his building the AI model, which could recognize individual birds simply from a photograph of their backs while they were busy with nest-building.

The model must be trained with thousands of labeled images in order to accurately identify individuals, but acquiring labeled photographs of birds is very difficult. Ferreira and other researchers finally overcamethis challenge. They built feeders with camera traps and sensors. Most birds in the study populations already carried a tag (标签) similar to microchips which are put in cats and dogs. Antennae (天线) on the bird feeders read the identity of the bird from these tags and touched off the cameras.

According to Ferreira, the use of the Al model could save on expensive fieldwork and spare researchers from procedures such as catching and marking animals. “It removes the need for the human to be a data collector, so researchers can spend more time thinking about the questions instead of collecting the data,” said Ferreira.

The AI model is currently only able to re-identify individuals that have been shown to it before. The researchers are now trying to build more powerful AI models that can identify larger study groups of birds, and distinguish individuals even if the AI models have never seen them before. This would enable new individuals without tags to be recorded by the cameras and recognized by the computers.

In order to help discover spoilage(变质）and reduce food waste for supermarkets and consumers,researchers have developed new low-cost,smart phone-linked, eco-friendly spoilage sensors for meat and fish packaging.

One in three UK consumers throw away food just because it reaches the use-by date(保存期），but 60%(4.2 million tonnes)of the £12.5 billion-worth of food we throw away each year is safe to eat.

The researchers,whose findings were published in ACS Sensors,say the sensors could also eventually replace the use-by date-a widely used indicator of being fresh and eatable.

The sensors cost two US cents each to make.Known as"paper-based electrical gas sensors(PEGS)",they detect spoilage gases like ammonia(a poisonous gas with a strong unpleasant smell)in meat and fish products.The information provided by the electronic nose is received by a smart phone,and then you can know whether the food is fresh and safe to eat.

The Imperial College London researchers who developed PEGS made the sensors by printing carbon electrodes(电极）onto a special type of paper.The materials are eco-friendly and harmless,so they don't damage the environment and are safe to use in food packaging.The sensors,combined with a tiny electronic system,then inform nearby mobile devices,which identify and understand the data about spoilage gases.

Lead author Dr Firat Guder,of Imperial's Department of Bioengineering,said, ＂Although they're designed to keep us safe,use-by dates can lead to eatable food being thrown away.They don't always reflect its actual freshness.In fact,people often get sick from foodborne diseases due to poor storage,even when an item is within its use-by date.

＂These sensors are cheap enough so we hope to see supermarkets using them within three years.Our goal is to use PEGS in food packaging to reduce unnecessary food waste."

The authors hope that PEGS could have applications beyond food processing,like sensing chemicals in agriculture,air quality,and detecting disease markers in breath like those involved in kidney disease.

Some Chinese students born in the 1990’s created a microsatellite that took a stunning picture of the moon and the Earth in space, an image praised as one of the best of the two celestial bodies and one that Science magazine also used on Feb.15, 2019.

The photograph was done by the camera on Longjiang 2, a microsatellite developed by students at the Harbin Institute of Technology in Northeast China's Heilongjiang province.

Before its publication in the leading scientific journal, the picture had been widely circulated on mainstream media worldwide, with some foreign media outlets even praising it as one of the best photos of the moon and the earth so far.

Wei Mingchuan, born in 1991, was the chief designer responsible for the sub-system of Longjiang 2, a microsatellite launched last year together with Queqiao, the relay satellite for the Chang'e 4 lunar probe. Wei successfully developed LilacSat 1 -- the country's first microsatellite independently designed, developed and controlled by students -- and has been dubbed China's "youngest chief designer", despite being the oldest in his team.

Tai Mier, one of Wei's team members who was born in 1996, designed the camera on Longjiang 2 that captured the amazing picture of the moon and the Earth. Tai was only 20 years old when work on the project began.

The youngest member in the team is 20-year-old Huang Jiahe, who handled software design for the Longjiang 2 ground station. Beginning in primary school, Huang would buy all kinds of electronic components and parts. At first, he just simply disassembled and assembled the parts, but now he can design and create things. This young man, considered a genius by some, said that he is just curious and passionate about aerospace technology.

In their dormitory, aerospace-related items are everywhere. A postcard on the wall is printed with the words "we were born for our dreams", a reminder that the future has just begun for this youngest team of the Chang'e 4 mission.