Guangzhou, 5 May -- Experts and scholars from Yangcheng "transformed" into a marine expert in science and technology appeared at the popular science forum to convey the voice of science and technology
China News Network reporter Wang Jian
Yangcheng is a sea of science and technology recently.
The Greater Bay Area Science Forum was held in Guangzhou from 20 to 22, attracting many experts and scholars in the field of science and technology. Guangzhou Science and Technology Week also opened from the 20th to the 31st, with more than 400 science and technology activities to create a popular science feast for the public.
On the 20th, the 2023 Guangzhou Science and Technology Activity Week and Gezhidao Bay Area, hosted by Guangzhou Science and Technology Bureau and undertaken by Guangzhou Internet of Things Research Institute, took the lead in opening the curtain of popular science feast for Yangcheng residents and tourists.
Photo courtesy of Guangzhou Internet of Things Research Institute at the event
With the theme of "Smart Choice", from face recognition to deep submersible design, from nano-mercury measurement to stingless fish improvement, from animal behavior evolution to carnivore community research, the event invited 6 experts, scholars and researchers in the field of science and technology to share their self-reliant and self-reliant scientific research stories, attracting nearly <>,<> people to watch.
"If you can generate an expressionless picture of the front from any face image, you can solve these problems well." Zhu Xiangyu, an associate researcher at the Institute of Automation, Chinese Academy of Sciences, said that he and his team spent nearly 10 years to develop a full-pose three-dimensional face reconstruction and correction algorithm, on the one hand, they generated realistic three-dimensional faces through synthetic analysis, and on the other hand, they trained neural networks to make them automatically learn, improving the speed of face generation.
Zhu Xiangyu said that by reconstructing the three-dimensional human face, the face photo is normalized in the three-dimensional space and re-imaged to generate a realistic human face. In addition, Zhu Xiangyu also mentioned that the potential of the human face is not only here, but also the physiological and mental health conditions can be measured through the face in the future, and artificial intelligence will become more humane.
Just as challenging as facial recognition technology is deep diving. In November 2020, the "Striver" manned deep submersible successfully dived 11,10909 meters to the bottom of the sea, setting a new record for manned deep diving in China, which is also the first time in the world to bring three people to the deepest part of the ocean at the same time.
The manned module is one of the key components of the deep submersible and an important guarantee for the life safety of submariners. In recent years, the international 64,2-meter deep submersible uses titanium 3 alloy, but only <> people can dive at a time. So, what kind of material can ensure that the manned cabin does not crack and is not flattened, but also achieve <> people and multiple dives? Domestic new titanium alloy materials are the key to solving this problem.
After 12 years of continuous optimization and thousands of tests, the new titanium alloy independently developed by the titanium alloy team of the Institute of Metal Research, Chinese Academy of Sciences, where researcher Ma Yingjie is located, finally came out, and successfully cast the "Striver" manned ball cabin. At present, the "Striver" deep submersible has achieved dozens of tens of thousands of meters of diving, and China has become the country with the largest number of <>,<> meters and people.
"The development of the manned cabin has relied on the strength of many scientific research units across the country, from material preparation, forging, welding to forming, driving the development of the entire domestic titanium industry, and it is really 'titanium' strong!" Ma Yingjie said.
Tang Yezhong, a researcher at the Chengdu Institute of Biology, Chinese Academy of Sciences, revealed the choice of the brain behind the magical animal behavior.
The instinct of animals is to survive and reproduce, but in fact, animals often do some abnormal behaviors beyond instinct, such as "moths to the fire" and "elephants drunk"... Tang Yezhong has long thought about the adaptive significance of these behaviors.
Tang Yezhong introduced that the occurrence of behavior is directly controlled by the nervous system, and the internal homogeneity of the nervous system leads to competition between each other due to the homogeneity of structure and function. For example, the network of neurons used to store information is composed of neurons through chemical synapses, which are strengthened when stimulated and attenuated when not stimulated. If not stimulated for a long time, the network of neurons will break down and cannot store information to form memories.
Tang Yezhong said that the formation of nerve synapses is slow, but animals can form instantaneous memory, which indicates that there is a large number of pre-reserved redundant networks in the brain. These redundant networks also require continuous, appropriate stimuli that appear purposeless, random, or accidental.
Based on this, Tang Yezhong originally proposed the theory of "indirect selection" of neural networks, which also has enlightenment for early childhood education. He believes that children's brains should be stimulated in all directions, rather than starting a single arithmetic and word practice too early.
In addition, Yao Meng, associate researcher of the Ecological Research Center of Peking University, explained the eating rules of plateau carnivores, Chen Peipei, researcher of the National Center for Nanoscience, popularized mercury nanosensors, and Gao Zexia, professor of the College of Fisheries of Huazhong Agricultural University, shared the mystery of making fish "thornless" and "thornless" by regulating genes. (End)