NASA is seeking help from the public to train AI

NASA is seeking help from the public to identify important images taken by Mars explorers. This project aims to help artificial intelligence systems identify different scientific elements in pictures.
Artificial intelligence plays an important role in NASA's processing of data collected by its spacecraft and exploration vehicles. Nasa has developed artificial intelligence algorithms to identify and classify elements with scientific value. But these algorithms must be trained by humans to work.
NASA said that Perseverance can send hundreds of pictures to Earth on any given day. Scientists and engineers examined these images while studying the different geological properties of the Martian surface.

These images can also help guide Perseverance and other rovers on Mars. But the time faced by teams on earth is very short.NASA said in a recent statement, "After these images have been transmitted millions of miles from Mars to Earth, team members have a few hours to formulate the next set of instructions."
Vivian Sun is a scientist at NASA's Jet Propulsion Laboratory (JPL) in California. She helps oversee the daily operations of Perseverance. She also advises the operators of a public project called AI4Mars."It is impossible for any scientist to carefully check all the downstream images every day in such a short period," Sun said in an online statement."If there is an algorithm that can say, ‘I think I see rock veins or nodules here’, then the scientific team can observe these areas in more detail, which will save us time."
The AI4Mars project aims to build a large, reliable data set from thousands of images. Researchers can use these images to quickly discover interesting geological features.
For other artificial intelligence researchers, it is much easier to train their algorithms with many pictures available on Earth. But NASA does not have such a huge collection of Mars images for research. Citizen scientists who wish to participate in the AI4Mars project can visit a special website to get started. There, users can find an online tool that can be used to draw lines around features of the image (such as sand and different kinds of rocks).
Once these features are marked, users can choose from descriptions that match the content of the picture.NASA said that the latest version of the AI4Mars system allows people to choose a more detailed description.
Annie Didier is a scientist at the Jet Propulsion Laboratory (JPL) and she is working on the Perseverance version of AI4Mars. She said that creating a powerful data set can serve several important purposes. She said: "With this algorithm, the lunar rover can automatically select the scientific target to drive towards."She added that it can also allow rovers on Mars to store a large number of images and then send back only the images that are of interest to scientists.NASA said that AI4Mars is a continuation of another project launched last year that uses images collected by the Curiosity probe. In this project, people can outline and mark features such as sand and rocks from nearly 500,000 pictures.
The result of this effort is an algorithm called soil attribute and object classification.NASA stated that the system can correctly identify the trained features 98% of the time.

Inquiry us

Our Latest Products

Nano Aluminium Oxide Powder Al2O3 Nanoparticles

Nano Aluminium Oxide Powder Al2O3 Nanoparticles has a stable crystal phase, high hardness, and good dimensional stability, which can be widely used in various plastic, rubber, and ceramic products. About Nano Aluminium Oxide Powder Al2O3 Nanoparti...…

TROX-300 Series Molecular Sieve

TROX-300 Series Molecular Sieve has extremly exceptional performance at low stress.Concerning TROX-300 Series Molecular Screen:Technical Criterion of TROX-300 Series Molecular Filter:< tableborder=" 1" cellpadding=" 0" cellspacing=" 0" design=" borde...…

HPEG 2400 Polycarboxylate Superplasticizer Monomer

HPEG is most commonly used for the production superplasticizer products of polycarboxylic Acid. HPEG2400 1. Scope of HPEG2400 1. HPEG is most commonly used for the manufacturing of superplasticizer polycarboxylic acid products. 1. Mix the oil direc...…

0086-0379-64280201 skype whatsapp