Aug 15, 2017

Researchers use moss to overcome machine-learning limitations

(Beth J. Harpaz / AP)

Deep-learning algorithms can identify objects and faces better than humans in some cases but are limited in their ability to recognize amorphous forms, like grasses and trees, that can take different shapes and sizes and are continually changing. But a team, led by Takeshi Ise at the Kyoto University in Japan, has developed a new technique that will help machines overcome these limitations, per the MIT Technology Review. The method centers on teaching machines to recognize different types of moss, a plant that doesn't take a well-defined, distinctive shape.

Why it matters: The method could be used to better recognize trees and crops in aerial photos, which would be valuable for monitoring agriculture and for conservation and land management efforts.

How they did it: Ise and his team started by photographing three different types of moss, both individually and in places where they were grouped with other plants. They then worked to create an algorithm that could distinguish the different types of moss by meticulously labeling the different versions and feeding it into the deep-learning machine.

The results: The algorithm quickly learned to identify each type of moss when grouped among others in a single image. "The model correctly classified test images with accuracy more than 90%," the team said.

Limitations of study: The algorithm recognized some types of moss better than others. For example, the types of moss that are more amorphous with less defined forms of growth were harder to identify than those that had a more distinctive shape. But the team said they plan to continue improving the accuracy of their method by experimenting with how the photos are taken.

Go deeper

Coronavirus dashboard

Illustration: Sarah Grillo/Axios

  1. Global: Total confirmed cases as of 2:30 p.m. ET: 905,279 — Total deaths: 45,371 — Total recoveries: 190,710Map.
  2. U.S.: Total confirmed cases as of 2:30 p.m. ET: 199,092 — Total deaths: 4,361 — Total recoveries: 8,362Map.
  3. Federal government latest: Vice President Mike Pence said that White House modeling suggests "Italy may be the most comparable area to the United States" in terms of coronavirus impact.
  4. State updates: Florida and Pennsylvania are the latest to issue stay-at-home orders, totaling 34 states and Washington, D.C.
  5. 2020 update: West Virginia is latest state to delay its primary — Bernie Sanders urged Wisconsin to delay April 7 primary.
  6.  🎧 Podcast: The Defense Production Act
  7. What should I do? Answers about the virus from Axios expertsWhat to know about social distancingQ&A: Minimizing your coronavirus risk.
  8. Other resources: CDC on how to avoid the virus, what to do if you get it.

Subscribe to Mike Allen's Axios AM to follow our coronavirus coverage each morning from your inbox.

West Virginia is latest state to delay primary due to coronavirus

Photo: Barbara Davidson/Getty Images

West Virginia Gov. Jim Justice rescheduled the state's May 12 primary election to June 9 on Wednesday, citing fears surrounding the coronavirus pandemic, AP reports.

Why it matters: 23 other states and the District of Columbia haven't held primaries yet. The White House is recommending, for now, that Americans practice social distancing and gather in groups of no more than 10 people — while many states have issued stay-at-home orders.

Go deeperArrowUpdated 1 hour ago - Politics & Policy

Florida governor issues stay-at-home order

Photo: Joe Raedle/Getty Images

Florida Gov. Ron DeSantis ordered a statewide stay-at-home order on Wednesday after the number of coronavirus cases in the state rose to nearly 7,000.

Why it matters: DeSantis has been criticized for declining to order any statewide mandates to curb the spread of coronavirus as Florida — home to a significant elderly population — has increasingly become a hotspot. The order will go into effect Thursday at midnight and last for 30 days.