Posts Tagged: Google
Google Weed View? Professor trains computer to spot invasive weed
Algorithm for AI enables low-cost tracking of invasive plant
To manage johnsongrass, a noxious weed that crowds out cotton and sickens horses, farmers have tried herbicides, burning and hand-pulling. Now, researchers at University of California, Davis, have developed a more high-tech weapon against the invasive weed: artificial intelligence and machine learning.
Using photos from Google's Street View database, UC Davis researchers have tracked down over 2,000 cases of johnsongrass in the Western United States for a fraction of the cost and time that it would take to do drive-by or other in-person surveys. They call their tool Google Weed View.
The advancement could help land managers easily and quickly survey for other problem plants.
“Once the model is trained, you can just go and run it on millions of images from Google Street View,” said Mohsen Mesgaran, an assistant professor in the Department of Plant Sciences at UC Davis. “We have huge flexibility, and its capability can be scaled up very quickly.”
The technique can easily be extended to other plant species. All that is needed is to label the new item in Street View photos and train the algorithm to identify that object in the images.
By providing location information, Google Weed View also offers an opportunity to examine how climate affects the growth and spread of weeds and invasive plants at very large scales.
“I think it can be both useful for management and for people with interests in more basic questions in ecology,” Mesgaran said.
A colleague's query
Mesgaran began looking at using Google's photo database of roadways, streets and highways after Kassim Al-Khatib, a professor of Cooperative Extension in the same department, asked if he could survey Western states for johnsongrass.
Al-Khatib studies where johnsongrass grows, ways to manage it and how this perennial has evolved to be so prevalent and resilient. He's also working with scientists at the University of Georgia to decode the genome of johnsongrass, which is one of the top 10 most invasive weeds worldwide.
Johnsongrass can crowd out native plants, harbor pathogens and affect agriculture. It grows up to 7 feet tall with flowers that are green, violet, dark red or purplish brown depending on maturity, according to a UC Statewide Integrated Pest Management Program briefing page.
“Johnsongrass is a major weed not just in California but worldwide,” Al-Khatib said. “It's very difficult to control. It's a problem on vineyards. It's a problem for cultivated crops. It's a problem on orchards.”
Google Weed View allows for rapid, convenient scanning. It is continuously updated via everyday users with compatible cameras and images collected by Google. “Instead of a day of in-person driving, we can use AI to determine if johnsongrass is in a county or not,” Al-Khatib said.
Setting the parameters
To find the weeds, Mesgaran went to Google Street View, which hosts billions of panoramic photos. It didn't take long to find johnsongrass.
“The pictures are really good quality,” he said. “You can see plants and flowers.”
Street View's photos offer a 360-degree view, so in his request Mesgaran set parameters, based on street direction (bearing), to only see the side view. He also specified latitude and longitude, and other factors. To train the deep, or machine learning model, he chose Texas, where johnsongrass is prevalent.
A student sorted through over 20,000 images from that request to find pictures with johnsongrass and drew rectangular shapes around the weeds. They located 1,000 images.
The labeled photos were fed into a computer to train a deep learning algorithm capable of identifying johnsongrass in Google's images. The model was run again to capture potentially more images containing johnsongrass. These additional images were then labeled and used to further refine the model. With each iteration, the algorithm learned and became more accurate.
“This deep learning model was trained by these images,” Mesgaran said. “Once we had a semi-working model, we ran it against about 300,000 images.”
For Al-Khatib's request, researchers focused on 84,000 miles of main roads in California, Nevada, Oregon and Washington states. The team discovered 2,000 locations with johnsongrass.
Google Weed View cost less than $2,000 to purchase the images and teach the model. A traditional car survey to cover the same area would cost an estimated $40,000 in gas, hotel, food and other costs.
“In a matter of months, we came up with 2,000 records and I can do it for the whole U.S.,” Mesgaran said.
Next up? The entire United States.
This story was originally published on the UC Davis College of Agricultural and Environmental Sciences news site.
GOOGLE Food Lab tours Pinnacle Organically Grown Produce with Phil and Katherine Foster in Hollister, CA!
April 25, 2023
Phil and Katherine Foster host GOOGLE Food Lab for tour of Pinnacle Organically Grown Produce!
About 53 folks who were part of the GOOGLE Food Lab out of Mountain View, CA were treated to a half-day walking tour and lunch at the beautiful organic farm of Phil and Katherine just east of Hollister, CA on April 25, 2023. The GOOGLE Food Lab describes itself as "an invitation-only community of food leaders, change agents and visionaries, creating positive change across food systems." Pinnacle Organically Grown Produce was chosen to host part of the group's 2023 tour because of the farm's long-term leadership and pioneering progress on organic vegetable production systems that use a wide range of beneficial practices that contribute to the farm's diversity, soil health, employee welfare, and economic success. The multitude of the Fosters' practices including use of 'single-line' cover crops to conserve water, on-farm composting and routine field application of compost, hedgerows, and reduced disturbance tillage, were on display during the two-hour walking tours that Phil Foster along with Jeff Mitchell took the visitors on. The tour was coordinated by Eva Antczak of GOOGLE following an earlier visit to the farm by Michiel Bakker, GOOGLE's Vice President of Global Workplace Programs in 2022.
IMG 0294
IMG 0290
Highlights from Google’s 2021 Geo for Good Summit
by Annie Taylor
If you attended any of the Google Earth Engine workshops I've led with IGIS, then you know that I love talking about the tool and hearing what others are doing with it. Last month, I attended my fifth Geo for Good Summit, which is quite possibly my favorite annual event. I'll provide a recap of the event and some of my highlights, and then links for you to check out the summit for yourself or explore other resources.
Google's Geo for Good Summit is an annual conference where folks from nonprofits, government agencies, and scientists hear about the latest updates to Google's mapping tools and get to share their work with each other. Geo for Good, also abbreviated as G4G, has been held since 2012 and was an in-person Bay Area conference until 2020, when it was first held virtually. G4G continued in that virtual format in 2021, with some exciting twists on the usual virtual format.
G4G always kicks off with an inspiring community video that showcases different people and use cases, and this year was no exception.
Next, we got updates on the latest improvements to Google's geospatial tools, which is my personal favorite session. Here's a rough summary of my highlights.
What's new in Google Earth?
In case you haven't heard, Google Earth is now a gorgeous web application. Google Earth Pro (the desktop application) is still supported as they migrate all of its tools and functionality to the web version. There's also a mobile application for your phone! These updates relate to the web application:
- You can now export tile overlays from EE and import them into Earth so that they drape on top of the 3D planet – very cool visualization for your data.
- You can collaborate on Earth projects just like you would a Google Doc using the Cloud Project option.
- The date of the basemap imagery (when available) now pops up when you move your cursor. Also, the scale bar shows the scale bar relative to where your cursor is, which is a great addition for 3D viewing.
What's new in Earth Engine?
- Landsat and Sentinel images are now ingested and loaded into EE in under 12 hours from their capture – enabling near real time analysis in EE.
- The Dynamic World dataset – a global land cover probability map at 10m resolution available every five days from 2017 to 2021 – will soon be available in the EE Data Catalog.
- Additions to the API (new machine learning models)
- Additions to the Data Catalog and more Python API example scripts
Other Geospatial Tools by Google
I won't go into these here, so check out the recorded sessions for more information:
- Google My Maps
- Google Earth Studio
- Google Street View and Maps API
G4G was very creative in their approach to the virtual conference. For example, they organized poster sessions and virtual mingling in a custom space they created using Gather, where you could see and hear the people ‘nearest' to you in the ‘room.' They also hosted user-created Meet Ups, where members of the community can break off to talk about their specific application, tool, or part of the world. A final highlight of the conference were the Office Hours hosted by the engineers working on each of these tools, which is an incredible opportunity to talk through your specific idea or challenge.
If you missed it, you can find all of the recorded sessions on the conference website or on Google Earth Outreach's YouTube site, which are both linked below.
I hope to see you next fall at G4G 2022!
Annie
Conference Website (with recordings)
Geo for Good YouTube Channel Playlist
Earth and Earth Engine Blog
/h3>/h3>/h3>
Game Changers at Google
For the second year in a row, the Computer Science Pathway Team of the University of California...
4-H National Youth Science Day Grows in Santa Clara County
To celebrate the National Youth Science Day (NYSD) challenge in October, the 4-H Youth Development...