總網頁瀏覽量

【○隻字片羽○雪泥鴻爪○】



○○○○○○○○○○○○○○○○○○

既然有緣到此一訪,
何妨放鬆一下妳(你)的心緒,
歇一歇妳(你)的腳步,
讓我陪妳(你)喝一杯香醇的咖啡吧!

這裡是一個完全開放的交心空間,
躺在綠意漾然的草原上,望著晴空的藍天,
白雲和微風嬉鬧著,無拘無束的赤著腳,
可以輕輕鬆鬆的道出心中情。

天馬行空的釋放著胸懷,緊緊擁抱著彼此的情緒。
共同分享著彼此悲歡離合的酸甜苦辣。
互相激勵,互相撫慰,互相提攜,
一齊向前邁進。

也因為有妳(你)的來訪,我們認識了。
請讓我能擁有機會回拜於妳(你)空間的機會。
謝謝妳(你)!

●●●●●●●●●●●●●●●●●●



2019年4月27日 星期六

Transforming fishing through transparency and technology

https://globalfishingwatch.org/map-and-data/technology/

How it works

Transforming fishing through transparency and technology

Just a decade ago, building an accurate picture of the commercial fishing across the globe would have been impossible. Today, thanks to advances in satellite technology, cloud computing and machine learning, Global Fishing Watch is making it a reality. So how do we do it?

1. Harvesting the data

The process starts with vessel tracking data. While Global Fishing Watch uses several vessel tracking systems, we start with the automatic identification system (AIS), a GPS-like device that  large ships use to broadcast their position in order to avoid collisions. The International Maritime Organization and many national governments require larger boats including many commercial fishing vessels to use AIS. Each year, more than 300,000 unique AIS devices broadcast the location of a vessel along with other information showing its identity, course and speed. Ground stations and satellites pick up this information, meaning a ship’s movements can be followed even in the remotest parts of the ocean.
While only a small fraction of the world’s roughly 2.9 million fishing boats carry AIS, they are responsible for a disproportionate amount of the fish caught, especially far from shore. It’s estimated vessels with AIS account for over half the fishing effort more than 100 nautical miles from shore, and as much as 80% of the fishing in the high seas.

2. Processing the information

AIS provides vast amounts of publicly available data – it’s far too much for any human being to make sense of, and only part of it is from fishing boats.
Global Fishing Watch runs this data through two neural networks using computer algorithms to learn and look for patterns in large data sets. More than 60 million points of information per day from more than 300,000 vessels are fed through machine-learning classifiers to determine the type of ship (e.g., cargo, tug, sail, fishing), its size, what kind of fishing gear (e.g. longline, purse seine, trawl) it’s using, and where and when it’s fishing based on its movement patterns. To do this, our research partners and fishery experts have manually classified thousands of vessel tracks to “teach” our algorithms what fishing looks like.
By using cloud computing to spread the load over thousands of machines in parallel, we’re able to apply that learning to the entire dataset producing 37 billion points over five years.

3. Sharing the results

Global Fishing Watch makes this vessel tracking information available to all through our interactive online map and downloadable data. Anyone with an internet connection can trace the movements of about 60,000 commercial fishing boats, along with their name and flag state, in near real time: our data shows all activity from 1 January 2012 until 72 hours ago.
You don’t need to be an expert to use the platform, any more than you need to know about complex algorithms to use a search engine: it’s aimed at members of the public and journalists as much as researchers, campaigners and governments.
Users can create heat maps to see patterns of commercial fishing activity, view tracks of individual vessels, and overlay information like the locations of marine protected areas or different countries’ exclusive economic zones (EEZ).

 Catching out the identity cheats

Most large fishing vessels are assigned a unique Maritime Mobile Service Identity (MMSI) number, but in practice some vessels use a number that is not assigned to them  either a false number (like 123456789) or the number of another vessel. This means that, throughout the ocean, multiple vessels are simultaneously broadcasting the same MMSI number making them indistinguishable from one another without closer inspection. Vessels can also manipulate their GPS location by tampering with the system (“spoofing”).
Our machine-learning algorithms automatically separate signals coming from multiple vessels using the same MMSI, and also detect when the broadcast location is inconsistent with the location of the satellite that received the signal. We can’t always determine the true identity of the spoofing vessel, but our algorithms can still detect the vessel’s behaviour and put it on a map.

Making an impact

The technology powering the Global Fishing Watch map may be impressive, but the really exciting stuff happens when people use it:  
  • Governments can identify and take action against boats that aren’t authorised to fish in their waters, or that are fishing illegally in protected areas
  • Seafood suppliers and retailers can see where and how fish are caught and ensure they only source from boats that are operating legally and responsibly
  • Researchers can study the impacts of fishing on ocean health, identify vulnerable areas, investigate how environmental changes influence where fish go, or evaluate the effectiveness of conservation and fisheries policies
  • NGOs and journalists can identify and investigate suspicious vessels, and advocate for stronger protection for important ecosystems
  • Fishers can show that they are operating legally and responsibly, giving them a market advantage by enabling them to sell their catch to customers who demand sustainable, traceable seafood
Every time I show the live map to somebody, they tell me something I didn’t know. In five seconds it can tell stories that never could have been told before.
Brian SullivanCo-founder of Global Fishing Watch and Senior Program Manager for Google Earth Outreach

沒有留言: