INNOTRENDZ | Knowledge Base for Everyone
Enter Valid Email
Show Error Password

Forgot your password?

Enter your first name
Enter your last name
Enter your valid email
Show Enter your password
Please Accept Terms

Lost your password? Please enter your email address. You will receive a unique code to create a new password.

Enter Valid Email

Back to log-in

An innovative way to recognize tree with machines

An innovative way to recognize tree with machines

Innovation 602 0 1 Download
Share :

In Summary

An innovative way towards the management of trees in urban areas by utilizing smart satellite technology.

Editor Posted by Sanchita

Managements of trees in urban areas will be much simpler and more data specific now with a new technique used in recognizing street trees and thus their management. One can find a detailed description of the technique in the 2016 IEEE conference on computer vision and pattern recognition, held in Las Vegas last year. Every 20 – 25 years, arborists are put to work to take a count of the trees in various cities along with certain specific details of trees. This was a manual task which used to be extremely arduous and time consuming.

How it works

A team of researchers from Caltech initiated their survey with the tree population of Pasadena. They used the automatic “look” method to count the number of trees of a certain area using the street level image from Google maps. Google allowed Caltech to use the data without any fee as that would help in further advancement of the society. An algorithm was development with certain patterns, which would help enable a machine to identify and distinguish a tree. The most interesting point about Perona’s research group is that they use artificial neural network, algorithms which is generally used by a human brain to learn and to recognize objects in images. The next task was to train the machines in identifying the different species of trees. The research group took a strategic decision to partner with a Tree management company, The Davey Resource Group in 2013. This partnership helped them with the tree inventory which has led to a survey which include, identification of species, measurements and geographical location of approximately 80,000 trees. The above information helps the researchers train the machine with around 18 of the 200 species of trees present in Pasadena. It was found that Perona’s team of researchers with the help of Google maps could identify the trees accurately up to 80%.

Advantages of this newer technique

Once this technique comes to use in the mainstream, it can be used for various reasons which will be of great help for the humankind. Urban forest management has been an ambitious project since long in most of the developed and developing nations. This smart way to identify and measure the various tree species will help several Government departments to take adequate steps towards increasing greenery and creating a pollution free green environment. With the help extended by Google, NASA and the Naval research, the Perona’s research team will help take several critical decision in various countries in terms of environment protection, ecological balance and urban forestation.