Humans continue to produce garbage and waste at unprecedented rates, and sadly much of it ends up in our oceans which promptly deposit great amounts of it on our beaches. Not only are garbage-covered beaches unsightly, but they pose very real health risks to our coastal wildlife like seabirds, fish or sea turtles that can often mistakenly consume it thinking it is a source of food. Combating all of this trash ending up on our beaches has prompted several international organizations and some of the most affected countries to put systems in place to monitor all of the garbage. Unfortunately many of these groups use volunteer workers to physically count the garbage found on the beach, a process that both is slow and can produce inconsistent data.
But a group of researchers at East China Normal University in Shanghai have developed a method of using high-powered 3D scanning technology to capture high-quality data and identify and count the garbage automatically. Not only could this process completely revolutionize the way that coastal garbage is tracked and analyzed, but it would lead to better, more efficient methods of preventing the garbage in the first place. Not only is the system that lead researcher Zhijun Dai and his team developed more accurate when documenting the garbage, it is capable of reducing the amount of time required to count and sort the garbage in a matter of minutes rather than the hours it would take to be do it manually.
The specific type of 3D scanning technology used is called LIDAR, a process of ranging and light detection that fires off laser pulses and measures the time it takes for the light to bounce off of its surroundings to develop a detailed point cloud. This point cloud can then be analyzed to reveal all sorts of detailed information about the surroundings. Each material and surface will bounce back slightly differently, and these differences can be isolated and recorded so objects can be identified and cataloged. The data can also be manipulated to exclude irrelevant data, or data that is not required, such as foliage, trees, grass or even people.
Dai and his team wanted a system that would allow them to send a 3D scanner to a beach and have it instantly be able to count and catalog all of the trash found while ignoring the rest of the immediate surroundings. In order to do this the team went to a clean beach and took 87 different types of garbage with them and scattered it around. The 3D scanner was then aimed at the beach with the newly deposited trash from about 100 meters away, and in a little over 10 minutes it had collected a dense point cloud with 96 million points. The researchers then removed all of the irrelevant data and developed an algorithm that would be able to recognize what is and is not trash enough to count it, and even differentiate between different types of garbage.
Once the algorithm was created, the team then moved on to a beach popular with tourists, and full of garbage, called Beihai. They fired the LIDAR at the beach three different times on three different days. The 3D scanner was working for about 20 minutes each day, and was able to capture a huge point cloud of data. In order to compare the results from the 3D scanner, Dai and his team physically counted as much of the trash on the beach as possible, including tin cans, plastic bags, styrofoam takeout containers and even odd debris like mismatched shoes. The algorithm was able to accurately count and document about 75% of the trash but was completely unable to detect glass garbage. However the system was able to detect paper, clothing, metal, and plants without much problem. The team published their results, “Semi-automatic recognition of marine debris on beaches”, in the scientific journal Nature. Authors include Zhenpeng Ge, Huahong Shi, Xuefei Mei, Zhijun Dai and Daoji Li.
While the results were not ideal, Dai and his team of researchers believe that they can improve on the process, and that it will save a tremendous amount of time and manpower. The 3D scan was able to capture its data in a matter of 20 minutes, while the hand counting took almost three hours to complete. Not only would this allow researchers studying the way that trash is deposited along coastal regions to capture data faster, it could be used to predict how and where trash out to sea can be intercepted and collected. Dai also believes that the LIDAR could be mounted on a robot or a drone that would automatically monitor the beaches, which would remove a human from the process almost entirely. You can read the entire report on the team’s findings online at Nature. Discuss further in the 3D Scanning to Calculate Beach Trash forum over at 3DPB.com.
You May Also Like
3D Printing News Briefs, October 24, 2020: nTopology & Etteplan, DSM, CAR3D Project, MELD Manufacturing
In today’s 3D Printing News Briefs, we’re talking about a partnership between nTopology and Etteplan, a new material from DSM, CAR3D’s COVID-19 protection equipment, and a pretty cool 3D printed...
Ansys and EOS Partner for Metal 3D Printing Simulation
One of the most exciting trends currently taking place in additive manufacturing (AM) is the development of simulation software for improving the predictability and repeatability of 3D printing technology. An...
Sigma Labs & Northwestern Partner for In-Process Quality Assurance of DED Metal 3D Printing
This spring, AM quality assurance software developer Sigma Labs, Inc. (NASDAQ: SGLB) launched the new Production Series of its PrintRite3D software for the commercial 3D metal printing industry, not long after signing an...
Velo3D Launches New 8-Laser Sapphire XC Metal 3D Printer
Velo3D announced today that it has launched the new Sapphire XC metal 3D printing system. The company states that it already has 13 advanced orders for the machine, touting a...
View our broad assortment of in house and third party products.