From Hobart, to London, to Dhaka: using cameras and AI to build an automatic litter detection system



Shutterstock

Arianna Olivelli, CSIRO and Uwe Rosebrock, CSIRO

It’s estimated about two million tonnes of plastics enter the oceans from rivers each year. But our waterways aren’t just conveyor belts transporting land waste to the oceans: they also capture and retain litter.

Currently, the most common method for monitoring litter relies on humans conducting on-ground visual counts. This process is labour-intensive and makes it difficult to monitor many locations simultaneously or over extended periods.

As part of CSIRO’s research to end plastic waste, we’ve been developing an efficient and scalable environmental monitoring system using artificial intelligence (AI).

The system, which is part of a larger pilot with the City of Hobart, uses AI-based image recognition to track litter in waterways.

Global insights help build a reliable model

The technology is underpinned by two branches of AI: computer vision and deep learning. Computer vision involves training computers to understand and interpret images and videos, whereas deep learning imitates how our brains process data.

Drawing on these capabilities, we worked in partnership with Microsoft (using its Azure cloud computing services) to develop an automated system for monitoring river litter.

We have been detecting and classifying items floating on the surface of Hobart’s stormwater channels, the River Thames in the UK and the Buriganga River in Bangladesh.

We’ve remotely analysed the amount of litter, the type of litter and how this changes across locations.

CSIRO research scientist Chris Wilcox setting up a fixed camera to monitor litter in Hobart.

Major damage from food packaging and bottles

Our work relies heavily on two applications of computer vision. These are “object detection” and “image classification”.

Object detection specifies the location of a particular object in an image and assigns it a label. Image classification assigns one or more labels to the image as a whole.

Before either of these models can be applied reliably, however, they have to be trained, tested and validated using a large number of labelled images. For this, we drew from our footage of river litter collected from Hobart, London and Dhaka.

Our dataset now contains more than 6,100 images with 14,500 individual items. The items are labelled across more than 30 categories including plastic bottles, packaging, beverage cans, paper and plastic cups.

Our data revealed food packaging, beverage bottles and cups were by far the most frequently spotted litter items across all three countries.

Aeriel view of the Buriganga River in Dhaka, Bangladesh.
The Buriganga river flows by Dhaka. It’s one of Bangladesh’s most polluted rivers due to the ongoing dumping of industrial waste (such as from leather tanneries) and human waste.
Shutterstock

Fake images aren’t always harmful

To build a well-performing machine learning model, we needed a balanced set of training images featuring all item categories — even if certain categories are more frequent in real life.

Introducing synthetic (computer generated) images to our dataset was a game changer.

These images were generated by Microsoft’s synthetics team based in Seattle. They rendered various objects and superimposed them over backgrounds obtained from our field photos.

Once the digital objects were created, the superimposition process was automatic. Thus, the team managed to produce thousands of synthetic pictures over just a few weeks, rapidly expanding our training dataset.

In this synthetic image, the transparent cup, face mask and aerosol container are digital renderings superimposed over an original photo taken by one of our cameras.

How are objects identified?

There are a few steps by which our system identifies litter objects in photos. First, the photos are all scored against a single-label (“trash”) object detector. This identifies items of litter in the frame and stores their coordinates as annotations.

These coordinates are then used to isolate the items and score them against an image classifier which includes all the litter categories.

Finally, the model presents the category it thinks the item most likely belongs to, along with a suggested probability for how accurate this guess is.

Here’s an example of the system detecting a water bottle and packaging as trash, and then placing both items into their respective categories. Probabilities are provided for the likely accuracy of the system’s guess regarding an item’s classification.

An AI-driven approach to litter management allows a quicker response than a manual system. But when it comes to litter, the major challenge lies in creating a model that can account for millions of different shapes, colours and sizes.




Read more:
As cities grow, the Internet of Things can help us get on top of the waste crisis


We wanted to build a flexible model that could be transferred to new locations and across different river settings, including smaller streams (such as Hobart’s stormwater system) and large urban rivers (such as the River Thames or the Buriganga River).

This way, rather than building new models for each location, we only have to deploy more cameras. Data retrieved could help identify litter hot spots, implement better waste-related policies and improve waste management methods to make them safer, smarter and relatively cheaper.

Keeping an eye on Hobart’s litter

We’ve also been collaborating with the City of Hobart to develop an autonomous sensor network to monitor gross pollutant traps, such as floating barriers or litter socks.

These structures, integrated into Hobart’s stormwater drainage system, are supposed to prevent solid waste such as cans, bottles, tree branches and leaves from reaching the estuary and ocean.

We currently have a network of sensors and six cameras installed under bridges tracking litter in the traps. The system can inform an operator when a trap requires emptying, or other maintenance.

Once in full use, the technology will provide almost real-time monitoring of litter around Hobart — assisting efforts to reduce environmental harm caused by stagnant, and potentially hazardous, waste lost to the environment.




Read more:
How sensors and big data can help cut food wastage


The Conversation


Arianna Olivelli, Research Affiliate, CSIRO and Uwe Rosebrock, Senior Software Engineer, CSIRO

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Forget sharks… here’s why you are more likely to be injured by litter at the beach



File 20181220 45400 1uwiqym.jpg?ixlib=rb 1.1
Sadly, people plus beach equals litter, so be careful out there.
Wikimedia Commons

Marnie Campbell, Murdoch University; Cameron McMains, Murdoch University; Chad Hewitt, Murdoch University, and Mariana Campos, Murdoch University

Our beaches are our summer playgrounds, yet beach litter and marine debris injures one-fifth of beach users, particularly children and older people.

Our research, published in the journal Science of the Total Environment, found more than 7,800 injuries on New Zealand beaches each year – in 2016, some 595 of them were related to beach litter. The most common injuries caused by litter were punctures and cuts, but they also included fractured limbs, burns, head trauma, and even blindness.

Children under 14 suffered 31% of all beach litter injuries, and were injured by beach litter at twice the rate compared with other locations in New Zealand. Beach litter injury claims exceeded NZ$325,000 in 2016, representing a growing proportion of all beach injury claims. Beach injury claims changed from 1.2% of the total in 2007 to 2.9% in 2016.




Read more:
This South Pacific island of rubbish shows why we need to quit our plastic habit


Our study relied on reported injury insurance claims in New Zealand, and thus probably underestimates the true injury rate, particularly for minor wounds. Our 2016 survey of beachgoers in Tasmania found that 21.6% of them had been injured by beach litter at any time previously – even on the island state’s most picturesque beaches.

Alarmingly, most beach users in the Tasmanian survey did not consider beach litter an injury risk, despite the high rate of self-reported injuries.

Awash with danger

As more debris washes ashore and our recreational use of our coasts increases, it is more likely than ever before that we will encounter beach litter, even on remote and “pristine” beaches.

Global studies have found up to 15 items of debris per square metre of beach, even in remote locations. On Henderson Island – a supposedly pristine South Pacific outpost miles from anywhere – some 3,570 new pieces of litter arrive every day on one beach alone.

Beach litter typically includes a huge range of items, such as:

  • broken glass
  • sharp and rusted metal such as car bodies, food cans, fish hooks, and barbed wire
  • flammable or toxic materials such as cigarette lighters, flares, ammunition and explosives, and vessels containing chemicals or rotten food
  • sanitary and medical waste such as used syringes, dirty nappies, condoms, tampons and sanitary pads
  • bagged and unbagged dog faeces and dead domestic animals.

The health hazards posed by beach litter include choking or ingesting poisons (particularly for young children), exposure to toxic chemicals, tripping, punctures and cuts, burns, explosions, and exposure to disease.

Degrading plastic can also produce toxins that contaminate seafood, potentially entering human or ecological food chains.

Rubbish knowledge

Despite the potential severity of these hazards our understanding and study of human health impacts from beach litter is poor. We know more about the impacts of beach litter and marine debris on wildlife than on humans.

Two of our previous studies in Australia and New Zealand have found beach litter that can cause punctures and cuts at densities 227 items per 100 square metres of beach, and choking hazards at densities of 153 items per 100 square metres of beach. These exposures to beach litter hazards in Australia and New Zealand may be 50% higher than global averages (based on preliminary data).




Read more:
How much plastic does it take to kill a turtle? Typically just 14 pieces


Even “clean” beaches can be hazardous, and may even increase the likelihood of injury. Visitors to a recently cleaned or supposedly “pristine” beach may be less vigilant for hazards. What’s more, European studies have found that actively cleaned beaches can still have hazardous debris items.

The risk of injury will continue to increase without concerted efforts to prevent addition of new debris and the active removal of existing rubbish. Besides watching where we tread when at the beach and participating in beach cleanups, we also need to make sure we deal with rubbish thoughtfully, so litter doesn’t end up there in the first place.The Conversation

Marnie Campbell, Chevron Harry Butler Chair in Biosecurity and Environmental Science, Murdoch University; Cameron McMains, PhD Candidate, Harry Butler Institute, Murdoch University; Chad Hewitt, Professor and Director, Murdoch Biosecurity Research Centre, Murdoch University, and Mariana Campos, Lecturer and researcher, Murdoch University

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Sustainable shopping: take the ‘litter’ out of glitter


Jennifer Lavers, University of Tasmania

Shopping can be confusing at the best of times, and trying to find environmentally friendly options makes it even more difficult. Welcome to our Sustainable Shopping series, in which we ask experts to provide easy eco-friendly guides to purchases big and small. Send us your suggestions for future articles here.


Scientists often get a bad rap as party poopers. As a case in point, my colleagues and I have provided data on the impacts of balloon releases on marine wildlife.

So when glitter – a highly visible and easily obtained microplastic – comes under the microscope, you might be tempted to groan. The good news is that we’re not out to ruin the fun: with Mardi Gras around the corner (bringing a ubiquity of sparkling Instagrams), here’s how to find ecologically friendly glitter.




Read more:
Mixing glitter and protest to support LGBTQ rights


All glitter goes to the ocean

When something fun or common is revealed to be destructive it should be a point of pride in our society that we adjust, adapt and move on to safer alternatives.

It therefore makes sense to investigate what data exist for glitter, and to consider whether it’s time for a change in attitude. So, what is glitter?

Glitter is typically made from polyethylene, the same plastic found in plastic bags and a host of other products. Despite glitter’s popularity in everything from cosmetics and toothpaste to crafts and clothes, remarkably little is known about the distribution or impacts of glitter on our environment. As a scientist, that worries me. Glitter is incorporated into consumer products without any real knowledge of its safety.




Read more:
Ten ‘stealth microplastics’ to avoid if you want to save the oceans


In contrast, there are dozens of scientific papers on micro-bead scrubbers (tiny plastic beads), which originate from many of the same products (such as cosmetics and toothpaste).

Research on micro-beads suggests that around 8 trillion beads are released into aquatic habitats every day in the United States alone.

Data for glitter are not available, but given its widespread use the situation is likely to be similarly alarming. It’s far too small for waste treatment facilities to capture, so glitter goes straight into your local river and out into the ocean. Because glitter particles are typically 1 millimetre in size or smaller, they can be ingested by a range of creatures, including mussels.

Again, data on micro-beads can tell us why we should be worried about this: a recent study from Australia showed that toxic chemicals associated with micro-beads can “leach” into the tissues of marine creatures, contaminating their bodies. If mussels, fish and other animals are ingesting glitter and micro-beads, these contaminants likely also pose a risk to humans that consume them.




Read more:
Eight million tonnes of plastic are going into the ocean each year


Thankfully, science is here to help. A range of compostable, vegan, 100% plastic-free “bio-glitters” have been created and are readily available online. So, at your next event, you can celebrate in glorious, sparkly style while also educating passers-by about ocean conservation. (I assure you, this is very popular; I do it all the time and I’m the life of the party.)

What to look for

Mica, a naturally occurring sparkling mineral, is often offered as a non-plastic alternative to glitter. However, some brands, such as Lush, are now using “synthetic mica” (made in a lab) because mica mining has been associated with child labour, especially in India.

Some plastics labelled “bio-degradable” will only break down in industrial composting units, at temperatures over 50℃. This is very unlikely to happen in the ocean, so look for terms like “compostable” and “organic” instead. (For more information on the difference between bio-degradable, compostable and everything in between, this United Nations report is very comprehensive – just read the summary if you’re in a hurry).

Fortunately, eco-friendly glitter is becoming much easier to find around the world, and more suppliers are turning to cellulose and other plant-derived bases for their product. Wild Glitter‘s founder, like many in the industry, cites “watching a weekend’s worth of plastic glitter wash down the plughole after a festival” as the impetus to sell an “ethical, eco-friendly, cruelty-free way to sparkle”.

Eco Glitter Fun is a member of the Plastics Ocean Foundation, a global non-profit; Glo Tatts makes beautiful temporary glitter tattoos; and, for an Australian twist, Eco Glitter make their product from Eucalyptus cellulose.




Read more:
Film review: A Plastic Ocean shows us a world awash with rubbish


Bio-glitter can be incorporated into any product. Tasmanian soap maker Veronica Foale switched to bio-glitter last year and hasn’t looked back – if a small business in a rural area can do it, you can too!

The ConversationThis is the key to success in the battle against litter: not all changes are difficult and affordable alternatives do exist. Once you’ve mastered bio-glitter, embrace the next challenge – a bamboo toothbrush perhaps, or reusable Onya produce bags? Never stop learning. Go forth and sparkle responsibly.

Jennifer Lavers, Research Scientist, Institute for Marine and Antarctic Studies, University of Tasmania

This article was originally published on The Conversation. Read the original article.