recognition software

FISH FACE - successful installation on-board!

The Fish Face project is in its final phase, and the first Fish Face unit has now been successfully installed on-board a fishing boat in Kupang, Indonesia! Amir from Refind struggled with everything from expected challenges like lack of electricity and internet connectivity availability, to more unexpected problems like cable-chewing mice and timing of the tide.

kupang_installation.JPG

Since July the Fish Face Unit has been used at a local fishery site in Kupang to collect images of different fish species in order to build up a large database of the local fish species. Thousands of images have been collected and there is a first “classifier” installed now - able to recognise about 65 different fish species. This software will be updated continuously along with an increased number of collected images.

The Fish Face Unit was, as planned, installed on the deck on a local fishing boat. Deck space is very limited and so is the access to constant electricity or internet. We have solved this by adding a power supply and a cloud based data storage solution to which information can be uploaded when possible. The activity of classifying fish is not real-time dependant and can be performed when possible. The geographical location tagging will however be done at the actual time when the photo is taken, so the GPS data collected is not depending on internet access.

Image Today at 19_21_10.JPG

All classifications are stored with their species name and GPS tags, and we have also added the possibility to do quality checks by connecting to a webpage where the latest classifications can be viewed. This will give both fishermen and the staff of The Nature Conservancy a transparent and easy process to overview and improve the performance.

Screenshot 2018-11-16 at 11.04.41.png

In the table above the classifications are listed and an image is shown from the actual Fish Face unit next to the label image of the classification performed. Later on a segmentation functionality will be added, where the fish length will be calculated. The system also allows for handling input from many different image sources (boats).

The next phase of the project will focus on evaluating the equipment and overall solution, and aim to expand the project into installing more camera units on more boats and expand the operations.

Refind sells battery sorter to CMA Ecocycle in Australia!

We are proud to announce that our technology is entering the southern hemisphere and more specifically Melbourne in Australia, where the leading Australasian electronics recycling company CMA Ecocycle is located. CMA Ecocycle are growing their business and are in the process of building a new facility, where the battery sorting process will take place. The OBS500 will be the first automatic battery sorting line installed in Australia.

“The automatic sorting is an important step for us to make the battery handling more effective, both in terms of sorting quality and direct costs”, says Doug Rowe, the managing director of CMA Ecocycle, “and it will enable us to safely handle our material, which we believe will increase in the time to come. We are happy to bring the latest advanced sorting technology to Australia with support from the government.”

CMA Ecocycle is originally a mercury recovery and recycling company, that handles lighting equipment, medical waste, electronic waste and batteries. The company has a high level of automation and always aims to work with state-of-the-art equipment to get the most out of the material.

For us at Refind, it is of great value to expand into yet another country and region of the world. However, it has also raised concerns in terms of maintenance availability and the battery database update, but we have seen the issues as opportunities by going straight at them. For example, we have now developed the scanning software so that most of the battery scanning process, that is needed for adding new batteries to the system, will be possible to perform by using the battery sorter.

The OBS500 will arrive to its new home in Melbourne towards the end of the year. The official opening of the new process line at CMA Ecocycle will take place in the 1st quarter 2019.

Picture CMA Ecocycle entrance.png

Wasteless Future blog post about Refind!

Antonis Mavropoulos from Wasteless Future contacted us for an interview recently and the result was a nice blog post about Refind, our technology and our visions. Read it here, along with a lot of other nice articles!

FISH FACE project goes on board!

We are happy to announce that we are launching the next phase of the FISH FACE project together with The Nature Conservancy! Until now we have verified that our software can handle fish species recognition by collecting fish images taken in a on-land fish photo booth. The next step is to design, build and install an on-board photo capturing device to be used on some of the small fishing boats of Indonesia.

The purpose of collecting the images and recognise them automatically is to understand the fish populations better by gathering information about the species, the age, the size and the geographical origin in a more efficient way. We work closely with The Nature Conservancy who are involved in many different fish-related projects - and hope to make a large impact on the fishing industry and fishing research by introducing artificial intelligence and automation. This phase of FISH FACE is planned to take about a year and will involve several field trips to Indonesia for testing and try-outs. 

The new fish photo booth should accommodate fish sizes up to 80 cm long, and later on even larger fish. Given the rather rough conditions on these types of boats, the device must be robust and foot print efficient as well as not cause any extra handling for the fishermen. Quite a challenge - stay tuned! 

Most frequently asked questions - answered!

We have compiled the most frequently asked questions we receive from you. Hopefully, they provide some more insight into our operations.

1. What kind of objects are Refind's systems capable of identifying?

We have been asked to identify all kinds of objects: both crazy and normal stuff. Body parts in trash, coconut grading as the unexpected objects and used electronics and defects in furniture as more on the normal side. Important questions in return that we ask the customers are WHY (why automated and not manual) and HOW (how do you tell it apart), and then we weigh that against what different sensors would be able to perform.

2. What kind of sensors do you utilize?

We use all kinds of sensors and combine them on our technical software platform; the neural networks can use different sensors as data input sources. However, to have everything takes time and usually costs more than it is worth. Also, a limitation of sensors usually increases the processing speed. The most important sensor for our current applications is an RGB camera for taking images.  We also use laser sensors for identification of sizes or location detection.

3. How do you select the right kind of sensors for each system?

The sensors are selected based on the material to be sorted. We can use material specific sensors (like NIR, X-ray, LIBS or something else that can only tell what material it is). This makes sense when you are dealing with objects with homogeneous material. There are many companies already doing this.

3.1 Following up on that, how does Refind differ from other companies?

We focus on objects with complex material structures, like whole products, used electronics, where a material sensor does not make sense. You need to understand what model something is for it to be useful information for the customer. Then a camera is the best thing.

4. What kind of camera do you utilize?

So far, we are doing fine with our RGB camera, but it is similar to a one-eyed person that cannot tell depth. By adding a 3D camera, we have basically added another eye, allowing for depth check, which makes a big difference for recognizing items in a co-mingling environment.

5. How does your classification system work?

A computer program is fed with images of known identity for example "this is a picture of a computer", it then learns what to look for in these images in order to correctly classify them. Once the system is trained, the learning process stops and it is used for real-time classification of the objects that are being sorted by our machines.

6. How accurate are the classifications?

It depends on how many images the system has got access to for training itself for the task. More images give better accuracy. The battery sorters produce fractions of 97 - 99% purity.

7. How many different kind of objects can you recognize? AND How many images do you need per object to be recognized?

The system can recognize as many objects as have example images of. The number of images needed is different depending on the objects and kinds of images we get, from 30 and up to several thousands depending on the object.

Do you have a question for us? Don't hesitate to email us at info@refind.se or give us a call at 073-310 03 62.

 

Meet and listen to us in Salzburg 17-20 January 2017!

As usual, Refind will exhibit at the International Electronics Recycling Congress in Salzburg on January 17-20! New for this year is that our CEO Johanna Reimers will be speaking about our sorting technology in one of the seminars - the "Sorting technology" session on Thursday January 19th in the afternoon. Don't miss the speech called "Is More Technology Really the Solution to the Challenges within Circular Economy?".

Check out the entire program here.

Also new for this year is that we will bring our latest machine - a small desktop sorting unit including a robotic pick arm. As of now it can sort phones and fish - come and have a look at it in our booth!

Vote for Fish Face - finalist in the Google Impact Challenge!

We are very excited to announce that our Fish Face project is one of 10 finalists in the Google Impact Challenge: Australia! With your vote the project can become one of the winners and receive an additional 500 000 AUD to develop our technology further!

A world without fish is a world without food for half of the world's population. By collecting more data about fish species, age and behaviour in an automated way, we will know more about fish populations and enable a more sustainable fishing. The world needs it - we need it!

The Fish Face project is mainly run by The Nature Conservancy, a world wide environmental organisation, who has access to the fisheries and the expertise knowledge about the fish. The other two project members are we at Refind - proud providers of recognition technology and app development for the automated data collection - and Seth Heine, an entrepreneurial leader with sustainability as the main interest.

Voting is now underway to choose an overall winner to receive extra prize money from Google to further develop FishFace. If you’d like to support a move to sustainable fisheries around the world and keep fish in the sea, vote for FishFace now and help it win. Voting ends at midnight (ADST) October 25, 2016.

We’re also really encouraged that the Australian Government’s Australian Fisheries Management Authority (AFMA) has agreed to share its data to help build the machine learning engine that will be the power behind FishFace.

Recycling International writes about the CEO of Refind

As a part of their recurring theme 'Women in Recycling', Recycling International has now discovered our woman at Refind - Johanna Reimers. Read the interview here, and look at their website for all other news within the recycling business!

RI AUG issue Interview.jpg

Meet the desktop grader!

Desktop_grader

Our latest product family member is born - the desktop grader! It is a smaller and more flexible version of the e-grader equipment, narrowed down to a smartphone and a light tunnel which enables instant identification through a mobile app.

As a customer, you install the desktop grader and help building up the image database for your needs. Once the classifier is ready for use, you pay a license fee based on number of classifications. The desktop grader can also be turned into a sorting equipment similar to the e-grader, by adding conveyor belts and separation actuators.

Read more in this product sheet, have a look at the video below:

 

FISH FACE - Refind software enables sustainable fishing

 

We are happy to announce that Refind are involved in a large fishing data collection project, Fish Face, together with The Nature Conservancy. The software from Refind will be used to identify fish species.

Without proper data, fish can't be sustainable managed. But a new technology could change all of that.

Fish stocks around the world are declining—with an estimated 90 percent of the world’s fisheries over or fully exploited. In developing countries, like Indonesia, the decline of a fishery can have severe consequences—as nearly 40 percent of the Indonesian population lives just above the poverty line, fishing is a way of life and provides an important food source for millions of people.

A key challenge in addressing overfishing is the lack of data on just how many fish still exist. Especially in complex multi-species fisheries, like the ones in Indonesia and in many other tropical developing countries, data just doesn’t exist on all types and sizes of individual fish, making sound management almost impossible. In fact, some 90% of fisheries globally are lacking in stock assessment data. Traditional stock assessment methods are prohibitively expensive, and in the majority of fisheries in the developing world, the condition of stocks is not known.

Facial recognition for fish

The Nature Conservancy’s Indonesia Fisheries program is working with Refind Technologies to identify fish. The project is called Fishface and the ultimate goal is to build this technology into a smartphone app that could be used on fishing boats throughout the region and eventually be deployed around the globe. Through the use of affordable image recognition software that will detect species from photos, much faster and more accurate sorting of fish will be possible at the processing plant, or even as it is landed on the boat.

Ultimately, the pilot of the Fishface technology will offer a low-cost assessment of fish stocks—providing the essential data needed to assess and manage fisheries that are struggling around the world.

The framework envisioned will be applied across these types of fisheries in a multitude of geographies, with the potential to impact the some 260 million people who depend on fish for income and food.

Read this and more here on the The Nature Conservancy website.

The Nature Conservancy (TNC) is the leading conservation organization working around the world to protect ecologically important lands and waters for nature and people. They are present in over 35 countries around the world.