News / Blog

Plastic waste is a major problem on earth because it causes pollution to the environment. Even so, plastic waste is large and easily visible. Then what about microplastics?

Microplastics are Threats to the Environment

Microplastics are very small pieces of plastic and can pollute the environment. Its extremely small size (less than 5mm) makes it difficult for the naked eye to catch. Microplastics scattered in the oceans is a threat to marine life.

For many years microplastics have been the cause of marine pollution worldwide. Its small size can be swallowed up by marine life such as fish. Fish enter the human food chain so that these microplastics also affect human health.

Therefore, it is necessary to know the actual conditions of microplastics in the sea by analyzing the number, size, and type of microplastics in each area, including identifying the source, route, and direction of the microplastic flow.

Japanese Company Develops AI Software to Count Microplastic

Until now, the method used to take microplastic samples is by using a fine net, then put the sample under a microscope and analyze the microplastic manually. Imagine how long it would take to do all this?

The other problem is, the manual method is considered to underestimate the smaller microplastic particles, with sizes below 300μm that can pass through fine nets.

Therefore, a Japanese company, NEC Corp together with JAMSTEC (Japan Agency for Marine-Earth Science and Technology) has developed a system to automatically detect microplastics in seawater and sediment samples using the help of AI.

NEC Corp and JAMSTEC revealed that the development of this system aims to improve the sophistication of microplastic measurement techniques to evaluate the dynamics and environmental impacts of marine plastic waste.

Using AI to Speed Up the Process and Improves Accuracy

JAMSTEC is conducting comprehensive research on technology that can automatically detect and analyze microplastics in the ocean. Then, NEC Corp creates a system capable of detecting microplastics with high speed and accuracy.

The technology capable of doing this is image recognition technology based on RAPID Machine-Learning, which combines deep learning technology. Specifically, NEC Corp develops software leveraging JAMSTEC’s expertise in R&D in the microplastics.

The process starts by coloring microplastics and then taking the video using a fluorescence microscope while flowing seawater samples. The video footage is then collected. Software developed by NEC Corp will then automatically detect the microplastic images from the video. 

The software will automatically classify microplastics based on shape and size using the help of AI. This process is very fast at 60 per minute.

Dr. Masashi Tsuchiya as Deputy Group Leader of Marine Plastics Research Group, JAMSTEC hopes this project can contribute to the development of appropriate regulations. Meanwhile, NEC Corp is committed to continuing to develop advanced technology with AI for environmental management.

Reference:

Japantoday

Photo by Brian Yurasits on Unsplash