Garbage Classification: Man-machine Matching, Not Tired of Work

  Computer vision

The people of Shanghai have really been having a rough time recently. It’s all rubbish.

You should have known that Shanghai’s “garbage sorting campaign” is real. According to the regulations to be implemented on July 1, if the classification is not in place, both units and individuals will be fined, even tourists.

条例中规定,对垃圾不分类并拒不改正的个人罚款 50 元至 200 元
The regulations stipulate that individuals who refuse to classify garbage and refuse to correct it will be fined from 50 yuan to 200 yuan.

As a result, the most important garbage classification standard has become the topic of conversation for all Shanghainese in the near future. Like going through the college entrance examination, everyone has picked up the corresponding classification of all kinds of garbage.

For this reason, Hungry, Alipay also quickly launched a garbage disposal service … It seems that garbage classification may lead to a number of innovative projects.

Garbage classification is no longer an armchair strategist.

In China, garbage classification was put on the agenda as early as 2000, and Beijing even had a pilot project earlier.

But in the past 18 years, all this has remained in propaganda. It was not until this month that he really entered the life of the Shanghai public. However, Beijing, Guangzhou, Shenzhen and other cities will also enact laws one after another to enter the “mandatory era” of garbage classification.

1996 年 12 月,北京大乘巷小区家委会发出《致居民的一封信》宣布小区垃圾将分类投放,成为北京第一个试点垃圾分类的小区
In December 1996, the Home Affairs Committee of Beijing Dachengxiang Residential Quarter issued a “Letter to Residents” announcing that residential waste would be classified and put into use, becoming the first residential quarter in Beijing to pilot waste classification.

Facing the “strictest” garbage classification in history, Shanghai residents are complaining bitterly. It is really a difficult task to remember the garbage categories corresponding to 104 items.

For example, the classification of dry and wet garbage has made many people dizzy: wet tissue is dry garbage, while dry mushroom is wet garbage. Plastic cups belong to dry garbage, while milk tea belongs to wet garbage, so what kind of garbage does half a cup of milk tea belong to?

However, we need not rush into this. Take Japan, which has the most careful and strict classification of garbage, for example, it did not reach the current level overnight.

Japan: How to Implement the Most Strict Garbage Classification in the World

Since the 1970s, Japan has been facing serious environmental pollution problems and began to implement garbage classification measures. It took 27 years to finally realize the national garbage classification.

Japan’s garbage classification and careful to strict degree, yokohama city alone has printed 27 pages containing 518 categories of manuals.

One page in the waste sorting manual of yokohama city, Japan

Nowadays, Japanese people carefully classify the garbage into different kinds. Different kinds of garbage are packed in designated bags and thrown into garbage stations on designated dates. This has become a part of their life.

Therefore, by comparison, Shanghai people may feel very happy.

In Japan, different kinds of garbage need to be thrown away at different recycling times. If they are missed, they have to wait another week or even a month.

As the first step to turn waste into treasure, garbage classification is of great significance. Garbage classification can bring social, ecological and economic benefits. If the resource potential contained in the recycled domestic waste can be fully exploited, Beijing alone can obtain an economic benefit of 1.1 billion yuan per year.

Now, we have taken the first step in our action against garbage. At the same time, artificial intelligence is also taking action.

AI classifies garbage accurately and quickly.

In the waste recycling process, artificial waste classification is only the first step. After that, the garbage will be sent to the sorting center, where the sorting workers will further refine the classification. This is a really dirty and tiring job.

In the variety show “running men”, members experienced the sorting of kitchen waste.

Fortunately, technology is helping us finish this unpleasant job.

As artificial intelligence technology becomes more and more mature, 20,000 tons of recyclable materials can be saved from landfills every year by means of intelligent technology. Using technology, we can greatly improve the garbage recovery rate, and the efficiency is far higher than that of manual labor.

Robot Arm for “Vision” Classification: Image Recognition+Neural Network

At present, a more mainstream method of garbage disposal is to use image recognition in computer vision and equip intelligent sorting robots to classify garbage.

U.S. company BHS( Bulk Handling Systems) has invested nearly 60 intelligent sorting machines MAX-AI in the world. MAX-AI uses visual recognition and multi-layer neural network technology to cooperate with conveyor belt to deliver garbage, and finally realizes classification of different materials.

The MAX-AI sorting robot is like an inverted tripod with suction cups at the end.

MAX-AI can identify and classify objects with an accuracy of 90%, almost the same as that of human beings, but at twice the speed of human beings.


Another company, ZenRobotics, uses mechanical arms in conjunction with image recognition to classify and pick up building debris.

The company’s largest robot classifier, called Heavy Picker, has an oversize pincers at the end of its arm, which can lift 60-pound objects, but it is extremely flexible with heavy loads and can carry out classification work quickly.


Sorting robots used by ZenRobotics are classified by the following steps: scanning the garbage flow through camera images, then using machine learning to identify materials, controlling mechanical arms to grab, and finally classifying garbage of the same material.

Taalas, chief executive of ZenRobotics, said that in addition to being able to sort building materials, their robots can also play a role in paper mills, plastic recycling and raw material recycling.

Robot Arm Classified by “Tactile”: Tactile Sensor

The Massachusetts Institute of Technology’s Computer Science and Artificial Intelligence Laboratory (CSAIL) has developed an intelligent robot that uses the “touch” of sensors to classify different waste materials.

Different from the robot based on vision principle, this intelligent robot mainly relies on the sensor fixture on the mechanical arm, which is completely driven by electric power. Through contact, it can detect the difference between paper, metal and plastic. For example, metal objects can be judged by sensing their conductivity.

The robot arm has a soft grip that can remove objects from the conveyor belt and identify garbage materials by touching them.

The fixture components are a pair of air cylinders and high deformation capacitive pressure and strain sensors. Earlier this month, the lab released a video showing working robots grabbing boxes, cups and cans. According to videos, paper cups, empty boxes and soda bottles can be classified by pressure sensors.

Garbage sorting guide: home → recycling point

In order to raise people’s awareness of garbage classification and recycling, Urbika and Greenpeace Russia are cooperating in a project: design a garbage collection map to help people find the nearest garbage collection point through the map.

As long as the user inputs the address, Recyclemap will pass different colors.
Shows the nearest drop-in point and drop-in garbage category.

Garbage is classified according to materials: plastic, glass, paper, metal and other materials, and they are sent to recycling. Recyclemap will show citizens handing over garbage that has been sorted and ready for recycling.

At present, the product has been popularized in 69 cities all over Russia. Every year, nearly 500,000 Russian users query the location information of garbage disposal points through the map.

Heart+action is the key to solving the problem.

The garbage sorting campaign launched in Shanghai has received many complaints and puzzles. However, I believe that with the advancement of all parties and the popularization of environmental protection knowledge, the public’s awareness in this regard will continue to increase.

Although there are many cases in which artificial intelligence technology has been successfully applied to garbage classification in this era, we cannot rely entirely on technology. People’s actions are still indispensable in order to really do this job well.

Whether it is a city or an individual, its beauty is not only reflected in the appearance of glamour, but also we should see the hidden dangers behind the glamour and use actions to change it.

Friends who can’t wait for AI help can look for them first (not advertisements)