Machine vision is an important branch of artificial intelligence. Machine vision has a wide range of applications and can be applied in intelligent manufacturing and many intelligent life fields; it is technically unique and is a cutting-edge technology for non-contact recognition and measurement of objects ; It has cost economy in hardware and will not cause cost pressure on the cost of the product. Extensiveness and uniqueness make it part of the core competitiveness of products in many fields; and economy can free the product from the shackles of hardware, so as to be more flexible in product design and customer demand, and make it more powerful. profitability.

1: Face Recognition

Face recognition is a popular application in the field of artificial intelligence vision and imagery. In February this year, MIT Technology Review released the list of the "Top Ten Breakthrough Technologies in the World in 2017". " is listed among them. This is a technological breakthrough from China in the 16 years since the list was created.

At present, face recognition technology has been widely used in finance, justice, military, public security, border inspection, government, aerospace, electric power, factories, education, medical and other industries. According to industry analysts, my country's face recognition industry is in strong demand, and the demand pushes companies to dare to invest. At present, the technology has the conditions for large-scale commercial use, and will grow rapidly in the next three to five years. This year, this technology is expected to usher in a big explosion in the fields of finance and security.

2: Video/surveillance analysis

Video/surveillance analytics is the second most popular application in AI vision and imaging.

Artificial intelligence technology can quickly retrieve and query structured video content information such as people, vehicles, and objects. This application makes it possible for the public security system to find criminals in the complex surveillance video. In transportation hubs where a large number of people flow, this technology is also widely used for crowd analysis, prevention and control early warning, etc.

The video/surveillance field has wide profit potential and various business models. It can not only provide overall solutions for the industry, but also sell integrated hardware equipment. The application of technology to video and surveillance is forming a trend among AI companies, and this technology application will be the first to set off a wave of applications in industries such as security, transportation and even retail.

3: Image recognition analysis

Still image recognition application popularity ranks third in the field of vision and imagery. The number of application companies that use artificial intelligence technology solely for image recognition and analysis is not as many as expected. There may be several reasons:

1. At present, there is a large profit space in the direction of video surveillance, and many companies focus on the field of video surveillance;

2. Face recognition is an application scenario of image recognition. Most companies that do face recognition are also providing image recognition services, but the sales results are not good, and the main profit point is face recognition;

3. Most commercial scenarios of image recognition still belong to the blue ocean, and the potential remains to be developed;

4. Most of the picture data is mastered by large Internet companies, and there are few data resources for start-up companies.

4: Driving assistance/intelligent driving

With the popularity of automobiles, automobiles have become a very large application direction of artificial intelligence technology, but for now, there is still a long way to go before the technology matures to fully realize automatic driving/unmanned driving.

However, using artificial intelligence technology, there are more and more functions and applications of driving assistance in automobiles. Most of these applications are based on computer vision and image processing technology.

5: 3D Image Vision

3D image vision is mainly for the recognition of 3D objects, and is used in 3D visual modeling, 3D mapping and other fields.

6: Industrial Visual Inspection

Machine vision can quickly acquire large amounts of information and process them automatically. In the process of automated production, machine vision systems are widely used in the fields of working condition monitoring, finished product inspection and quality control.

The characteristic of machine vision system is to improve the flexibility and automation of production. It is used in some dangerous working environments or occasions where artificial vision is difficult to meet the requirements; in addition, in the process of large-scale industrial production, machine vision inspection can greatly improve production efficiency and production automation.

7: Medical Imaging Diagnosis

More than 90% of medical data comes from medical images. The field of medical imaging has massive data that nurtures deep learning. Medical imaging diagnosis can assist doctors and improve the efficiency of doctors' diagnosis.

8: Text Recognition

Computer text recognition, commonly known as optical character recognition, uses optical technology and computer technology to read the text printed or written on paper and convert it into a format that can be accepted by computers and understood by humans. This is a key technology to achieve high-speed text entry.

9: Image and Video Editing

In 2016, Google held an exhibition of paintings by artificial intelligence writers. Through an art generator called DeepDream, Google can transmit neural networks from the inside to the outside. Instead of recognizing images, creating images. Some call the paintings made by these machines the dream of machines.

At present, there are many applications and machine learning algorithms on the market to process images, which can automatically repair, beautify, and transform images. And more and more favored by users.

Machine vision has gradually broken through industrial detection in application scenarios, and its application boundaries have gradually expanded to the field of intelligent life.