Application of machine vision in the Internet of Things

   The Internet of Things is an important part of the new generation of information technology. As the name suggests, the Internet of Things is the Internet where things are connected. Its implementation is mainly through various information sensing devices to collect in real time any objects or processes that need to be monitored, connected, and interacted with. All kinds of needed information are combined with the Internet to form a huge network. Its purpose is to realize the connection between things and things, things and people, and all things and the network, so as to facilitate identification, management and control. FORRESTER, an authoritative American consulting organization, predicts that by 2020, the world's Internet of Things business will reach 30:1 compared with the human-to-human communication business. Therefore, the "Internet of Things" is called the next trillion-level business. Communication business, all signs indicate that the world has begun to enter the Internet of Things era.

The components of the Internet of Things can be summarized into the following four parts:

1. The item coding and identification system is the basis of the Internet of Things; 2. The automatic information acquisition and perception system solves the problem of the source of information; 3. The network system solves the problem of information interaction; 4. The application and service system, which is the building material purpose of networking. The IoT architecture is shown in the figure below.

At the basic layer of the Internet of Things, information collection is mainly achieved by sensors, of which visual sensors are the most important and widely used. The discipline that studies the application of visual sensors is machine vision. Machine vision is equivalent to the human eye and is mainly used to detect some complex pattern recognition tasks. Nowadays, more and more projects require the use of such detection, such as landmark point recognition on AOI, appearance defect detection of electronic equipment, quality traceability of food and drugs, and visual navigation on AGV. These areas are all major areas of machine vision. A place with a purpose. At the same time, as the Internet of Things technology continues to ferment, the application of machine vision in this field is attracting widespread attention.

In the current Internet of Things technology, automatic recognition is generally achieved through image processing based on images collected by image sensors. Barcode reading, biometric identification (face, voice, fingerprint, vein), image recognition, OCR optical character recognition, etc., all use machine vision image acquisition equipment to collect target images, and then use software to analyze and compare the texture features in the image, etc. , to achieve automatic identification. At present, among domestic machine vision manufacturers, Xinxiwang has the most visual products used in the Internet of Things industry. The main application directions of its products in this industry include: automatic identification of fabric tissue based on image processing technology, automatic identification of fingerprints, and stripe traces. Image processing automatic identification, animal hair and plant fiber microscopic automatic identification, etc.

According to estimates by authoritative departments, with the research and development of Internet of Things technology and industrial development, China's Internet of Things market size will reach 489.6 billion yuan in 2013. By 2015, this scale will reach 750 billion yuan, and its development prospects will exceed those of computers, the Internet, Traditional IT fields such as mobile communications. As the third revolution in the development of the information industry, the Internet of Things covers more and more fields, and its concepts are becoming increasingly mature. Addressable, communicable, controllable, ubiquitous and open models are gradually becoming the evolution of the development of the Internet of Things. Target. For the construction of "smart cities", the Internet of Things extends information exchange to the category of things. The extremely rich value information and ubiquitous intelligent processing will become an important means for city managers to solve problems. Because of its close connection with the Internet of Things industry, the machine vision industry will once again experience a rapid take-off process. All practitioners in the vision industry must improve their professionalism, wait for the arrival of this century, and grow together with the development of the industry.