One article to understand the working principle, application and selection of vision sensors

Vision sensor technology is one of the seven categories of sensor technology. Vision sensor refers to the image processing of the image captured by the camera to calculate the feature quantity (area, center of gravity, length, position, etc.) of the object, and output Sensors for data and judgment results. Vision sensors are the direct source of information for the entire machine vision system, and are mainly composed of one or two graphics sensors, sometimes accompanied by light projectors and other auxiliary equipment. The main function of a vision sensor is to acquire enough raw images for a machine vision system to process.

Vision sensor technology is one of the seven categories of sensor technology. Vision sensor refers to the image processing of the image captured by the camera to calculate the feature quantity (area, center of gravity, length, position, etc.) of the object, and output Sensors for data and judgment results. Vision sensors are the direct source of information for the entire machine vision system, and are mainly composed of one or two graphics sensors, sometimes accompanied by light projectors and other auxiliary equipment. The main function of a vision sensor is to acquire enough raw images for a machine vision system to process.

One article to understand the working principle, application and selection of vision sensors

How Vision Sensing Works

Vision originates from a way for the biological world to obtain external environmental information. It is the most effective way for natural creatures to obtain information, and is one of the core components of biological intelligence. 80% of human information is obtained by vision. Based on this inspiration, researchers began to install “eyes” for machines, so that machines can obtain external information by “seeing” like humans, thus giving birth to a new discipline – computer vision , People imitate the production of machine vision system through the study of biological vision system, although it is very different from the human vision system, but this is a breakthrough progress for sensor technology. The essence of visual sensor technology is image processing technology, which is presented to researchers by intercepting the signal on the surface of the object and drawing it into an image.

Vision sensors have thousands of pixels that capture light from an entire image. The clarity and detail of an image is usually measured by resolution, expressed in the number of pixels. After capturing an image, the vision sensor compares it to a baseline image stored in memory for analysis. For example, if a vision sensor is programmed to recognize a machine part with eight bolts correctly inserted, the sensor knows to reject a part with only seven bolts, or a part with misaligned bolts. In addition, no matter where the machine part is located in the field of view, whether or not the part is rotated within a 360-degree range, the vision sensor can make a judgment. The problem of inoperability is thus widely welcomed by the industrial manufacturing community.

Visual sensing technology includes 3D visual sensing technology, 3D visual sensor has a wide range of uses, such as multimedia mobile phones, network cameras, digital cameras, robot vision navigation, automotive security systems, biomedical pixel analysis, human-machine interface, virtual reality, surveillance , industrial testing, wireless long-distance sensing, microscope technology, astronomical observation, marine autonomous navigation, scientific instruments, etc. These various applications are based on 3D vision image sensor technology. In particular, 3D imaging technology has urgent applications in industrial control and autonomous vehicle navigation.

Intelligent vision sensing technology is also a kind of visual sensing technology. The intelligent vision sensor under the intelligent vision sensing technology, also known as the smart camera, is a new technology developing the fastest in the field of machine vision in recent years. A smart camera is a small machine vision system with image acquisition, image processing and information transfer functions, and is an embedded computer vision system. It integrates image sensors, digital processors, communication modules and other peripherals into a single camera, reducing system complexity and improving reliability due to this all-in-one design. At the same time, the size of the system is greatly reduced, which broadens the application field of vision technology.

The advantages of intelligent vision sensors are easy to learn, use, maintain, and install, and can build a reliable and effective visual inspection system in a short period of time, which has led to the rapid development of this technology. The image acquisition unit of the vision sensor is mainly composed of a CCD/CMOS camera, an optical system, an illumination system and an image acquisition card, which converts the optical image into a digital image and transmits it to the image processing unit.

Application of Vision Sensing Technology

1. Vehicle body visual inspection system

Body forming is one of the key processes in automobile manufacturing. It has strict requirements on various indicators of the body, and requires 100% inspection of the body. The traditional method of car body inspection is to use a three-coordinate measuring machine, which is complicated in operation, slow in speed and long in construction period, and can only be checked by spot. Usually, the key dimensions of the body are mainly the size of the windshield, the position of the edge where the door is installed, and the position of the positioning hole. Therefore, vision sensors are distributed near these locations to measure the spatial location dimensions of their corresponding edges, holes, and surfaces. A measurement station is designed on the production line. After the body is positioned, it is placed in a frame. The frame is composed of metal columns and rods distributed vertically and horizontally. Vision sensors can be flexibly installed on the frame as required. According to the number of measurement points, a corresponding number of vision sensors can be installed, (usually each vision sensor measures a measured point), according to different forms of sensors, including binocular stereo vision sensors, contour sensors and other types.

The working process of the measurement system is as follows: the body is transported from the production line to the measurement station for accurate positioning, and then the sensors start to work in the required order, the computer collects and processes the images of the detection points, and calculates the three-dimensional coordinates of the measured points. The calculated value and the standard value By comparison, the test results are obtained, and the body is sent to the measuring station.

2. Online visual measurement system of section size

In industrial production, seamless steel pipe is an important industrial product, and its quality parameters are important data for manufacturing. The straightness and cross-sectional area of ​​the steel pipe are the main geometric parameters, which control the manufacturing quality of seamless steel pipes. critical, but the measurement of the parameter is difficult for the following reasons:

1. The seamless steel pipe adopts non-contact measurement, and the manufacturing site environment is harsh;

2. The space size of seamless steel pipes is large, which also requires the detection system to have a large measurement space. The emergence of visual sensing technology solves the above problems. Visual sensing technology adopts non-contact measurement and has a large measurement range.

The measurement system consists of multiple structured light sensors. The light plane projected by the structured light projector on the sensor intersects the steel pipe under test to obtain a partial arc on the circumference of the steel pipe section. The sensor measures the position of the partial arc in space. Each sensor in the system realizes the measurement of a partial arc on a section. Through appropriate mathematical methods, the section size and the spatial position of the section center are obtained by arc fitting, and the straightness parameter is obtained from the spatial envelope of the section center distribution. Under the control of the computer, the measurement system can complete the measurement within a few seconds to meet the real-time requirements.

3. Three-dimensional topography visual measurement

The three-dimensional topography digital measurement technology is the basic support technology for reverse engineering and product digital design, management and manufacturing. It realizes the mechanism of 3D topography digital measurement by combining visual non-contact, fast measurement and the latest high-resolution digital imaging technology. Since the objects to be measured are mostly large objects with complex surfaces, the measurement is usually divided into two parts: local three-dimensional information acquisition and overall splicing. First, the visual scanning sensor is used to measure each local area of ​​the measured topography, and then the splicing technology is used to The topography of each part is stitched together to obtain a complete image.

The sensor’s vision scanning probe is designed using the principle of local binocular stereo vision measurement. The overall splicing of the topography is essentially to put the collected data on the common coordinates, so that the overall data description can be obtained. The high-resolution digital camera is used to collect data from the top of the measurement space at different angles and positions. The spatial coordinates of the control points are obtained by using the principle of beam orientation intersection adjustment and a global coordinate system is established. Finally, each coordinate system is used to associate, Convert to complete data stitching.

How to choose a vision sensor?

At present, how to choose a machine vision sensor is more and more widely used in contemporary applications. How to choose a machine vision sensor is worth learning. Now we will have a deep understanding of how to choose a machine vision sensor. The camera is the eyes of the machine vision system, and the heart of the camera is the image sensor. Sensor selection depends on accuracy, output, sensitivity, cost of the machine vision system, and a good understanding of the application requirements. A basic understanding of the sensor’s main properties can help developers quickly narrow their search to the right sensor.

Most users of machine vision systems recognize that the camera is a key element of the system, often thinking of it as the “chip” of the vision system. The camera itself is a complex system: including the lens, signal processor, communication interface, and the most core part – the device that converts photons into electrons: the image sensor. The lens and other components work together to support the function of the camera, and the sensor ultimately determines the highest performance of the camera.

Much of the discussion in the industry has focused on process technology and the relative merits of CMOS and CCD sensors. Both technologies have their advantages and disadvantages, and the processed sensors have different performances. What the end user cares about is not “how” the sensor is made, but how it will perform in the end application.

In a given application, three key elements determine sensor selection: dynamic range, speed, and responsiveness. Dynamic range determines the quality of the image the system is able to capture, also known as the ability to capture detail. Sensor speed refers to how many images the sensor can produce per second and the amount of image output the system can receive. Responsiveness refers to how efficiently the sensor converts photons into electrons, and it determines the level of brightness the system needs to capture a useful image. The technology and design of the sensor together determine the above characteristics, so system developers must have their own criteria when selecting a sensor, and a detailed study of these characteristics will help make a sound judgment.

The Links:   NL6448BC33-10 RM500UZ-M

Related Posts