Five typical architectures of machine vision

What is machine vision ? What are its five typical architectures? The editor will explain it to you below:

1. Lighting
I still remember that when I attended a conference on machine vision technology, a technical expert specifically emphasized the importance of lighting in machine vision. It is understood that lighting is an important factor affecting the input of the machine vision system, which directly affects the quality of the input data and the application effect.
Light sources can be divided into visible light and invisible light. Several commonly used visible light sources are incandescent lamps, fluorescent lamps, mercury lamps and sodium lamps. The disadvantage of visible light is that light energy cannot remain stable. How to make light energy stable to a certain extent is an urgent problem that needs to be solved in practice. On the other hand, ambient light may affect the quality of the image, so a protective screen can be used to reduce the impact of ambient light.
Lighting systems can be divided into back lighting, forward lighting, structured light and strobe lighting according to their illumination methods. Among them, back-illumination places the object under test between the light source and the camera. Its advantage is that it can obtain high-contrast images. Forward illumination means that the light source and camera are located on the same side of the object being measured, which facilitates installation. Structured light illumination projects gratings or linear light sources onto the object to be measured, and demodulates the three-dimensional information of the object under test based on the distortion they produce. Strobe lighting irradiates high-frequency light pulses onto objects, and camera shooting requires synchronization with the light source.
2. Lens
The lens is equivalent to the human eyeball. It is mainly responsible for beam modulation and signal transmission in the machine vision system. It is understood that most lenses on the market so far can meet the needs of machine vision applications, but more professional machine vision systems may require customized lenses and coatings. Fortunately, many lens manufacturers with in-house production capabilities are ready to customize lenses to meet the needs of these applications. Of course, these custom lenses can be expensive, so they are usually only used in specialized imaging systems that are not cost-sensitive (such as military applications), or in production lines for high-volume consumer products.
3. Industrial cameras
The most essential function of industrial cameras in machine vision systems is to convert optical signals into electrical signals. Compared with ordinary cameras, it has higher transmission power, anti-interference ability and stable imaging capabilities. There are many classifications according to different standards: according to the output signal mode, it can be divided into analog industrial cameras and digital industrial cameras; according to different chip types, it can be divided into CCD industrial cameras and CMOS industrial cameras.
4. Image capture card
Although the image capture card is only a component of the complete machine vision system, it plays a very important role and directly determines the camera interface: black and white, color, analog, digital, etc. Typical ones include PCI capture cards, 1394 capture cards, VGA capture cards and GigE Gigabit network capture cards. Some of these capture cards have built-in multi-channel switches that can connect multiple cameras and capture multiple channels of information at the same time.
5. Machine vision software
Machine vision software is a key component of automated processing in the machine vision system. According to specific application requirements, the software package is re-developed to automatically complete image collection, display, storage and processing. Through early transmission to dedicated image processing software, it is converted into digital signals based on pixel distribution, brightness, color and other information; the machine vision software then performs various operations on these signals to extract the characteristics of the target, and then controls the scene based on the results of the discrimination. device action.