Robot vision systems can be divided into the following structures:
1. Optical imaging module.
This module can be further divided into two parts: lighting system design and lens optical system design.
Lighting design involves studying the optical characteristics, distance, size, and background characteristics of the object being measured, and then rationally designing the intensity, color, uniformity, structure, and size of the light source, as well as designing a reasonable optical path, to obtain relevant structural information about the target.
The lens is the main component that projects spatial information from the object space onto the image space. Lens design primarily involves selecting the focal length and aperture range based on the lighting conditions and target characteristics. After determining the lens model, the rear-end mounting structure of the lens is designed.
2. Image sensor module
This module is primarily responsible for the photoelectric conversion of information and is located on the image plane at the rear of the lens. Currently, mainstream image sensors can be divided into two categories: CCD (Charge-coupled Device) and CMOS image sensors. Because it is the source of electrical signals, a reliable and stable circuit drive is crucial for the design of this module.
3. Image processing module
This module is mainly responsible for image processing and information parameter extraction, and can be divided into two levels: hardware structure and software algorithm.
The hardware layer is generally a CPU-centric circuit system. PC-based machine vision uses the PC's CPU and related peripherals; embedded system-based intelligent cameras with independent data processing capabilities rely on onboard information processing chips such as DSPs, ARMs, and FPGAs.
The software component includes a complete image processing and decision-making scheme, comprising a series of algorithms. Advanced image systems integrate data algorithm libraries to facilitate system portability and reuse. When the algorithm library is large, it is invoked through a graphical interface.
4. I/O module
The I/O module is responsible for outputting the computational results and data of the machine vision system. PC-based machine vision systems can be divided into internal and external interfaces. The internal interface is only responsible for transmitting signals to the PC's high-speed communication port, while the external interface handles communication and information exchange with other systems or users. Smart cameras typically utilize general-purpose I/O and high-speed Ethernet to perform all their functions.
5. Display module
The display module can be considered a special type of user I/O, allowing users to more intuitively observe the system's operation. In PC-based machine vision systems, system data can be directly transmitted to the graphics card via the PCI bus and then displayed on the computer screen via the VGA interface. Independently processed smart cameras typically achieve image visualization through an extended LCD screen and an image display control chip.
