The workflow of a machine vision inspection system is mainly divided into three parts: image information acquisition, image information processing, and electromechanical system execution of inspection results. In addition, parameters can be set and adjusted in real time through a human-machine interface as needed by the system.
When the object being detected moves to a predetermined position, it is detected by a position sensor. The position sensor sends an electrical pulse signal indicating "object detected" to the PLC controller. The PLC controller calculates when the object will move to the acquisition position of the CCD camera and then accurately sends a trigger signal to the image acquisition card. After acquiring this trigger signal, the CCD camera is immediately requested to acquire an image. The acquired image of the object is sent to the industrial control computer in BMP file format. Then, dedicated analysis software is used to analyze and process the image, determining whether the detected object meets the preset requirements. Based on the "qualified" or "unqualified" signal, the system will take corresponding actions to process the detected object. The system works in this loop, continuously processing the queue of detected objects. (See the diagram below.)
Working principle of machine vision inspection system
The main working process of a complete machine vision inspection system is as follows:
① The workpiece positioning sensor detects that the object being inspected has moved close to the center of the field of view of the machine vision camera system and sends a trigger pulse to the image acquisition unit of the machine vision inspection system.
②The image acquisition unit of the machine vision inspection system sends trigger pulses to the camera and lighting system respectively according to the pre-set program and delay.
③ The machine vision camera stops the current scan and starts a new frame scan, or the machine vision camera is in a waiting state before the trigger pulse arrives, and starts a frame scan after the trigger pulse arrives.
④ The electronic shutter is opened before the machine vision camera starts scanning a new frame, and the exposure time can be set in advance.
⑤ Another trigger pulse turns on the lighting, and the lighting time should match the exposure time of the machine vision camera.
⑥ After the machine vision camera is exposed, the scanning and output of a new frame of image officially begins.
⑦ The image acquisition unit of the machine vision inspection system receives analog video signals and digitizes them through an A/D converter, or directly receives the digital video signals digitized by the machine vision camera.
⑧ The processing results control the operation of the production line, perform positioning, and correct motion errors, etc.
As can be seen from the above workflow, machine vision inspection systems are relatively complex systems. Most monitored and inspected objects are moving, making the matching and coordination between the system and the moving objects particularly important. This places stringent requirements on the action time and processing speed of each part of the system. In certain application areas, such as robotics and guidance of flying objects, there are also strict requirements on the weight, volume, and power consumption of the entire system or parts thereof.
Although machine vision applications vary, they generally involve the following processes:
① Image acquisition: The optical system acquires images, converts them into digital format, and transfers them to the computer's memory.
② Image processing: The processor uses different algorithms to improve image factors that affect detection.
③ Feature extraction: The processor identifies and quantifies key features of the image, such as location, quantity, and area. This data is then transmitted to the control program.
④ Judgment and Control: The processor's control program makes conclusions based on the received data. For example, whether the position conforms to specifications, or how the actuator should move to pick up a certain part.
Machine vision inspection system
The image above shows a typical machine vision inspection system in engineering applications. On the production line, when a part reaches the trigger via the conveyor belt, the machine vision camera unit immediately turns on the illumination to capture an image of the part. The image data is then transmitted to the processor, which performs calculations based on pixel distribution, brightness, color, and other information to extract the target's features: area, length, quantity, position, etc. The processor then outputs results based on preset criteria, such as size, angle, offset, number, pass/fail, present/existing, etc. The system communicates with the PLC via a fieldbus to control the actuator (such as an air pump) to eject non-conforming products.
