Machine vision camera selection
First, clarify the requirements.
First, the accuracy requirements of the tested products must be determined.
Second, you need to determine the size of the field of view that the camera needs to see.
Third, the speed of the object being detected needs to be determined.
Fourth, it is necessary to determine whether it is dynamic or static detection.
Then determine the hardware type.
1. Determining the pixel size of the image plane
Currently, most software on the market offers near-zero precision, often referred to as sub-pixel accuracy. However, while the software is error-free, hardware errors are unavoidable. Therefore, most machine vision systems on the market guarantee an error of one pixel. This leads to the following calculation formula:
Accuracy = Field of view (length or width) ÷ Camera pixels (length or width)
With the above formulas and our first and second requirements, it's not difficult to determine the camera's pixel size.
Assuming a field of view of 10mm and a required accuracy of 0.02mm, then the camera's pixel count would be 10 ÷ 0.02 = 500 pixels. Therefore, a camera with 300,000 pixels (640*480) would be sufficient.
2. Determining the Camera Transmission Method
Currently available camera transmission methods include:
1) Analog camera (PCI acquisition card) can be selected if speed requirements are not high.
Advantages: Stable, cost-effective. Disadvantages: Low frame rate, generally only reaching 25-30 frames per second.
2) USB interface camera: This option is suitable for systems using only a single camera, and is also suitable for applications requiring high speed.
Advantages: Does not occupy a PCI slot, high frame rate, high cost-performance ratio. Disadvantages: Utilizes system CPU resources.
3) 1394 interface camera: When the system uses multiple cameras, the appropriate camera can be selected first; high speed is also a priority when selecting the appropriate camera.
Advantages: Does not consume system CPU resources, high frame rate; Disadvantages: Occupies a PCI slot, expensive.
3. Selection of camera triggering method
1) Continuous acquisition mode: Selectable for static detection; also selectable when the product is in continuous motion and cannot generate a trigger signal.
2) Software trigger mode: Selectable for dynamic detection; selectable for continuous product movement that can provide a trigger signal.
3) Hardware triggering mode: Selectable for high-speed dynamic detection; selectable for triggering signals provided by continuous high-speed product movement.
