Machine vision inspection technology is usually classified according to its application, principle or method. The following are common classifications:

1. Classification by Imaging Technology
2D machine vision: Utilizes a single industrial camera to capture two-dimensional images of a product for inspection and analysis. Suitable for most traditional inspection tasks, it primarily targets dimensional measurement, defect detection, appearance inspection, and character recognition for flat objects.
3D machine vision: Utilizes technologies such as laser triangulation, structured light projection, and binocular vision to acquire three-dimensional information (depth, height, and shape) about an object. It is commonly used for more complex dimensional measurement, volume calculation, and 3D reconstruction of objects.
Spectral imaging (hyperspectral/multispectral): Utilizes specific wavelengths of light to detect material properties that are invisible to the human eye, such as for composition analysis and agricultural product maturity testing.
Thermal imaging (infrared imaging): Utilizes infrared cameras to detect the heat distribution of an object. It is commonly used for circuit board defect detection and monitoring of heat-generating components.

2. Classification by Function
Appearance Inspection: Inspects the appearance quality of an object, such as surface cracks, stains, and deformation.
Dimensional Measurement: Used to accurately measure the size, position, and shape of an object, commonly found on industrial production lines.
Defect Detection: Automatically identifies various defects, including cracks, holes, and deformation.
Classification and Recognition: Utilizes deep learning and other technologies to classify and identify objects, commonly found in applications such as facial recognition and object recognition.

3. Classification by Technical Method
Traditional image processing: Image analysis and processing based on traditional methods such as edge detection, template matching, and color analysis.
AI-based detection: This utilizes algorithms such as deep learning and convolutional neural networks (CNNs) for more complex pattern recognition and classification, particularly suited for complex image recognition tasks.