A brief discussion on edge detection in machine vision

Machine vision has the advantages of high speed, high precision, repeatability, objectivity, etc. After adding machine vision to automated equipment, its detection and assembly efficiency and accuracy will be significantly improved compared to manual work. The field of machine vision has now become one of the cutting-edge research hotspots. Edge detection is an essential part of machine vision and an important image preprocessing technology.

Since edges are the result of discontinuous grayscale values, this discontinuity can often be easily detected using derivatives. Generally, first-order and second-order derivatives are selected to detect edges. In machine vision inspection, this method is usually called edge detection local operator method. For the detection of image edges, the Canny algorithm is used to process and segment the image. The basic steps of the specific algorithm are as follows:

1. Filtering: The edge detection algorithm is mainly based on the first-order and second-order derivatives of the image intensity, but the calculation of the derivatives is very sensitive to noise, so filters must be used to improve the performance of the edge detector related to noise. It should be noted that most filters also cause a loss of edge strength while reducing noise, so there is a trade-off between edge enhancement and noise reduction.
2. Enhancement: The basis of edge enhancement is to determine the change value of the neighborhood intensity of each point in the image. The enhancement algorithm can highlight points with significant changes in neighborhood (or local) intensity values. Edge enhancement is generally accomplished by calculating gradient amplitude.
3. Detection: There are many points in the image with relatively large gradient amplitudes, and these points are not all edges in specific application fields, so some method should be used to determine which points are edge points. The simplest edge detection criterion is the gradient amplitude threshold criterion.
4. Positioning: If an application requires determining the edge position, the edge position can be estimated at sub-pixel resolution, and the edge orientation can also be estimated.

In edge detection algorithms, the first three steps are very commonly used. This is because in most cases, the edge detector is only required to point out that the edge appears near a certain pixel in the image, but there is no need to point out the precise location or direction of the edge.

These four steps are essential when using machine vision for dimensional measurement, especially the precise location and orientation of the edge. Machine vision inspection technology, with its powerful performance advantages, standardizes product quality, has fast inspection speed, reliable and stable inspection results and can be inspected for a long time, and is widely used in various fields.