Welcome to the website of Shenzhen Nuoxinde Technology Co., Ltd!

Shenzhen Nuoxinde Technology Co., Ltd

Home >Information >Company News
Understand the working principle of AOI optical detectors and improve product quality

Time:2022-12-06     Browse:1571

Working principle of aoi optical detector
The general process of aoi detection is the same, mostly through graphic recognition. Compare the standard digital images stored in the AOI system with the actual detected images to obtain the detection results. For example, when detecting a solder joint, a standard digital image is established based on a intact solder joint, and compared with the measured image. Whether the detection result passes or fails depends on the standard image, resolution, and the detection program used. Various algorithms are used in graphic recognition, such as finding the proportion of black to white, color, composition, averaging, summing, subtracting, finding planes, and finding edges and corners.

AOI's light irradiation includes two types of devices: white light and color light. White light uses 256 levels of grayscale, while color uses red, green, and blue light. The light shines on the surface of solder/component, and then reflects back to the lens, producing a three-dimensional display of a two-dimensional image to reflect the height and color difference of the solder/component. People see and recognize objects based on the amount of light reflected back, with most reflections being bright and less reflections being dark. The principle of AOI is the same as that of human judgment.

AOI has single lens and multiple lenses in terms of lens quantity, which is only a choice for technical solution implementation. It is difficult to say which method is necessarily good, because a single lens can also obtain good detection images through different angles of multiple light sources. Especially for lead-free welding, the surface is relatively rough, which can produce solder joints with different shapes, easily forming bubbles, and prone to the feature of one end of the part being tiltedAOI detection equipmentAdaptable hardware and algorithm updates have also been carried out.

Currently, PCB AOI detection algorithms can be roughly divided into three categories:

1. Design rule-based detection algorithm

2. Detection algorithm based on image processing

3. The combination of the above two

There are generally two types of logic for AOI reporting shortcomings: design specification (DRC) and motherboard (CAM reference) comparison.

design code

AOI determines based on the parameters you set that reporting parameter violations is a disadvantage.

The main drawbacks of this type are * small line width, * small spacing, pinholes, and copper slag.

Motherboard comparison

AOI compares the motherboard data learned from CAM data with the actual scanned images, and reports any defects that are not within the tolerance range.

The main drawbacks of this type are open circuits, short circuits, solder pads, and pinholes.

Integrate design specifications and motherboard comparison organically to report only the shortcomings you want to identify and minimize false points.

The above content refers to some online content, just to illustrate the working principle of AOI, provide a guide for quality control, reduce misjudgment, improve production efficiency, and improve product quality.