Ultraviolet Lighting for Reasonable Use of Invisible Light Illumination to Improve Visual Inspection Efficiency

Machine vision is traditionally defined as a system that uses machines instead of human eyes to make various measurements and judgments, and has been very maturely applied to various fields of industry and national production. In machine vision applications, lighting is one of the very key technologies. By adjusting the information such as the shape, illumination angle, and brightness of the light source, it can significantly highlight the features we want to detect or measure, and shield out unwanted information. Interference to the system.

In the past, machine vision technology relied on many types of light sources to capture images, including fluorescent lamps, quartz halogen lamps, LEDs, metal halides (mercury), and xenon. Mature, are various types of visible light LED light sources. However, what we are discussing today are two invisible light lighting technologies that are being used in machine vision systems and solving many practical problems, namely ultraviolet and infrared lighting, in addition to visible light LEDs. Reasonable use of invisible light in some vision systems can greatly improve system efficiency. This article mainly introduces the application of UV imaging.

Ultraviolet (UV) is electromagnetic radiation with a wavelength of 10 ~ 400nm, divided into three different wavelength bands. 300 ~ 400nm is called near-ultraviolet band, and this band is divided into two sub-bands: UV-A (315 ~ 400nm) and UV-B (280 ~ 315nm). The UV-C band is below 300nm and covers a wavelength of 100 ~ 280nm. The ultraviolet wavelengths commonly used in machine vision are in the UV-A band, the more commonly used are 365nm and 395nm.

Ultraviolet light is commonly used in machine vision applications to detect features that cannot be detected with visible light. Because ultraviolet light is absorbed by many materials, it can capture images of the product surface, and because ultraviolet light has a shorter wavelength than visible light, it can be scattered by the surface characteristics of the product. UV lighting can be applied to machine vision systems in two different ways. In reflective UV imaging applications, an object is illuminated with UV light and the image is captured using a monochrome or color camera that is sensitive to UV light. In ultraviolet-fluorescence imaging, the surface of an object is irradiated with ultraviolet light. In coatings, plastics, printing inks, and dyes added with fluorescent whitening agents, these fluorescent materials will absorb ultraviolet radiation and then emit longer-wavelength fluorescence. . The difference between the absorption peak wavelength and the emission peak wavelength is called the Stokes shift, as shown in the figure below.

1527561958501034.jpg

The typical application case of using this technology is the detection of diaper sutures. The ordinary visible light camera cannot clearly show the difference between diapers and sutures. However, in the imaging of a 365nm UV light source with a filter, the characteristics of the sutures can be clearly seen. The quality inspection of the product is completed accordingly. The following figure shows the contrast between conventional shooting and shooting of diaper sutures using a Vita camera, lens, and a custom-made UV light source:

1527561973413304.jpg

Although UV-fluorescence imaging is commonly used in many fields, reflective UV imaging can also be used (not emitting fluorescence) in displaying product defects. In reflective UV imaging applications, the product under test is irradiated with UV light and the emitted UV light is captured. For example, in the application of detecting the presence of air bubbles on the product label, you can use UV lighting to highlight any air bubbles that may exist. The following figure shows the actual effect of detecting air bubbles on the label by using Vesvision image products with UV lighting:

1527561992337264.jpg

Currently, UV lighting has been deployed in many industrial inspection applications, both in fluorescence imaging mode and reflected UV imaging mode. Although UV lighting is still in its infancy, as the cost of UV LEDs decreases and developers integrate UV lighting, off-the-shelf machine vision hardware and image processing software in their production environments, this will lead to many new UV lighting applications .