With competitive price and timely delivery, Getian sincerely hope to be your supplier and partner.
Six Essential Considerations for
Machine Vision Lighting
Image processing software generates intelligence from data captured by the camera sensor. Even the most sophisticated software cannot deliver information that is not available in the original image, which means that the quality of system illumination is crucial. This white paper describes the six essential considerations for enhancing machine vision lighting.
Free 6-page white paper
The image contrast and signal-to-noise ratio are key to producing an optimum image and these two measures are determined by the amount and quality of light reaching the camera. To obtain a good, consistent image the light must be both sufficiently bright and stable. Variable light intensity can result in poor measurement repeatability with major repercussions in a manufacturing environment, leading to excessive waste, defective product reaching the end user or even line stoppages.
LED lighting is often specified by voltage but the brightness is actually determined by the current through the LEDs and not the supply voltage. This means that accurate control of light intensity requires good current control. LEDs are remarkably reliable devices but the light output reduces as the device ages and, in shorter timescale, the increase in temperature after the light is switched on will change the brightness. Switching the light on only when an image is being acquired by pulsing the light minimises the amount of heat produced and significantly improves LED lifetime. Strobing an LED light also brings the major benefit of allowing the light to be &#;overdriven&#; to intensities much higher than the continuous light rating.
Find out more
It is common for machine vision requirements to change. To guard against the unexpected, system flexibility should be built in so that existing systems can be adapted to changing requirements and environment. The inclusion of variable lighting control allows the intensity to be adjusted to suit an altered application so that the same vision system can be used with different light settings for different product batches. Making use of overdrive offers even more flexibility because, with an appropriate lighting controller, LEDs can safely be overdriven at up to 10x the published maximum. A common occurrence is where an established system becomes adversely affected by extraneous light sources. If the light is already running at maximum rating in continuous mode, then overdrive enables increased brightness to significantly reduce ambient light interference.
LED brightness is proportional to the current running through the device. In fact, a very small change in LED drive voltage will result in a much larger change in light output intensity. This is critical since even for simple vision calliper measurements a 10% change in light level can cause a change of 0.5% in the measured values. All Gardasoft LED controllers regulate the LED current to produce a stable, tightly controlled and highly repeatable light output.
Download white paper now
A well-designed lighting controller can bring significant benefits to the setup and use of machine vision systems. Opto-isolated trigger inputs can connect to all common signal sources and a digital, button front panel allows easy and instant configuration of lighting and a controller should incur minimal delay between the trigger signal and the light pulse. Ethernet compatibility with inbuilt web pages can provide live, measured values of current and voltage to determine the performance of the light.
GigE Vision and GenICam compatibility bring enhanced connectivity between components from different manufacturers and enable the controller to be discoverable on the network. Ethernet communications provides remote access to all the lighting within a system to indicate whether a light is connected, disconnected, or short-circuit. It also offers the possibility of remote troubleshooting of a vision system and allowing the configuration of a lighting controller to be changed to suit different batches of product.
An application such as the Gardasoft Vision Utility allows the machine vision control software to select a batch type and directly communicate with the controller to set the configuration required for that batch.
Download Free paper
Accurate and consistent lighting control has a massive influence on the quality of the captured images and will ensure that machine vision software receives all the data that is potentially available.
If you would like advice on your specific application please contact Gardasoft or .
Gardasoft sends occasional newsletters with useful information on many applications. If you would like to receive these please subscribe to our newsletters. You can unsubscribe at any time.
Are you interested in learning more about IR LEDs for Machine Vision? Contact us today to secure an expert consultation!
Essential Considerations on Machine Vision Lighting
Proper lighting is crucial to the success of machine vision applications. Without it, images will not show enough visual information about the important features of an item.
Common machine vision applications in which illumination is part of the solution involve inspecting products for defects before they leave the manufacturing plant or reading barcodes or other alphanumeric data on products or packages as they move through the logistics pipeline.
As engineers working in imaging and vision know, designing a reliable lighting solution is not a simple proposition.
There are thousands of lighting options to choose from including those that are halogen, incandescent, fluorescent, compact-fluorescent, plasma-type, high-intensity discharge (HID), Xenon, light-emitting-diode (LED), fiber optic, and structured light.
Of these, LEDs are the most common choice for machine vision applications because they have long life spans, can be left on continuously, and come in many sizes, shapes and colors. LED lighting choices also span the visible, ultraviolet (UV), narrow-band and broad-band visible and infrared (IR) spectrum.
To choose from so many options, it is important to consider the characteristics of the object to be imaged, such as size, shape and color; the spatial relationship between the object, light and camera; and how the light is reflected or absorbed by the object and background. (Vision Systems Design offers more explanation here.)
With so much to consider, missteps abound. To get insights on how to avoid, or solve, common pitfalls in illumination planning and execution, Vision Systems Design asked experts at a handful of lighting companies to share their thoughts. We&#;ve included their comments here.
Numerous experts agreed on the most common pitfall: Waiting until the tail end of the design process to address lighting for a machine vision application.
As Brett Thrailkill, application support and inside sales representative at Advanced Illumination (Rochester, VT, USA), explains, &#;Unfortunately, many users have the misconception that proper machine vision lighting doesn&#;t need to be considered until the end of the system design process. This often leads to cost overruns, inferior lighting, and delays in production timelines. To prevent this, we encourage all application engineers to consider lighting options early in the development process.&#;
Lindsey Sullivan, technical marketing manager at CCS America (Woburn, MA, USA), agrees, saying, &#;It is important to test and solve the application at the beginning of the design process. Once the required lighting is understood, the design of the inspection system can account for the positioning of the sample, size, working distance, color (and) geometry of the light.&#;
Light can be reflected or absorbed differently, depending on the color of the light and the object being imaged.
In general, according to Advanced Illumination, a light that is of a similar color as an object will reflect and thus brighten the object while a light on the opposite side of the color wheel will darken the object.
Sometimes the best choice is to darken the object&#;particularly if the goal is to read printed codes or words. By illuminating red print on a white envelope with a green light, you&#;ll get the clearest image of the print, Advanced Illumination says on its website.
However, brightness would be an advantage if the goal is to brighten or enhance the contrast between an object and its background, such as to detect wrapped candy on a conveyor belt.
This is why engineers choose white lights, which comprise both warm and cool colors, for many applications, and especially those involving a color camera. Colored lights are used with monochrome cameras.
David Hardy, vice president and design engineer for Spectrum Illumination (Montague, MI, USA), says that white lights also provide the most flexibility over time if a product&#;s color changes in a future iteration. &#;For example, the part under inspection is red or white, and a red light is chosen. But if a part color of green or blue is added later, it can appear quite dark as green and blue colors are somewhat opposite red on the color wheel. In this case, using a white light initially would have been better, or even an infrared light to remove the color.&#;
Simon Stanley, director of technology at ProPhotonix (Salem, NH, USA) also provides an example of how a customer in the medical device industry addressed the issue of color. &#;The customer was inspecting data matrix codes on different colored capsules. They were limited to using a high-resolution, but monochrome, Cognex In-Sight camera in line-scan mode; and a white line light and found that for some of the colored capsules, the images did not have adequate contrast.&#;
To solve the problem, he says ProPhotonix sold the company the RGB-White COBRA MultiSpec LED, which has individual color and intensity controls that engineers access through a graphical user interface (GUI). This allows them to &#;tune the light to optimize contrast across the whole grayscale image resulting in accurate image acquisition regardless of the color of the capsule.&#;
The line light has up to 12 different wavelength outputs, ranging from 365 nm- m.
(More information from Vision Systems Design on addressing lighting challenges in medical device and pharmaceutical inspection is available here.)
Want more information on UV LEDs for 3D Printing? Feel free to contact us.