Aufrufe
vor 1 Jahr

Automation Technologies 4/2016

  • Text
  • Automation
  • Industrial
  • Manufacturing
  • Thermal
  • Technologies
  • Sensors
  • Products
  • Engineering
  • Imaging
  • Measurement
Automation Technologies 4/2016

02 Thermal imaging

02 Thermal imaging cameras allow you to 'see' the heat MULTIMEDIA CONTENT External Video: Outstanding infrared performance http://bit.ly/2aUFhuv MACHINE VISION Therefore, the spot size calculates as IFOV (in mrad) divided by 1000 and multiplied by the distance to the target: ⎛IFOV ⎞ Spot Size = ⎜ ⎟ x Distance to target (1) ⎝ 1000 ⎠ Where the spot size and distance to target are expressed in cm and IFOV in mrad. For a distance of 100 cm and an IFOV of 1 mrad the spot size would be 0.1 cm. If a spot size of 0.1 cm can be measured at a distance of 100 cm, then the distance at which a spot size of 1 cm can be measured is 1 000 cm. This means that the spot size ratio is 1:1 000. If we put the above calculations into a formula to express the SSR as 1 : x, with 1 standing for the spot size and x for the distance, then the formula for x would be as follows (the IFOV is expressed in mrad): Ideal and real optics Using the formula you can calculate that a camera with an IFOV of 1.4 mrad will have a theoretical SSR of 1:714, so in theory you should be able to measure an object of 1 cm in diameter at a distance of more than 7 meters. However, as was stated previously this theoretical value does not correspond to real life situation, because it does not take into account the fact that real life optics are never completely perfect. The lens that projects the infrared radiation onto the detector can cause dispersion and other forms of optical aberration. You can never be sure that your target is exactly projected onto one single detector element. Projected infrared radiation can also ‘spill over’ from neighbor detector elements. In other words: the temperature of the surfaces surrounding the target might influence the temperature reading. Just like with a spot pyrometer, where the target should not only cover the spot size entirely but should also cover a safety margin around the spot size, it is advisable to employ a safety margin when using a microbolometer thermal imaging camera for temperature measurements. This safety margin is captured in the term Measurement Field of View (MFOV). The MFOV describes the real measurement spot size of a thermal camera, in other words: the smallest measureable area for correct temperature readings. It is usually expressed as a multitude of the IFOV, the field of view of a single pixel. A commonly used guideline for microbolometer cameras is that the target needs to cover an area at least 3 times the IFOV to take into account optical aberrations. This means that in the thermal image the target should not only cover one pixel, which in an ideal situation would have been sufficient for the measurement, but also the pixels around it. Detect temperatures from a distance Even when the factor of ideal versus realistic optics is taken into account the difference between thermal imaging cameras and spot pyrometers in measuring distance is huge. Most spot pyrometers cannot be held any farther away than 10 to 50 cm, assuming a 1 cm target. Most thermal imaging cameras can accurately measure the temperature of a target of this size (1 cm) from several meters away. Even the Flir E40 thermal imaging camera, with an IFOV of 2.72 mrad, can measure the temperature of a spot of this size (1 cm) from more than 120 centimeters away. The Flir T1030sc thermal imaging camera, one of Flir’s most advanced models for industrial inspections, can measure temperature of a target of this size at a distance of more than seven meters with a standard 28 ° lens. These values are calculated assuming that the standard lens is used. Many of the more advanced thermal imaging cameras feature interchangeable lenses. When a different lens is used this changes the IFOV, which in turn affects the spot size ratio. For the Flir T1030sc thermal imaging camera, for instance, Flir not only offers the standard 28 ° lens, but also a 12 ° telephoto lens. With this lens, the IFOV of a Flir T1030sc thermal imaging camera is 0.20 mrad. With this lens, the same thermal imaging camera can accurately measure the temperature of the same size target from a distance of almost 17 meters. See whether you need to move in closer Thermal imaging cameras clearly outperform spot pyrometers when it comes to the SSR values, but the SSR values only refer to the distance from which an accurate temperature measurement can be made. In real life detecting a hot spot does not always require an accurate temperature reading. The hot spot can be discernible in the thermal image even when the target covers only one pixel in the thermal image. The temperature reading might not be perfect, but the hot spot is detected and the operator can move closer to make sure that the target covers more pixels in the thermal image, ensuring that the temperature reading is correct. Spot pyrometers also have challenges with measuring temperature on small objects. This capability is increasingly important for electronics inspection. Conclusion As devices continue to get faster in processing speed, yet are required to fit into smaller packages, finding ways to dissipate the heat and identify hot spots is a real problem. A temperature gun can effectively detect and measure temperature, but its spot size is simply too large. However, thermal cameras with close-up optics can focus down to less than 5 μm per pixel spot size. This allows engineers and technicians to make measurements on a very small scale. www.flir.com AUTOMATION TECHNOLOGIES 4/2016

Visions for the future Whether on land, in water or in air, industrial image processing has managed to reach virtually every sector imaginable. Nevertheless, the manufacturers in this industry segment are largely unknown to the public, staying behind the scenes as the provider of the technology in the final products. One such “hidden champion” is Matrix Vision. When Gerhard Thullner and Werner Armingeon set up the company on June 24 th 1986, little did they know that 30 years later they would have close to 100 employees and a depth and breadth of technological expertise that is unusual in the industry. The young company did not begin with image processing directly. Rather, at the start it developed software for atomic absorption spectrometers. In the early years of the PC, Atari products were very popular with early adopters like universities, and among users with graphicsrelated tasks. Thullner and Armingeon had their first business ventures in this field and developed a graphic controller for largescreen Atari computers. Matrix Vision became the global market leader with the Atari graphic controller. They briefly debated whether to apply their graphic controller expertise to the booming PC market. However, there were already established companies in the PC market, so that the company started developing frame grabbers for industrial applications. With these electronic components that are used to digitize analog video signals, the company had entered the still-young image processing sector. From smart cameras to letter sorting systems This move opened up a new area for Armingeon who felt that the standard image processing solution, comprising a frame grabber, camera and PC, was too complex and unreliable. By integrating all the components into a single product, the smart camera was born. This idea became the foundation for many solutions based on smart cameras, such as traffic flow surveillance in Great Britain; smart sewing machines, glass lens grinding machines and letter sorting systems in France. More than 210 camera versions Once interfaces such as USB and Ethernet had become established on the market, standard cameras joined frame grabbers in the company’s standard product line in 2004. Today, Matrix Vision has a portfolio of over 210 camera versions, which have secured a firm position on the market thanks to the special features they offer. For example, the FPGA used in all the cameras performs a whole range of processing jobs, thereby reducing the load on the host system. The internal image memories ensure reliable image transmission without any data loss. All of these features are requirements that have become increasingly important in recent years with regard to green automation and continuous process monitoring. Image processing in agricultural applications Everyone’s going organic. The demand for organic produce has increased enormously in recent years. Many conventional methods, such as the use of pesticides and herbicides, are not permitted in the cultivation and production of organic products. For cost reasons, however, manual labor - such as in the case of weeding - is not an option. A French startup company uses Matrix Vision cameras for autonomous weeding robots. The cameras give the robots three-dimensional vision so they can make their way through the rows of plants on their own. New innovations In 2015 the company received an innovation award for the innovative mvBlueSirius 6D industrial camera which, in addition to capturing static 3D data, also perceives the movement and color of objects in a room. Marking its 30 th anniversary, the company achieved its next masterstroke in early 2016 with the mvBlueGemini smart camera. Featuring mvImpact Configuration Studio software, the smart camera has become an easy-to-use but powerful component for automation. The software opens up the world of image processing both to users without programming knowledge and developers without image processing knowledge. In the context of Industry 4.0, in which industrial image processing plays a key role, smart cameras can now be easily integrated into an automation solution as complex programming tasks are now a thing of the past. This is very much in keeping with the company’s slogan celebrating 30 years of Matrix Vision – “We Change Your Vision”. Photographs: Matrix Vision www.matrix-vision.com About Matrix Vision Matrix Vision develops components and customised solutions for industrial image processing and is one of the leading companies in this field. The focus is on the distribution of digital and intelligent cameras for various sectors of manufacturing and non-manufacturing industries. Based on 30 years of experience as a vision technology pioneer and currently backed by almost 100 employees, Matrix Vision is shaping the industrial image processing future as an active member of the GigE-Vision, the USB3 Vision and the GenICam standard commitees. AUTOMATION TECHNOLOGIES 4/2016