Cameras to move towards zero-defect production

With this goal of increasing the quality of products and processes, production processes are constantly changing. The pressure related to time and costs encourages increased production rates and reduced rejects. A lot of energy is then devoted to the automation of the control steps.

Without mechanical assistance, man is unable to achieve these goals. With the increasing automation of quality assurance, necessary but time-consuming and cost-intensive laboratory tests can increasingly be replaced by “direct on line” product controls. Modern camera technology thus keeps an eye on the objects to be checked and their properties during production at any time and almost everywhere, without actively intervening in the process. Combined with image processing, the use of camera images offers many advantages and possibilities.

“Optical measurement and control techniques” allow quality controls to be carried out very quickly, directly during the manufacturing process, and absolutely wear-free. The replacement of simple barcode readers by industrial cameras makes it possible to determine and evaluate the content of the code, but also the quality of the code, alignment, color, size, environment, etc. This not only expands the possibilities of control, but also its quality while further simplifying the automation of quality assurance. The cameras can detect already existing errors early, even before subsequent errors occur, which then need little or no correction. However, what camera technology is capable of meeting the high demands of measurement and control technology, of enabling machines to best assist people, or even replace them, with their own decisions?

“Processing 3D images can significantly improve the flexibility and precision of measurement and control processes , ” says Dr. Martin Hennemann, Ensenso Product Manager at IDS Imaging Development Systems GmbH.

Current 3D camera technology allows robots to analyze their environment in order to respond to different events on their own. In addition to the dimensions and the position of the objects in the workshop, these data allow precise conclusions to be drawn about any deviations or imperfections compared to reference objects. It is then possible to carry out checks on the line indicating whether the “right” components of the product are assembled.

In the automotive industry in particular, where production includes many variations, there is a high potential for error. Customers can build the car of their dreams. The selection includes different tires, engine versions, chassis or interior equipment. For this reason, production personnel particularly benefit from automated systems, which trigger an alarm at the appropriate time, before a faulty or defective part is assembled. This reduces the need for subsequent checks, or worse yet, the risk of production downtime due to errors.

“The high degree of customization in vehicle configuration inevitably leads to a great variety of variants, which are difficult to understand and control without the aid of camera technology , ” emphasizes Martin Hennemann.

In April 2018, at the international trade fair for quality assurance “Control 2018” held in Stuttgart, IDS demonstrated an example of “3D object verification”. The demo system reconstructed objects with an Ensenso N35 3D camera and then performed a data comparison using Halcon image processing. The generated 3D data was also compared to a reference model in order to detect any discrepancies or imperfections. 3D image processing identifies anomalies or tiny deviations on test objects, which are not even visible to the human eye. According to IDS, the quality of the artifacts can be dramatically improved for subsequent process steps through high-precision scanning with Ensenso 3D cameras and subsequent image processing to check the objects.

For Martin Hennemann, 3D technology is a beneficial addition to 2D camera technology. Together, these technologies make it possible to solve more and more applications. Each of the two technologies has advantages for defined applications. 3D data is used to determine shapes, surface structures or the position of objects in space. 2D data, on the other hand, accurately detects outlines and colors. They are also able to read markings. In the example of the IDS demonstration, the 3D data is also used to compare objects. Existing part codes can also help with identification. They are perfectly decoded with 2D camera images. Car manufacturers also use control techniques with cameras combining 3D and 2D technologies during the final inspection, for example to look for deviations on an already assembled body and to check or measure them precisely according to prescriptions.

In general, camera technology offers many possibilities for moving towards zero defect production. To effectively limit production errors, each required control criterion can use one or the other technology, or both. Camera systems also support manual processes during manufacturing. They monitor manual sequences at workstation level. Image analysis checks whether the correct parts have been entered, whether the required number of screws matches or whether a part has been missed. A consecutive system indicates by an audible or visual warning on the monitors or by projection the correct or incorrect execution of the work. Quality control is thus guaranteed during manufacture.

Technologies related to cameras and production continue to develop. But new camera technology does not immediately replace existing technology. Young 3D technology, which has grown enormously in recent years, visualizes other characteristics of an object. It opens the way to new applications and complements, but does not replace, 2D camera technology. Cameras are also evolving from “simple” image providers to flexible and “intelligent” image processing solutions. Along with 2D and 3D cameras, machine vision systems, such as the IDS NXT platform, when they effectively combine cameras, sensors, intelligence, communication and adaptability, also greatly contribute to the prevention of errors in insurance. quality. A "perfect" production without rework or defect is today like yesterday an ideal representation impossible to implement, even with a lot of money and the most modern technologies. The use of camera and analysis technology has, however, become essential for anyone who wants to quickly approach the goal of zero-defect production.

Find out more: www.ids-imaging.com