On the face of it, there are not many parallels between vehicle production and packaging machinery. However, the challenges are very similar, particularly in the increasing need to offer manufacturing versatility, dealing efficiently with different products. Mass production used to be a synonym for standardisation – “you can have any colour as long as it’s
It’s fairly well accepted that manufacturing processes now have to take sustainability into account along with increased productivity. Put simply, production machinery must offer minimum resource consumption along with maximum efficiency. And one of the key technologies that will help deliver this is image processing.
Energy consumption and the use of raw materials can be considerable during manufacturing, so it follows that if a defect is identified early in the production cycle, further processing can be stopped, reducing waste of both materials and energy. Defective parts can often be recycled in the production process, to reduce wastage still further.
One of the ways of delivering reduced costs and improved productivity is through industrial image processing, since if the technology is deployed effectively, it is possible to intervene in production processes at the earliest possible stage. Image processing also enable 100% quality control, seamless documentation and thus traceability of individual steps in the production process.
So why is image-processing suddenly the Next Big Thing? One reason is the development in recent years of true-colour processing. All 256 graduations of each RGB (red, green, blue) colour are processed directly, enabling more than 16 million colours to be detected.
Vision sensors and image processing systems can thus be set up to “read” a precisely defined colour – or, if appropriate, a narrow colour band – converting the colour data into digital RGB values. If, for example, a colour is registered as R=225, G=150, B=130, a different threshold can be determined for each of these values. Where a product shows any variations from this norm, it is rejected and, as appropriate, the production machine adjusted accordingly.
True-colour detection provides reliable on-the-fly inspection, even in difficult lighting conditions, shiny surfaces, or where there is low contrast between object and background. It’s particularly well suited to food processing and packaging, where decisions are often based on the colour characteristics of the product being made.
Ruling out uncertainties
One leading biscuit manufacturer, for example, uses image processing sensors in production conveyor belts to determine precisely when a biscuit is fully baked. In addition to information such as temperature and conveyor belt speed, the recipe for the biscuits includes RGB data from the image processing sensor. This allows closed-loop control to ensure the required baking quality, ruling out uncertainties associated with the subjective assessment by individual workers – it’s also much faster, so machines can operate at higher speeds.
The biscuit manufacturer has been able to reduce significantly its consumption of flour, sugar, dairy products, water and energy. This not only represents an improvement in business terms, but is also a contribution to environmental protection. The opportunities for other packaging and production processes are considerable: what’s required is creative thinking about ways of applying the technology.