X-Ray-Based Machine Vision – Part 2 - Applications in Industries Other Than Electronics

The first part of this article concentrated on reviewing applications of machine vision-based X-Ray systems in the electronic industry. In this part we will cover some of the applications in other industries.

Real-time X-Ray systems have been used for many years as an integral part of a non-destructive evaluation program where the risk of failure has been high. Thanks to lawyers the cost of these risks has escalated in recent years. Theoretically this should make the argument for adopting machine vision-based X-Ray systems even more compelling. Certainly this is the case for those failure modes associated with products that pose a danger to the user.

As suggested in Part 1 of this article, there are various degrees of 'machine vision-ness.' Many applications in industry still rely on real-time X-Ray systems. These are no more than closed-circuit TV systems where the image is derived from an X-Ray of the scene. In these cases, the operator makes the decision regarding integrity of the object being X-Rayed, so there is still substantial subjectivity in the decision process.

The next iteration is one that incorporates machine vision-based software to enhance the X-Ray image to reduce noise, increase the contrast separation between acceptable and unacceptable conditions and/or enhance edges. Some of these packages come with additional tools such as 'calipers' to aid an operator in making dimensional checks on artifacts such as solder voids. Some come with specific tool sets for specific objects/materials being inspected. Some of these systems are further 'automated' in that their stages can be programmed to follow a specific path or sequence to take multiple views of an object systematically.

The ultimate class of these X-Ray systems are those that incorporate machine vision-based automated defect detection capabilities as well as machine vision-based image processing and enhancement. These do not require any operator intervention to determine if the product or material being X-Rayed is acceptable or not. These systems have typically been used in an offline scenario but some are emerging that are fast enough to be used for many online applications as well.

Applications
Food Industry
Following the electronic industry, the industry that is the next largest user of machine vision-based X-Ray systems is the food industry. It is likely that every hamburger patty and chicken fillet served by a fast food franchise establishment has been X-Rayed to make sure there are no bones or other foreign objects. Systems can handle either cooked or raw products. Typically these systems inspect these products in bulk form, either delivered in slurry through a pipe or on a conveyor. In addition to the meat products cited, these bulk inspection systems are also being applied to inspect nuts, potatoes, cookies/crackers, salty snack foods, tobacco, dried vegetables, cereal, etc. Both metallic and non-metallic foreign objects can be detected: foil wrappers, stainless steel, glass, mineral stones, rubber, high-density plastics, etc. Sensitivity, that is size of foreign object detected, usually depends on speed, specific product and overall area or volume.

Many of the suppliers of these products have also adapted them to inspection of discrete objects, e.g. packages. In these instances they not only inspect for metallic and non-metallic foreign objects but they also can detect presence and completeness of contents in the package as well as incorrect counts, misshapen product, excessive settling. These systems can be used by fresh and frozen food, beverage, dairy, baked goods, snack foods, confectionary, pharmaceutical, health foods, liquids, soaps and sauces, loose powders and granular materials and seafood producing companies. These systems can inspect product in a variety of packages - glass, metal, plastic, paper. Companies also offer versions adapted for case inspection for presence/absence, etc.

In consumer goods packaging and beverage markets another application is fill height validation. An X-Ray-based system offers advantages over other high-energy detection approaches, such as gamma rays, because the X-Ray source can be turned off when not in use posing less of a hazard. These systems can detect underfill as well as overfill in steel, aluminum, glass, plastic and paper containers.

Table 1 depicts some of the companies offering machine vision-based X-Ray products for the North American consumable goods market.

Other Industries
In other industries associated more with durable goods manufacturing X-Ray-based machine vision systems are used wherever safety and reliability are an issue. This includes the inspection of castings used in the aerospace and automotive industries, tires for both automotives and aircraft, assemblies such as air bag assemblies used in the automotive industry. Among the more interesting applications are those in the wood industry where X-Ray-based machine vision systems are being used to grade wood based on density changes as well as to maximize the wood yield by identifying internal defects and their location. Table 2 depicts some of the companies in the North American market whose products have addressed these applications.

Questions to Ask Prospective Vendors of X-Ray-Based Machine Vision Systems
As you examine the products from different vendors you will find most make the same claims. It is clearly important to get them to put their claims in writing. The following questions are meant to provide the framework for a systematic analysis of the competitive landscape. The answers given should be consistent with the application requirements anticipated. This list is not meant to be complete. 

1. For which specific applications does your company offer products?
2. Regarding your system:
   1. Is your system an on-line or offline system? NOTE: If both, please answer the following questions for each style.
   2. What does the system do and what are the specs? 
   3. Size of anomalies system can reliably detect? 
   4. Do you have a recommended calibration procedure to demonstrate accuracy of the system or the reliability of detection you specify and, if so, what is it?
   5. Is system based on area camera or line scan camera?
   6. Can you describe the fundamental underlying principles for capturing defect image data?
   7. How long does it take to train the system on a new product? Or different model of the same product line?
   8. What is the changeover time where different products have been previously trained?
   9. What is the throughput at what specific pixel size or anomaly size?
   10. What is your false reject rate? Escape rate? How have these been demonstrated?
   11. Is there an action that takes place if there are 'x' number of consecutive rejects? 
   12. Does the system have the ability to archive detected anomalous conditions? If so how many can it archive?
   13. Is the system design based on your own proprietary hardware or commercially available products such as frame grabbers or vision processors or is it a host-based processing system?
   14. Do you offer an upgrade patch for future generation products?
   15. How many cameras does your system have and why? 
   16. What type of camera is used and why? (line scan, area scan, amorphous silicon)
   17. Does your system have the ability to adapt field-of-view/resolution as a function the model size being inspected?
   18. Does your system have Internet trouble-shooting compatibility?
   19. What is the price range of your systems?
   20. What options, if any, are offered for your system?

This checklist is meant to be an example and not necessarily comprehensive. For your application there are likely to be other questions that should be asked and whose answers should be the basis of determining the appropriate vendor. The key is to use a systematic approach when evaluating vendors, one that minimizes any bias that might creep in because of the 'smoothness' of the salesperson or a vendor's halo effect.