HALCON uses a classification approach for optical character recognition which requires segmentation of individual characters from their background in an image and then reading each character using a pre-trained classifier. HALCON includes many pre-trained fonts for classification, but also enables users to train their own custom fonts. The OCR assistant in HALCON 11’s HDevelop IDE provides developers with a graphical user interface tool designed to automatically determine or optimize the appropriate parameters for character segmentation, test and optimize different classifiers for desired results, train custom fonts, and automatically generate HALCON code for OCR applications.
As seen in the image , the OCR assistant provides a quick setup tool for rapidly configuring an OCR solution. To use the tool, users simply load or acquire a sample image then highlight the image region containing the characters, define the characters that are in the region, and describe basic information about their appearance in the image. The setup tool then automatically determines parameter settings for segmenting and classifying the characters.
Since the quick setup tool only uses basic image information, further optimization may be needed for more challenging applications. The OCR assistant supports this by providing a graphical interface for manually selecting the best parameters for segmentation. Users can adjust settings like character appearance, size, shape features, orientation range, fragmentation, and layout (number of lines of text, expected number of characters, etc). Results of the parameter settings are displayed in real-time on the sample image.
Once the proper segmentation parameters have been selected, the OCR assistant allows users to choose an effective classifier from either a pre-trained HALCON classifier, a classifier previously trained by the user, or the ability to train a new classifier.
Training a new classifier is easy thanks to the powerful teaching tool in the OCR assistant. The teaching tool allows users to define the correct character for each of the segmented characters in the image. This can be done across several sample images to increase the amount of training data and improve the robustness of the classifier.
Additionally, the OCR assistant provides a training file browser window that enables users to view all the images used for training, edit their designated character definitions and add or remove training data. This is a very efficient way to ensure that no characters are trained incorrectly. The browser also allows users to create additional training data by automatically generating modified versions of the segmented character images. These modifications can include noise, rotation, slant, stroke width, radial deformation, and local deformation.
To further improve performance, the OCR assistant allows users to define or restrict the possible results based on features like known syntax or expressions, and a dictionary file of allowable words or strings.
Finally, once the desired performance is achieved, the OCR assistant can automatically generate and insert the corresponding HALCON code into the HDev program.
Additional information on using the OCR assistant can be found in section 7.5 of the HDevelop User’s Guide available on the MVTec website.
Because process plants and packaging lines are required to be ever more resourceful with their continuous improvement programmes, the trend towards automation is now almost omnipresent. However, one of the common obstacles encountered when looking to implement automation is a distinct lack of space.
“Size was a major factor when we were sourcing a suitable infrared thermal imaging camera to partner with HALCON software,” states Multipix Imaging director, Julie Busby. “However, not only is the FLIR A5 extremely compact and good value for money, it also offers the plug-and-play compatibility of a GigE Vision interface, which means we can link directly to HALCON without any need for a separate frame grabber.”
GigE Vision is a well established camera interface standard developed using the Gigabit Ethernet communication protocol. It is the first industrial standard to allow for fast image transfer using low cost standard cables, even over long distances. With GigE Vision, hardware and software from different vendors can interoperate seamlessly over GigE connections.
Automated infrared radiation imaging offers a vast potential for improving a host of industrial production applications, including process monitoring and control, and quality assurance. For instance, take a robotic gluing operation on a clay-coated paperboard or plain corrugated packaging line. Using FLIR thermal imaging technology it is now possible for a camera to detect glue beads and verify seams.
Before now, this was virtually impossible because adhesives for the packaging industry are almost always clear or opaque, making them difficult to view using standard vision cameras as these require a contrast between materials. Conversely, infrared radiation is emitted by all objects at temperatures above absolute zero. FLIR cameras convert that infrared radiation into a visual image that depicts thermal variations across an object or process.
Efficient ways of testing glued together boxes on high speed packaging lines are scarce – most tend to involve destructive testing on randomly selected samples, which is slow and cumbersome. In addition, glue application has a good deal of variability that is best monitored and recorded using statistical quality control routines. Using a thermal imaging camera, images can be digitised and stored in a database for trend analysis and equipment monitoring. Ultimately, temperature is used to replace mechanical methods of inspection and testing.
A thermal image can easily detect incorrectly sealed boxes. It can also detect the correct positioning of glue spots and indicate failed boxes to remove from the line. Furthermore, a pass/fail process can be implemented that sounds an alarm in the event of pre-defined failures, thus helping prevent unscheduled downtime or the introduction of poor quality products into distribution. Similarly, because many manufacturers place more adhesive on their packages than needed for insurance purposes, the use of such a system allows tighter adhesive specifications and therefore reduced material costs.
Using the measurement tools in HALCON, the list of potential benefits is expanded even further. For instance, it is possible to not only check the presence and integrity of a continuous glue bead, but also its position, width, height, straightness and more – this can all be achieved within milliseconds. If there is a problem with the glue gun, for example, this feedback-based system can stop the process before an entire batch of packaging is completed incorrectly and has to be subsequently scraped.
The system can also help identify trends for maintenance purposes. Incorrect bead size, position and occasional gaps can be detected and recorded, permitting scheduled preventative maintenance instead of unscheduled corrective maintenance and the associated production downtime this incurs.
Packaging lines today are an integral part of modern manufacturing. Carton building, tray making, case erection and sealing have all become highly automated, complex operations. Growing innovation and intricate designs have become typical, placing increased requirements on the previously simple adhering and sealing process. What’s more, recycled materials, complex coatings and finishes have seen the introduction of specialist hot melt adhesives, which on occasions can make adhesive application set-up and operation more difficult.
Among additional capabilities of the HALCON-FLIR solution are pattern matching, grading applications and adhesive temperature checking – many types of glue only activate within upper and lower temperature bands. All packaging line users are set to benefit, including the food industry, cosmetics and electronics to name but a few – most primary and secondary packaging materials give good results.
Ease-of-use is another principal factor likely to score well. Using HALCON’s Image Acquisition Assistant, plugging in a FLIR A-Series camera will prompt the user to ‘detect’ and ‘connect’. Two clicks are all that’s needed to get up and running.
“The use of FLIR thermal imaging for industrial automation projects could potentially be revelatory in many applications,” says Julie Busby. “It is even possible to combine thermal and 3D vision cameras in the same system. This way, data from the former can overlay data from the latter. The result is a powerful solution that can monitor glue beads and seals using infrared, and deploy 3D vision for shape and content checking.”
The benefit of a combined camera verification system such as this is that a single set-up by the operator would be enough to permit the two systems to operate in conjunction with each other: are the glue runs in the correct place? are there any gaps? is the machine feed correct? If a manufacturing log is required for the benefit of customers, then an integrated camera solution is the answer.
For any process or packaging line with a reputation to protect, the Multipix combination of HALCON vision software and FLIR A5 thermal imaging camera provides not only peace of mind but potentially untold savings in rejections from the retailer and/or end user dissatisfaction.
Beyond the immediate target of the process and packaging marketplace, Multipix is already scrutinising the electronics sector, particularly battery manufacturers in the automotive segment.
Multipix Imaging has been the exclusive UK distributor of HALCON from MVTec Software GmbH for the past 15 years. This comprehensive software library for machine vision offers an integrated development environment (IDE) that is used worldwide. The flexible architecture facilitates rapid development of machine vision and image analysis applications.
The FLIR A5 thermal imaging camera (80 x 64 pixels, 60Hz) is a compact and affordable unit designed specifically for machine vision. Its small form factor, measuring only 40 x 43 x 106mm, is ideal for tight spaces with no loss of performance. With <50mK thermal sensitivity, the FLIR A5 captures the finest image and temperature difference information. Other A-Series cameras being offered by Multipix to complement HALCON are the FLIR A15 (160 x 128 pixels, 60Hz), FLIR A35 (320 x 256 pixels, 60Hz) and FLIR A65 (640 x 612 pixels, 9Hz) models. The FLIR A5 camera will allow thermal imaging to be widely used in industrial applications, solving historic issues of thermal imaging being prohibitively expensive and camera size large and obtrusive.
Once the proper segmentation parameters have been selected, the OCR assistant allows users to choose an effective classifier from either a pre-trained HALCON classifier, a classifier previously trained by the user, or the ability to train a new classifier.
To further improve performance, the OCR assistant allows users to define or restrict the possible results based on features like known syntax or expressions, and a dictionary file of allowable words or strings.