Frequently Asked Questions

Q: Can we try before we buy?

A: Yes, we offer a 45-day free trial license to all interested clients. Please fill out our Contact Form and we will contact you for the the next steps.

Alternatively, you can try out the ConPDS OCR Demo app or search Google Play for “ConPDS OCR Demo”. This app features the ConPDS OCR Engine, and you will quickly be able to see how it performs.

Q: Is the ConPDS OCR Engine a Cloud Service?

A: No, the ConPDS OCR Engine is executed locally on the device on which it is launched.

Q: How is the ConPDS OCR Engine licensed?

A: We differentiate between two different licenses: a) Developer License and b) Runtime Licenses. Here are a couple of examples:

Example 1

OCR Engine for the Android and iOS Operating Systems on 90 devices.
Customer should purchase 2x “Get Started” packages (one for Android and one for iOS). This includes 2x Developer Licenses and  2x 50 Runtime Licenses, 100 in total, and can be combined into one mixed pool of Runtime Licenses, which can be shared between Android and iOS devices.

Example 2

OCR Engine for Windows Operating System and 10 stationary cameras.
Customer should purchase 2x “Get Started” packages which includes one Developer License and Runtime Licenses for up to 50 cameras. A Runtime License is required per stationary camera.

Example 3

OCR Engine for Android and 3000 devices.
Customer should purchase 1x “Site License” package, which includes one Developer License and unlimited number of Runtime Licenses.

If you have special scenarios or requirements feel free to contact us. We are flexible!

Q: Is the ConPDS OCR Engine only for Windows platform?

A: No, the ConPDS OCR Engine is available for Android, iOS, Linux (x64 and ARM), Windows (x86 and x64), and it is the exact same OCR engine for any of them.

Should you have special requirements feel free to contact us. We can compile for almost any operating system or processor architecture.

Q: Are your licenses perpetual?
A: Yes, all licenses are perpetual, lifetime licenses.
Q: What's the price for future OCR engine upgrade?

A: As long as customers pay the yearly Maintenance Fee, access to all future upgrades of the ConPDS OCR Engine – including support – are completely free. Included in the yearly Maintenance Fee is also the option to have our OCR engine trained with your or your customers’ own images.

Q: For the Android platform, is it an SDK that I can include and compile in my app?

A: No and Yes.

No, the ConPDS OCR Engine is not an SDK but an out-of-the-box solution. And yes, it can be compiled into your own app or any software project. Get more details here.

Note: The same answer applies to all other Operating Systems. Get more details here.

Q: Can your OCR engine identify the container ISO type, such as 22G1/45G1?
A: Yes, the ConPDS OCR Engine does recognize ISO Size and Type codes according to the ISO 6346 standard.
Q: Can your OCR engine detect the crack on the surface of the container?

A: No, we only focus on recognizing the BIC and ILU container codes. For BIC containers also ISO Size and Type codes are recognized. Recognition of cracks, hazardous logos etc. is not a part of an OCR engine.

Q: I need to integrate OCR engine into my C# project. Is it possible?

A: Yes, the ConPDS OCR Engine is available as a library which can be loaded into C# projects and also for other.Net languages.

Q: I want to run OCR engine on Linux. What are the requirements?

A: Ubuntu and Debian Linux distributions (64bit) are supported, but most likely the ConPDS OCR Engine will run on others also.
Hardware requirements: Minimum 1 (v)CPU, 512MB RAM, 1 NIC (for license activation), and 50MB HDD.
Other prerequisites: Java Runtime environment. Get more details here.

Q: What is the expected output of your OCR engine?

A: The output of OCR processing is a JSON response.

The JSON response will have information about processing time, recognized Container Code and ISO Size and Type (BIC only), recognition result (success/failure) based on comparing recognized and calculated digit, confidence factor and a subset of data for other potential candidates.

Sample JSON respone:
{“processing_time_ms”: 62, “recognition_type”: “BIC”, “result”: “MSKU5844165”, “checksum”: 1, “iso_size_type”: “22G1”, “confidence”: 91.23, “recognition_area”: [{“ul_x”: 456,”ul_y”: 292}, {“lr_x”: 880,”lr_y”: 599}], “candidates”: [{“recognition_type”: “BIC”, “result”: “MSVU5844165”, “checksum”: 0, “iso_size_type”: “22G1”, “confidence”: 90.98}, {“recognition_type”: “BIC”, “result”: “MSKU5849416”, “checksum”: 0, “iso_size_type”: “22G1”, “confidence”: 90.98},{“recognition_type”: “BIC”, “result”: “MSVU5849416”, “checksum”: 0, “iso_size_type”: “22G1”, “confidence”: 90.73}], “image_metadata”: {“image_resolution”: {“width”: 1280, “height”: 960}}}

Get more details here.

Q: What is optimal image resolution?

A: The most optimal image resolutions are 1280×1024 or 1440×1280 in either landscape or portrait mode.
Much lower resolution can result in inaccurate recognition and higher resolutions do actually not improve recognition significantly, and will only result in a bigger volume of data.

Image quality (focused and including the full container code) is also equally important even though the ConPDS OCR Engine as part of image pre-processing phase are doing Image Noise reduction, Binarization, Image emphasis, and so on, it will make a better starting point for the detection and correct recognition of characters.

Get more details here.

Q: Why should we choose your ACCR engine (OCR engine)?
A: There are numerous reasons and let us list some of them here:

Our ACCR software is not based on an SDK or open source OCR solution. It is built from scratch, written in C and C++ and, based on a very simple concept: Image in – Container Code out.

Once our OCR engine is added to your software project you will get instant and guaranteed result, and you can focus on more important tasks like business workflows and optimizing business processes.

The importance of a well-performing OCR engine, not least when used on a handheld device, is massive. The ConPDS OCR (ACCR) engine performs fast and accurately. This is simply because it is created for the sole purpose of recognizing container codes. We believe that a single purpose OCR engine outperforms a multipurpose OCR engine on many parameters. Here is why:

  • A limited character set. Our ACCR engine only looks for the characters and fonts that a container code can contain. This means higher accuracy and faster processing.
  • Tailored image processing. To achieve high accuracy at bad conditions (low light/visibility, varying perspectives & distances, etc.), an OCR engine requires great image processing. Because our solution is built for a single purpose only, image processing has been tailored accordingly.
  • Continuous training. We constantly improve the performance of our ACCR engine, by training it with new instances of container codes (different/mixed formatting, horizontal/vertical reading combinations, codes on rugged container surface, etc.). Similarly, the ACCR Engine is trained with new instances of images taken in challenging conditions. This way, you are assured that our ACCR engine will continue to outperform other OCR solutions.
  • Because of the above, our ACCR engine can recognize characters as small as 14×14 pixel (optimal is 20x20px or 40x40px), meaning that container codes can be recognized from a distance, even if the container is stacked several layers high.
  • Horizontal and vertical text recognition. The ConPDS ACCR engine recognizes both 1 and 2-line horizontal container numbers, as well as vertical container numbers.
  • For BIC Container Codes (ISO 6346) the ACCR engine also does recognition of ISO Size and Types.