Comparative analysis of barKoder's PDF417 scanning performance against other barcode libraries
PDF417 Barcode Scanning Tests
PDF417 is a two-dimensional stacked barcode symbology capable of encoding large amounts of data in a compact space. More info for this type of barcode can be found on our site in the barcode types section and further more by clicking here.
The PDF417 barcode by design has stacked data rows columns enclosed by Start and Stop patterns which are mandatory by specification for detecting its location within an image. Following the Start pattern and all the way to the Stop pattern are the “Row Indicator” columns where all necessary info for correctly decoding PDF417 is encoded, like the number of rows and columns, the error correction level and so on.
The barKoder Barkode Scanner SDK is an exceptional software-based barcode decoding API that can read PDF417 barcodes on all types of official documents, like US & Canadian driver's licenses and military IDs, including US Uniformed Services Identification Cards or Common Access Cards. These IDs by specification must have standard or truncated barcodes printed on them. The same goes for Argentinian IDs (Documento Nacional de Identidad), Colombian IDs (Cédula de Ciudadanía), as well as a range of documents from South Africa, like South African Driver's Licenses, Vehicle Registration Disks and Smart ID cards. These IDs have barcodes either on their front or back sides.
Powered by its PDF417-LineSight® algorithm, barKoder API's accuracy and reading rate are unsurpassed not even by competitor offerings in the software market for barcode scanners, but even by hardware-based barcode scanners. Remember when we talked above how PDF417 barcodes demand by design to have Start and Stop patterns? Well that is not an issue when it comes to barKoder, as it can read not only PDF417 barcodes with damaged or completely lacking Start and Stop patterns, but even those lacking its Stop Row Indicator and entire data columns! You can read more about the PDF417-LineSight's awesomeness here: https://barkoder.com/barcode-scanner-sdk/pdf417-linesight
In order to establish the above statement, we made a series of internal tests in which we compared the reading rate of most providers in our competitive landscape, utilizing the most damaged and hard-to-read PDF417 barcodes from real-life samples that we have in our database. A total of 22 samples were utilized, part of them that can't be shared due to privacy concerns. The results were summarized in the graph bellow:
The final result puts barKoder API firmly in the lead with a 90.4% reading rate, outscoring the nearest competitor by almost 50%. Following were Scanbot (52.38%), Scandit (47.61%), Dynamsoft (42,85%), Viziotix (38.10%) and Cognex (23.80%).
And when it comes to open-source solutions, there's simply no competition, since the latter's performance goes into single digits. You can read about this comparison in our previous blog post here: https://barkoder.com/blog/comparison-of-barkoder-sdk-with-open-source-barcode-scanner-sdks
The new version of our SDK can be obtained by registering on the barKoder Developer Portal and visiting the Repository subpage. We support broad range of platforms, including native Android & iOS, Web, Flutter, React Native, Xamarin, .NET Maui, Capacitor and from recently, Cordova and NativeScript. Each registered account can utilize the self-service for generating a completely free trial license for an initial duration of 30 days that will grant you a full run to evaluate the capabilities of our mobile & web barcode scanner SDK for up to 25 devices. If your project has somewhat different requirements, we'll be happy to hear from you.
