Blog
Processing Human Readable Text That Accompanies a Linear Barcode
Human Readable Barcode
Compared with 2d barcodes, linear barcodes are less robust and error-prone due to the lack of or low level of error detection and error correction mechanisms. For this reason, we often print text that encodes the same data as the barcode. Most often, the text string is put beneath the barcode. In demanding situations where 100% accuracy of barcode reading is a must, we usually use OCR to read the text to cross-check the barcode result.
Solution
Dynamsoft offers a comprehensive solution for processing human-readable text accompanying linear barcodes. Dynamsoft Label Recognizer SDK, alongside Dynamsoft Barcode Reader SDK, provides extensive OCR capabilities for text extraction and recognition. The main idea is that Dynamsoft Barcode Reader identifies the barcode quadrilateral and provides its location information as a reference for Dynamsoft Label Recognizer.
At the heart of Dynamsoft’s Capture Vision Suite architecture is a router class that orchestrates all functional modules, including the camera module, barcode reader, label OCR module, document normalizer, and code parser. Let’s explore how Dynamsoft’s barcode reader and label OCR module collaborate.
How to Read Human Readable Barcode
The process of handling accompanying text is divided into two major stages: text detection and text recognition.
Text Detection
Dynamsoft’s Capture Vision Suite features a modular architecture, allowing all functional modules to be seamlessly pluggable and interoperable. These modules work in a cascading sequence, where the output of one task dynamically adjusts the region of interest for the next. Task settings can be conveniently configured using a JSON template file.
A straightforward example of this architecture in action is reading human-readable interpretations. In this workflow, Dynamsoft Barcode Reader serves as the initial task, identifying the quadrilateral of the barcode. This quadrilateral, combined with an offset setting, defines the TargetROIDef (target region of interest) for the label OCR task. The label OCR task then proceeds to detect text blocks.
For more details on TargetROIDef, refer to the Dynamsoft documentation.
Text Recognition
For each detected text block, Dynamsoft Label Recognizer SDK leverages a highly trained neural network model to perform text recognition. This SDK can be tailored to extract alphanumeric characters and standard symbols from images with different fonts, background colors, and text sizes, making it ideal for applications such as vehicle identification numbers (VINs), ID cards, inventory labels, price tags, etc. The detected characters are then returned as the OCR result, which can be compared against the barcode decoding output to ensure data integrity.
Customer Stories: Decoding Text Beyond Human-Readable Interpretation
Customer No. 1: Extracting Contextual Information Alongside Barcodes
A customer using Dynamsoft Barcode Reader JavaScript SDK (DBRJS) faced challenges in associating scanned barcodes with specific data fields due to insufficient contextual information.
Integrating Dynamsoft Label Recognizer (DLR) with DBRJS enables:
- Mapping barcodes to specific fields
- Contextual data interpretation using regex-based templates
- Real-time recognition through video streaming
The integration ensured accurate context-aware barcode recognition, enhancing operational efficiency.
Customer No. 2: Achieving Efficiency with Barcode and Label Scanning Integration
A customer tested the integration of Dynamsoft Barcode Reader (DBR) and Dynamsoft Label Recognizer (DLR) for combined barcode and label scanning and reported highly positive results:
- Combined Scanning Success: The solution effectively scans both barcodes and labels, ensuring accurate data capture even in scenarios where label extraction faced challenges.
- Scalability: The customer is confident in DLR’s ability to handle new label types with minimal adjustments, showcasing its adaptability.
- Performance Excellence: DBR maintained its fast scanning speed even when combined with DLR, meeting the customer’s expectations for efficiency.
The tests demonstrated exceptional accuracy, with the data consistently captured in most scenarios. DLR performed quickly and accurately with high-quality images.
The customer is impressed with the solution’s reliability and performance and looks forward to future enhancements from Dynamsoft to optimize their workflows further.
Why Use Dynamsoft Label Recognizer SDK?
- Enhanced Accuracy: Cross-checking barcode results with OCR enhances reliability, particularly in critical applications such as healthcare and inventory management.
- Sophisticated Image Pre-Processing Algorithms: Dynamsoft Label Recognizer SDK employs advanced algorithms while tackling poor-quality images to enhance contrast, reduce noise, and refine image quality. These algorithms also leverage text characteristics, such as curvature, to correct distortions.
- Seamless Integration: Dynamsoft Label Recognizer SDK complements Dynamsoft Barcode Reader SDK, creating a unified solution for barcode reading as well as text recognition.
Try Dynamsoft’s OCR and Barcode Reading SDKs
If you’re at the stage where you’re testing different options for barcode decoding and OCR integration, Dynamsoft’s solutions are a great starting point. Get in touch with our team of seasoned experts to discuss your specific use case.
Try Online Demo: