01
Inkjet date inspection
Inspect date, batch number, expiry date and production code presence, clarity and position.
Back to products
OCR Traceability
Character inspection and OCR recognition solution
Inspect presence, clarity, position, content and traceability data for inkjet codes, laser marks, label text, serial numbers and traceability codes.
Object
Text
Inkjet codes, laser marks, serial numbers, dates and traceability codes.
Capability
OCR
Presence, recognition, clarity and missing-character judgement.
Data value
Trace
Results, images, batches and time can be stored together.
Scenarios
For stations that need batch traceability and content checks
Character inspection must judge not only readability, but also missing, blurry, shifted, duplicated or business-inconsistent content.
02
Laser mark recognition
Recognize and judge laser characters on metal, plastic and electronic components.
03
Traceability binding
Read barcodes, QR codes or serial numbers and bind them with batch, equipment and inspection images.
System
Connect OCR recognition with site business data
OCR projects usually cover imaging, character localization, recognition, business rule validation and data interfaces.
01
Imaging and text localization
Select coaxial, ring, bar or back lighting according to material and character process.
02
OCR and rule validation
Use OCR, Deep OCR, segmentation and format rules to judge content, missing, blur and wrong codes.
03
Traceability interfaces
Connect databases, MES, ERP or file interfaces to store recognition results and image evidence.
Process
Define acceptable characters before recognition accuracy
OCR projects need normal, blurry, missing, contaminated and wrong-code samples to define stable acceptance.
01
Character sample collection
Collect images across batches, positions, clarity levels and abnormal types.
02
Imaging validation
Test lighting, exposure, contrast, background interference and motion blur.
03
Recognition and business checks
Run localization, OCR, format validation, allowlists or batch-rule checks.
04
Data retention and review
Store original images, recognition result, confidence, batch, equipment and review records.
Deliverables
Make character results readable, searchable and reviewable
OCR delivery connects recognition results with quality traceability so every judgement can be verified later.
01
Character rules and sample set
Character format, abnormal samples, misrecognition boundaries and review rules.
02
Recognition interface and fields
Recognition result, confidence, image path, batch and equipment fields.
03
Exception handling process
Strategies for low confidence, unreadable text, format errors and business mismatches.
FAQ
Boundaries to confirm before OCR projects
Recognition quality depends on printing, material, lighting, font, speed and contamination.
Can tiny or reflective characters be recognized?
Camera resolution, lens magnification and lighting must be evaluated first. Small text and reflective material usually need sample tests.
Can it compare with batch data in a database?
Yes. Business rules can be read through database, API or file interfaces and compared with OCR results.
How are recognition errors traced?
Store original images, text, confidence, time, equipment and batch for later manual review.
Sample Review
An OCR route can start from a set of text samples
Send normal samples, abnormal samples, character rules and takt time for an initial OCR route and risk review.