About this barcode reader
This barcode reader decodes the bars or modules in a local image and shows their stored value. It is useful for checking a product label, a shipping label, or a barcode screenshot before entering the number into another system. The bundled ZXing library recognizes common retail and logistics formats, including EAN, UPC, Code 128, Code 39, and several two-dimensional symbols.
Select your image and press Read barcode. The picture is drawn locally and the decoder checks the pixel pattern; it does not send the image or decoded number to a catalog service. The result includes the text and the detected format, so you can distinguish an EAN product identifier from a free-form Code 128 label. Several selected files are processed individually.
A successful decode is not a guarantee that an item is genuine or that a database contains that identifier. This reader performs no product lookup. Keep the entire barcode, including clear space at each end, inside the image. Motion blur, glossy packaging, clipped bars, unusual rotations, and low resolution can prevent a result. A single image returns the first recognized symbol rather than a complete list of every label on a sheet.
Which controls should I check first?
The Barcode Reader form labels its controls as Choose image files (up to 20, 40 MiB each). Use those labels to keep the source and settings separate. A sample argument that selects a mode is not additional source text. Choose the setting matching your intended operation before copying a result, and change it explicitly when comparing another mode. The labels describe what this form accepts; the worked examples below demonstrate only the stated supplied values.
How should I interpret this specific operation?
The result identifies the stored value and detected symbology, without querying a product catalog. A correct EAN check digit does not prove authenticity. Include clear quiet margins at both ends of a linear barcode and avoid glare or clipped bars. A helper's checksum example can confirm validation arithmetic; recognition of a photograph requires a separate decoder observation.
Worked helper values
These values come from the reproducible calculation examples supplied with this tool. Each row identifies the complete supplied arguments and the returned value. A visible ↵ denotes a line break and ⇥ denotes a tab in text examples. Settings appear alongside source data so that a different mode is not mistaken for an identical input. These are helper calculations, not observations of a live device, fetched target, image decoder, browser preview or downloaded file. Use the stated method above to relate the returned fields to the actual operation.
| Supplied values | Calculated return value |
|---|---|
| 4006381333932 | false |
| {"rawValue": "123456"} | 123456 |
Which information is present in the source pixels?
Recognition works from the actual image rather than from its filename. Keep the intended text or symbol fully visible, including quiet borders around machine-readable codes. Blurred edges, glare, compression artifacts and a small subject inside a large picture can reduce the information available to the decoder. Crop to a relevant area while retaining the boundaries the format needs. Enlargement can make pixels easier to inspect but cannot recover detail that the source never captured. Check the actual image before changing recognition settings, because a correct setting cannot repair missing letters or clipped bars.
What does a returned value establish?
A returned string is the recognizer’s interpretation of the supplied picture or the helper’s supplied result. It is not automatically a verified fact. Compare names and numbers with the original image, and treat decoded links as data before opening them. A retail identifier does not establish product authenticity, and OCR confidence does not independently verify every character. Preserve the result and relevant source crop together when accuracy matters. The helper example below illustrates cleaning, validation or display logic under explicit inputs; it does not demonstrate recognition of a real image in a browser.
How should I prepare a useful comparison?
Choose a sharp source containing one clearly visible target and start with the default recognition settings. Inspect the whole result before processing a larger batch. If a result is missing or incorrect, change one factor at a time: crop, image quality, scale or the language selection where applicable. Keep the original file so that an aggressive crop does not become the only available source. For a sheet containing several symbols, check whether this reader promises one recognized result per image. A first successful result should not be described as a complete inventory of every target on that sheet.
Why can text require more review than a symbol checksum?
Ordinary text recognition makes character and word guesses without a universal checksum. A machine-readable format can include error correction or a check digit, but those rules still do not authenticate the content. A valid code can point to a wrong destination, and a plausible OCR word can contain a mistaken letter. Use the format-specific result to detect transcription issues while preserving the need to review meaning and context. Numbers, identifiers and dates deserve an exact comparison with the source because a one-character error can change their practical use.
Does this identify the product?
It reads the encoded identifier only. No catalog, price, ownership, or authenticity lookup is performed.
Which barcode formats are supported?
The local ZXing multi-format reader supports common EAN, UPC, Code 128, Code 39, ITF, QR, and other standard symbols.
Why does a label fail to scan?
Include the full symbol and quiet margins, avoid reflections, and use a clear image with sufficient resolution.