Hi everyone, Could you please answer the question "Should I use QR codes or barcodes to digitize samples and manage my plant collection?
There is no one size fits all answer to this question, both serve a similar purpose. It will depend on what information you are looking to store in the label, and how you plan to retrieve that information at a later point in time.
QR codes can hold a lot more information than a barcode, and have some advantages when it comes to error correction in the event that part of the code is damaged. Barcodes are more fit for purpose when you have a single short number that you need to store, and may work better with older software. If you have an existing inventory of barcode scanners, those will likely not work with QR codes.
If you are looking to encode hyperlinks, QR codes are most likely the solution that you are looking for. This could enable you to link a specimen to, say, the occurrence page on GBIF via a unique identifier, or to your collection management system.
1 Like
thanks. Do you think Barcodes are generally considered more secure?
I work in a plant herbarium. By the way, can you provide information on whether QR codes or barcodes are more commonly used and prioritized in herbaria?
Can you elaborate on what you mean by secure (what are you trying to secure agaisnt?) Data loss? Physical damage? Cybersecurity?
QR codes have built in error correction capabilities which can help protect against partial damage, which is configured at the point of generation. For a very in depth analysis of how these different error correcting levels work, I would refer to QR error correction helps and hinders scanning | Huon on the internet.
I do not know the prevalence of barcodes compared to QR codes in herbaria overall, however I can say that we use 1-D barcodes in the museum which I work at. I would postulate that many herbaria use 1-D barcodes over QR codes not because they are technically superior, but because QR codes were not invented until the 1990’s, at which point many herbaria likely already had a barcode system in place.
In 2024, if there is no existing barcode system in place, or no software that can only work with 1-D codes, my personal recommendation would be to use QR codes.
1 Like
I got it, thank you.
@vnmngan, please also consider DataMatrix barcodes: see this blog post if you are only storing registration numbers or databasing IDs.
2 Likes
Besides what @mark-pitblado has said, here is an extensive resource about collection digitisation: https://dissco.github.io/
You can find some quick recommendations about barcodes in the Pre-Digitisation Curation Checklist, Suplies section, as well as a link to a couple of references about barcodes: Pre-Digitisation Curation | Digitisation Guides
2 Likes
I’ll add one thing that I don’t think anyone has mentioned and that is that 1D (conventional) handheld barcode scanners are more common in general and are less expensive than 2D barcode readers. This is probably a factor that will become less and less important as 2D barcodes become more prevalent. I’ve never seen a QR code or other 2D barcode used on an herbarium specimen for the catalog number or unique ID so I’d say that most herbaria are accustomed to 1D barcodes (I’ve seen 2D used for supplemental data). I’ll also add that I think it’s important for a barcode label to include human readable characters with the barcode. With 1D barcodes that’s typically just a string of characters (usually no more than 13) but with QR codes you may be tempted to have longer strings.
2 Likes
I’ll chime in with another recommendation to take a look at data matrix codes. They are more compact (in absolute size) than QR codes, and have full industrial support. As long as you’re not supporting a big hardware investment in 1D-only scanners, 2D scanners are very available today at prices very comparable to 1D scanners. Any newer 2D scanner will be able to read 1D barcodes, and a variety of 2D barcodes (including both QR and data matrix).
2 Likes
I also want to support the notion of human readable labels that hold the same information as in the barcode or datamatrix. I also believe the advantages of QR should be weighed against their disadvantages:
- readable under multiple orientations
- error correction
- higher data density
Against:
- Requirement for higher resolution scanners
- Temptation to store more data than is available in human readable form
- overkill for short strings, take up a lot of space on the specimen, increasing the chance that the code gets covered or damaged.
And I wonder, if QR was completely superior to other data matrices and barcodes, I suspect QR would be more dominant.
Barcodes have a limited orientation in which they can be read and don’t hold a lot of data, but have very wide adoption. Data matrices can be very compact, also hold more data than you’d want in this case and also support error correction.
Honestly, I feel you’d want to keep it as simple as possible really.
1 Like
Thank you for your information
Hi joaquimsantos
The information and practical examples are very helpful.
If there is a video tutorial or online course… it would be very helpful for me to practice. It’s quite difficult to keep up because I only do classification research and know little about bioinformatics. By the way, do you know free specimen scanning Sofware and herbarium database software?
Thank you for your information. 1-D code still needs to buy a software to manage? and need to buy a scanner? To create a 1-D code, we must apply to a barcode creation organization to request a code.
Hi @vnmngan,
You do not need to apply to a barcode creation organization to create a code, they aren’t unique identifiers that are managed by a centralized entity[1] (unless you mean to request printing services). It is not proprietary technology, so there is no software that you need to buy in order to create them (although I am sure there are commercial products that create barcodes). For example, I created the following ean13 with 4 lines of python code. This could be put in a loop to generate 10’s, 100’s, or 1000’s of barcodes for free.
import barcode
from barcode.writer import ImageWriter
ean = barcode.get("ean13", "123456789102", write=ImageWriter())
filename = ean.save("ean13_barcode")
# Generates a file called ean13_barcode.png, shown below
Most barcode scanners will allow the user to put their cursor into a box, press the button on the scanner, and the contents of the barcode are then input into that textbox (similar to how a paste operation works) without any additional software required. Smartphones can also scan barcodes just fine. If you don’t use a smartphone, then yes, you would need to purchase the scanner. The price will be dependent on what features you want it to have (battery, wireless etc)
The identifier itself may be centrally managed (such as ean13 in manufacturing), but the ability to create the barcode from and identifier is something that anyone can do. ↩︎
1 Like
I’m not aware of any full video tutorial, but I’m sure you will be able to find short videos where digitisation of specimens is addressed. They might lack detailed discussions of processes and alternatives though.
You don’t need specific software to scan specimens. The manufacturer software will be fine. Nowadays, many institutions use photographic cameras, which is are a lot quicker than scanners. You just need to be aware that the detail should be compliant with the purpose, so the fist question you should ask is: why do I need an image of this specimen? Just for documenting its existence => low detail (or no image at all); To make taxonomy => high detail, or even microscopic detail of certain parts.
About databasing, my personal opinion is that there should exist by now something really good, but I’m afraid the perfect Collection Management System doesn’t exist yet. I would recommend you to explore Specify, which can be used for free and is widely used (not very user friendly though): https://www.specifysoftware.org. IDigBio has compiled a list of CMS software here: Collections Management Systems - iDigBio. If you don’t have the time to learn, you could just record the specimens on a spreadsheet (you could then migrate all to a proper CMS later on, as many have done). It all depends on your purpose and available resources. Nevertheless, you should always have in mind that the ultimate goal of digitisation is not having a digital list, but to achieve a digital transformation of all the processes involved in the management of the specimens (have a look to the Digital Extended Specimen concept: Digital Extended Specimens: Enabling an Extensible Network of Biodiversity Data Records as Integrated Digital Objects on the Internet | BioScience | Oxford Academic). Regarding the fields to record, I would recommend you to become familiar with darwinCore terms (Darwin Core Quick Reference Guide - Darwin Core), and maybe Minimum Information about a Digital Specimen (MIDS), since you might the digitisation to happen by stages (mids/current-draft /MIDS-definition-v0.17-13Jul2023.md at working-draft · tdwg/mids · GitHub)
1 Like
Hi joaquimsantos, thank you very much!