{"id":7264,"date":"2024-12-10T11:20:37","date_gmt":"2024-12-10T00:20:37","guid":{"rendered":"https:\/\/weeds.org.au\/?page_id=7264"},"modified":"2026-04-10T14:51:27","modified_gmt":"2026-04-10T04:51:27","slug":"weedscan-noticeboard","status":"publish","type":"page","link":"https:\/\/weeds.org.au\/weedscan-noticeboard\/","title":{"rendered":"WeedScan noticeboard"},"content":{"rendered":"\n
30 June 2025 | WeedScan 2.0: Progress behind the scenes<\/summary>\n

WeedScan is evolving\u2014but unlocking its full potential means tackling some tough accuracy challenges head-on.<\/p>\n\n\n\n

Our pioneering WeedScan app continues to develop, with steady progress underway to strengthen the app\u2019s functionality and performance behind the scenes.<\/p>\n\n\n\n

WeedScan 2.0 will feature an expanded number of species and a suite of targeted enhancements, such as fixes to image shredding, improved handling of low certainty thresholds, and adjustments to reduce confusion between similar-looking plants. These upgrades have been validated through extensive user acceptance testing (UAT), and we\u2019re incredibly grateful to everyone who took part\u2014your contributions have been essential.<\/p>\n\n\n\n

Importantly, version 2.0 has not yet launched \u2014 so users won\u2019t see these improvements just yet. While we\u2019re keen to roll them out, we\u2019re taking the time to address some remaining hurdles. UAT highlighted that consistent identification accuracy is still a challenge. In response, we\u2019ve launched a deep technical audit to better understand the underlying limitations and map out the way forward. This work is already revealing valuable insights and helping shape the next phase of development. None of this is unexpected\u2014WeedScan is pioneering technology, and breaking new ground always comes with complexity.<\/p>\n\n\n\n

\"WeedScan
WeedScan 2.0 will bring more functionality to the app.<\/figcaption><\/figure>\n\n\n\n

In the meantime, the current version of WeedScan 1.0 remains a vital tool. Ongoing use of the app is helping us train the AI model further, and every scan and image submission contributes to building a stronger, more reliable tool for the future. We encourage all users to continue using WeedScan 1.0\u2014it\u2019s playing a critical role in shaping what comes next.<\/p>\n<\/details>\n\n\n\n

14 October 2025 | Help shape the future of WeedScan 2.0<\/summary>\n

The Centre for Invasive Species Solutions (CISS) is looking for volunteers to help build WeedScan 2.0<\/strong>, Australia’s next-generation weed identification app.<\/p>\n\n\n\n

We\u2019ve nearly doubled the number of species to around 950 plants<\/strong> and gathered over 900,000 images<\/strong> to train our AI system. Before training begins, we\u2019re conducting a structured quality-control review of this image set – screening for low-quality, duplicate, or off-topic images – so the AI learns from clean, reliable data.<\/p>\n\n\n\n

If you\u2019re detail-oriented, reasonably tech-savvy, know a bit about plants<\/strong> (scientists \u2013 especially botanists or other scientists and keen naturalists are encouraged to apply) and can spare a few solid days (or even a week or two)<\/strong> over the next month, we\u2019d love your help!<\/p>\n\n\n\n

You can join us remotely from anywhere in Australia<\/strong> (provided you have a laptop, reliable access to the internet and your own Google or Microsoft account) or in person at our Canberra office. Training and support are provided \u2014 and you\u2019ll get a sneak peek of WeedScan 2.0 before launch<\/strong>!<\/p>\n\n\n\n

\"\ud83d\udce9\" Interested? Email weeds@invasives.com.au<\/a> with the subject line ‘WeedScan 2.0 Volunteer’.<\/p>\n<\/details>\n\n\n\n

4 November 2025 | Two ways you can help shape WeedScan 2.0! \ud83c\udf3f<\/summary>\n

We\u2019re gearing up for the next phase of WeedScan 2.0, Australia\u2019s next-generation weed identification app – and we\u2019d love your help! \ud83e\udd1d

To build a smarter, more accurate AI model, we\u2019re calling for two kinds of volunteers:

1\ufe0f\u20e3 Image Curators \u2013 Help us prepare the training datasets so the AI learns from only the best material. You\u2019ll filter out low-quality, duplicate or off-topic images and spot species that need more photos so we can achieve the minimum threshold of quality images per plant species.

2\ufe0f\u20e3 Expert Botanists & Plant Specialists \u2013 We\u2019re also inviting volunteers with expert botanical knowledge to assist with some of the more challenging species – those that are tricky to separate because of similar lookalikes. Your expertise will help validate identifications and strengthen the model\u2019s accuracy in distinguishing between closely related species. \ud83c\udf31

Whether you\u2019re a botanist, scientist, or passionate naturalist, this is a great opportunity to contribute to national weed management. You can join from anywhere in Australia (you\u2019ll just need a laptop and internet) or in person at our Canberra office.<\/p>\n\n\n\n

\"\ud83d\udce9\" Interested? Email weeds@invasives.com.au<\/a> with the subject line ‘WeedScan 2.0 Volunteer’.<\/p>\n<\/details>\n\n\n\n

30 January 2026 | A big ‘thank you!’ to our volunteers \"Fist<\/summary>\n

CISS and the entire WeedScan team send heartfelt thanks to the volunteers who responded to our call last year.<\/p>\n\n\n\n

Their time, attention to detail, and commitment allowed the team to review and curate thousands of images: a critical step in preparing high-quality datasets for our AI\/machine learning (ML) model. This support has directly improved the scientific integrity of WeedScan and strengthened its capacity to correctly identify weed species across Australia.<\/p>\n\n\n\n

These efforts mean we are now well on our way toward releasing a more robust and reliable WeedScan 2.0, incorporating nearly 1,000 weed and lookalike species.<\/p>\n\n\n\n

Sending a big ‘thank you!’ to our volunteers for advancing Australia\u2019s biosecurity and weed management capacity via this cutting-edge citizen science tool.<\/p>\n\n\n\n

<\/p>\n<\/details>\n\n\n\n

<\/p>\n","protected":false},"excerpt":{"rendered":"

Check in for the latest updates, user tips and top weed observations on the WeedScan app.<\/p>\n","protected":false},"author":2,"featured_media":669,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"templates\/gutenberg-page.php","meta":{"_acf_changed":false,"footnotes":""},"class_list":["post-7264","page","type-page","status-publish","has-post-thumbnail","hentry"],"acf":[],"_links":{"self":[{"href":"https:\/\/weeds.org.au\/wp-json\/wp\/v2\/pages\/7264","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/weeds.org.au\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/weeds.org.au\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/weeds.org.au\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/weeds.org.au\/wp-json\/wp\/v2\/comments?post=7264"}],"version-history":[{"count":26,"href":"https:\/\/weeds.org.au\/wp-json\/wp\/v2\/pages\/7264\/revisions"}],"predecessor-version":[{"id":7510,"href":"https:\/\/weeds.org.au\/wp-json\/wp\/v2\/pages\/7264\/revisions\/7510"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/weeds.org.au\/wp-json\/wp\/v2\/media\/669"}],"wp:attachment":[{"href":"https:\/\/weeds.org.au\/wp-json\/wp\/v2\/media?parent=7264"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}