Geoalert Team random weekend. Board games with family members

Towards the global #Buildings model!

You may have noticed that our models perform differently in different regions. The reason is that land cover and urban patterns are so different on our Earth that no existing model can comprehend them all.
Today we step forward in our Buildings model development, introducing our new region-specific model selection! When you launch the processing, the actual regional model is selected based on your area of interest.
Currently, there are 3 global regions and 3 corresponding buildings segmentation models, which will be further improved.

All models are now acting as a single "🏠 Buildings" model at both #mapflow web and QGIS. Try it - you can reprocess your area if you weren't satisfied with results - and don't forget to leave us your feedback from ⭐️ to ⭐️⭐️⭐️⭐️⭐️

If you need to customize model further to even better meet your requirements, contact us at [email protected] for a demo or a custom plan

The new story from National Institute of Urban Management about the use of #Mapflow for Regional Planning and Mapping in Telangana. Thanks for sharing 🙏👍

Good news for everyone who wants quick and easy statistics without using GIS tools!

Starting from today, every Building, Construction site, and Field geometry in #Mapflow will have an area tag with object area in square meters.

Mapflow – QGIS 1.8.0 released. 🎉

Upgrade to the newest version if you already use this plugin.

What's new?

* Updated UI. We added a sidebar which makes it easier to set up processing. Yes, we know it would be better to dock this widget in QGIS - looking forward to enabling it soon
* "Rate processing" function - the same as in the Web Ui. Will make user feedback more consistent.
* We removed "Sentinel Fields" model as it was too experimental and provided unstable results. Please apply to us if you need this or some custom models with Mapflow
* Some minor improvements with user data validation
* Enabled zoom to raster XYZ layer given its AOI extent
* Enabled necessary features to work under team accounts (Are you using Mapflow for your professional work? Maybe it's time to contact us and request to create an account for your organization 😉)

👉 We encourage you to join our open-source repo on Github.

Sometimes Geoalert office is visited by young family members. Next time investors ask us about our strategic plan we will show them this drawing. 🧑‍🎓

Our article "Being an AI analytics SaaS start-up in the Earth observation market" was published in GIM International 3 2023. 🎉
The full issue can be downloaded from GIM's web-site

Time to launch #starship

Geoalert #Team has been updated 😃
Daria joined our RnD team as a product design Engineer 👏

Many thanks to all #Mapflow users who leave their feedback. Last week results make 3.6 all model rating on average. We look forward to increasing user rating by improving Mapflow's ready-to-use models.
Last year we wrote a feedback review and organised our own helpdesk feedback in return for your comments and assessments.

Организаторами воркшопа выступают:  Агентство космических исследований и технологий при Министерстве цифровых технологий Республики Узбекистан, компании “INT-TECH”, “ГеоАлерт” при поддержке профессионального сообщества GISGeo и ОЮЛ “Ассоциация компаний в области ДЗЗ”.  

Вот сколько организаторов GIS workshop, который наша компания проведет 25го апреля в Ташкенте. Главное, чтобы участников собралось не меньше 😅

ПРИГЛАШАЕМ ВСЕХ!

Сообщаем Вам о семинаре которые пройдет в г. Ташкент 25 Апреля в Trilliant Tashkent (Алайский базар).

В рамках конференции SPACE TECHNOLOGY CONFERENCE Central Eurasia состоится международный воркшоп: «Дистанционное зондирование Земли и Геоинформационные технологии. Продукты, сервисы, решения».

Тема семинара: «Дистанционное зондирование Земли и Геоинформационные технологии. Продукты, сервисы, решения»
Локация: Trilliant Tashkent - г. Ташкент, Отель International Hotel Tashkent, проспект Амира Темура, 107А, 11 этаж
Дата: 25 Апреля 2023г.
Время: 14:00 - 18:00
Регистрация обязательна: https://shorturl.at/ijptM

Вход свободный!

This week we have updated our forest heights model.

As before, it is available via #Mapflow #API and QGIS plugin, but the new height model is more sensitive to local height changes, and has increased overall accuracy. Forest segmentation mask remains the same.

If you perform forest growth monitoring, check it out!

An interesting showcase of using #Mapflow platform for #Powerline monitoring and management.
The #forest height model can provide great visual results in QGIS2threejs when accompanied by DTM and object presets.

Just some pretty accurate vegetation mask from the user's feedback.
#Mapflow #Forest model

👉 Note that you can run a Team account to invite collaborators and set up #Mapflow usage limits.
Currently you need to reach out to us to create a Team and assign an Admin role to your active Mapflow account.
After the Team is created the Admin user can invite new team members and manage their limits in the Mapflow settings. Only Admin user gets acces to the Team profile.

Building height estimation in a single satellite image is one of our most demanded #Mapflow services, as it helps reduce costs and time when one needs to get 3D #urban planning, detailed population density maps or design the Radio Frequency Network.

Traditional tools for 3D data generation include manual stereo-photogrammetry workflow or computationally expensive DSM creation. Our approach is based on AI/ML algorithms and provides accuracy to 4 meters.

Just today we have updated the walls and shadows segmentation model which is an important step for more accurate height estimation.
This is not a breaking change, but the walls segmentation results became better, so the heights (and footprint locations) will be more stable and accurate. See footprint outlines at the picture: new (green) vs old (red)

For more details about the technology see our blog article.