Automation of Fire Door Surveys Using LiDAR and Point Cloud Classification
LiDAR (Light Detection and Ranging) is an advanced technology that uses light in the form of a pulsed laser to measure ranges (variable distances) to objects. By scanning the indoor environment, LiDAR can generate a high-resolution three-dimensional point cloud, a set of data points in space that represents the scanned environment.
Fire door surveys can be automated through the application of machine learning algorithms to classify these point cloud objects or using images. This involves training an algorithm to recognize the specific characteristics of fire doors – such as their size, shape, and location in relation to other objects. Once the algorithm has been trained, it can then automatically identify fire doors in the point cloud data generated by a LiDAR scan.
This automation process eliminates the need for manual surveys, which can be time-consuming and prone to human error. It also enables a more comprehensive and accurate assessment of fire doors, including their number, locations, and potential deficiencies, which are critical for fire safety compliance.
Processing Fire Doors and Fire Signage Using Image Recognition
In addition to LiDAR, image recognition technology can further enhance the automation of fire door surveys. By using high-resolution cameras in conjunction with LiDAR scanners, detailed images of the surveyed environment can be captured. These images can then be processed using image recognition algorithms to identify specific features such as fire doors and fire signage.
Image recognition algorithms are trained on large datasets of images, learning to recognize patterns and features that define different objects. In the context of fire safety surveys, these algorithms can be trained to recognize fire doors and signage based on their distinctive characteristics. For example, fire doors typically have specific features such as intumescent seals and self-closing devices, while fire signage is usually highly distinctive in its colors, shapes, and symbols.
More context
In short: LiDAR scanners and cameras capture a building interior as a point cloud and images; machine learning then classifies objects such as fire doors and fire signage, so surveyors get a list of locations and possible defects instead of walking every corridor with a clipboard. A person still has to inspect the doors themselves, but the inventory and the paperwork become much faster and easier to keep up to date.
LiDAR capture is now more accessible, with handheld scanners and some phones and tablets carrying LiDAR sensors. The scan becomes most useful when it feeds a digital twin that is updated after each round of works. See the related posts on digital twins and on building 3D models from floor plans below.
Related
- Importance of Digital Twins and Regular 3D Scans of Buildings
- Digital twins in the UK: building hotels in 3D from floor plans
- Collection of my NeRFs. Neural Radiance Field.
- 3 practical use cases for #3DGS Gaussian Splatting 2023-12
- AR, Generative ML in construction and fire asset inspections 2023
Translations
Polski
Wpis opisuje automatyzację przeglądów drzwi przeciwpożarowych. Skanery LiDAR i kamery rejestrują wnętrze budynku jako chmurę punktów i zdjęcia, a uczenie maszynowe klasyfikuje obiekty, takie jak drzwi przeciwpożarowe i oznakowanie ewakuacyjne. Dzięki temu inspektor dostaje listę lokalizacji i możliwych usterek, zamiast obchodzić każdy korytarz z notesem. Same drzwi nadal musi sprawdzić człowiek, ale inwentaryzacja i dokumentacja są znacznie szybsze. Skan jest najbardziej przydatny, gdy zasila cyfrowego bliźniaka budynku aktualizowanego po każdych pracach remontowych.
日本語
この記事は防火扉調査の自動化について解説しています。LiDARスキャナーとカメラで建物内部を点群と画像として記録し、機械学習で防火扉や避難標識などを分類します。これにより、調査員は廊下を一つずつ歩いて記録する代わりに、位置や不具合の可能性の一覧を得られます。扉そのものの点検は人が行う必要がありますが、台帳作成や書類作業は大幅に速くなります。スキャンは、改修のたびに更新される建物のデジタルツインに取り込むことで最も役立ちます。
Filipino
Inilalarawan ng post na ito ang awtomatikong survey ng fire door. Kinukuha ng LiDAR scanner at camera ang loob ng gusali bilang point cloud at mga larawan, at kinikilala ng machine learning ang mga bagay gaya ng fire door at fire signage. Dahil dito, nakakakuha ang surveyor ng listahan ng lokasyon at posibleng depekto sa halip na isa-isahin ang bawat pasilyo. Kailangan pa ring personal na suriin ang mga pinto, pero mas mabilis ang imbentaryo at dokumentasyon. Pinakakapaki-pakinabang ang scan kapag ipinapasok ito sa digital twin ng gusali na ina-update pagkatapos ng bawat renovation.
العربية
تشرح هذه التدوينة أتمتة مسوحات أبواب الحريق. تلتقط ماسحات LiDAR والكاميرات المبنى من الداخل في صورة سحابة نقاط وصور، ثم يصنّف التعلم الآلي العناصر مثل أبواب الحريق ولافتات السلامة. وبذلك يحصل المفتش على قائمة بالمواقع والعيوب المحتملة بدلاً من التجول في كل ممر وتدوين الملاحظات يدوياً. لا يزال فحص الأبواب نفسها مهمة بشرية، لكن الجرد والتوثيق يصبحان أسرع بكثير. ويكون المسح أكثر فائدة عندما يغذي توأماً رقمياً للمبنى يُحدَّث بعد كل أعمال صيانة.
2 thoughts on “Fire Door detection, Building scans , Digital Twins and fire safety.”