Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Pile length real-time detection method based on self-attention mechanism deep learning
The invention discloses a pile length real-time detection method based on self-attention mechanism deep learning, and the method comprises the following steps: S1, installing a plurality of cameras at a construction site, and collecting site pictures and videos; s2, constructing a target recognition model to recognize the precast pile; s3, constructing a target tracking model to detect and track the precast pile in real time; and S4, the length of the precast pile is measured. The method has the beneficial effects that compared with traditional manual measurement, the multiple cameras are installed on a construction site to collect site pictures and videos, the target recognition model recognizes the precast pile, and the target tracking model detects and tracks the precast pile in real time; and a long-short-term self-attention mechanism is introduced, so that the model adapts to a complex construction site environment, and the pile length is monitored in real time in the pile sinking process of the precast pile.
本发明公开了一种基于自注意力机制深度学习的桩长实时检测方法，包括以下步骤：S1：将若干摄像头安装到施工现场，并采集现场图片和视频；S2：构建目标识别模型对预制桩进行识别；S3：构建目标追踪模型对预制桩进行实时检测跟踪；S4：进行预制桩桩长测量。本发明的有益效果是：本发明相较于传统的人工测量，通过将若干摄像头安装到施工现场采集现场图片和视频，目标识别模型对预制桩进行识别以及目标追踪模型对预制桩进行实时检测跟踪，并引入长短期自注意力机制使模型适应复杂的施工现场环境，实现预制桩沉桩过程中对桩长实时监测。
Pile length real-time detection method based on self-attention mechanism deep learning
The invention discloses a pile length real-time detection method based on self-attention mechanism deep learning, and the method comprises the following steps: S1, installing a plurality of cameras at a construction site, and collecting site pictures and videos; s2, constructing a target recognition model to recognize the precast pile; s3, constructing a target tracking model to detect and track the precast pile in real time; and S4, the length of the precast pile is measured. The method has the beneficial effects that compared with traditional manual measurement, the multiple cameras are installed on a construction site to collect site pictures and videos, the target recognition model recognizes the precast pile, and the target tracking model detects and tracks the precast pile in real time; and a long-short-term self-attention mechanism is introduced, so that the model adapts to a complex construction site environment, and the pile length is monitored in real time in the pile sinking process of the precast pile.
本发明公开了一种基于自注意力机制深度学习的桩长实时检测方法，包括以下步骤：S1：将若干摄像头安装到施工现场，并采集现场图片和视频；S2：构建目标识别模型对预制桩进行识别；S3：构建目标追踪模型对预制桩进行实时检测跟踪；S4：进行预制桩桩长测量。本发明的有益效果是：本发明相较于传统的人工测量，通过将若干摄像头安装到施工现场采集现场图片和视频，目标识别模型对预制桩进行识别以及目标追踪模型对预制桩进行实时检测跟踪，并引入长短期自注意力机制使模型适应复杂的施工现场环境，实现预制桩沉桩过程中对桩长实时监测。
Pile length real-time detection method based on self-attention mechanism deep learning
一种基于自注意力机制深度学习的桩长实时检测方法
LI HONGWU (Autor:in) / WU WEI (Autor:in) / LU SILIANG (Autor:in) / HAN CHAO (Autor:in) / TANG ZIQIANG (Autor:in) / GU KAIXIN (Autor:in) / LIU YONGHAO (Autor:in) / FAN ZHOU (Autor:in) / SUN KE (Autor:in) / KONG SHUOYING (Autor:in)
01.03.2024
Patent
Elektronische Ressource
Chinesisch
IPC:
E02D
FOUNDATIONS
,
Gründungen
/
G01B
MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS
,
Messen der Länge, der Dicke oder ähnlicher linearer Abmessungen
/
G06N
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
,
Rechnersysteme, basierend auf spezifischen Rechenmodellen
/
G06T
Bilddatenverarbeitung oder Bilddatenerzeugung allgemein
,
IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
/
G06V
Foundation pile breakage and pile length detection method
| Europäisches Patentamt | 2020
| Europäisches Patentamt | 2023
Pile foundation over-filling real-time detection equipment and method
| Europäisches Patentamt | 2024