CN-122021846-A - Space-time knowledge graph database system for construction environment monitoring

CN122021846ACN 122021846 ACN122021846 ACN 122021846ACN-122021846-A

Abstract

The invention discloses a space-time knowledge graph database system for construction environment monitoring, which relates to the technical field of construction monitoring and is used for solving the problem of low analysis efficiency of environmental change causes, acquiring construction area environmental image data through remote sensing image recognition, screening and dividing the environmental image data, matching the environmental image data with a spectrum database to obtain environmental spectrum characteristics, extracting ground object types and environmental change attributes based on the environmental spectrum characteristics, evaluating environmental evolution trend through analyzing area evolution states and combining the ground object types, acquiring construction activity data from the construction database, analyzing the space-time matching characteristics, screening marks for related construction activities, detecting the construction change states of the marked construction activity data after setting evaluation time, evaluating causal relation indexes through combined environmental evolution trends, deciding whether to write the construction activity data into the knowledge graph database according to index results, and improving the intelligent level and decision efficiency of environment supervision through establishing a dynamic relation model of the environment evolution characteristics and the construction activity states.

Inventors

ZHOU YI
CHEN FEI
ZHANG YING
LI HAO
CAO LEI
LIU YINGYING
SHI JIANBO
DU DEJUN
ZHOU ZHIQIANG
GU JUN

Assignees

国网湖北省电力有限公司神农架供电公司
国网湖北省电力有限公司电力科学研究院

Dates

Publication Date: 20260512
Application Date: 20251208

Claims (9)

1. The space-time knowledge graph database system for construction environment monitoring is characterized by comprising an environment recognition module, an evolution detection module, a space-time matching module and a cause and effect analysis module, wherein the functions of the modules are as follows: the environment recognition module recognizes environment image data of a construction area to be detected through the remote sensing image, screens and divides the environment image area according to the environment image data, and matches corresponding environment spectrum characteristics after the environment image area is imported into the spectrum database and is imported into the evolution detection module; The evolution detection module extracts the ground object type and the environment change attribute of the environment image area by utilizing the environment spectrum characteristics, analyzes the area evolution state according to the environment change attribute, evaluates the evolution trend of the environment image area by combining the ground object type and transmits the evolution trend to the causal analysis module; The space-time matching module is used for accessing the construction database to obtain construction activity data of a construction area to be detected, analyzing space-time matching characteristics according to the construction activity data, screening and marking the construction activity data by utilizing the space-time matching characteristics, and transmitting the construction activity data to the causal analysis module; the causal analysis module is used for setting evaluation time, detecting construction change states of the marked construction activity data, evaluating causal association indexes by combining the construction change states and evolution trend of the environment image area, and judging whether the knowledge graph database is written according to the causal association indexes.
2. The spatiotemporal knowledge graph database system for construction environment monitoring of claim 1, wherein: In an environment recognition module, acquiring a remote sensing image of a construction area to be detected through unmanned aerial vehicle near-field aerial photography carrying a remote sensing sensor; Preprocessing a remote sensing image of a construction area to be detected, and performing unsupervised semantic segmentation recognition on the construction area to be detected based on a deep convolutional neural network model to generate environment image data of the construction area to be detected; an image dividing unit recorded in the environmental image data is used as a preliminary environmental image candidate region.
3. The spatiotemporal knowledge graph database system for construction environment monitoring of claim 2, wherein: in the environment recognition module, if the coverage area of each preliminary environment image area candidate is greater than or equal to a preset area threshold value, judging that the preliminary environment image area candidate is an environment image area; Otherwise, judging the environment image candidate area as a noise area; Matching each environment image area with a spectrum database to obtain environment spectrum characteristics corresponding to each environment image area; The environmental spectral feature refers to a spectral parameter obtained by counting the pixel reflectivity of an environmental image region in a multispectral band.
4. The spatiotemporal knowledge graph database system for construction environment monitoring of claim 1, wherein: in the evolution detection module, calculating to obtain the similarity between the environmental spectral characteristics of each environmental image area and the spectral characteristic standard value of each ground object type in the spectral characteristic standard database by a spectral angle matching method; obtaining the ground object type of each environment image area based on the similarity; matching each environment image area with a historical environment database to obtain a coverage area of each environment image area during last monitoring; And taking the absolute value of the subtraction of the current coverage area and the coverage area of the last detection of one environment image area as the environment change attribute of the environment image area.
5. The spatiotemporal knowledge graph database system for construction environment monitoring of claim 4, wherein: In the evolution detection module, dividing the environmental change attribute of each environmental image area by the coverage area in the last detection to obtain the area evolution state of each environmental image area; carrying out standardization processing on the area evolution state of each environment image area to obtain an area evolution factor; And calculating a preset ground feature type factor corresponding to the ground feature type of each environment image area by the comprehensive area evolution factor to obtain the evolution trend of each environment image area.
6. The spatiotemporal knowledge graph database system for construction environment monitoring of claim 1, wherein: in the space-time matching module, matching the to-be-constructed area with a construction database to obtain construction occurrence time stamps and construction disturbance center coordinates of all construction activities in the to-be-constructed area; Matching the construction area to be detected with an environment image database to obtain geographic boundary coordinates and change identification time stamps of the environment image area in the construction area to be detected; Calculating the space distance between each construction disturbance center coordinate and the geographic boundary coordinate of each environment image area according to the Euclidean distance calculation method, and taking the space distance as the space proximity between each construction activity and each environment image area; And taking the difference value between the construction occurrence time stamp of the construction activity and the change identification time stamp of the environment image area as the time association degree.
7. The spatiotemporal knowledge graph database system for construction environment monitoring of claim 6, wherein: In the space-time matching module, the spatial proximity and the time correlation of each construction activity and each environment image area are standardized to obtain a spatial proximity factor and a time proximity factor; The space-time matching characteristics are obtained by integrating the space proximity factors and the time proximity factors; If the space-time matching characteristic is greater than or equal to a preset matching characteristic threshold, judging that space-time correlation exists between the construction activity data and the environment image area, and marking the construction activity data; otherwise, judging that the construction activity data and the environment image area have no space-time correlation, and not marking.
8. The spatiotemporal knowledge graph database system for construction environment monitoring of claim 1, wherein: In the causal analysis module, taking the absolute value of the difference value between the construction occurrence time stamp marked with construction activity data and the change identification time stamp of the corresponding environment image area as the evaluation time; counting the number of changes of construction disturbance center coordinates of marked construction activity data in the evaluation time; Dividing the number of times of change of the construction disturbance center coordinates of the marked construction activity data in the evaluation time by the evaluation time to obtain a construction change state of the marked construction activity data; Carrying out standardized processing on the construction change state of the marked construction activity data and the evolution trend of the environment image area to obtain a construction change factor and an evolution trend factor; And calculating the causal relation index by combining the construction change factor and the evolution trend factor.
9. The spatiotemporal knowledge graph database system for construction environment monitoring of claim 8, wherein: in the causal analysis module, if the causal association index is greater than or equal to a preset association index threshold, causal relation exists between the marked construction activity and the evolution trend of the corresponding environment image area, and the marked construction activity data and the environment image data of the corresponding environment image area are written into the knowledge graph database; If the causal association index is smaller than the preset association index threshold, no causal relation exists between the construction activity and the evolution trend of the corresponding environment image area, and the knowledge graph database is not written.

Description

Space-time knowledge graph database system for construction environment monitoring Technical Field The invention relates to the technical field of construction monitoring, in particular to a space-time knowledge graph database system for construction environment monitoring. Background Along with the continuous expansion of the construction scale of the infrastructure, the influence of large-scale construction activities on surrounding natural environment, ecological structure and land utilization pattern presents a trend of complicacy, dynamization and regionalization, and the traditional construction environment monitoring mainly depends on site inspection, fixed point position monitoring equipment and artificial experience judgment, and has limited monitoring range and insufficient timeliness, especially in engineering scenes with long construction period and multiple construction links. The prior art has the following defects: At present, the prior art is difficult to uniformly fuse and dynamically associate and model the multisource remote sensing image, the spectral characteristics and the construction activity data, the space-time matching and causal association analysis between the environment evolution process and the construction activity state cannot be realized, and the quantitative evaluation mechanism for the causal relationship among the environment change attribute, the ground object type evolution trend and the construction dynamic state is lacked, so that the analysis efficiency of the environment change cause is low, the construction influence is difficult to realize automatic traceability and interpretable expression, and therefore, the space-time knowledge map database system for the construction environment monitoring is provided. The above information disclosed in the background section is only for enhancement of understanding of the background of the disclosure and therefore it may include information that does not form the prior art that is already known to a person of ordinary skill in the art. Disclosure of Invention In order to overcome the above-mentioned drawbacks of the prior art, embodiments of the present invention provide a space-time knowledge graph database system for construction environment monitoring, which solves the problems set forth in the above-mentioned background art by applying remote sensing image recognition, spectral feature extraction, multi-source construction data space-time matching, and environmental evolution and construction activity causal analysis methods. In order to achieve the above purpose, the invention provides a technical scheme that a space-time knowledge graph database system for construction environment monitoring comprises an environment recognition module, an evolution detection module, a space-time matching module and a causal analysis module, wherein the functions of each module are as follows: the environment recognition module recognizes environment image data of a construction area to be detected through the remote sensing image, screens and divides the environment image area according to the environment image data, and matches corresponding environment spectrum characteristics after the environment image area is imported into the spectrum database and is imported into the evolution detection module; The evolution detection module extracts the ground object type and the environment change attribute of the environment image area by utilizing the environment spectrum characteristics, analyzes the area evolution state according to the environment change attribute, evaluates the evolution trend of the environment image area by combining the ground object type and transmits the evolution trend to the causal analysis module; The space-time matching module is used for accessing the construction database to obtain construction activity data of a construction area to be detected, analyzing space-time matching characteristics according to the construction activity data, screening and marking the construction activity data by utilizing the space-time matching characteristics, and transmitting the construction activity data to the causal analysis module; the causal analysis module is used for setting evaluation time, detecting construction change states of the marked construction activity data, evaluating causal association indexes by combining the construction change states and evolution trend of the environment image area, and judging whether the knowledge graph database is written according to the causal association indexes. In a preferred embodiment, in the environment recognition module, acquiring a remote sensing image of a construction area to be detected through near-to-ground aerial photographing of an unmanned aerial vehicle carrying a remote sensing sensor; Preprocessing a remote sensing image of a construction area to be detected, and performing unsupervised semantic segmentation recognition on the construction area to be detected ba