阅读说明：本技术 一种基于视觉感知的桥梁挠度的监测方法、装置及系统 (Method, device and system for monitoring bridge deflection based on visual perception ) 是由 康春光 赵文一 陈宇轩 何显银 宋杰 张亮 胡辉 于 2019-09-26 设计创作，主要内容包括：本发明公开了一种基于视觉感知的桥梁挠度的监测方法、装置、系统及存储介质,该方法包括：获取设定时间段内各个时间节点视觉传感器采集的桥梁的监测图像；识别各个所述监测图像的目标特征点,并计算各个所述监测图像的目标特征点的初始挠度信号,以获得初始挠度信号序列；将设定个数的互不相同的高斯白噪声分别与所述初始挠度信号序列叠加,以获取设定个数的叠加挠度信号序列；基于希尔伯特-黄变换对各个所述叠加挠度信号序列进行滤波,以获取所述桥梁在所述设定时间段内各个时间节点的挠度。本发明实施例的技术方案,实现了桥梁挠度的自动监测,有效减少了由于相机振动、温度等引起的测量误差,提高了监测精度。(The invention discloses a method, a device and a system for monitoring bridge deflection based on visual perception and a storage medium, wherein the method comprises the following steps: acquiring monitoring images of the bridge acquired by each time node vision sensor in a set time period; identifying target characteristic points of each monitoring image, and calculating an initial deflection signal of the target characteristic points of each monitoring image to obtain an initial deflection signal sequence; superposing a set number of different Gaussian white noises with the initial deflection signal sequence respectively to obtain a set number of superposed deflection signal sequences; and filtering each superposed deflection signal sequence based on Hilbert-Huang transform to obtain the deflection of each time node of the bridge in the set time period. According to the technical scheme of the embodiment of the invention, the automatic monitoring of the bridge deflection is realized, the measurement errors caused by camera vibration, temperature and the like are effectively reduced, and the monitoring precision is improved.)

1. A method for monitoring bridge deflection based on visual perception is characterized by comprising the following steps:

acquiring monitoring images of the bridge acquired by each time node vision sensor in a set time period;

identifying target characteristic points of each monitoring image, and calculating an initial deflection signal of the target characteristic points of each monitoring image to obtain an initial deflection signal sequence;

superposing a set number of different Gaussian white noises with the initial deflection signal sequence respectively to obtain a set number of superposed deflection signal sequences;

and filtering each superposed deflection signal sequence based on Hilbert-Huang transform to obtain the deflection of each time node of the bridge in the set time period.

2. The method of claim 1, wherein the monitoring image comprises a two-dimensional code pattern.

3. The method of claim 2, wherein the identifying target feature points of each of the monitored images comprises:

and identifying target characteristic points of the two-dimensional code pattern of each monitoring image.

4. The method of claim 1, wherein the filtering each of the superimposed deflection signal sequences based on Hilbert-Huang transform to obtain the deflection of each time node of the bridge in the set time period comprises:

performing empirical mode decomposition on each superposed deflection signal sequence to obtain an eigenmode function set of each superposed deflection sequence, wherein the empirical mode decomposition is performed for multiple times, and the eigenmode function set consists of eigenmode functions obtained by each decomposition;

calculating the average eigenmode function of each eigenmode function obtained by each decomposition;

performing Hilbert transform on each average eigenmode function to obtain a Hilbert spectrum which is the relation among time, frequency and amplitude of the average eigenmode function;

and screening the average eigenmode function according to the Hilbert spectrum to acquire the deflection of each time node of the bridge in the set time period.

5. The method of claim 4, wherein the step of screening the average eigenmode function according to the Hilbert spectrum to obtain the deflection of each time node of the bridge in the set time period comprises:

according to the Hilbert spectrum, removing all average eigenmode functions of which the frequency is lower than a set frequency and the amplitude fluctuation does not accord with a set rule to obtain a high-frequency eigenmode function;

and superposing the high-frequency eigenmode functions to obtain the deflection of each time node of the bridge in the set time period.

6. The method of claim 1, after obtaining the deflection of the bridge at each time node within the set time period, further comprising:

and carrying out abnormal value zeroing treatment on the deflection of each time node according to the amplitude of the deflection.

7. The method of claim 6, wherein said performing an outlier zeroing process on the deflection of each of said time nodes based on the magnitude of said deflection comprises:

calculating the average value and the standard deviation of the amplitude values of the deflection of all time nodes in the set time;

and when the difference between the amplitude of the deflection of the current time node and the average value is more than 3 times of the standard deviation, carrying out abnormal value zeroing processing on the deflection of the current time node.

8. The utility model provides a monitoring devices of bridge amount of deflection based on vision perception which characterized in that includes:

the monitoring image acquisition module is used for acquiring monitoring images of the bridge acquired by the vision sensors at each time node within a set time period;

the initial deflection calculation module is used for identifying the target characteristic points of the monitoring images and calculating initial deflection signals of the target characteristic points of the monitoring images so as to obtain an initial deflection signal sequence;

the superposed deflection acquisition module is used for superposing a set number of different Gaussian white noises with the initial deflection signal sequence respectively to acquire a set number of superposed deflection signal sequences;

and the deflection filtering module is used for filtering each superposed deflection signal sequence based on Hilbert-Huang transform so as to obtain the deflection of each time node of the bridge in the set time period.

9. A system for monitoring bridge deflection based on visual perception, the system comprising: the system comprises a vision sensor and a signal processing module, wherein the vision sensor is used for acquiring a monitoring image of the bridge; the signal processing module is used for executing the monitoring method of bridge deflection in any one of claims 1-7.

10. A storage medium containing computer executable instructions for performing the method of monitoring bridge deflection of any one of claims 1-7 when executed by a computer processor.

Technical Field

The embodiment of the invention relates to the technical field of image processing, in particular to a method, a device and a system for monitoring bridge deflection based on visual perception and a storage medium.

Background

With the rapid development of the traffic industry, various infrastructures generate aging and diseases with different degrees, and the regular detection and maintenance of bridges serving as important components of traffic infrastructures also become increasingly important. Generally speaking, the deflection is the most direct parameter for representing the performance of the bridge, and therefore, the monitoring of the deflection of the bridge becomes an important link for operation and maintenance and safety evaluation of the bridge.

The monitoring of bridge amount of deflection can be realized through setting up the vision sensor, however because the special operating mode of bridge monitoring, the erect position of vision sensor is far away from the monitoring point usually, and to the bridge that the span is big, the working radius can reach more than 500 meters. When a telephoto camera is used for long-distance imaging, relatively obvious image jitter is easy to occur between adjacent frames, and high temperature at noon can also cause image distortion, so that the problems can not be solved optically. Meanwhile, the camera itself generates a certain low frequency vibration in the monitoring process, which generates a large monitoring error.

Disclosure of Invention

The invention provides a method, a device, equipment and a storage medium for monitoring bridge deflection based on visual perception, which are used for realizing automatic monitoring of bridge displacement, effectively removing monitoring environmental errors and errors generated by camera vibration and improving monitoring precision.

In a first aspect, an embodiment of the present invention provides a method for monitoring bridge deflection based on visual perception, where the method includes:

acquiring monitoring images of the bridge acquired by each time node vision sensor in a set time period;

identifying target characteristic points of each monitoring image, and calculating an initial deflection signal of the target characteristic points of each monitoring image to obtain an initial deflection signal sequence;

superposing a set number of different Gaussian white noises with the initial deflection signal sequence respectively to obtain a set number of superposed deflection signal sequences;

and filtering each superposed deflection signal sequence based on Hilbert-Huang transform to obtain the deflection of each time node of the bridge in the set time period.

In a second aspect, an embodiment of the present invention further provides a device for monitoring bridge deflection based on visual perception, where the device includes:

the monitoring image acquisition module is used for acquiring monitoring images of the bridge acquired by the vision sensors at each time node within a set time period;

the initial deflection calculation module is used for identifying the target characteristic points of the monitoring images and calculating initial deflection signals of the target characteristic points of the monitoring images so as to obtain an initial deflection signal sequence;

the superposed deflection acquisition module is used for superposing a set number of different Gaussian white noises with the initial deflection signal sequence respectively to acquire a set number of superposed deflection signal sequences;

and the deflection filtering module is used for filtering each superposed deflection signal sequence based on Hilbert-Huang transform so as to obtain the deflection of each time node of the bridge in the set time period.

In a third aspect, an embodiment of the present invention further provides a system for monitoring bridge deflection based on visual perception, where the system includes:

the system comprises a vision sensor and a signal processing module, wherein the signal processing module is used for executing the monitoring method for the bridge deflection provided by any embodiment of the invention.

In a fourth aspect, an embodiment of the present invention further provides a storage medium containing computer-executable instructions, where the computer-executable instructions are used to execute the method for monitoring bridge deflection provided in any embodiment of the present invention when executed by a computer processor.

According to the technical scheme of the embodiment of the invention, the monitoring image of the bridge is acquired through the visual sensor, the monitoring is easy to realize, and the influence on the bridge is small by adopting a non-contact mode for monitoring; the bridge deflection is determined by identifying the characteristic points of the monitoring image, the identification precision is high, and the bridge deflection is not easily interfered by the environment; the deflection signal and the set number of Gaussian white noises are superposed, and the superposed signal is filtered based on Hilbert-Huang transform, so that low-frequency interference generated by camera vibration is effectively reduced, and the deflection monitoring precision is further improved.

Drawings

FIG. 1 is a flowchart of a method for monitoring bridge deflection based on visual perception according to a first embodiment of the present invention;

FIG. 2 is a flow chart of a method for monitoring bridge deflection based on visual perception in a second embodiment of the invention;

FIG. 3 is a schematic view of a device for monitoring bridge deflection based on visual perception in a third embodiment of the invention;

fig. 4 is a schematic diagram of a beam deflection monitoring system based on visual perception in the fourth embodiment of the invention.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.