CN-121982325-A - Method, device, system and medium for evaluating pretreatment effect of optical fiber sensing signal

CN121982325ACN 121982325 ACN121982325 ACN 121982325ACN-121982325-A

Abstract

The invention provides a method, a device, a system and a medium for evaluating the preprocessing effect of an optical fiber sensing signal, wherein the method comprises the steps of obtaining an original space-time waterfall diagram; the method comprises the steps of obtaining at least one filtering space-time waterfall diagram which is obtained by filtering an original optical fiber sensing signal based on at least one filtering algorithm, carrying out binarization processing on the original space-time waterfall diagram to obtain an original binarization image, identifying and marking a region of interest on the original binarization image based on morphological operation, determining a first signal characteristic region in the original space-time waterfall diagram and a second signal characteristic region in the filtering space-time waterfall diagram based on the region of interest, determining a multi-dimensional evaluation index of the filtering space-time waterfall diagram based on the first signal characteristic region and the second signal characteristic region, and determining a comprehensive score based on the multi-dimensional evaluation index. According to the invention, through the multi-dimensional evaluation index, the problem of single blind area and dimension in subjective evaluation is solved, and the accuracy of comprehensive scoring is improved.

Inventors

JIN DEFA
DONG ZHIYONG
ZHAO PENG
YANG YUE

Assignees

武汉理工光科股份有限公司

Dates

Publication Date: 20260505
Application Date: 20251231

Claims (10)

1. The method for evaluating the preprocessing effect of the optical fiber sensing signal is characterized by comprising the following steps of: acquiring an original space-time waterfall diagram corresponding to an original optical fiber sensing signal acquired by an optical fiber sensing system; Acquiring at least one filtering space-time waterfall diagram corresponding to at least one filtering signal obtained by filtering the original optical fiber sensing signal based on at least one filtering algorithm; Performing binarization processing on the original space-time waterfall diagram to obtain an original binarized image, and identifying and marking a region of interest on the original binarized image based on morphological operation; Determining a first signal characteristic region in the original space-time waterfall diagram and a second signal characteristic region in the filtered space-time waterfall diagram based on the region of interest; determining a characteristic intensity retention index, a characteristic space focusing index, a background noise suppression index, a cross interference suppression index and a characteristic space-time continuity index of the filtered space-time waterfall map based on the first signal characteristic region and the second signal characteristic region; And determining a comprehensive score for evaluating the filtering algorithm based on the characteristic intensity retention index, the characteristic space power index, the background noise suppression index, the cross interference suppression index and the characteristic space-time continuity index.
2. The method for evaluating the preprocessing effect of the optical fiber sensing signal according to claim 1, wherein the step of obtaining the original space-time waterfall diagram corresponding to the original optical fiber sensing signal acquired by the optical fiber sensing system comprises the steps of: Acquiring an original optical fiber sensing signal acquired by an optical fiber sensing system; taking the time sequence as a vertical axis, taking the spatial distribution of the optical fiber sensors in the optical fiber sensing system as a horizontal axis, and converting the original optical fiber sensing signals into a two-dimensional matrix; and visualizing the two-dimensional matrix into a gray level image or a pseudo color image to form the original space-time waterfall diagram.
3. The method of claim 1, wherein determining a first signal feature region in the original spatiotemporal waterfall plot and a second signal feature region in the filtered spatiotemporal waterfall plot based on the region of interest comprises: Determining the first signal feature region based on the coordinate values of the region of interest; And mapping the interested region into the filtering space-time waterfall diagram based on the coordinate mapping relation of the original space-time waterfall diagram and the filtering space-time waterfall diagram to obtain the second signal characteristic region.
4. The method for evaluating the preprocessing effect of the optical fiber sensing signal according to claim 1, wherein the characteristic intensity retention index is: the characteristic space focusing index is as follows: the background noise inhibition index is as follows: the cross interference suppression degree index is as follows: the characteristic space-time continuity index is as follows: In the formula, Reserving a degree index for the characteristic intensity; Is the sum of all pixel gray values in the first signal characteristic region; Is the sum of all pixel gray values in the second signal characteristic region; Is a characteristic space focusing index; An average value of continuous pixel widths with gray values larger than a preset threshold value in each row in the first signal characteristic region; an average value of continuous pixel widths with gray values larger than a preset threshold value in each row in the second signal characteristic region; is a background noise inhibition index; setting the sum of the pixel gray values of the residual areas in the original binarized image after the pixel gray values of all the pixels in the first signal characteristic area are set to zero; Filtering the sum of the pixel gray values of the residual areas in the binarized image after setting all the pixel gray values in the second signal characteristic areas to zero; Is a cross interference suppression degree index; an average of the ratios of the second most energy sum to the maximum energy sum in each sensor channel for the first signal feature region; an average of the ratios of the second most energy sum to the maximum energy sum in each sensor channel for the second signal feature region; is a characteristic space-time continuity index; a total number of rows of pixels occupied by the second signal feature region; Is the total number of breaks in the second signal characteristic region.
5. The method according to claim 1, wherein determining the composite score for evaluating the filtering algorithm based on the characteristic intensity retention index, characteristic spatial power index, background noise suppression index, cross-interference suppression index, and characteristic spatiotemporal continuity index comprises: Calculating a weighted sum of the characteristic intensity retention index, the characteristic space focusing index and the characteristic space-time continuity index to obtain a characteristic local index; Calculating a weighted sum of the background noise suppression degree index and the cross interference suppression degree index to obtain a noise global index; And calculating the weighted sum of the characteristic local index and the noise global index to obtain the comprehensive score.
6. The method according to claim 5, wherein weights of the characteristic intensity retention index, the characteristic spatial focusing index, the characteristic spatiotemporal continuity index, the background noise suppression index, the cross interference suppression index, the characteristic local index, and the noise global index are determined based on performance requirements of an application scene.
7. The method for evaluating the preprocessing effect of the optical fiber sensing signal according to claim 1, characterized in that the method further comprises: and displaying the filtering parameters and the corresponding comprehensive scores of the filtering algorithms on a human-computer interaction interface.
8. An optical fiber sensing signal preprocessing effect evaluation device is characterized by comprising: the original space-time waterfall diagram acquisition unit is used for acquiring an original space-time waterfall diagram corresponding to an original optical fiber sensing signal acquired by the optical fiber sensing system; a filtering space-time waterfall diagram obtaining unit, configured to obtain at least one filtering space-time waterfall diagram corresponding to at least one filtering signal obtained after filtering the original optical fiber sensing signal based on at least one filtering algorithm; the interesting region labeling unit is used for carrying out binarization processing on the original space-time waterfall diagram to obtain an original binarized image, and identifying and labeling interesting regions on the original binarized image based on morphological operation; A signal feature region determining unit configured to determine a first signal feature region in the original spatiotemporal waterfall diagram and a second signal feature region in the filtered spatiotemporal waterfall diagram based on the region of interest; A multidimensional index determining unit, configured to determine a characteristic intensity retention index, a characteristic space degree index, a background noise suppression degree index, a cross interference suppression degree index, and a characteristic space-time continuity index of the filtering space-time waterfall map based on the first signal characteristic region and the second signal characteristic region; And the comprehensive scoring unit is used for determining the comprehensive score for evaluating the filtering algorithm based on the characteristic intensity retention index, the characteristic space focusing index, the background noise suppression index, the cross interference suppression index and the characteristic space-time continuity index.
9. An optical fiber demodulation system is characterized by comprising an optical fiber sensing system, a memory and a processor, wherein, The optical fiber sensing system is used for collecting original optical fiber sensing signals; the memory is used for storing programs; The processor, coupled to the memory and the optical fiber sensing system, is configured to execute the program stored in the memory, so as to implement the steps in the optical fiber sensing signal preprocessing effect evaluation method according to any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer-readable program or instructions that, when executed by a processor, is capable of implementing the steps in the optical fiber sensing signal preprocessing effect evaluation method according to any one of claims 1 to 7.

Description

Method, device, system and medium for evaluating pretreatment effect of optical fiber sensing signal Technical Field The invention relates to the technical field of intelligent transportation, in particular to a method, a device, a system and a medium for evaluating pretreatment effect of an optical fiber sensing signal. Background The optical fiber sensor has been widely used for monitoring the running track of vehicles in intelligent traffic to collect weak dynamic signals such as speed and acceleration due to the characteristics of electromagnetic interference resistance, small volume, light weight, long transmission distance, high sensitivity and the like. However, the actually collected optical fiber signals often overlap with various environmental noises such as temperature drift, mechanical vibration, light source intensity fluctuation and the like, and transmission attenuation and device thermal noise, so that important track characteristics are covered, and the accuracy of subsequent demodulation, positioning and traffic situation analysis is affected. To obtain high quality sensing data, the raw fiber optic signal must be filtered to suppress noise and highlight useful features. However, different filtering algorithms such as IIR, FIR, kalman filtering and the like and parameter settings thereof have trade-offs between noise suppression and useful signal retention, wherein excessive smoothing of parameters can cause loss of signal details to influence dynamic response, and incorrect parameter selection can cause serious noise residues to influence discrimination accuracy. At present, the evaluation of the pretreatment effect mainly depends on the experience of engineers to subjectively score or locally evaluate through single indexes such as signal-to-noise ratio, root mean square error and the like. These methods have certain disadvantages. The method has the advantages that the subjectivity is strong, the reproduction is difficult, the manual scoring is greatly influenced by experience and personal judgment, different engineers are difficult to agree, the evaluation dimension is single, the noise suppression capability, the useful information retention degree and the dynamic characteristic response cannot be considered at the same time, the filtering performance is difficult to comprehensively reflect, the optimization efficiency of the guide parameters is low, the multidimensional feedback is lacked, the adjustment direction of the filtering parameters cannot be accurately positioned, and the follow-up algorithm tuning and automatic deployment are influenced. Therefore, it is needed to provide a method, a device, a system and a medium for evaluating the preprocessing effect of the optical fiber sensing signal, which can realize the comprehensive and accurate evaluation of the filtering effect of the filtering algorithm and provide scientific basis for the selection and the self-adaptive adjustment of the filtering algorithm. Disclosure of Invention In view of the foregoing, it is necessary to provide a method, a device, a system and a medium for evaluating the preprocessing effect of an optical fiber sensing signal, which are used for solving the technical problems that in the prior art, the filtering effect of a filtering algorithm is evaluated by relying on a single-dimension evaluation parameter, the evaluation is inaccurate, and therefore scientific basis cannot be provided for selection and self-adaptive adjustment of the filtering algorithm. In order to solve the above technical problems, in a first aspect, the present invention provides a method for evaluating a preprocessing effect of an optical fiber sensing signal, including: acquiring an original space-time waterfall diagram corresponding to an original optical fiber sensing signal acquired by an optical fiber sensing system; Acquiring at least one filtering space-time waterfall diagram corresponding to at least one filtering signal obtained by filtering the original optical fiber sensing signal based on at least one filtering algorithm; Performing binarization processing on the original space-time waterfall diagram to obtain an original binarized image, and identifying and marking a region of interest on the original binarized image based on morphological operation; Determining a first signal characteristic region in the original space-time waterfall diagram and a second signal characteristic region in the filtered space-time waterfall diagram based on the region of interest; determining a characteristic intensity retention index, a characteristic space focusing index, a background noise suppression index, a cross interference suppression index and a characteristic space-time continuity index of the filtered space-time waterfall map based on the first signal characteristic region and the second signal characteristic region; And determining a comprehensive score for evaluating the filtering algorithm based on the characteristic inten