Remote Automatic Pressure Gauge Data Acquisition System and Method

Abstract

The present application discloses a remote automatic pressure gauge data acquisition system and method. By obtaining the image data of the pointer-type pressure gauge collected by a camera, optimizing the extraction of the pressure gauge image features through polar coordinate transformation, and further enhancing the accuracy and robustness of pointer recognition by local-global feature phase alignment, the reading accuracy of the pressure gauge reading value is improved. Furthermore, a dynamic analysis model is constructed based on the time series distribution of the pressure gauge reading value to distinguish real equipment failures from environmental interferences and reduce the false alarm rate. In this way, the system can significantly improve the accuracy of the pressure gauge reading under complex working conditions, and through an anomaly detection mechanism that combines dynamic thresholds with statistical laws, achieve reliable early warning of potential hazards, while reducing the need for manual inspections, enhancing the real-time and intelligent level of equipment status monitoring, and providing efficient technical support for industrial site safety management.

Description

Technical Field

[0001] This application relates to the field of intelligent acquisition, and more specifically, to a remote automatic acquisition system and method for pressure gauge data. Background Art

[0002] As a key instrument for monitoring pressure status in industrial sites, pressure gauges are widely used in fields such as electric power, chemical industry, and energy. However, traditional pressure gauges usually only provide local readings and lack the ability of real-time remote data transmission and abnormal warning, resulting in a high dependence on manual inspections for equipment status monitoring. Although there are remote meter reading systems in the prior art, such as monitoring schemes based on dual-channel communication, their functions are limited to data acquisition and transmission, and they cannot perform abnormal diagnosis on the dynamic changes of pressure data, especially it is difficult to distinguish environmental interference from real equipment failures under complex working conditions.

[0003] For pointer-type pressure gauges, existing image recognition technologies mostly rely on feature extraction in a planar coordinate system and are easily affected by perspective deviation or dial deformation, resulting in insufficient reading accuracy; in addition, existing systems mostly use static thresholds for abnormal determination and lack dynamic analysis based on statistical laws, with a high false alarm rate. These problems not only increase the labor cost of equipment management but also may lead to hidden dangers not being processed in time, triggering safety accidents.

[0004] Therefore, an optimized remote automatic acquisition method for pressure gauge data is expected. Summary of the Invention

[0005] To solve the above technical problems, this application is proposed. Embodiments of this application provide a remote automatic acquisition system and method for pressure gauge data, which obtain image data of a pointer-type pressure gauge collected by a camera, optimize the extraction of pressure gauge image features through polar coordinate transformation, and further enhance the accuracy and robustness of pointer recognition by aligning local-global feature phases, thereby improving the reading accuracy of pressure gauge readings. Furthermore, a dynamic analysis model is constructed based on the temporal distribution of pressure gauge readings to distinguish real equipment failures from environmental interference and reduce the false alarm rate. In this way, the system can significantly improve the accuracy of pressure gauge readings under complex working conditions, and through an abnormal detection mechanism that combines dynamic thresholds with statistical laws, achieve reliable early warning of hidden dangers, while reducing the need for manual inspections and enhancing the real-time and intelligent level of equipment status monitoring, providing efficient technical support for industrial site safety management.

[0006] According to one aspect of this application, a remote automatic acquisition method for pressure gauge data is provided, which includes:

[0007] Obtaining image data of a pointer-type pressure gauge collected by a camera;

[0008] Perform coordinate system transformation on the image data of the pointer pressure gauge to transform the image data of the pointer pressure gauge from the planar coordinate system to the polar coordinate system to obtain the image data of the pointer pressure gauge after coordinate system transformation;

[0009] Input the image data of the pointer pressure gauge after coordinate system transformation into the image recognition network to obtain the reading value of the pressure gauge, including: determining the reading value of the pressure gauge based on the phase alignment degree of the local-global features in the image data of the pointer pressure gauge after coordinate system transformation;

[0010] Based on the time series distribution of the reading value of the pressure gauge, determine whether the pressure data is in an abnormal state.

[0011] According to another aspect of the present application, there is provided a remote automatic acquisition system for pressure gauge data, which includes:

[0012] An image acquisition module for acquiring the image data of the pointer pressure gauge collected by the camera;

[0013] A coordinate system transformation module for performing coordinate system transformation on the image data of the pointer pressure gauge to transform the image data of the pointer pressure gauge from the planar coordinate system to the polar coordinate system to obtain the image data of the pointer pressure gauge after coordinate system transformation;

[0014] An image recognition module for inputting the image data of the pointer pressure gauge after coordinate system transformation into the image recognition network to obtain the reading value of the pressure gauge;

[0015] An abnormality judgment module for determining whether the pressure data is in an abnormal state based on the time series distribution of the reading value of the pressure gauge.

[0016] Compared with the prior art, a remote automatic acquisition system and method for pressure gauge data provided by the present application acquire the image data of the pointer pressure gauge collected by the camera, optimize the extraction of the pressure gauge image features through polar coordinate system transformation, and further enhance the accuracy and robustness of pointer recognition by using the phase alignment of local-global features, thereby improving the reading accuracy of the reading value of the pressure gauge. Furthermore, a dynamic analysis model is constructed based on the time series distribution of the reading value of the pressure gauge to distinguish real equipment failures from environmental interferences and reduce the false alarm rate. In this way, the system can significantly improve the accuracy of the pressure gauge reading under complex working conditions, and through the abnormal detection mechanism combining dynamic thresholds and statistical laws, realize reliable early warning of potential hazards, while reducing the need for manual inspections, enhancing the real-time and intelligent level of equipment status monitoring, and providing efficient technical support for industrial site safety management. Description of the Drawings

[0017] The above and other objects, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are used to provide a further understanding of the embodiments of the present application, and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application, and do not constitute a limitation to the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0018] Figure 1 FIG. is a flowchart of a method for remotely and automatically collecting pressure gauge data according to an embodiment of the present application;

[0019] Figure 2 FIG. is a schematic diagram of data flow of a method for remotely and automatically collecting pressure gauge data according to an embodiment of the present application;

[0020] Figure 3 FIG. is a flowchart of sub-step S3 of a method for remotely and automatically collecting pressure gauge data according to an embodiment of the present application;

[0021] Figure 4 FIG. is a flowchart of sub-step S32 of a method for remotely and automatically collecting pressure gauge data according to an embodiment of the present application;

[0022] Figure 5 FIG. is a block diagram of a system for remotely and automatically collecting pressure gauge data according to an embodiment of the present application. Detailed Embodiments

[0023] Hereinafter, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described herein.

[0024] As shown in the present application and the claims, unless the context clearly indicates otherwise, words such as "a", "an", "one", and / or "the" are not specifically singular and may also include plural. Generally speaking, the terms "including" and "comprising" only indicate the inclusion of the clearly identified steps and elements, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0025] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0026] In this application, flowcharts are used to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the operations described above or below do not necessarily have to be performed precisely in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more steps can be removed from these processes.

[0027] Next, exemplary embodiments according to this application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited by the exemplary embodiments described herein.

[0028] In the technical solution of this application, a method for remotely and automatically collecting pressure gauge data is proposed. Figure 1 It is a flowchart of the method for remotely and automatically collecting pressure gauge data according to the embodiments of this application. Figure 2 It is a schematic diagram of data flow of the method for remotely and automatically collecting pressure gauge data according to the embodiments of this application. As Figure 1 and Figure 2 shown, the method for remotely and automatically collecting pressure gauge data according to the embodiments of this application includes the steps: S1, obtaining the image data of the analog pressure gauge collected by the camera; S2, performing coordinate system transformation on the image data of the analog pressure gauge to transform the image data of the analog pressure gauge from the planar coordinate system to the polar coordinate system to obtain the image data of the analog pressure gauge after coordinate system transformation; S3, inputting the image data of the analog pressure gauge after coordinate system transformation into the image recognition network to obtain the pressure gauge reading value; S4, judging whether the pressure data is in an abnormal state based on the time series distribution of the pressure gauge reading value.

[0029] Specifically, in step S1, the image data of the analog pressure gauge collected by the camera is obtained. Among them, the image data of the analog pressure gauge carries key information such as the dial scale, pointer angle, identification text, and environmental background. Considering that traditional manual inspection relies on the naked eye to read the pressure value and cannot achieve real-time remote monitoring, and the image data, as a digital carrier, can completely retain the instantaneous state of the pressure gauge and provide the original input for subsequent automatic processing.

[0030] Specifically, in step S2, coordinate system transformation is performed on the image data of the pointer-type pressure gauge to transform the image data of the pointer-type pressure gauge from a planar coordinate system to a polar coordinate system, so as to obtain the image data of the pointer-type pressure gauge after coordinate system transformation. It should be understood that in an industrial site, due to installation position or angle limitations of the camera, the image of the pointer-type pressure gauge captured may have problems such as inclination and perspective distortion, and it is difficult for image processing in the traditional planar coordinate system to effectively eliminate such geometric distortions. Therefore, in the technical solution of this application, coordinate system transformation is performed on the image data of the pointer-type pressure gauge to transform the image data of the pointer-type pressure gauge from a planar coordinate system to a polar coordinate system, so as to obtain the image data of the pointer-type pressure gauge after coordinate system transformation. It should be understood that the dial of the pointer-type pressure gauge is essentially a circular structure, and its scale distribution and pointer movement both expand at an angle centered on the center of the circle. The right-angle characteristics of the planar coordinate system cannot directly adapt to such circular features, which is likely to cause the scale lines to bend or the pointer angle calculation to deviate. By converting the image from a planar coordinate system to a polar coordinate system, the dial can be expanded as a two-dimensional mapping of angle - radius with the center of the circle as the origin, so that the elliptical dial originally caused by the viewing angle inclination is restored to a standard circular distribution, and the position information of the scale lines and the pointer is recalibrated to a radial linear relationship. The image data of the pointer-type pressure gauge after coordinate system transformation can more clearly present the spatial correlation between local features (such as the position of the pointer tip) and global features (such as the complete scale ring), providing a geometric consistency basis for phase alignment, thereby enhancing the robustness of the image recognition network for judging the pointer angle, reducing the risk of misrecognition caused by dial occlusion, light reflection or mechanical vibration, ensuring the stability and accuracy of the pressure reading under complex working conditions, and providing reliable underlying data support for dynamic anomaly analysis.

[0031] Specifically, in step S3, the image data of the pointer-type pressure gauge after coordinate system transformation is input into the image recognition network to obtain the pressure gauge reading value. That is, in the technical solution of this application, the pressure gauge reading value is determined based on the phase alignment degree of the local - global features in the image data of the pointer-type pressure gauge after coordinate system transformation. Specifically, in a specific example of this application, as Figure 3 shown, step S3 includes: S31, respectively extracting the local feature and the global feature of the image data of the pointer-type pressure gauge after coordinate system transformation to obtain the local encoded feature of the pointer-type pressure gauge image and the global encoded feature of the pointer-type pressure gauge image; S32, performing image local - global joint perception alignment encoding on the local encoded feature of the pointer-type pressure gauge image and the global encoded feature of the pointer-type pressure gauge image to obtain the local - global joint significant perception encoded feature of the pointer-type pressure gauge image; S33, determining the pressure gauge reading value based on the local - global joint significant perception encoded feature of the pointer-type pressure gauge image.

[0032] Specifically, the S31 extracts the local features and global features of the pointer pressure gauge coordinate system transformed image data respectively to obtain the pointer pressure gauge image local coding features and the pointer pressure gauge image global coding features. In an embodiment of the present application, first, the pointer pressure gauge coordinate system transformed image data is passed through an image local feature extractor based on a hole convolutional neural network model to obtain a pointer pressure gauge image local feature coding map as the pointer pressure gauge image local coding feature; it should be understood that although the polar coordinate system image of the pointer pressure gauge corrects the geometric distortion of the dial through transformation, its local details (such as the tip of the pointer, the edge of the scale line, the stains on the dial or the reflective area) still have complex noise interference and multi-scale feature distribution. Traditional convolution kernels are difficult to balance fine-grained feature extraction and context information association under a fixed receptive field, while the dilated convolution expands the receptive field range by interval sampling, which can enable the network to penetrate the local noise layer while maintaining a high-resolution feature map, and capture the topological correlation between the pointer edge and the scale line edge in the radial-angular space, such as the tiny displacement of the pointer tip, the local break of the scale line and other key details, while correlating the redundant information in the dial background (such as the edge of the protective cover or the mounting bracket) to distinguish effective features. Therefore, in the technical solution of the present application, the coordinate system transformation image data of the pointer pressure gauge is passed through an image local feature extractor based on a dilated convolutional neural network model to mine the expression ability of the local features of the coordinate system transformation image of the pointer pressure gauge, and obtain a local feature encoding map of the pointer pressure gauge image. The generated local feature encoding map of the pointer pressure gauge image not only retains the geometric information of the pointer angle, but also suppresses the interference caused by illumination changes or local occlusion through the cross-layer receptive field of the dilated convolution, so that the image recognition network has stronger noise resistance and detail resolution under complex working conditions, thereby ensuring high-precision output of pressure readings and laying a reliable local feature foundation for dynamic anomaly analysis.

[0033] Furthermore, the coordinate system transformed image data of the pointer type pressure gauge is passed through an image global feature extractor based on a transformer to obtain a pointer type pressure gauge image global feature encoded map as the pointer type pressure gauge image global encoded feature. It should be understood that although the perspective distortion has been eliminated in the pressure gauge image after polar coordinate transformation, its global features (such as the overall scale distribution of the dial, the relative position of the pointer to the full scale, and the dial background noise) still need to be semantically understood through context association. Traditional convolutional operations are limited by the local receptive field and are difficult to model the continuous distribution characteristics of the scale lines in the circular structure of the dial or the geometric relationship between the pointer angle and the overall range. However, the transformer can dynamically capture the long-range dependence relationship between any two points in the image through the self-attention mechanism, such as identifying the angular correspondence between the pointer tip and the zero scale and the full scale, or the spatial isolation feature between the dial stain area and the effective scale. Therefore, in the technical solution of this application, in order to establish an abstract representation of the global semantics of the dial, the coordinate system transformed image data of the pointer type pressure gauge is passed through an image global feature extractor based on a transformer to obtain a pointer type pressure gauge image global feature encoded map. The generated pointer type pressure gauge image global feature encoded map can represent the integrity of the dial structure, dynamically focus on key areas (such as the pointer swing interval) through attention weights, suppress irrelevant background interference, so that the image recognition network can still accurately infer the global rationality of the pressure value under conditions of mechanical vibration, uneven illumination or partial occlusion, thereby improving the ability to distinguish environmental interference from real faults in abnormal state discrimination and providing a highly reliable global semantic basis for the dynamic analysis model.

[0034] Specifically, in S32, local-global joint perception alignment coding is performed on the local coding features and global coding features of the analog pressure gauge image to obtain the local-global joint saliency perception coding features of the analog pressure gauge image. It should be understood that although the local feature coding map of the analog pressure gauge image can capture the slight movement of the pointer tip or the wear of the scale lines, it lacks the understanding of the overall range distribution of the dial; while the global feature coding map of the analog pressure gauge image can perceive the circular structure and marking position of the dial, but it is easy to ignore the pixel-level changes in the key areas. A single feature perspective is difficult to comprehensively describe the state of the pressure gauge. For example, the pointer jitter caused by mechanical vibration needs to be analyzed by combining local displacement and global swing range, and the interference of dial stains needs to distinguish the semantic relevance between local noise and the overall background. Therefore, in order to construct a cross-level feature interaction network to complement and enhance the high-resolution details of the local features and the context semantics of the global features of the analog pressure gauge, in the technical solution of this application, local-global joint perception alignment coding is performed on the local coding features and global coding features of the analog pressure gauge image to obtain the local-global joint saliency perception coding features of the analog pressure gauge image. By dynamically searching and aligning the phase relationship between local and global features, the system can break through the problem of isolated feature space in traditional methods. For example, the angle of the pointer tip (local) and the full-scale position (global) are geometrically correlated in the polar coordinate system, and at the same time, the redundant coupling between the edge of the dial protective cover (global interference) and the effective scale line (local key information) is decoupled. The generated local-global joint saliency perception coding features of the analog pressure gauge image not only retain the geometric accuracy of the pointer position, but also incorporate the integrity constraint of the dial structure, enabling the image recognition network to adaptively focus on the key feature areas (such as the pointer swing trajectory) under the interference of uneven illumination, local occlusion or mechanical vibration, suppressing the misleading influence of environmental noise, thereby improving the robustness of pressure readings under complex working conditions and the credibility of abnormal discrimination, and providing a high-quality feature expression with both detail sensitivity and global consistency for the dynamic analysis model. In a specific example of this application, as Figure 4 shown, S32 includes: S321, performing feature decoupling on the local feature coding map and global feature coding map of the analog pressure gauge image to obtain a set of local monomer feature coding matrices and a set of local associated feature coding matrices of the analog pressure gauge image; S322, performing adaptive aggregation analysis based on local-global feature phase alignment on the set of local monomer feature coding matrices and the set of local associated feature coding matrices of the analog pressure gauge image to obtain the local-global joint saliency perception coding map of the analog pressure gauge image as the local-global joint saliency perception coding features of the analog pressure gauge image.

[0035] More specifically, in S321, the local feature encoding map and the global feature encoding map of the pointer-type pressure gauge image are decoupled to obtain a set of local monomer feature encoding matrices of the pointer-type pressure gauge image and a set of local associated feature encoding matrices of the pointer-type pressure gauge image. It should be understood that the processing of pointer-type pressure gauge image data faces multiple challenges such as perspective shift, dial deformation, and illumination interference. When traditional image recognition methods directly operate on the global feature map, it is easy to cause semantic confusion due to too high feature coupling degree. For example, the pointer contour feature interferes with the dial scale background, which in turn affects the reading accuracy. Therefore, in the technical solution of this application, the local feature encoding map and the global feature encoding map of the pointer-type pressure gauge image are decoupled to obtain a set of local monomer feature encoding matrices of the pointer-type pressure gauge image and a set of local associated feature encoding matrices of the pointer-type pressure gauge image. By deconstructing the spatial coupling relationship of the global features, the high-dimensional local feature encoding map and global feature encoding map of the pointer-type pressure gauge image are transformed into a set of local feature matrices of the pointer-type pressure gauge image with clear semantic boundaries, thus breaking through the limitations of traditional methods at the feature representation level. This decoupling is not simply cutting the feature map into isolated fragments, but disassembling the feature map into local monomer feature matrices of the pointer-type pressure gauge image representing local structural information and associated feature matrices describing the associated patterns between the features of the pointer-type pressure gauge image through topological analysis, forming a feature primitive library that can be dynamically combined. In a specific example of this application, the following decoupling formula is used to decouple the local feature encoding map and the global feature encoding map of the pointer-type pressure gauge image to obtain a set of local monomer feature encoding matrices of the pointer-type pressure gauge image and a set of local associated feature encoding matrices of the pointer-type pressure gauge image; where the decoupling formula is:

[0036]

[0037] Wherein, is the local feature encoding map of the pointer-type pressure gauge image, is the global feature encoding map of the pointer-type pressure gauge image, is for performing the feature decoupling operation, and are respectively the first, the second, the th, and the th local monomer feature encoding matrices in the set of local monomer feature encoding matrices of the pointer-type pressure gauge image, and are respectively the first, the second, the th, and the A local correlation feature encoding matrix of a pointer pressure gauge image.

[0038] More specifically, in S322, an adaptive aggregation analysis based on image local-global feature phase alignment is performed on the set of local monomer feature encoding matrices of the pointer-type pressure gauge image and the set of local associated feature encoding matrices of the pointer-type pressure gauge image to obtain a local-global joint saliency perception encoding map of the pointer-type pressure gauge image as the local-global joint saliency perception encoding feature of the pointer-type pressure gauge image. In the embodiments of the present application, first, based on the feature phase alignment degree between any two local monomer feature encoding matrices of the pointer-type pressure gauge image and local associated feature encoding matrices of the pointer-type pressure gauge image in the set of local monomer feature encoding matrices of the pointer-type pressure gauge image and the set of local associated feature encoding matrices of the pointer-type pressure gauge image, a local-global feature phase dynamic search alignment is performed on the set of local monomer feature encoding matrices of the pointer-type pressure gauge image and the set of local associated feature encoding matrices of the pointer-type pressure gauge image to obtain a set of phase-aligned {local monomer feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pairs. It should be understood that when the pressure gauge undergoes slight deformation of the dial due to mechanical vibration, or the camera has a non-normal viewing angle due to installation conditions, traditional static feature matching methods often fail due to the invalidation of the fixed corresponding relationship in space position, resulting in misjudgment of the reading. For example, a certain scale area may have geometric feature compression due to perspective distortion in the plane coordinate system, and although the radial deformation is alleviated after the conversion to the polar coordinate system, the topological relationship between different scale segments and the pointer may still have dynamic offsets due to slight warping of the dial. At this time, if only relying on the preset spatial position corresponding rules, the system cannot adaptively capture the feature association under this non-rigid deformation. Therefore, in the technical solution of the present application, based on the feature phase alignment degree between any two local monomer feature encoding matrices of the pointer-type pressure gauge image and local associated feature encoding matrices of the pointer-type pressure gauge image in the set of local monomer feature encoding matrices of the pointer-type pressure gauge image and the set of local associated feature encoding matrices of the pointer-type pressure gauge image, a local-global feature phase dynamic search alignment is performed on the set of local monomer feature encoding matrices of the pointer-type pressure gauge image and the set of local associated feature encoding matrices of the pointer-type pressure gauge image to obtain a set of phase-aligned {local monomer feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pairs. That is, by dynamically matching the decoupled local monomer features (such as the texture and edges of independent scale segments) with the local associated features (such as the angular relationship between the scale and the pointer), the system can break through the dependence of traditional methods on the strong constraint of spatial position rigidity. For example, when the monomer features in a certain area are distorted in shape due to contamination, the system can dynamically screen other associated features (such as the relative angle change trend of adjacent scale segments) that have a strong logical association with this area through the analysis of the phase alignment degree of the associated features, so as to reconstruct a credible pointer-scale mapping relationship.In this way, the spatial reasoning ability for the true position of the pointer is improved, thereby providing a highly robust feature pair input for subsequent attention weight allocation, and ultimately supporting the stable recognition of pressure readings and the reliable discrimination of abnormal states under complex working conditions. In a specific example of this application, the following alignment formula is used to perform image local-global feature phase dynamic search alignment on the set of local monomer feature encoding matrices of the pointer-type pressure gauge image and the set of local associated feature encoding matrices of the pointer-type pressure gauge image to obtain a set of phase-aligned {local monomer feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pairs; where the alignment formula is:.

[0039]

[0040] Wherein, is the transpose operation, is to calculate the Frobenius norm, is and the feature phase alignment degree between, is to return the value corresponding to the maximum value, is to find the position of the maximum approximate matching value in the set of local associated feature encoding matrices of the pointer-type pressure gauge image, is the trace metric value of the matrix, is the phase-aligned {local monomer feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pair, is and the trace metric value of the pointer-type pressure gauge image between.

[0041] Next, each phase-aligned {local monomer feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pair in the set of phase-aligned {local monomer feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pairs is input into the local-global feature joint perception attention network of the pointer-type pressure gauge image to obtain a set of local-global feature joint perception attention weights of the pointer-type pressure gauge image. It should be understood that under the complex working conditions in the industrial field, the semantic association between feature pairs in the pointer-type pressure gauge image often shows dynamic offsets due to mechanical vibrations, installation tilts, or environmental interferences. When the local monomer feature matrix (such as the pointer tip) and the associated feature matrix (such as the scale line distribution) are phase-aligned, although a preliminary matching relationship is established, there are significant differences in the credibility of different feature pairs in complex deformation or occlusion scenarios. For example, the radial stretching caused by the thermal expansion of the dial may weaken the feature association strength in some scale regions, while the oil stain occlusion may completely destroy the structural consistency of the local features. At this time, if all phase-aligned feature pairs are equally aggregated, the noise features will interfere with the accuracy of reading recognition. More importantly, the local image blurring caused by common oil stains or condensed water mist in the industrial field may form pseudo-alignment relationships between some associated features (such as the adjacent scale spacing relationship) and monomer features (such as the edge gradient of the blurred area), directly threatening the reliability of reading recognition. Therefore, in the technical solution of this application, through the in-depth analysis of the phase-aligned feature pairs by the attention network, the system can identify the hierarchical dependence relationships hidden in the feature space. For example, when a certain monomer feature encoding matrix (such as the pointer axis region) is phase-aligned with multiple associated feature encoding matrices (such as the scale distributions at different radial angles), the attention mechanism can automatically determine the association strength between the axis feature and the scale at a specific angle according to the structural rules in the historical data, and suppress the redundant associations caused by dial fouling. This dynamic weight allocation essentially constructs a self-verification mechanism for the feature contribution degree, enabling the system to maintain the stability of reading inference through the probabilistic coupling relationship between features when the physical state of the dial is abnormal. In this way, when facing local deformation of the dial, the attention network can autonomously enhance the feature weights of the areas not affected by the deformation by analyzing the trace metric parameters of the feature pairs. For example, if the phase of the associated feature is offset due to the warping of the dial edge caused by mechanical stress, but the monomer feature (such as the pointer color contrast) and the associated feature (such as the angle of the centrally symmetric scale) in the central region still maintain high consistency, the system will strengthen the decision-making influence of the feature pairs in the central region through the attention weights.

[0042] Preferably, by introducing an attention mechanism based on the number of biconnectivity and manifold compactification, the topological correlation strength of feature pairs is deeply analyzed. Specifically, after calculating the L1 / L2 distance between feature matrices, the distribution stationarity of feature pairs is quantified by the number of biconnectivity α and β, revealing the structural stability of phase-aligned feature pairs in the manifold space. For example, when a single feature encoding matrix (such as the pointer axis region) is phase-aligned with multiple associated feature encoding matrices (such as the scale distributions at different radial angles), the manifold compactification operation optimizes the embedding of the original matrix through an exponential function, effectively suppressing the divergence of the feature manifold caused by the warping of the dial. By embedding the trace metric parameters optimized by the manifold, the system can autonomously identify highly robust feature pairs: when the edge of the dial undergoes an asymmetric deformation due to mechanical stress, the single feature (such as the pointer color difference contrast) in the central region and the associated feature (such as the angular relationship of symmetric scales) will exhibit a higher number of biconnectivities, triggering the attention network to strengthen the decision weight of the feature pair in this region. This attention mechanism based on manifold topology not only effectively filters out the pseudo-feature correlations caused by local deformations, but also ensures the logical consistency of reading inference under extreme working conditions by preserving the stability of long-range correlations (such as the continuity of scale angles across the deformation region), providing a highly reliable data basis for subsequent dynamic anomaly detection based on temporal distribution and significantly improving the intelligent level of industrial site safety management. Specifically, the specific steps of inputting each phase-aligned {local single feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pair in the set of phase-aligned {local single feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pairs into the local-global feature joint perception attention network of the image are as follows: calculating the correlation matrix between each phase-aligned {local single feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pair in the set of phase-aligned {local single feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pairs to obtain a set of feature pair scale correlation matrices of the pointer-type pressure gauge image; performing trace metric on each feature pair scale correlation matrix in the set of feature pair scale correlation matrices of the pointer-type pressure gauge image to obtain a set of trace metric values of the pointer-type pressure gauge image; and obtaining a set of local-global feature joint perception attention weights of the pointer-type pressure gauge image based on the set of trace metric values of the pointer-type pressure gauge image.Among them, based on the set of trace metric values of the pointer-type pressure gauge image, a set of joint perception attention weights of the local-global features of the pointer-type pressure gauge image is obtained, including: based on the set of trace metric values of the pointer-type pressure gauge image, for each phase-aligned {local monomer feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pair in the set of feature pairs, matrix manifold optimization based on the number of biconnected components is performed to obtain a set of phase-optimized aligned {local monomer feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pairs; based on the set of phase-optimized aligned {local monomer feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pairs, a set of joint perception attention weights of the local-global features of the pointer-type pressure gauge image is obtained.

[0043] Specifically, here, for each and for which matrix trace metric operations are performed, let and , and by calculating the number of eigenvalues satisfying the corresponding and between distance and distance , the number of biconnected components and between and is obtained, that is, the distribution map smoothness measure between the local monomer feature encoding matrix of the pointer-type pressure gauge image and the local associated feature encoding matrix used to represent phase alignment.

[0044] Then, in the discrete manifold representation of the local monomer feature encoding matrix and the local associated feature encoding matrix of the pointer-type pressure gauge image, embedding optimization of the trace metric operations of the local monomer feature encoding matrix and the local associated feature encoding matrix of the pointer-type pressure gauge image is performed through embedding manifold compactification based on the number of biconnected components, so as to improve the calculation accuracy of the joint matrix trace metric on the basis of enhancing the long-range association connectivity robustness:

[0045]

[0046] Among them, is the optimized phase-optimized aligned local monomer feature encoding matrix of the pointer-type pressure gauge image, is The optimized phase-optimized alignment pointer pressure gauge image local correlation feature coding matrix.

[0047] That is, based on the initial and calculated , and In the case of, substitute , , , the initial and for optimization, and then calculate Calculate the optimized trace metric.

[0048] In a specific example of the present application, each phase alignment {pointer pressure gauge image local monomer feature coding matrix, pointer pressure gauge image local correlation feature coding matrix} feature pair in the set of phase alignment {pointer pressure gauge image local monomer feature coding matrix, pointer pressure gauge image local correlation feature coding matrix} features is input into the image local-global feature joint perception attention network according to the following joint perception attention formula to obtain a set of pointer pressure gauge image local-global feature joint perception attention weights; wherein, the joint perception attention formula is:

[0049]

[0050] Wherein, is the feature joint perception attention weight calculation, is the phase-optimized alignment {pointer pressure gauge image local monomer feature coding matrix, pointer pressure gauge image local correlation feature coding matrix} feature pair, is the optimized phase-optimized alignment pointer pressure gauge image local monomer feature coding matrix, is the optimized phase-optimized alignment pointer pressure gauge image local correlation feature coding matrix, is and the optimized pointer pressure gauge image trace metric value between, is the normalization function, is and the pointer pressure gauge image local-global feature joint perception attention weight between.

[0051] Furthermore, based on the set of joint perception attention weights of the local-global features of the pointer-type pressure gauge image, a set of phase-aligned {local monomer feature encoding matrix of the pointer-type pressure gauge image, local correlation feature encoding matrix of the pointer-type pressure gauge image} feature pairs is subjected to attention-driven saliency aggregation to obtain a local-global joint saliency perception encoding map of the pointer-type pressure gauge image. It should be understood that non-rigid deformation of the dial caused by mechanical vibration may lead to misalignment of the association between local features and global features, while perspective shift or uneven illumination will cause significant fluctuations in the feature response intensity of some key regions (such as the pointer tip and scale lines). Traditional feature aggregation methods use equal weights or fixed rules to fuse features and cannot adaptively distinguish high-confidence features from noise interference. For example, when the dial undergoes radial stretching due to thermal expansion, the phase alignment relationship between some scale line features and the pointer may be weakened due to the deformation. If all feature pairs are still aggregated with fixed weights at this time, the low-consistency features in the deformed area will dilute the contribution of effective information, resulting in reading errors. Therefore, in the technical solution of this application, based on the set of joint perception attention weights of the local-global features of the pointer-type pressure gauge image, a set of phase-aligned {local monomer feature encoding matrix of the pointer-type pressure gauge image, local correlation feature encoding matrix of the pointer-type pressure gauge image} feature pairs is subjected to attention-driven saliency aggregation to obtain a local-global joint saliency perception encoding map of the pointer-type pressure gauge image. That is, by introducing an attention-driven saliency aggregation mechanism, the problem of uneven feature contribution degrees under complex working conditions is solved through dynamic weight allocation, breaking through the technical bottleneck of traditional methods being sensitive to noise. In this process, the system can quantify the importance of each phase-aligned feature pair in the global feature expression. For example, when there is oil stain occlusion on the dial, the phase alignment degree between the monomer feature matrix (such as blurred scale lines) and the correlation feature matrix (such as the relative position between the pointer and the scale line) in the contaminated area is low, and the attention network will automatically reduce the weight of such features and instead strengthen the accurately matched features in the non-contaminated area. This dynamic weight allocation is not based on preset physical rules, but through the correlation matrix and trace metric values between feature pairs, the internal dependence relationship between features is mined from a data-driven perspective, enabling the local-global feature interaction in the key area (such as the precise correspondence between the pointer tip and the accurate scale) to dominate the final aggregation process, thereby forming a high-fidelity joint feature expression. In this way, by focusing on high-weight feature pairs, the stability of pointer pointing interpretation can be maintained when there are local deformations or environmental noises on the dial. For example, in a scenario where sudden illumination changes cause the dial to reflect light, the attention mechanism effectively avoids the interference of artifacts on reading recognition by suppressing the feature contribution of the reflective area. At the same time, the manifold compactification process optimized based on the double-connected number further strengthens the robustness of long-range feature association, enabling the aggregated feature map to accurately reflect the true state of the pressure gauge under dynamic working conditions.This intelligent feature fusion strategy not only improves the accuracy of reading recognition but also provides high-quality data input for anomaly diagnosis based on statistical laws.

[0052] In a specific example of this application, the following aggregation formula is used to perform attention-driven significant aggregation on the set of {local monomer feature encoding matrix of the pointer-type pressure gauge image, local correlation feature encoding matrix of the pointer-type pressure gauge image} feature pairs to obtain the local-global joint significant perception encoding map of the pointer-type pressure gauge image; where the aggregation formula is:

[0053]

[0054] Where and are the first joint perception weight matrix and the second joint perception weight matrix, is subtraction by position points, is addition by position points, is and The combined pointer-type pressure gauge image feature joint perception matrix, that is, the th pointer-type pressure gauge image feature joint perception matrix in the set of pointer-type pressure gauge image feature joint perception matrices, , and are respectively the 1st, 2nd, and the rd pointer-type pressure gauge image feature joint perception matrices in the set of pointer-type pressure gauge image feature joint perception matrices, is the local-global joint significant perception encoding map of the pointer-type pressure gauge image.

[0055] Specifically, in S33, based on the local-global joint saliency perception coding features of the pointer-type pressure gauge image, the pressure gauge reading value is determined. That is, in the technical solution of this application, the local-global joint saliency perception coding map of the pointer-type pressure gauge image is input into the pressure gauge reading recognition engine based on a classifier to obtain the pressure gauge reading value. It should be understood that although the local-global joint saliency perception coding map of the pointer-type pressure gauge image contains deep features such as the position of the pointer tip, the distribution of scale lines, and their dynamic correlation, these abstract representations still need to be converted into operable engineering values. Traditional methods directly use regression models to map the relationship between feature vectors and reading values. However, in working conditions with perspective shift, dial deformation, or uneven illumination, such methods are prone to reading jumps or deviations due to the non-linear distortion of the feature space. Therefore, in the technical solution of this application, a reading recognition engine based on a classifier is introduced to transform the continuous pressure value range into a robust classification decision problem through discrete category space modeling, thereby effectively resisting the damage to the continuity of readings caused by feature space perturbations. In this process, the classifier maps the multi-dimensional features in the joint saliency perception coding map to the most matching category label through predefined discrete intervals of pressure values (such as 0.1 MPa as the minimum classification unit). In this way, the false alarm rate caused by environmental factors is reduced, enabling the system to maintain high performance under various conditions, improving the real-time and intelligent level of equipment status monitoring, and providing strong technical support for safety management in industrial sites.

[0056] Particularly, in S4, based on the temporal distribution of the pressure gauge reading values, it is determined whether there is an abnormal state in the pressure data. In one example, basic statistics such as the mean and variance can be calculated to understand the general behavior of the pressure gauge data; autoregressive integrated moving average models (ARIMA), seasonal decomposition of time series analysis (STL), or time series prediction models in machine learning can also be applied to capture the long-term trends and periodic changes in the data. In addition, control charts or dynamic thresholds can be introduced, and reasonable upper and lower limits can be set according to historical data. In response to the current data point exceeding the limit, it is considered that an abnormal situation may exist. It is worth mentioning that considering that some types of faults may not immediately cause the readings to deviate significantly from the normal values but rather show a slow change process, therefore, when determining whether there is an abnormal state, not only the readings at a single time point exceeding the normal range need to be considered, but also the relationships between multiple consecutive time points need to be concerned. Through a detailed analysis of the temporal distribution of the pressure gauge reading values, not only many problems existing in traditional monitoring methods, such as high false alarm rates and low efficiency, are solved, but also strong support is provided for safety management and equipment maintenance in industrial sites.

[0057] In summary, the method for remotely and automatically collecting pressure gauge data according to the embodiments of the present application is elucidated. It obtains the image data of the pointer-type pressure gauge collected by the camera, optimizes the extraction of the pressure gauge image features through polar coordinate system transformation, and further enhances the accuracy and robustness of pointer recognition by aligning the local-global feature phases, thereby improving the accuracy of reading the pressure gauge reading value. Furthermore, a dynamic analysis model is constructed based on the time-series distribution of the pressure gauge reading value to distinguish real equipment failures from environmental interferences and reduce the false alarm rate. In this way, the system can significantly improve the accuracy of the pressure gauge reading under complex working conditions, and through the anomaly detection mechanism that combines dynamic thresholds with statistical laws, achieve reliable early warning of potential hazards, while reducing the need for manual inspections, enhancing the real-time and intelligent levels of equipment status monitoring, and providing efficient technical support for industrial site safety management.

[0058] Furthermore, a system for remotely and automatically collecting pressure gauge data is also provided.

[0059] Figure 5 It is a block diagram of the system for remotely and automatically collecting pressure gauge data according to the embodiments of the present application. As Figure 5 shown, the system 300 for remotely and automatically collecting pressure gauge data according to the embodiments of the present application includes: an image acquisition module 310 for acquiring the image data of the pointer-type pressure gauge collected by the camera; a coordinate system transformation module 320 for performing coordinate system transformation on the image data of the pointer-type pressure gauge to convert the image data of the pointer-type pressure gauge from a planar coordinate system to a polar coordinate system to obtain the image data of the pointer-type pressure gauge after coordinate system transformation; an image recognition module 330 for inputting the image data of the pointer-type pressure gauge after coordinate system transformation into an image recognition network to obtain the pressure gauge reading value; and an anomaly judgment module 340 for judging whether the pressure data is in an abnormal state based on the time-series distribution of the pressure gauge reading value.

[0060] As described above, the system 300 for remotely and automatically collecting pressure gauge data according to the embodiments of the present application can be implemented in various wireless terminals, such as a server with the algorithm for remotely and automatically collecting pressure gauge data. In a possible implementation manner, the system 300 for remotely and automatically collecting pressure gauge data according to the embodiments of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the system 300 for remotely and automatically collecting pressure gauge data can be a software module in the operating system of the wireless terminal, or can be an application program developed for the wireless terminal; of course, the system 300 for remotely and automatically collecting pressure gauge data can also be one of the many hardware modules of the wireless terminal.

[0061] Alternatively, in another example, the pressure gauge data remote automation acquisition system 300 and the wireless terminal may also be discrete devices, and the pressure gauge data remote automation acquisition system 300 can be connected to the wireless terminal through a wired and / or wireless network and transmit interaction information in accordance with a predefined data format.

[0062] The embodiments of the present disclosure have been described above. The above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations are obvious to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The choice of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to technologies in the market, or to enable other ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for remotely and automatically collecting pressure gauge data, characterized in that, Including: Obtaining the image data of the pointer-type pressure gauge collected by the camera; Conducting coordinate system transformation on the image data of the pointer-type pressure gauge to transform the image data of the pointer-type pressure gauge from a planar coordinate system to a polar coordinate system to obtain the image data of the pointer-type pressure gauge after coordinate system transformation; Inputting the image data of the pointer-type pressure gauge after coordinate system transformation into an image recognition network to obtain the reading value of the pressure gauge, including: determining the reading value of the pressure gauge based on the phase alignment degree of the local-global features in the image data of the pointer-type pressure gauge after coordinate system transformation; Judging whether the pressure data is in an abnormal state based on the temporal distribution of the reading values of the pressure gauge; Among them, determining the reading value of the pressure gauge based on the phase alignment degree of the local-global features in the image data of the pointer-type pressure gauge after coordinate system transformation includes: Respectively extracting the local features and global features of the image data of the pointer-type pressure gauge after coordinate system transformation to obtain the local encoded features of the pointer-type pressure gauge image and the global encoded features of the pointer-type pressure gauge image; Conducting image local-global joint perception alignment coding on the local encoded features of the pointer-type pressure gauge image and the global encoded features of the pointer-type pressure gauge image to obtain the local-global joint significant perception encoded features of the pointer-type pressure gauge image, including: decoupling the features of the local feature encoding map of the pointer-type pressure gauge image and the global feature encoding map of the pointer-type pressure gauge image to obtain a set of local monomer feature encoding matrices of the pointer-type pressure gauge image and a set of local associated feature encoding matrices of the pointer-type pressure gauge image; conducting adaptive aggregation analysis based on the phase alignment of the local-global features of the image on the set of local monomer feature encoding matrices of the pointer-type pressure gauge image and the set of local associated feature encoding matrices of the pointer-type pressure gauge image to obtain the local-global joint significant perception encoding map of the pointer-type pressure gauge image as the local-global joint significant perception encoded features of the pointer-type pressure gauge image; Determining the reading value of the pressure gauge based on the local-global joint significant perception encoded features of the pointer-type pressure gauge image.

2. The method for remotely and automatically collecting pressure gauge data according to claim 1, wherein Respectively extracting the local features and global features of the image data of the pointer-type pressure gauge after coordinate system transformation to obtain the local encoded features of the pointer-type pressure gauge image and the global encoded features of the pointer-type pressure gauge image, including: Passing the image data of the pointer-type pressure gauge after coordinate system transformation through an image local feature extractor based on a dilated convolutional neural network model to obtain the local feature encoding map of the pointer-type pressure gauge image as the local encoded features of the pointer-type pressure gauge image; Passing the image data of the pointer-type pressure gauge after coordinate system transformation through an image global feature extractor based on a transformer to obtain the global feature encoding map of the pointer-type pressure gauge image as the global encoded features of the pointer-type pressure gauge image.

3. The method for remotely and automatically collecting pressure gauge data according to claim 2, wherein Conducting adaptive aggregation analysis based on the phase alignment of the local-global features of the image on the set of local monomer feature encoding matrices of the pointer-type pressure gauge image and the set of local associated feature encoding matrices of the pointer-type pressure gauge image to obtain the local-global joint significant perception encoding map of the pointer-type pressure gauge image, including: Based on the feature phase alignment degree between any two local monomer feature coding matrices and local associated feature coding matrices of the pointer-type pressure gauge image in the set of local monomer feature coding matrices of the pointer-type pressure gauge image and the set of local associated feature coding matrices of the pointer-type pressure gauge image, perform image local-global feature phase dynamic search alignment on the set of local monomer feature coding matrices of the pointer-type pressure gauge image and the set of local associated feature coding matrices of the pointer-type pressure gauge image to obtain a set of phase-aligned {local monomer feature coding matrix of the pointer-type pressure gauge image, local associated feature coding matrix of the pointer-type pressure gauge image} feature pairs; Input each phase-aligned {local monomer feature coding matrix of the pointer-type pressure gauge image, local associated feature coding matrix of the pointer-type pressure gauge image} feature pair in the set of phase-aligned {local monomer feature coding matrix of the pointer-type pressure gauge image, local associated feature coding matrix of the pointer-type pressure gauge image} feature pairs into the image local-global feature joint perception attention network to obtain a set of local-global feature joint perception attention weights of the pointer-type pressure gauge image; Based on the set of local-global feature joint perception attention weights of the pointer-type pressure gauge image, perform attention-driven significant aggregation on the set of phase-aligned {local monomer feature coding matrix of the pointer-type pressure gauge image, local associated feature coding matrix of the pointer-type pressure gauge image} feature pairs to obtain a local-global joint significant perception coding map of the pointer-type pressure gauge image.

4. The method for remotely and automatically collecting pressure gauge data according to claim 3, characterized in that, Input each phase-aligned {local monomer feature coding matrix of the pointer-type pressure gauge image, local associated feature coding matrix of the pointer-type pressure gauge image} feature pair in the set of phase-aligned {local monomer feature coding matrix of the pointer-type pressure gauge image, local associated feature coding matrix of the pointer-type pressure gauge image} feature pairs into the image local-global feature joint perception attention network to obtain a set of local-global feature joint perception attention weights of the pointer-type pressure gauge image, including: Calculate the correlation matrix between each phase-aligned {local monomer feature coding matrix of the pointer-type pressure gauge image, local associated feature coding matrix of the pointer-type pressure gauge image} feature pair in the set of phase-aligned {local monomer feature coding matrix of the pointer-type pressure gauge image, local associated feature coding matrix of the pointer-type pressure gauge image} feature pairs to obtain a set of feature pair scale correlation matrices of the pointer-type pressure gauge image; Perform trace metric on each feature pair scale correlation matrix in the set of feature pair scale correlation matrices of the pointer-type pressure gauge image to obtain a set of trace metric values of the pointer-type pressure gauge image; Based on the set of trace metric values of the pointer-type pressure gauge image, obtain a set of local-global feature joint perception attention weights of the pointer-type pressure gauge image.

5. The method for remotely and automatically collecting pressure gauge data according to claim 2, wherein, Based on the set of trace metric values of the pointer-type pressure gauge image, obtain a set of local-global feature joint perception attention weights of the pointer-type pressure gauge image, including: Based on the set of trace metric values of the pointer-type pressure gauge image, perform matrix manifold optimization based on the number of biconnectivities for each phase-aligned {local monomer feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pair in the set of phase-aligned {local monomer feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pairs to obtain a set of phase-optimized aligned {local monomer feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pairs; Based on the set of phase-optimized aligned {local monomer feature encoding matrix of the pointer-type pressure gauge image, local associated feature encoding matrix of the pointer-type pressure gauge image} feature pairs, obtain a set of local-global feature joint perception attention weights of the pointer-type pressure gauge image.

6. The method for remotely and automatically collecting pressure gauge data according to claim 2, wherein Based on the local-global joint saliency perception encoding features of the pointer-type pressure gauge image, determine the pressure gauge reading value, including: Input the local-global joint saliency perception encoding map of the pointer-type pressure gauge image into the pressure gauge reading recognition engine based on the classifier to obtain the pressure gauge reading value.

7. A remote automatic pressure gauge data acquisition system for implementing the remote automatic pressure gauge data acquisition method described in claim 1, characterized in that, Including: An image acquisition module for acquiring pointer-type pressure gauge image data collected by a camera; A coordinate system transformation module for performing coordinate system transformation on the pointer-type pressure gauge image data to transform the pointer-type pressure gauge image data from a planar coordinate system to a polar coordinate system to obtain pointer-type pressure gauge coordinate system transformed image data; An image recognition module for inputting the pointer-type pressure gauge coordinate system transformed image data into an image recognition network to obtain the pressure gauge reading value; An anomaly judgment module for judging whether the pressure data is in an abnormal state based on the temporal distribution of the pressure gauge reading values.

Images