Light field data dimension reduction compressed sensing method and system applied to ultra-high voltage transformer substation

Abstract

The invention relates to a light field data dimension reduction compressed sensing method and system applied to an ultra-high voltage transformer substation. The method comprises the following steps: constructing an image array based on an image obtained by use and an image generation position; determining a monitoring object in the image; constructing a compression area on the image based on the monitoring object; comparing the contents of the monitoring objects in the adjacent images to obtain a content comparison result, and determining a monitoring object reference image; constructing a reconstruction matrix according to the content comparison result; performing sparse sampling on a non-monitoring object area outside the compression area to obtain a sampling result; and outputting the reference image of the monitoring object, the reconstruction matrix and the sampling result as a compressed sensing result. According to the light field data dimensionality reduction compressed sensing method and system applied to the ultra-high voltage transformer substation disclosed by the invention, real-time monitoring based on the light field data is realized by carrying out regional processing on the light field data and balancing a conversion processing mode of a monitored object among early-stage data compression, data transmission and data decompression.

Description

Technical Field

[0001] The present invention relates to the technical field of image data compression, and in particular to a method and system for reducing the dimension of light field data compression perception applied to an ultra-high voltage substation. Background Art

[0002] In the power grid system, substations, as the core hub of the network, are the core of the real integration of electricity into the network. Substation operation risks may lead to large-scale power outages. Due to the scale and complexity of substations, traditional video surveillance systems may not be able to provide comprehensive coverage and high-quality images, which may lead to monitoring blind spots and affect the real-time monitoring of key equipment and operating areas. The main reasons are as follows: The limited installation location leads to monitoring blind spots; The existing monitoring system displays the existing camera images in a single-screen layout. The chaotic structure of the images and the weak correlation between the images make it impossible to fully understand the operation status of the entire station through the monitoring images. The monitoring contents of each monitoring image are highly similar, making it difficult to quickly confirm the monitoring target. Then, conducting on-site inspections is cumbersome. When an abnormality appears on the screen, it is impossible to immediately confirm the location of the abnormality, which affects the rapid response and handling of the abnormal status on site.

[0003] Existing Hawkeye surveillance cameras achieve simple splicing of surveillance images through multiple cameras arranged horizontally. However, the surveillance images are severely distorted, details are blurred, and the images are fragmented. Except for the center part that can be viewed normally, important details of the areas on both sides are difficult to obtain, and the real surveillance scene cannot be intuitively displayed.

[0004] Light field imaging is an important means to achieve three-dimensional imaging. Light field imaging technology captures the distribution information of light in space (including the position and direction of light), providing a rich data foundation for subsequent image processing and three-dimensional reconstruction.

[0005] However, light field imaging has the problem of huge data volume, which means huge data processing volume for real-time monitoring. The current wireless transmission method and the network are difficult to meet the demand. The use of compressed sensing technology can over-compress the image, but this means that over-recovery calculations are required before playback. How to balance data compression and image quality requires further research. Summary of the invention

[0006] The present invention provides a method and system for reducing the dimension of light field data and compressing the perception of it for ultra-high voltage substations. By performing regional processing on light field data and converting the processing mode of monitoring objects, a balance is achieved between early data compression, data transmission and data decompression, thereby realizing real-time monitoring based on light field data.

[0007] The above object of the present invention is achieved through the following technical solutions: In a first aspect, the present invention provides a method for reducing the dimension of light field data and compressing the sensing method applied to an ultra-high voltage substation, comprising: An image array is constructed based on the obtained images and the image generation positions, wherein the images in the image array are arranged in an MxN matrix form, where M and N are both natural numbers greater than zero; Identify or determine the monitored object in the image according to the received instruction; A compression region is constructed on the image based on the monitored object, and the monitored object is located inside the compression region; Comparing the contents of the monitored object in adjacent images, obtaining a content comparison result and determining a baseline image of the monitored object; Construct a reconstruction matrix based on the content comparison results; Sparse sampling is performed on the non-monitored object area outside the compressed area to obtain a sampling result; The monitored object reference image, reconstruction matrix and sampling results are output as compressed sensing results.

[0008] In a possible implementation manner of the first aspect, constructing a reconstruction matrix according to the content comparison result includes: The two monitoring objects in the adjacent images are respectively recorded as the standard monitoring object and the substitute monitoring object; Determine a set of change directions of the monitored object according to changes in the outlines of the standard monitored object and the alternative monitored object; Determine the change content of the standard monitoring object based on the contour change; Determine a set of spatial change modes according to a set of change directions; Determine light analysis areas in the surrounding directions of the standard monitoring object and the substitute monitoring object and calculate the intensity value of each light analysis area; Construct a set of light impact parameters using the positions of the light analysis areas and the intensity values ​​of each light analysis area; The set of spatial variation methods and the set of light influence parameters are used as a reconstruction matrix.

[0009] In a possible implementation manner of the first aspect, the light analysis area surrounds a standard monitoring object position, and distances from the light analysis area to a designated position on the standard monitoring object are all equal.

[0010] In a possible implementation manner of the first aspect, sparse sampling is performed on the added content of the standard monitoring object.

[0011] In a possible implementation manner of the first aspect, performing sparse sampling includes: Establishing expression curves in the horizontal or vertical direction of the image; The expression curve is transferred into the sparse domain for expression and the peak value in the sparse domain is determined, and the number of the peak value is multiple; Calculate the correlation value of two adjacent peaks; When the correlation value of any two adjacent peaks is less than or equal to the set threshold, the expression curve is randomly subsampled.

[0012] In a possible implementation manner of the first aspect, the method further includes: When the correlation value between any two adjacent peaks is greater than the set threshold, the expression curve is decomposed using a filter to obtain a sub-expression curve; Check the correlation values ​​of adjacent peaks of the sub-expression curve in the sparse domain to ensure that the correlation values ​​of adjacent peaks are less than or equal to the set threshold; Sub-expression curves were randomly subsampled; When the correlation value between any two adjacent peaks on the sub-expression curve is greater than a set threshold, the filter is adjusted to regenerate the sub-expression curve or the sub-expression curve is decomposed again.

[0013] In a possible implementation manner of the first aspect, the method further includes: Use color channels to decompose the image into red, green, and blue images; Expression curves are established in the horizontal direction or the vertical direction of the red image, the green image, and the blue image, respectively.

[0014] In a second aspect, the present invention provides a light field data dimension reduction and compression sensing device applied to an ultra-high voltage substation, comprising: An image processing unit, used for constructing an image array based on the obtained image and the image generation position, wherein the images in the image array are arranged in a matrix form of MxN, where M and N are both natural numbers greater than zero; A content recognition unit, used to recognize or determine the monitored object in the image according to the received instruction; A region construction unit, used for constructing a compression region on the image based on the monitored object, wherein the monitored object is located inside the compression region; A content comparison unit, used to compare the contents of the monitored object in adjacent images, obtain a content comparison result and determine a baseline image of the monitored object; A matrix construction unit, used for constructing a reconstruction matrix according to the content comparison result; A sampling unit, used for performing sparse sampling on the non-monitoring object area outside the compression area to obtain a sampling result; The result output unit is used to output the monitored object reference image, reconstruction matrix and sampling result as compressed sensing results.

[0015] In a third aspect, the present invention provides a light field data dimension reduction compressed sensing system applied to an ultra-high voltage substation, the system comprising: one or more memories for storing instructions; and One or more processors, used to call and run the instructions from the memory to execute the method as described in the first aspect and any possible implementation of the first aspect.

[0016] In a fourth aspect, the present invention provides a computer-readable storage medium, the computer-readable storage medium comprising: Program, when the program is executed by a processor, the method described in the first aspect and any possible implementation of the first aspect is executed.

[0017] In a fifth aspect, the present invention provides a computer program product, comprising program instructions, and when the program instructions are executed by a computing device, the method described in the first aspect and any possible implementation of the first aspect is executed.

[0018] In a sixth aspect, the present invention provides a chip system comprising a processor for implementing the functions involved in the above aspects, for example, generating, receiving, sending, or processing the data and / or information involved in the above methods.

[0019] The chip system may be composed of chips, or may include chips and other discrete devices.

[0020] In a possible design, the chip system also includes a memory, which is used to store necessary program instructions and data. The processor and the memory can be decoupled and respectively set on different devices, connected by wired or wireless means, or the processor and the memory can also be coupled on the same device. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a schematic flow chart of the steps of a light field data dimensionality reduction and compressed sensing method provided by the present invention.

[0022] Figure 2 It is a schematic diagram of the introduction of a two-plane light field representation method provided by the present invention.

[0023] Figure 3 It is a schematic diagram of the principle of image acquisition by a camera array provided by the present invention.

[0024] Figure 4 It is a schematic diagram for comparing contour changes of a standard monitoring object and a substitute monitoring object provided by the present invention.

[0025] Figure 5It is a schematic diagram of replacing a monitoring object and increasing or decreasing content based on a standard monitoring object provided by the present invention.

[0026] Figure 6 It is a schematic diagram of establishing a light analysis area provided by the present invention.

[0027] Figure 7 It is a schematic diagram of dividing an image array into a plurality of sub-image arrays provided by the present invention. DETAILED DESCRIPTION

[0028] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings.

[0029] The present invention discloses a method for reducing the dimension of light field data and compressing the sensing method applied to ultra-high voltage substations. Figure 1 In some examples, the light field data dimension reduction and compressed sensing method for ultra-high voltage substation disclosed in the present invention includes the following steps: S101, constructing an image array based on the obtained images and image generation positions, wherein the images in the image array are arranged in an MxN matrix form, where M and N are both natural numbers greater than zero; S102, identifying or determining a monitoring object in the image according to the received instruction; S103, constructing a compression region on the image based on the monitored object, wherein the monitored object is located inside the compression region; S104, comparing the contents of the monitored object in adjacent images, obtaining a content comparison result and determining a monitored object reference image; S105, constructing a reconstruction matrix according to the content comparison result; S106, performing sparse sampling on the non-monitored object area outside the compressed area to obtain a sampling result; S107, outputting the monitored object reference image, the reconstruction matrix and the sampling result as a compressed sensing result.

[0030] First, let's introduce the light field. The light field is the distribution of a physical quantity of light in space. Simply put, the light field describes the intensity of light in any direction at any point in space, and the full light function that fully describes the light field is a 7-dimensional function, including the position of any point (x, y, z), any direction (Θ, Φ in polar coordinates), wavelength (λ) and time (t).

[0031] In practical applications, information in color and time dimensions is usually represented by RGB channels and different frames, so for light fields, we only need to focus on the direction and position of light, which reduces the dimension from 7 to 5. In general, in most imaging systems, light propagates in a limited optical path, so a simpler way of representing the light field with two planes is introduced, such as Figure 2 shown.

[0032] See also Figure 3 The device used to collect the light field is a camera array. Each image sensor in the camera array will collect images. Because the positions of the image sensors are different, there are differences between the images in the obtained image group.

[0033] Taking a specific scenario as an example, ordinary equipment only collects one image, while the light field camera needs to collect 64 (8x8) images. By comparison, it can be seen that light field data has a natural disadvantage in terms of data volume. For monitoring, it is mainly manifested in two aspects: data transmission and data decompression.

[0034] In step S101 of the present invention, an image array is first constructed based on the obtained images and the image generation positions, and the image array is used to determine the position of each image. The images in the image array are arranged in an MxN matrix form, where M and N are both natural numbers greater than zero.

[0035] In step S102, the monitoring object in the image is identified or determined according to the received instruction. The monitoring object here refers to a specific device in the ultra-high voltage substation. The device can be obtained by using an identification method or specified according to the instruction.

[0036] In step S103, a compression area is constructed on the image based on the monitored object, and the monitored object is located inside the compression area, or it can be described that the edge of the compression area is the edge of the monitored object. The compression area on the image and the area outside the compression area are compressed using different methods.

[0037] Next, in step S104, the contents of the monitored object in adjacent images are compared to obtain a content comparison result and determine a monitored object reference image, and then a reconstruction matrix is ​​constructed according to the content comparison result, which is the content in step S105.

[0038] In step S106, sparse sampling is performed on the non-monitored object area outside the compressed area to obtain a sampling result. Finally, in step S107, the monitored object reference image, the reconstruction matrix and the sampling result are output as compressed sensing results.

[0039] In general, the light field data dimensionality reduction and compressed sensing method applied to ultra-high voltage substations provided by the present invention adopts a separate processing method to divide the images in the image array into baseline images and reference images. The baseline images correspond to images including the baseline images of the monitored objects, and the reference images correspond to images not including the baseline images of the monitored objects.

[0040] The image is divided into compressed areas and sparsely collected areas by using a region division method. Lossless compression or limited lossy compression is used for the compressed areas, while compressed sensing is used for the sparsely collected areas. The sparsely collected areas on the reference image are processed by compressed sensing, while the compressed areas on the reference image are generated based on the object reference image on the reference image.

[0041] The advantage of this method is that it greatly reduces the amount of data transmission, while retaining more content in the compressed area. In addition, the monitored object is displayed by generating an object reference image based on a reference image.

[0042] In some examples, the specific method of constructing a reconstruction matrix according to the content comparison result is: S201, recording two monitoring objects in adjacent images as a standard monitoring object and a substitute monitoring object respectively; S202, determining a set of change directions of the monitored object according to changes in the outlines of the standard monitored object and the substitute monitored object; S203, determining the change content of the standard monitoring object according to the contour change; S204, determining a set of spatial change modes according to the set of change directions; S205, determining light analysis areas in the surrounding directions of the standard monitoring object and the substitute monitoring object and calculating the intensity value of each light analysis area; S206, constructing a light impact parameter set using the position of the light analysis area and the intensity value of each light analysis area; S207, using the space variation mode set and the light influence parameter set as a reconstruction matrix.

[0043] In step S201 to step S207, firstly, two monitoring objects in adjacent images are recorded as standard monitoring objects and substitute monitoring objects respectively. Figure 4 ,Then, the change direction set of the monitored object is determined according to the ,contour changes of the standard monitored object and the alternative monitored ,object.

[0044] Adjacent images are generated based on two image sensors respectively. The positions of these two image sensors are different, so the outline of the standard monitoring object and the outline of the substitute monitoring object will also be different. When the outline of the standard monitoring object is converted into the outline of the substitute monitoring object, spatial movement is involved. There are six degrees of freedom specified here, namely X-axis movement, Y-axis movement, Z-axis movement, X-axis rotation, Y-axis rotation, and Z-axis rotation.

[0045] At this time, in step S102, the change direction set of the monitored object is determined according to the contour changes of the standard monitored object and the substitute monitored object. The specific method is to move the standard monitored object within the spatial range so that it can overlap with the contour of the substitute monitored object. At this time, the overlap will produce a moving vacant area, which means that the substitute monitored object has increased content or reduced content based on the standard monitored object, such as Figure 5 shown.

[0046] To move the standard monitoring object within the spatial range, it is necessary to refer to the relative position of the two image sensors. Since the relative position of the two image sensors is fixed, the movement of the standard monitoring object within the spatial range is also fixed. For example, if one image sensor is to the left of another image sensor, the movement of the standard monitoring object within the spatial range must first be performed by X-axis movement and Y-axis rotation, and then fine-tuned according to the contrast difference between the outline of the standard monitoring object and the outline of the alternative monitoring object.

[0047] At this time, the changed content of the standard monitoring object can be obtained, and at the same time, a set of spatial change methods can be obtained. The changed content of the standard monitoring object refers to the content that is added or reduced when the standard monitoring object is converted into a replacement monitoring object. The set of spatial change methods refers to the process of obtaining the outline of the replacement monitoring object based on the outline of the standard monitoring object.

[0048] In step S205, refer to Figure 6 , the light analysis area will be determined in the surrounding directions of the standard monitoring object and the alternative monitoring object and the intensity value of each light analysis area will be calculated. The light analysis area refers to an area of ​​a fixed size. The light analysis area is composed of multiple areas such that these areas are located on the same straight line.

[0049] The intensity value of the light analysis area is calculated by averaging the grayscale values ​​in the light analysis area.

[0050] Then, in step S206, the position of the light analysis area and the intensity value of each light analysis area are used to construct a light impact parameter set, which includes the position, intensity value and intensity value change of the light analysis area. Through this change, the color of the replacement monitoring object can be adjusted, because when using the standard monitoring object to generate the replacement monitoring object, it is not possible to only translate the content.

[0051] Finally, in step S207, the spatial change mode set and the light impact parameter set are used as a reconstruction matrix. In the subsequent recovery process, the spatial change mode set can be used as a basis for adding or reducing the content of the standard monitoring object, and the comparison of the two light impact parameter sets (standard monitoring object, replacement monitoring object) can be used as a basis for adjusting the color of the standard monitoring object.

[0052] In some possible implementations, the light analysis area surrounds the position of the standard monitoring object, and the distance from the light analysis area to the specified position on the standard monitoring object is equal. Of course, the same method is used to process the replacement monitoring object.

[0053] In some possible implementations, the additional content of the standard monitoring object is sparsely sampled. This is because if the additional content of the standard monitoring object is compressed using a lossless compression or a limited lossy compression method, it means that the standard monitoring object is fragmented and the data transmission volume is increased.

[0054] Furthermore, considering that the additional content replacing the monitored object is limited, a sparse sampling method is used for processing in the present invention.

[0055] At this time, another issue needs to be considered, which is to limit the amount of content added to replace the monitored object. One way to achieve this is to limit it based on the area of ​​the added area. When the area of ​​the added area that replaces the monitored object exceeds the set allowable value, a new standard monitored object will be added. That is, for the image array mentioned above, the image array will be divided into multiple sub-image arrays. For example, an image array (8x8) can be divided into four image arrays (4x4). Figure 7 shown.

[0056] The specific method of sparse sampling is as follows: S301, establishing an expression curve in the horizontal direction or the vertical direction of the image; S302, transferring the expression curve into a sparse domain for expression and determining a peak value in the sparse domain, where the number of peak values ​​is multiple; S303, calculating the correlation value of two adjacent peaks; S304: When the correlation value between any two adjacent peaks is less than or equal to a set threshold, the expression curve is randomly sub-sampled.

[0057] In step S304, when the correlation value between any two adjacent peaks is greater than a set threshold, the expression curve is decomposed using a filter to obtain a sub-expression curve; Check the correlation values ​​of adjacent peaks of the sub-expression curve in the sparse domain to ensure that the correlation values ​​of adjacent peaks are less than or equal to the set threshold; Sub-expression curves were randomly subsampled; When the correlation value between any two adjacent peaks on the sub-expression curve is greater than a set threshold, the filter is adjusted to regenerate the sub-expression curve or the sub-expression curve is decomposed again.

[0058] In the above method, you can also use the following method for pre-processing, as follows: Use color channels to decompose the image into red, green, and blue images; Expression curves are established in the horizontal direction or the vertical direction of the red image, the green image, and the blue image, respectively.

[0059] The purpose of decomposition is to increase the retained content, because sparse acquisition means more data loss. When red images, green images and blue images are mixed together, the sparse acquisition method has difficulty in distinguishing details. When the red images, green images and blue images are disassembled, it means that the monochrome image can be sparsely acquired, which can retain more details.

[0060] In the subsequent synthesis process, the red image, the green image and the blue image are restored first, and then the restored red image, the green image and the blue image are mixed together.

[0061] The present invention also provides a light field data dimension reduction and compression sensing device applied to an ultra-high voltage substation, comprising: An image processing unit, used for constructing an image array based on the obtained image and the image generation position, wherein the images in the image array are arranged in a matrix form of MxN, where M and N are both natural numbers greater than zero; A content recognition unit, used to recognize or determine the monitored object in the image according to the received instruction; A region construction unit, used for constructing a compression region on the image based on the monitored object, wherein the monitored object is located inside the compression region; A content comparison unit, used to compare the contents of the monitored object in adjacent images, obtain a content comparison result and determine a baseline image of the monitored object; A matrix construction unit, used for constructing a reconstruction matrix according to the content comparison result; A sampling unit, used for performing sparse sampling on the non-monitoring object area outside the compression area to obtain a sampling result; The result output unit is used to output the monitored object reference image, reconstruction matrix and sampling result as compressed sensing results.

[0062] Furthermore, it also includes: a marking unit, used for marking two monitoring objects in adjacent images as a standard monitoring object and a substitute monitoring object respectively; A change direction set processing unit, used to determine a change direction set of a monitoring object according to contour changes of a standard monitoring object and a substitute monitoring object; A change content processing unit, used to determine the change content of the standard monitoring object according to the profile change; A spatial variation mode set processing unit, used to determine a spatial variation mode set according to a variation direction set; A light content processing unit, used to determine light analysis areas in the surrounding directions of the standard monitoring object and the substitute monitoring object and calculate the intensity value of each light analysis area; A result building unit is used to build a light impact parameter set using the position of the light analysis area and the intensity value of each light analysis area; The result output unit is used to use the set of spatial variation modes and the set of light influence parameters as a reconstruction matrix.

[0063] Furthermore, the light analysis area surrounds the position of the standard monitoring object, and the distances from the light analysis area to the designated position on the standard monitoring object are all equal.

[0064] Furthermore, the added content of the standard monitoring objects is sparsely sampled.

[0065] Furthermore, it also includes: A first expression curve establishing unit, used to establish an expression curve in a horizontal direction or a vertical direction of the image; A sparse domain processing unit, used for transferring the expression curve into the sparse domain for expression and determining a plurality of peak values ​​in the sparse domain; A correlation value calculation unit, used for calculating the correlation value of two adjacent peaks; The first sub-sampling unit is used to perform random sub-sampling on the expression curve when the correlation value between any two adjacent peak values ​​is less than or equal to a set threshold.

[0066] Furthermore, it also includes: An expression curve decomposition unit is used to decompose the expression curve using a filter to obtain a sub-expression curve when the correlation value between any two adjacent peaks is greater than a set threshold; An expression curve checking unit is used to check the correlation values ​​of adjacent peaks of the sub-expression curve in the sparse domain to ensure that the correlation values ​​of adjacent peaks are less than or equal to a set threshold; A second subsampling unit, used for randomly subsampling the sub-expression curves; When the correlation value between any two adjacent peaks on the sub-expression curve is greater than a set threshold, the filter is adjusted to regenerate the sub-expression curve or the sub-expression curve is decomposed again.

[0067] Furthermore, it also includes: An image decomposition unit, used for decomposing the image into a red image, a green image and a blue image using color channels; The second expression curve establishing unit is used to establish expression curves in the horizontal direction or the vertical direction of the red image, the green image and the blue image respectively.

[0068] In one example, the unit in any of the above devices can be one or more integrated circuits configured to implement the above methods, such as: one or more application specific integrated circuits (ASICs), or, one or more digital signal processors (DSPs), or, one or more field programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0069] For another example, when the units in the device can be implemented in the form of a processing element scheduling program, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call a program. For another example, these units can be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0070] Various objects such as various messages / information / equipment / network elements / systems / devices / actions / operations / processes / concepts that may appear in the present invention are named. It can be understood that these specific names do not constitute a limitation on the relevant objects. The names assigned may change with factors such as scenarios, contexts or usage habits. The understanding of the technical meaning of the technical terms in the present invention should be mainly determined from the functions and technical effects embodied / executed in the technical scheme.

[0071] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0072] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0073] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0074] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0075] It should also be understood that in various embodiments of the present invention, the first, second, etc. are only used to indicate that multiple objects are different. For example, the first time window and the second time window are only used to indicate different time windows. They should not have any impact on the time window itself, and the first, second, etc. mentioned above should not impose any limitations on the embodiments of the present invention.

[0076] It should also be understood that in the various embodiments of the present invention, unless otherwise specified or there is any logical conflict, the terms and / or descriptions between the different embodiments are consistent and can be referenced to each other, and the technical features in the different embodiments can be combined to form new embodiments according to their internal logical relationships.

[0077] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a computer-readable storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned computer-readable storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0078] The present invention also provides a light field data dimension reduction compressed sensing system applied to an ultra-high voltage substation, the system comprising: one or more memories for storing instructions; and One or more processors are used to call and run the instructions from the memory to execute the method as described above.

[0079] The present invention also provides a computer program product, which includes instructions. When the instructions are executed, the terminal device and the network device perform operations of the terminal device and the network device corresponding to the above method.

[0080] The present invention also provides a chip system, which includes a processor for implementing the functions involved in the above content, such as generating, receiving, sending, or processing the data and / or information involved in the above method.

[0081] The chip system may be composed of chips, or may include chips and other discrete devices.

[0082] The processor mentioned in any of the above places can be a CPU, a microprocessor, an ASIC, or one or more integrated circuits for executing programs for controlling the above-mentioned feedback information transmission method.

[0083] In a possible design, the chip system also includes a memory, which is used to store necessary program instructions and data. The processor and the memory can be decoupled and respectively set on different devices, connected by wire or wireless means to support the chip system to implement various functions in the above embodiments. Alternatively, the processor and the memory can also be coupled on the same device.

[0084] Optionally, the computer instructions are stored in a memory.

[0085] Optionally, the memory is a storage unit within the chip, such as a register, a cache, etc. The memory can also be a storage unit within the terminal located outside the chip, such as a ROM or other types of static storage devices that can store static information and instructions, RAM, etc.

[0086] It can be understood that the memory in the present invention can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories.

[0087] The non-volatile memory may be a ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), or a flash memory.

[0088] The volatile memory may be a RAM, which is used as an external cache. There are many different types of RAM, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct memory bus RAM.

[0089] The embodiments of this specific implementation method are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for reducing the dimension of light field data and compressing the sensing method applied to ultra-high voltage substations, characterized in that: include: An image array is constructed based on the obtained images and the image generation positions, wherein the images in the image array are arranged in an MxN matrix form, where M and N are both natural numbers greater than zero; Identify or determine the monitored object in the image according to the received instruction; A compression region is constructed on the image based on the monitored object, and the monitored object is located inside the compression region; Comparing the contents of the monitored object in adjacent images, obtaining a content comparison result and determining a baseline image of the monitored object; Construct a reconstruction matrix based on the content comparison results; Sparse sampling is performed on the non-monitored object area outside the compressed area to obtain a sampling result; The monitored object reference image, reconstruction matrix and sampling results are output as compressed sensing results.

2. The method for reducing dimension and compressing sensing of light field data applied to ultra-high voltage substation according to claim 1 is characterized in that: Constructing a reconstruction matrix based on the content comparison results includes: The two monitoring objects in the adjacent images are respectively recorded as the standard monitoring object and the substitute monitoring object; Determine a set of change directions of the monitored object according to changes in the outlines of the standard monitored object and the alternative monitored object; Determine the change content of the standard monitoring object based on the contour change; Determine a set of spatial change modes according to a set of change directions; Determine light analysis areas in the surrounding directions of the standard monitoring object and the substitute monitoring object and calculate the intensity value of each light analysis area; Construct a set of light impact parameters using the positions of the light analysis areas and the intensity values ​​of each light analysis area; The set of spatial variation methods and the set of light influence parameters are used as a reconstruction matrix.

3. The method for reducing dimension and compressing sensing of light field data applied to ultra-high voltage substation according to claim 2 is characterized in that: The light analysis area surrounds the position of the standard monitoring object, and the distances from the light analysis area to the specified position on the standard monitoring object are equal.

4. The method for reducing dimension and compressing sensing of light field data applied to ultra-high voltage substation according to claim 2 is characterized in that: Sparse sampling of the additions to standard monitoring objects.

5. The method for reducing dimension and compressing sensing of light field data applied to ultra-high voltage substation according to claim 1 or 4, characterized in that: Sparse sampling includes: Establishing expression curves in the horizontal or vertical direction of the image; The expression curve is transferred into the sparse domain for expression and the peak value in the sparse domain is determined, and the number of the peak value is multiple; Calculate the correlation value of two adjacent peaks; When the correlation value of any two adjacent peaks is less than or equal to the set threshold, the expression curve is randomly subsampled.

6. The method for reducing dimension and compressing perception of light field data applied to ultra-high voltage substation according to claim 5 is characterized in that: Also includes: When the correlation value between any two adjacent peaks is greater than the set threshold, the expression curve is decomposed using a filter to obtain a sub-expression curve; Check the correlation values ​​of adjacent peaks of the sub-expression curve in the sparse domain to ensure that the correlation values ​​of adjacent peaks are less than or equal to the set threshold; Sub-expression curves were randomly subsampled; When the correlation value between any two adjacent peaks on the sub-expression curve is greater than a set threshold, the filter is adjusted to regenerate the sub-expression curve or the sub-expression curve is decomposed again.

7. The method for reducing dimension and compressing sensing of light field data applied to ultra-high voltage substation according to claim 5 is characterized in that: Also includes: Use color channels to decompose the image into red, green, and blue images; Expression curves are established in the horizontal direction or the vertical direction of the red image, the green image, and the blue image, respectively.

8. A light field data dimension reduction and compression sensing device applied to an ultra-high voltage substation, characterized in that: include: An image processing unit, used for constructing an image array based on the obtained image and the image generation position, wherein the images in the image array are arranged in a matrix form of MxN, where M and N are both natural numbers greater than zero; A content recognition unit, used to recognize or determine the monitored object in the image according to the received instruction; A region construction unit, used for constructing a compression region on the image based on the monitored object, wherein the monitored object is located inside the compression region; A content comparison unit, used to compare the contents of the monitored object in adjacent images, obtain a content comparison result and determine a baseline image of the monitored object; A matrix construction unit, used for constructing a reconstruction matrix according to the content comparison result; A sampling unit, used for performing sparse sampling on the non-monitoring object area outside the compression area to obtain a sampling result; The result output unit is used to output the monitored object reference image, reconstruction matrix and sampling result as compressed sensing results.

9. A light field data dimension reduction compressed sensing system applied to an ultra-high voltage substation, characterized in that: The system comprises: one or more memories for storing instructions; and One or more processors, configured to call and execute the instructions from the memory to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer readable storage medium comprises: The program, when the program is executed by a processor, the method according to any one of claims 1 to 7 is executed.

Images