WO2020215203A1 - 数据处理系统和方法 - Google Patents

数据处理系统和方法 Download PDF

Info

Publication number
WO2020215203A1
WO2020215203A1 PCT/CN2019/083874 CN2019083874W WO2020215203A1 WO 2020215203 A1 WO2020215203 A1 WO 2020215203A1 CN 2019083874 W CN2019083874 W CN 2019083874W WO 2020215203 A1 WO2020215203 A1 WO 2020215203A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
area
compression error
region
compression
Prior art date
Application number
PCT/CN2019/083874
Other languages
English (en)
French (fr)
Inventor
张鹏
闻博
范顺杰
Original Assignee
西门子股份公司
西门子（中国）有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 西门子股份公司, 西门子（中国）有限公司 filed Critical 西门子股份公司
Priority to EP19926607.3A priority Critical patent/EP3917014A4/en
Priority to PCT/CN2019/083874 priority patent/WO2020215203A1/zh
Priority to CN201980091284.0A priority patent/CN113383496A/zh
Priority to US17/432,971 priority patent/US11381251B2/en
Publication of WO2020215203A1 publication Critical patent/WO2020215203A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
    • H03M7/00—Conversion of a code where information is represented by a given sequence or number of digits to a code where the same, similar or subset of information is represented by a different sequence or number of digits
    • H03M7/30—Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
    • H03M7/60—General implementation details not specific to a particular type of compression
    • H03M7/6041—Compression optimized for errors
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
    • H03M7/00—Conversion of a code where information is represented by a given sequence or number of digits to a code where the same, similar or subset of information is represented by a different sequence or number of digits
    • H03M7/30—Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
    • H03M7/3059—Digital compression and data reduction techniques where the original information is represented by a subset or similar information, e.g. lossy compression
    • H03M7/3064—Segmenting

Definitions

  • the invention relates to a data processing system and method.
  • SDT Swing Door Trend
  • the present invention aims to solve the above and/or other technical problems and provide a data processing system and method that can reduce or minimize compression errors.
  • a data processing system includes: a data acquisition unit configured to acquire a plurality of data related to a target object; and a data processing unit configured to receive the A plurality of data, and a plurality of regions adjacent to each other are set in the two-dimensional spatial representation of the plurality of data according to the compression allowable error value, wherein the plurality of regions include adjacent second regions that respectively cover the plurality of data A region and a second region, wherein the data processing unit is further configured to expand the second region forward to obtain an expanded second region that overlaps the first region, and calculate the value of the data covered by the expanded second region Compression error, reset the first area according to the calculated compression error, and compress the data covered by the reset first area.
  • the data processing unit is configured to set multiple regions, calculate compression errors, and perform compression according to the rules of the revolving door SDT algorithm.
  • the data processing unit is configured to use data covered by an overlapping area where the first area and the expanded second area overlap each other as the end boundary of the reset first area.
  • the plurality of data includes data y1, y2, ..., yn-1, yn, yn+1, ..., yn+m, ... arranged sequentially in a two-dimensional space representation, wherein n and m are greater than 2, the data processing unit is configured to set the first area to be bounded by data y1 and data yn and to cover the data y1, y2, ..., yn-1, yn, and the data processing unit is configured to set the second area Set to take data yn and data yn+m as boundaries and cover data yn, yn+1,..., yn+m-1, yn+m, where the data processing unit is constructed as follows: The second area is expanded in a direction opposite to the direction along which the data is arranged to overlap the first area, thereby obtaining data yi, yi+1, ..., yn-1 covering the overlapped area, where, i is an integer and 1 ⁇ i ⁇ n-1, and the extended second region of data
  • the data processing unit is configured to calculate the compression error CE1_i of the data y1, y2,..., yi covered by the reduced first area bounded by the data y1 and the data yi, and the compression error CE1_i bounded by the data yi and the data yn+m
  • the compression error CEi_n+m of the data yi, yi+1,..., yn-1, yn, yn+1,..., yn+m-1, yn+m covered by the extended second area and determine the compression Whether the error CE1_i is less than or equal to the compression error CE1 of the first region and whether the compression error CEi_n+m is less than or equal to the compression error CE2 of the second region, wherein the data processing unit is configured to: when it is determined that the compression error CE1_i is less than or equal to the first region When the compression error CE1 of a region and the compression error CEi_n+m are less than or equal to the compression error CE2 of the second
  • the data processing unit is configured to: when it is determined that the compression error CE1_i+j-1 is greater than the compression error CE1 of the first region, or the compression error CEi+j-1_n+m is greater than the compression error CE2 of the second region, the calculation is based on the data y1
  • the data processing unit is configured to set the plurality of regions to include an Nth region and an N+1th region, wherein the Nth region and the N-1th region are adjacent to each other and cover a plurality of data, and the N+1th region
  • the Nth area is adjacent to each other and covers a plurality of data, where N is an integer and N>1, wherein the data processing unit is configured to expand the Nth area forward to Take the data as the end boundary of the reset N-1th area as the expanded Nth area of the start boundary; expand the N+1th area forward to obtain the expanded N+1th area that overlaps the expanded Nth area Area, calculate the compression error of the data covered by the expanded N+1th area, and reset the expanded Nth area according to the calculated compression error; compress the newly set expanded Nth area.
  • a data processing method may include: setting a plurality of regions adjacent to each other in a two-dimensional spatial representation of a plurality of data to be compressed according to a compression allowable error value, wherein the plurality of The area includes a first area and a second area adjacent to each other covering a plurality of data; the second area is expanded forward to obtain an expanded second area that overlaps the first area, and the area covered by the expanded second area is calculated
  • the data compression error and reset the first area according to the calculated compression error; compress the data covered by the reset first area.
  • the step of setting multiple regions, the step of calculating the compression error, and the step of performing compression can be performed according to the rules of the revolving door SDT algorithm.
  • the step of resetting the first area includes: using the data covered by the overlapping area where the first area and the expanded second area overlap each other as the end boundary of the reset first area.
  • the plurality of data includes data y1, y2, ..., yn-1, yn, yn+1, ..., yn+m, ... arranged sequentially in a two-dimensional space representation, wherein n and m are greater than An integer of 2, the first area uses data y1 and data yn as the boundary and covers data y1, y2,..., yn-1, yn, and the second area uses data y1 and data yn+m as the boundary to cover data yn, yn+ 1...., yn+m-1, yn+m, the step of expanding the second area forward includes: expanding the second area in a direction opposite to the direction along which the data is arranged in the two-dimensional space representation to be aligned with the first One area overlaps, thereby obtaining data yi, yi+1,..., yn-1 covering the overlapped area, where i is an integer and 1 ⁇ i ⁇ n-1, and is covered by the second area
  • the step of calculating the compression error of the data covered by the expanded second area includes: calculating the compression error CE1_i and the compression error CE1_i of the data y1, y2,..., yi covered by the reduced first area bounded by the data y1 and the data yi Data yi, yi+1,..., yn-1, yn, yn+1,..., yn+m-1, yn+m covered by the extended second area bounded by data yi and data yn+m And determine whether the compression error CE1_i is less than or equal to the compression error CE1 of the first region, and whether the compression error CEi_n+m is less than or equal to the compression error CE2 of the second region,
  • the step of resetting the first region according to the calculated compression error includes: when it is determined that the compression error CE1_i is less than or equal to the compression error CE1 of the first region, and the compression error CEi_n+m is less than or equal to the compression error CE2 of the second region,
  • the data y1 and the data yi are the boundary and the reduced first area covering the data y1, y2,... Yi is used as the reset first area.
  • the step of calculating the compression error corresponding to the expanded second region includes: when it is determined that the compression error CE1_i+j-1 is greater than the compression error CE1 of the first region, or the compression error CEi+j-1_n+m is greater than the compression error of the second region
  • calculate the compression error CE1_i+j of the data y1, y2,..., yi+j covered by the reduced first area bounded by the data y1 and data yi+j and the data yi+j and data yn +m is the compression error of the data yi+j,..., yn-1, yn, yn+1,..., yn+m-1, yn+m covered by the extended second area of the boundary CEi+j_n+m
  • the step of resetting the first region according to the calculated compression error includes: when it is determined that the compression error CE1_i+j is less than or equal to the compression error CE1 of the first region
  • the plurality of regions includes an Nth region and an N+1th region, wherein the Nth region and the N-1th region are adjacent to each other and cover a plurality of data, and the N+1th region and the Nth region are adjacent to each other and cover A plurality of data, where N is an integer and N>1, the method further includes: when the N-1th area is reset, the Nth area is expanded forward to be the end of the N-1th area that is reset The data of the boundary is used as the expanded Nth area of the start boundary; the N+1th area is expanded forward to obtain the expanded N+1th area overlapping with the expanded Nth area, and the expanded N+1th area is calculated to cover And reset the expanded Nth area according to the calculated compression error; compress the newly set expanded Nth area.
  • an electronic device may include: at least one processor; and a memory connected to the at least one processor, the memory having instructions stored therein, and the instructions are When executed by at least one processor, the electronic device executes the method described above.
  • a non-transitory machine-readable medium may store computer-executable instructions that, when executed, cause at least one processor to perform the method as described above.
  • a computer program may include computer-executable instructions that, when executed, cause at least one processor to perform the method as described above.
  • Fig. 1 is a schematic block diagram showing a data processing system according to an exemplary embodiment
  • FIG. 2 is a flowchart showing a data processing method according to an exemplary embodiment
  • Fig. 3 is a schematic diagram showing an area according to an exemplary embodiment
  • FIG. 4 is a schematic diagram showing a setting area according to an exemplary embodiment
  • FIG. 5 is a schematic diagram showing a result of resetting the first area in FIG. 4 according to an exemplary embodiment
  • FIG. 6 is a schematic diagram showing an electronic device according to an exemplary embodiment.
  • Fig. 1 is a schematic block diagram showing a data processing system according to an exemplary embodiment.
  • the data processing system may include a data acquisition unit 100 and a data processing unit 300.
  • the data processing system may obtain the data of the target object M through the data obtaining unit 100.
  • the target object M may be a motor.
  • the data acquisition unit 100 may be implemented as a driver that drives the operation of the motor and can acquire data related to the operation and configuration of the motor.
  • the data processing unit 300 may receive data related to the target object from the data acquisition unit 100, and may process the data, for example, compress it.
  • the processed (e.g., compressed) data can then be sent to the outside, e.g., the cloud.
  • the data processing unit 300 can be implemented as an edge device such as an industrial gateway of Siemens' Mind-Connect IOT 2040, Mind-Connect Nano and other products, and can Send the processed data to a cloud such as Siemens' MindSphere.
  • FIG. 2 is a flowchart showing a data processing method according to an exemplary embodiment.
  • the data processing unit 300 may execute the data processing method according to an exemplary embodiment as shown in FIG. 2. Therefore, hereinafter, the processing (compression) operation performed on data by the data processing unit 300 according to an exemplary embodiment will be described in detail with reference to FIG. 2.
  • the data may be real-time data related to the operation and configuration of machinery and equipment such as motors.
  • Such data may generally include time information and information indicating the operating state of the machine equipment at the point in time indicated by the time information. Therefore, for such data, it is possible to express such data in a two-dimensional coordinate system based on the sensed value of time and the operating state of the equipment.
  • the exemplary embodiment is not limited to this, and those skilled in the art should understand that the data processing method according to the exemplary embodiment can be implemented on any data that can be represented in a two-dimensional space. Two different dimensions on which the two-dimensional space representation is based can be selectively defined, such as the sensed value of the time and the operating state of the machine equipment described in the current embodiment, or any other desired dimension definition of the two-dimensional space representation the way.
  • the data processing unit 300 may set multiple regions adjacent to each other in the two-dimensional spatial representation of multiple data according to a certain compression allowable error value ⁇ E.
  • the representation of multiple data in a two-dimensional space according to the predefined dimensions can be obtained according to the predefined dimensions.
  • multiple regions can be set in the two-dimensional space according to a predetermined or selected compression allowable error value ⁇ E.
  • the data processing unit 300 may adopt the rules of the SDT algorithm to set multiple regions in the shape of a parallelogram in the two-dimensional spatial representation of the data.
  • the compression allowable error value ⁇ E for setting the area of each parallelogram shape may be different from each other.
  • FIG. 3 is a schematic diagram illustrating setting regions in a two-dimensional spatial representation of a plurality of data according to an exemplary embodiment.
  • Figure 3 it exemplarily shows a two-dimensional coordinate system defined by the time (X axis) and the sensed value (Y axis) of a certain operating state of the machine equipment at that time.
  • Data represented in the space y1, y2, y3, y4, y5, y6, y7, y8, y9.
  • the data processing unit 300 may set the compression allowable error value ⁇ E1 in advance, and may set the first area (parallelogram) A1 according to the compression allowable error value ⁇ E1 using the rules of the SDT algorithm.
  • the rules of the SDT algorithm can be used to set the first area A1, that is, the first area A1 can take the data y1 and y6 as the start boundary (start point) and end boundary (end point), and can cover the data y1, y2, and y3, y4, y5, y6.
  • the data processing unit 300 may set the compression allowable error value ⁇ E2 in advance, and may set the second area (parallelogram) A2 according to the compression allowable error value ⁇ E2 using the rules of the SDT algorithm.
  • the rules of the SDT algorithm can be used to set the second area A2
  • the second area (parallelogram) A2 can cover the data y6, y7, y8, y9, and the data y6 can be used as the starting boundary (starting point) and the data y9 serves as the end boundary (end point).
  • the compression allowable error value ⁇ E2 is shown as being smaller than the compression allowable error value ⁇ E1 in FIG. 3, according to an exemplary embodiment, the compression allowable error value ⁇ E2 may be equal to or greater than the compression allowable error value ⁇ E1.
  • the data processing unit 300 may only store and/or send only the boundary data y1, y6, y9 to the outside, so that the data y1,..., Y9 can be compressed.
  • the exemplary embodiment will additionally consider the compression error before compressing the data y1,... Y9, and reset the area according to the compression error, so that according to the reset The compression error when the area is compressed is minimized.
  • the data processing unit 300 may expand the second area forward to obtain an expanded second area that overlaps the first area, and then the data processing unit 300 may calculate the The expanded second area covers the compression error of the data, and the first area can be reset according to the calculated compression error.
  • forward expansion refers to: without changing the end boundary of the area, the beginning of the area is moved in the direction (forward direction) opposite to the direction along which the data is arranged (back direction).
  • the boundary moves in parallel to intersect the forward data.
  • the upper boundary and the lower boundary are extended to intersect the moved start boundary, thereby forming the moved start boundary, extended upper boundary, extended lower boundary, and unchanged End the new forward expansion area defined by the boundary.
  • a new reduced area can also be obtained, that is, an area obtained by shrinking the adjacent previous area before the forward expanded area.
  • the reduced area may be defined by the initial boundary of the previous area, the upper and lower boundaries of the previous area defined by the data of the initial boundary and the compression allowable error value, and the data as the start boundary of the forwardly expanded area. As the end boundary definition.
  • the rules of the SDT algorithm can be used to set multiple regions in the shape of a parallelogram. Therefore, although the description of forward expansion and contraction according to the exemplary embodiment of the present invention above defines the region The initial boundary, the upper boundary, the lower boundary and the end boundary of, but those skilled in the art can understand that the parallelogram-shaped area defined here and its boundary may have the same meaning as defined in the SDT algorithm.
  • the end boundary of the second area A2 that intersects the data y9 may not be changed, along the direction opposite to the X-axis direction along which the data is arranged (that is, -X Axis direction) Move the starting boundary from the original intersection with data y6 and parallel to data y5. At the same time, extend the upper and lower boundaries to intersect the moved starting boundary with data y5, thereby obtaining the expanded second Area A2, A3.
  • an overlapping area A3 in which the expanded second area A2 overlaps the first area A2 can be obtained, and the data y5 that falls in the overlapping area A3 farthest from the second area A2 can be used as the expanded second area
  • the extended second area with data y5 and data y9 as the boundary can be obtained.
  • the reduced first area can also be obtained, that is, without changing the upper and lower boundaries of the first area A1 defined by the data y1 of the start boundary of the first area A1 and the compression allowable error value ⁇ E1,
  • the data y5 that is the start boundary of the expanded second area serves as the end boundary of the reduced first area, so that the reduced first area with the data y1 and data y5 as the start boundary and the end boundary can be obtained.
  • the data processing unit 300 may calculate the compression error.
  • the compression error CE5_9 of the extended second region with the data y5 and y9 as the boundary can be calculated according to the rules of the SDT algorithm.
  • the compression error CE1_5 of the reduced first region with the data y1 and y5 as the boundary can be calculated according to the rules of the SDT algorithm.
  • the compression error CE5_9 and the compression error CE1_5 corresponding to the expanded second area can be compared.
  • the compression error CE1 of the first area A1 bounded by the data y1 and y6 and the compression error CE2 of the second area A2 bounded by the data y6 and y9 can be calculated. Then, it can be determined whether the compression error CE1_5 is less than or equal to the compression error CE1 of the first region, and whether the compression error CE5_9 is less than or equal to the compression error CE2 of the second region.
  • the compression error CE1_5 is less than or equal to the compression error CE1 of the first region
  • the compression error CE5_9 is less than or equal to the compression error CE2 of the second region
  • the data y1 and y2 are paired with the reduced first region and the expanded second region.
  • y9 can obtain a compression error less than or equal to the compression error obtained by compressing the first area A1 and the second area A2. Therefore, according to an exemplary embodiment, a reduced first area with a smaller compression error may be used as a new first area, that is, the first area may be reset.
  • the data processing unit 300 may compress the data y1,..., Y5 covered by the newly set first area.
  • the rules of the SDT algorithm are used for compression, that is, data y1, y5 can be stored without storing data y2, y3, and y4.
  • the first area can be reset with a smaller compression error, so that the compression error for data compression can be reduced or minimized.
  • the exemplary embodiment of resetting the first area is described above with reference to FIGS. 2 and 3, and the same or similar operations can also be performed on the second area adjacent to the first area and other areas to minimize compression. error.
  • the data to be compressed may include data y1, y2, ..., yn-1, yn, yn+1, ..., yn+m, ... which are sequentially arranged in a two-dimensional space representation.
  • n and m may be integers greater than 2.
  • the data processing unit 300 may set the first area to be the boundary between the data y1 and the data yn and cover the data y1, y2, ..., yn-1, yn, and may set the second area to be based on the data y1 and the data yn+m. It is the boundary and covers the data yn, yn+1,..., yn+m-1, yn+m.
  • the second area can be expanded forward in this way, that is, in the two-dimensional space representation, the second area is expanded in a direction opposite to the direction along which the data is arranged to overlap the first area, thereby covering the overlapped area Covered data yi, yi+1, ..., yn-1, and data yn, yn+1, ..., yn+m-1, yn+m covered by the second region are extended second regions.
  • i may be an integer and 1 ⁇ i ⁇ n-1.
  • the data processing unit 300 may calculate the compression error CE1_i of the data y1, y2, ..., yi covered by the reduced first area bounded by the data y1 and the data yi and the compression error CE1_i bounded by the data yi and the data yn+m
  • the data processing unit 300 may determine whether the compression error CE1_i is less than or equal to the compression error CE1 of the first region and whether the compression error CEi_n+m is less than or equal to the compression error CE2 of the second region.
  • the data y1 and the data yi can be used as the boundary and cover the data y1, y2. ,..., yi's reduced first area is used as the reset first area.
  • FIG. 4 and 5 show schematic diagrams of compressing data according to another compression method according to an exemplary embodiment, wherein FIG. 4 is a schematic diagram showing a setting area; FIG. 5 is a schematic diagram showing the first area in FIG. 4 Schematic diagram of the result of resetting.
  • the data to be compressed y1, y2,..., Y9 are represented in two-dimensional spaces of dimensions X (for example, time) and Y (for example, values sensed by a sensor).
  • the data processing unit 300 may use the rules of the SDT algorithm to set the first area A1_7 and the second area A7_9 according to a predetermined compression allowable error value ⁇ E.
  • the first area A1_7 may be bounded by the data y1 and y7, and cover the data y1, y2, y3, y4, y5, y6, and y7.
  • the second area A7_9 may be bounded by data y7 and data y9, and cover data y7, y8, and y9.
  • the data processing unit 300 may expand the second area A7_9 forward to obtain an expanded second area.
  • the expanded second area may be bounded by data y4 and y9 and cover data y4, y5, y6, y7, y8, and y9.
  • a reduced first area can be obtained, and the reduced first area can take the data y1 and y4 as the boundary and cover the data y1, y2, y3, and y4.
  • the data processing unit 300 may calculate the compression error CE4_9 of the data covered by the expanded second area and the compression error CE1_4 of the data covered by the reduced first area. Then it can be determined whether the compression error CE4_9 of the data covered by the expanded second area is less than or equal to the compression error CE7_9 of the data covered by the second area A7_9, and it can be determined whether the compression error CE1_4 of the data covered by the reduced first area is less than Or equal to the compression error CE1_7 of the data covered by the first area A1_7.
  • the data processing unit 300 may further calculate the compression errors CE1_5 of the data y1, y2,..., y5 covered by the reduced first area bounded by the data y1 and the data y5 and the compression errors CE1_5 and the data y5 and y5.
  • the data y9 is the compression error CE5_9 of the data y5,..., y9 covered by the extended second area of the boundary. Then, it can be determined whether the compression error CE5_9 is less than or equal to the compression error CE7_9, and it can be determined whether the compression error CE1_5 is less than or equal to the compression error CE1_7.
  • the data processing unit 300 may further calculate the compression errors CE1_6 of the data y1, y2,..., y6 covered by the reduced first area bounded by the data y1 and the data y6 and the compression errors CE1_6 and the data y6 and y6.
  • the data y9 is the compression error CE6_9 of the data y6,..., y9 covered by the extended second area of the boundary. Then it can be determined whether the compression error CE6_9 is less than or equal to the compression error CE7_9, and it can be determined whether the compression error CE6_5 is less than or equal to the compression error CE1_7.
  • the data processing unit 300 may reset the first area to the new first area A1_6 with the data y1 and y6 as the boundary and cover the data y1,..., Y6.
  • the data processing unit 300 can obtain the expanded second area A6_9 starting with the newly set data y6 bits of the end boundary of the first area A1_6.
  • the start boundary/end boundary of the new area can be changed until it is determined that the change The compression error of the data covered by the area of the start boundary/end boundary is less than or equal to the compression error position of the data covered by the original area.
  • the exemplary embodiment of resetting the first area is described above with reference to FIGS. 2 to 5. The same can also be done for the second area adjacent to the first area, the third area adjacent to the second area... Or similar operations to minimize the compression error.
  • the plurality of regions may include an Nth region adjacent to each other and an N-1th region including an Nth region and an N+1th region.
  • the Nth area and the N-1th area may be adjacent to each other and cover a plurality of data
  • the N+1th area and the Nth area may be adjacent to each other and cover a plurality of data.
  • N is an integer and N ⁇ 1.
  • N is equal to 1, the operation described here will be the same as the operation in the embodiment described above with reference to FIGS. 2 to 5.
  • the Nth area When the N-1th area is reset, the Nth area may be expanded forward into the Nth area with the data as the end boundary of the reset N-1th area as the start boundary.
  • the start boundary of the expanded Nth region (second region) when N is equal to 2, may be data y5.
  • the starting boundary of the expanded Nth region (second region) when N is equal to 2, may be data y6.
  • the N+1th area can be expanded forward in a manner consistent with the manner in which the Nth area is expanded forward before resetting the N-1th area above to obtain an expanded N+1th area that overlaps with the expanded Nth area. area.
  • the compression error of the data covered by the expanded N+1th area is calculated, and the expanded Nth area can be reset according to the calculated compression error.
  • the re-set extended Nth area can be compressed.
  • FIG. 6 is a block diagram showing an electronic device according to an exemplary embodiment.
  • the electronic device may include at least one processor 610 and a memory 630.
  • the processor 610 may execute at least one computer-readable instruction (ie, the above-mentioned element implemented in the form of software) stored or encoded in a computer-readable storage medium (ie, the memory 630).
  • computer-executable instructions are stored in the memory 630, which, when executed, cause at least one processor 610 to implement or execute the methods described above with reference to FIGS. 2 to 5.
  • a program product such as a non-transitory machine-readable medium.
  • the non-transitory machine-readable medium may have instructions (that is, the above-mentioned elements implemented in the form of software), which when executed by a machine, cause the machine to execute the various embodiments described above with reference to FIGS. 2 to 5 in the various embodiments of the present application.
  • instructions that is, the above-mentioned elements implemented in the form of software
  • a computer program product including computer-executable instructions, which when executed, cause at least one processor to execute the various embodiments of the present application. As described above with reference to FIGS. 2 to 5 Various operations and functions.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Compression, Expansion, Code Conversion, And Decoders (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)

Abstract

本发明提供了一种数据处理系统和方法。所述数据处理系统包括：数据获取单元，被构造为获取与目标对象相关的多个数据；数据处理单元，被构造为接收所述多个数据，并根据压缩允许误差值在所述多个数据的二维空间表示中设置彼此相邻的多个区域，其中，所述多个区域包括分别覆盖多个数据的彼此相邻的第一区域和第二区域，其中，所述数据处理单元还被构造为前向扩展第二区域以得到与第一区域交叠的扩展的第二区域，计算被扩展的第二区域覆盖的数据的压缩误差，根据计算的压缩误差重新设置第一区域，并对被重新设置的第一区域覆盖的数据进行压缩。根据示例性实施例的数据处理系统可以使数据压缩误差减小或最小化。

Description

数据处理系统和方法 技术领域
本发明涉及一种数据处理系统和方法。
背景技术
在诸如工业4.0、物联网(IOT)等的应用中，机器设备在实时运行中可能产生大量的数据。因此，为了减小这些数据的大小以更为便捷地进行传输，通常考虑对这些数据进行压缩处理。旋转门(Swing Door Trend，SDT)算法是一种可用的压缩数据的算法。因为SDT算法的执行效率较高，所以在对这样的大量的数据的传输和处理的过程中得到了广泛的应用。
发明内容
本发明在于解决上述和/或其他技术问题并提供一种可以减小或最小化压缩误差的数据处理系统和方法。
在一个示例性实施例中，提供了一种数据处理系统，所述数据处理系统包括：数据获取单元，被构造为获取与目标对象相关的多个数据；数据处理单元，被构造为接收所述多个数据，并根据压缩允许误差值在所述多个数据的二维空间表示中设置彼此相邻的多个区域，其中，所述多个区域包括分别覆盖多个数据的彼此相邻的第一区域和第二区域，其中，所述数据处理单元还被构造为前向扩展第二区域以得到与第一区域交叠的扩展的第二区域，计算被扩展的第二区域覆盖的数据的压缩误差，根据计算的压缩误差重新设置第一区域，并对被重新设置的第一区域覆盖的数据进行压缩。
数据处理单元被构造为根据旋转门SDT算法的规则来设置多个区域、计算压缩误差和进行压缩。
数据处理单元被构造为：将被第一区域与扩展的第二区域彼此交叠的交叠区域覆盖的数据作为重新设置的第一区域的结束边界。
所述多个数据包括在二维空间表示中顺序排列的数据y1、y2、……、yn-1、yn、yn+1、……、yn+m、……，其中，n和m为大于2的整数，数据处理单元被构造为将第一区域设置为以数据y1和数据yn为边界并覆盖数据y1、y2、……、yn-1、yn，数据处理单元被构造为将第二区域设置为以数据yn和数据yn+m为边界并覆盖数据yn、yn+1、……、 yn+m-1、yn+m，其中，数据处理单元被构造为：在二维空间表示中沿与数据的排列所沿的方向相对的方向扩展第二区域以与第一区域交叠，从而得到覆盖了被交叠的区域覆盖的数据yi、yi+1、……、yn-1，其中，i为整数且1＜i≤n-1，和被第二区域覆盖的数据yn、yn+1、……、yn+m-1、yn+m的扩展的第二区域。
数据处理单元被构造为：计算被以数据y1和数据yi为边界的缩小的第一区域覆盖的数据y1、y2、……、yi的压缩误差CE1_i和被以数据yi和数据yn+m为边界的扩展的第二区域覆盖的数据yi、yi+1、……、yn-1、yn、yn+1、……、yn+m-1、yn+m的压缩误差CEi_n+m，并确定压缩误差CE1_i是否小于或等于第一区域的压缩误差CE1、且压缩误差CEi_n+m是否小于或等于第二区域的压缩误差CE2，其中，数据处理单元被构造为：当确定压缩误差CE1_i小于或等于第一区域的压缩误差CE1、且压缩误差CEi_n+m小于或等于第二区域的压缩误差CE2时，将以数据y1和数据yi为边界且覆盖数据y1、y2、……、yi的缩小的第一区域作为重新设置的第一区域。
数据处理单元被构造为：当确定压缩误差CE1_i+j-1大于第一区域的压缩误差CE1、或压缩误差CEi+j-1_n+m大于第二区域的压缩误差CE2时，计算被以数据y1和数据yi+j为边界的缩小的第一区域覆盖的数据y1、y2、……、yi+j的压缩误差CE1_i+j和被以数据yi+j和数据yn+m为边界的扩展的第二区域覆盖的数据yi+j、……、yn-1、yn、yn+1、……、yn+m-1、yn+m的压缩误差CEi+j_n+m，其中，j为整数且1≤j＜n-i，其中，数据处理单元被构造为：当确定压缩误差CE1_i+j小于或等于第一区域的压缩误差CE1且压缩误差CEi+j_n+m小于或等于第二区域的压缩误差CE2时，将以数据y1和数据yi+j为边界的缩小的第一区域作为重新设置的第一区域。
数据处理单元被构造为将所述多个区域设置为包括第N区域和第N+1区域，其中，第N区域与第N-1区域彼此相邻并覆盖多个数据，第N+1区域与第N区域彼此相邻并覆盖多个数据，其中，N为整数且N＞1，其中，数据处理单元被构造为：当重新设置第N-1区域时，将第N区域前向扩展为以作为重新设置的第N-1区域的结束边界的数据作为开始边界的扩展的第N区域；前向扩展第N+1区域以得到与扩展的第N区域交叠的扩展的第N+1区域，计算被扩展的第N+1区域覆盖的数据的压缩误差，并根据计算的压缩误差重新设置扩展的第N区域；对重新设置的扩展的第N区域进行压缩。
在另一个示例性实施例中，一种数据处理方法可以包括：根据压缩允许误差值在待压缩的多个数据的二维空间表示中设置彼此相邻的多个区域，其中，所述多个区域包括分别覆盖多个数据的彼此相邻的第一区域和第二区域；前向扩展第二区域以得到与第一区域交叠的扩展的第二区域，计算被扩展的第二区域覆盖的数据的压缩误差，并根据计算的压缩 误差重新设置第一区域；对被重新设置的第一区域覆盖的数据进行压缩。这里，可以根据旋转门SDT算法的规则来执行设置多个区域的步骤、计算压缩误差的步骤、和进行压缩的步骤。
重新设置第一区域的步骤包括：将被第一区域与扩展的第二区域彼此交叠的交叠区域覆盖的数据作为重新设置的第一区域的结束边界。
所述多个数据包括在二维空间表示中顺序排列的数据y1、y2、……、yn-1、yn、yn+1、……、yn+m、……，其中，n和m为大于2的整数，第一区域以数据y1和数据yn为边界并覆盖数据y1、y2、……、yn-1、yn，第二区域以数据y1和数据yn+m为边界覆盖数据yn、yn+1、……、yn+m-1、yn+m，前向扩展第二区域的步骤包括：在二维空间表示中沿与数据的排列所沿的方向相对的方向扩展第二区域以与第一区域交叠，从而得到覆盖了被交叠的区域覆盖的数据yi、yi+1、……、yn-1，其中，i为整数且1＜i≤n-1，和被第二区域覆盖的数据yn、yn+1、……、yn+m-1、yn+m的扩展的第二区域。
计算被扩展的第二区域覆盖的数据的压缩误差的步骤包括：计算被以数据y1和数据yi为边界的缩小的第一区域覆盖的数据y1、y2、……、yi的压缩误差CE1_i和被以数据yi和数据yn+m为边界的扩展的第二区域覆盖的数据yi、yi+1、……、yn-1、yn、yn+1、……、yn+m-1、yn+m的压缩误差CEi_n+m，并确定压缩误差CE1_i是否小于或等于第一区域的压缩误差CE1、且压缩误差CEi_n+m是否小于或等于第二区域的压缩误差CE2，
根据计算的压缩误差重新设置第一区域的步骤包括：当确定压缩误差CE1_i小于或等于第一区域的压缩误差CE1、且压缩误差CEi_n+m小于或等于第二区域的压缩误差CE2时，将以数据y1和数据yi为边界且覆盖数据y1、y2、……、yi的缩小的第一区域作为重新设置的第一区域。
计算与扩展的第二区域对应的压缩误差的步骤包括：当确定压缩误差CE1_i+j-1大于第一区域的压缩误差CE1、或压缩误差CEi+j-1_n+m大于第二区域的压缩误差CE2时，计算被以数据y1和数据yi+j为边界的缩小的第一区域覆盖的数据y1、y2、……、yi+j的压缩误差CE1_i+j和被以数据yi+j和数据yn+m为边界的扩展的第二区域覆盖的数据yi+j、……、yn-1、yn、yn+1、……、yn+m-1、yn+m的压缩误差CEi+j_n+m，其中，j为整数且1≤j＜n-i，根据计算的压缩误差重新设置第一区域的步骤包括：当确定压缩误差CE1_i+j小于或等于第一区域的压缩误差CE1且压缩误差CEi+j_n+m小于或等于第二区域的压缩误差CE2时，将以数据y1和数据yi+j为边界的缩小的第一区域作为重新设置的第一区域。
所述多个区域包括第N区域和第N+1区域，其中，第N区域与第N-1区域彼此相邻 并覆盖多个数据，第N+1区域与第N区域彼此相邻并覆盖多个数据，其中，N为整数且N＞1，所述方法还包括：当重新设置第N-1区域时，将第N区域前向扩展为以作为重新设置的第N-1区域的结束边界的数据作为开始边界的扩展的第N区域；前向扩展第N+1区域以得到与扩展的第N区域交叠的扩展的第N+1区域，计算被扩展的第N+1区域覆盖的数据的压缩误差，并根据计算的压缩误差重新设置扩展的第N区域；对重新设置的扩展的第N区域进行压缩。
在另一个示例性实施例中，一种电子设备可以包括：至少一个处理器；以及与所述至少一个处理器连接的存储器，所述存储器具有存储于其中的指令，所述指令在被所述至少一个处理器执行时使所述电子设备执行如上所述的方法。
在又一个示例性实施例中，一种非暂时性机器可读介质可以存储有计算机可执行指令，所述计算机可执行指令在被执行时使至少一个处理器执行如上所述的方法。
在再一个示例性实施例中，一种计算机程序可以包括计算机可执行指令，所述计算机可执行指令在被执行时使至少一个处理器执行如上所述的方法。
附图说明
以下附图仅旨在于对本发明做示意性说明和解释，并不限定本发明的范围。其中，
图1是示出根据示例性实施例的数据处理系统的示意性框图；
图2是示出根据示例性实施例的数据处理方法的流程图；
图3是示出根据示例性实施例的区域的示意图；
图4是示出根据示例性实施例的设置区域的示意图；
图5是示出根据示例性实施例的对图4中的第一区域进行重新设置的结果的示意图；
图6是示出根据示例性实施例的电子设备的示意图。
附图标记说明：
100数据获取单元 M目标对象 300数据处理单元
610处理器 630存储器
具体实施方式
为了对本发明的技术特征、目的和效果有更加清楚的理解，现对照附图说明本发明的具体实施方式。
图1是示出根据示例性实施例的数据处理系统的示意性框图。如图1中所示，根据示例性实施例的数据处理系统可以包括数据获取单元100和数据处理单元300。数据处理系统可以通过数据获取单元100获取目标对象M的数据。当根据示例性实施例的处理系统应 用于物联网(IOT)时，目标对象M可以为电机。数据获取单元100可以被实施为驱动电机运行并可以获取与电机运行和配置相关的数据的驱动器。
数据处理单元300可以接收来自数据获取单元100的与目标对象相关的数据，并可以对数据进行处理，例如，压缩。然后可以将经处理的(例如，被压缩的)数据发送到外部，例如，云。这里，当目标对象M为IOT应用的电机时，数据处理单元300可以被实现为诸如西门子公司的Mind-Connect IOT2040、Mind-Connect Nano等产品的工业网关等的边缘设备(edge device)，并可以将经处理的数据发送到诸如西门子公司的MindSphere的云。
因为来自数据获取单元100的数据的量可能非常大，所以数据处理单元300可以在将数据发送到外部之前对数据进行处理，例如，压缩，以减小将被发送的数据量。例如，图2是示出根据示例性实施例的数据处理方法的流程图。数据处理单元300可以执行如图2中所示的根据示例性实施例的数据处理方法。因此，在下文中，将参照图2详细描述根据示例性实施例的数据处理单元300对数据执行的处理(压缩)操作。
如上所述，例如在IOT的应用中，数据可以与诸如电机等的机器设备的运行和配置相关的实时数据。这样的数据通常可以包括时间信息和表示与时间信息所表示的时间点时的机器设备的运行状态的信息。因此，对于这样的数据，可以在以时间和机器设备的运行状态的感测的值为二维坐标系中表示这样的数据。然而，示例性实施例不限于此，本领域技术人员应该理解的是，可以对任何能够进行二维空间表示的数据实施根据示例性实施例的数据处理方法。可以选择性地定义二维空间表示所依据的两个不同的维度，例如当前实施例所描述的时间和机器设备的运行状态的感测的值、或其他任何期望的二维空间表示的维度定义方式。
如图2中所示，在步骤S201，数据处理单元300可以根据一定的压缩允许误差值ΔE在多个数据的二维空间表示中设置彼此相邻的多个区域。为此，可以根据预先定义的维度得到多个数据的在根据预先定义的维度的二维空间中的表示。然后，可以根据预定的或选定的压缩允许误差值ΔE在该二维空间中设置多个区域。在一个示例性实施例中，数据处理单元300可以采用SDT算法的规则来在这些数据的二维空间表示中设置平行四边形形状的多个区域。这里，用于设置每个平行四边形形状的区域的压缩允许误差值ΔE可以彼此不同。
图3是示出根据示例性实施例的在多个数据的二维空间表示中设置区域的示意图。在图3中，示例性地示出了在以时间(X轴)和在该时间处的机器设备的某种运行状态的感测的值(Y轴)为坐标系为二度定义的二维空间中表示的数据y1、y2、y3、y4、y5、y6、y7、y8、y9。数据处理单元300可以预先设置压缩允许误差值ΔE1，并可以根据压缩允许 误差值ΔE1利用SDT算法的规则设置第一区域(平行四边形)A1。例如，可以采用SDT算法的规则来设置第一区域A1，即，第一区域A1可以以数据y1和y6为开始边界(开始点)和结束边界(结束点)，并可以覆盖数据y1、y2、y3、y4、y5、y6。
然后，数据处理单元300可以预先设置压缩允许误差值ΔE2，并可以根据压缩允许误差值ΔE2利用SDT算法的规则设置第二区域(平行四边形)A2。例如，可以采用SDT算法的规则来设置第二区域A2，第二区域(平行四边形)A2可以覆盖数据y6、y7、y8、y9，并可以以数据y6作为开始边界(开始点)、并以数据y9作为结束边界(结束点)。此外，虽然在图3中压缩允许误差值ΔE2被示出为小于压缩允许误差值ΔE1，但是根据示例性实施例，压缩允许误差值ΔE2可以等于或大于压缩允许误差值ΔE1。
此时，根据SDT算法的规则，数据处理单元300可以仅存储和/或仅向外部发送作为边界的数据y1、y6、y9，从而可以实现数据y1、……、y9的压缩。然而，如下面将进行详细描述的，示例性实施例在对数据y1、……、y9进行压缩之前，将另外地考虑压缩误差，并根据压缩误差对区域进行重新设置，以使得根据重新设置的区域进行压缩时的压缩误差最小化。
具体地讲，如图2中所示，在步骤S203，数据处理单元300可以前向扩展第二区域以得到与第一区域交叠的扩展的第二区域，然后，数据处理单元300可以计算被扩展的第二区域覆盖的数据的压缩误差，并可以根据计算的压缩误差重新设置第一区域。
在本发明的示例性实施例中，前向扩展是指：在不改变区域的结束边界的情况下、沿与数据排列所沿的方向(后向)相对的方向(前向)将区域的开始边界平行移动至与前向的数据相交，同时，延长上边界和下边界以与移动后的开始边界相交，从而形成由移动的开始边界、延长的上边界和延长的下边界、以及没有改变的结束边界所限定的新的前向扩展的区域。
在前向扩展的同时，还可以得到新的缩小的区域，即，对经前向扩展的区域之前的相邻的前一区域进行缩小而得到的区域。具体地讲，缩小的区域可以由前一区域的初始边界、由初始边界的数据和压缩允许误差值限定的前一区域的上边界和下边界、以及作为前向扩展的区域的开始边界的数据作为结束边界限定。
在本发明的示例性实施例中，可以采用SDT算法的规则来设置平行四边形形状的多个区域，因此，虽然在上面根据本发明示例性实施例的前向扩展和缩小的描述二限定了区域的初始边界、上边界、下边界和结束边界，但是本领域技术人员可以理解，这里限定的平行四边形形状的区域及其边界可以具有在SDT算法中限定的相同的含义。
下面将以图3中示出的示例性实施例为例进行具体地描述。当前向扩展第二区域A2 时，可以不改变第二区域A2的与数据y9(结束点)相交的结束边界的情况下、沿与数据排列所沿的X轴方向相反的方向(即，-X轴方向)将开始边界从原来的与数据y6相交平行移动至与数据y5相交，同时，延长上边界和下边界至与移动后的与数据y5相交的开始边界相交，从而得到扩展后的第二区域A2、A3。因此，可以得到扩展的第二区域A2与第一区域A2交叠的交叠区域A3，并可以将落入交叠区域A3中的距离第二区域A2最远的数据y5作为扩展的第二区域的开始边界，从而可以得到以数据y5和数据y9作为边界的扩展的第二区域。
此外，还可以得到缩小的第一区域，即，在不改变由第一区域A1的开始边界的数据y1和压缩允许误差值ΔE1限定的第一区域A1的上边界和下边界的情况下，由作为扩展的第二区域的开始边界的数据y5作为缩小的第一区域的结束边界，从而可以得到以数据y1和数据y5作为开始边界和结束边界的缩小的第一区域。
然后，数据处理单元300可以计算压缩误差。例如，可以根据SDT算法的规则计算以数据y5和y9作为边界的扩展的第二区域的压缩误差CE5_9。同时，可以根据SDT算法的规则计算以数据y1和y5作为边界的缩小的第一区域的压缩误差CE1_5。然后可以比较与扩展的第二区域对应的压缩误差CE5_9和压缩误差CE1_5。
具体地讲，可以计算以数据y1和y6为边界的第一区域A1的压缩误差CE1和以数据y6和y9作为边界的第二区域A2的压缩误差CE2。然后，可以确定压缩误差CE1_5是否小于或等于第一区域的压缩误差CE1、且压缩误差CE5_9是否小于或等于第二区域的压缩误差CE2。
当确定压缩误差CE1_5小于或等于第一区域的压缩误差CE1、且压缩误差CE5_9小于或等于第二区域的压缩误差CE2时，说明以缩小的第一区域和扩展的第二区域对数据y1、y2、……、y9进行压缩可以得到的压缩误差小于或等于以第一区域A1和第二区域A2进行压缩可以得到的压缩误差。因此，根据示例性实施例，可以将压缩误差更小的缩小的第一区域作为新的第一区域，即，重新设置第一区域。
然后，参照图2中的步骤S205，数据处理单元300可以对重新设置的第一区域覆盖的数据y1、……、y5进行压缩。例如，采用SDT算法的规则进行压缩，即，可以存储数据y1、y5而不存储数据y2、y3、y4。
这样，通过前向扩展第二区域可以得到压缩误差更小的重新设置第一区域，从而能够得使对于数据的压缩的压缩误差减小或最小化。
上面参照图2和图3描述了对于第一区域进行重新设置的示例性实施例，也可以对于与第一区域相邻的第二区域以及其他的区域进行相同或相似的操作，来最小化压缩误差。
例如，待压缩的数据可以包括在二维空间表示中顺序排列的数据y1、y2、……、yn-1、yn、yn+1、……、yn+m、……。这里，n和m可以为大于2的整数。
数据处理单元300可以将第一区域设置为数据y1和数据yn为边界并覆盖数据y1、y2、……、yn-1、yn，并可以将第二区域设置为以数据y1和数据yn+m为边界并覆盖数据yn、yn+1、……、yn+m-1、yn+m。
可以这样前向扩展第二区域，即，在二维空间表示中沿与数据的排列所沿的方向相对的方向扩展第二区域以与第一区域交叠，从而得到覆盖了被交叠的区域覆盖的数据yi、yi+1、……、yn-1，以及被第二区域覆盖的数据yn、yn+1、……、yn+m-1、yn+m的扩展的第二区域。这里，i可以为整数且1＜i≤n-1。
然后，数据处理单元300可以计算被以数据y1和数据yi为边界的缩小的第一区域覆盖的数据y1、y2、……、yi的压缩误差CE1_i和被以数据yi和数据yn+m为边界的扩展的第二区域覆盖的数据yi、yi+1、……、yn-1、yn、yn+1、……、yn+m-1、yn+m的压缩误差CEi_n+m。然后，数据处理单元300可以确定压缩误差CE1_i是否小于或等于第一区域的压缩误差CE1、且压缩误差CEi_n+m是否小于或等于第二区域的压缩误差CE2。当确定压缩误差CE1_i小于或等于第一区域的压缩误差CE1、且压缩误差CEi_n+m小于或等于第二区域的压缩误差CE2时，可以将以数据y1和数据yi为边界且覆盖数据y1、y2、……、yi的缩小的第一区域作为重新设置的第一区域。
然而，示例性实施例不限于此。图4和图5示出了根据示例性实施例的另一压缩方法对数据进行压缩的示意图，其中，图4是示出设置区域的示意图；图5是示出对图4中的第一区域进行重新设置的结果的示意图。
在图4中，示出了待压缩的数据y1、y2、……、y9以维度X(例如，时间)和Y(例如，传感器感测的值)的二维空间表示。首先，数据处理单元300可以根据预定的压缩允许误差值ΔE，采用SDT算法的规则来设置第一区域A1_7和第二区域A7_9。第一区域A1_7可以以数据y1和y7为边界，并覆盖数据y1、y2、y3、y4、y5、y6、y7。第二区域A7_9可以以数据y7和数据y9为边界，并覆盖数据y7、y8、y9。
根据示例性实施例，数据处理单元300可以前向扩展第二区域A7_9，以得到扩展的第二区域。虽然在图4中没有示出，但是扩展的第二区域可以为以数据y4和y9为边界并覆盖数据y4、y5、y6、y7、y8、y9。同时，可以得到缩小的第一区域，缩小的第一区域可以以数据y1和y4为边界并覆盖数据y1、y2、y3、y4。
然后，数据处理单元300可以计算被扩展的第二区域覆盖的数据的压缩误差CE4_9和被缩小的第一区域覆盖的数据的压缩误差CE1_4。然后可以确定被扩展的第二区域覆盖的 数据的压缩误差CE4_9是否小于或等于第二区域A7_9覆盖的数据的压缩误差CE7_9、同时可以确定被缩小的第一区域覆盖的数据的压缩误差CE1_4是否小于或等于第一区域A1_7覆盖的数据的压缩误差CE1_7。
在图4和图5中示出的示例性实施例中，CE1_4＞CE1_7，且CE4_9＞CE7_9。这时，根据示例性实施例，数据处理单元300可以进一步计算数据y1和数据y5为边界的缩小的第一区域覆盖的数据y1、y2、……、y5的压缩误差CE1_5和被以数据y5和数据y9为边界的扩展的第二区域覆盖的数据y5、……、y9的压缩误差CE5_9。然后可以压缩误差CE5_9是否小于或等于压缩误差CE7_9、同时可以确定压缩误差CE1_5是否小于或等于压缩误差CE1_7。
在图4和图5中示出的示例性实施例中，CE1_5＞CE1_7，且CE5_9＞CE7_9。这时，根据示例性实施例，数据处理单元300可以进一步计算数据y1和数据y6为边界的缩小的第一区域覆盖的数据y1、y2、……、y6的压缩误差CE1_6和被以数据y6和数据y9为边界的扩展的第二区域覆盖的数据y6、……、y9的压缩误差CE6_9。然后可以压缩误差CE6_9是否小于或等于压缩误差CE7_9、同时可以确定压缩误差CE6_5是否小于或等于压缩误差CE1_7。
在图4和图5中示出的示例性实施例中，CE1_6≤CE1_7，且CE6_9≤CE7_9。这时，如图5中所示，数据处理单元300可以将第一区域重新设置为以数据y1和y6为边界、并覆盖数据y1、……、y6的重新设置的第一区域A1_6。同时，数据处理单元300可以得到以重新设置的第一区域A1_6的结束边界的数据y6位开始边界的扩展的第二区域A6_9。
换句话说，当前向扩展后得到的新的区域覆盖的数据的压缩误差与原区域覆盖的数据的压缩误差相比增大时，可以改变新的区域的开始边界/结束边界，直到确定改变了开始边界/结束边界的区域覆盖的数据的压缩误差小于或等于原区域覆盖的数据的压缩误差位置。
上面参照图2-图5描述了对于第一区域进行重新设置的示例性实施例，也可以对于与第一区域相邻的第二区域、与第二区域相邻的第三区域……进行相同或相似的操作，来最小化压缩误差。
例如，在重新设置了第一区域之后，可以对后面的区域重复上面的操作。例如，所述多个区域可以包括彼此相邻的第N区域与第N-1区域包括第N区域和第N+1区域。第N区域与第N-1区域可以彼此相邻并覆盖多个数据，第N+1区域与第N区域可以彼此相邻并覆盖多个数据。这里，N为整数且N≥1。例如，当N等于1时，这里描述的操作将于上面参照图2-图5描述的实施例中的操作相同。
当重新设置了第N-1区域时，可以将第N区域前向扩展为以作为重新设置的第N-1区 域的结束边界的数据作为开始边界的扩展的第N区域。在图3中示出的示例中，当N等于2时，扩展的第N区域(第二区域)的开始边界可以为数据y5。在图4和图5中示出的示例中，当N等于2时，扩展的第N区域(第二区域)的开始边界可以为数据y6。
然后，可以在上面在重新设置第N-1区域之前前向扩展第N区域的方式一致的方式前向扩展第N+1区域以得到与扩展的第N区域交叠的扩展的第N+1区域。这时，计算被扩展的第N+1区域覆盖的数据的压缩误差，并可以根据计算的压缩误差重新设置扩展的第N区域。然后，可以对重新设置的扩展的第N区域进行压缩。上面已经描述了当N等于1时的具体细节。本领域技术人员应该理解当N大于1时的操作，并因此省略重复内容的冗余描述。
上面参照图2到图5对根据示例性实施例的数据处理方法，这样的方法可以由硬件、软件或者硬件和软件的组合来实现。图6是示出根据示例性实施例的电子设备的框图。当前的示例性实施例，电子设备可以包括至少一个处理器610和存储器630。处理器610可以执行在计算机可读存储介质(即，存储器630)中存储或编码的至少一个计算机可读指令(即，上述以软件形式实现的元素)。
在一个实施例中，在存储器630中存储计算机可执行指令，其当执行时使得至少一个处理器610来实现或执行上面参照图2到图5描述的方法。
应该理解，在存储器630中存储的计算机可执行指令当执行时使得至少一个处理器610进行各个实施例中结合图2-图5描述的以上各种操作和功能。
根据一个实施例，提供了一种诸如非暂时性机器可读介质的程序产品。非暂时性机器可读介质可以具有指令(即，上述以软件形式实现的元素)，该指令当被机器执行时，使得机器执行本申请的各个实施例中以上结合图2-图5描述的各种操作和功能。
根据一个实施例，提供了一种计算机程序产品，包括计算机可执行指令，所述计算机可执行指令在被执行时使至少一个处理器执行本申请的各个实施例中以上结合图2-图5描述的各种操作和功能。
应当理解，虽然本说明书是按照各个实施例描述的，但并非每个实施例仅包含一个独立的技术方案，说明书的这种叙述方式仅仅是为清楚起见，本领域技术人员应当将说明书作为一个整体，各实施例中的技术方案也可以经适当组合，形成本领域技术人员可以理解的其他实施方式。
以上所述仅为本发明示意性的具体实施方式，并非用以限定本发明的范围。任何本领域的技术人员，在不脱离本发明的构思和原则的前提下所作的等同变化、修改与结合，均应属于本发明保护的范围。

Claims (17)

  1. 数据处理系统，其特征在于，所述数据处理系统包括：
    数据获取单元(100)，被构造为获取与目标对象(M)相关的多个数据；
    数据处理单元(300)，被构造为接收所述多个数据，并根据压缩允许误差值在所述多个数据的二维空间表示中设置彼此相邻的多个区域，其中，所述多个区域包括分别覆盖多个数据的彼此相邻的第一区域和第二区域，
    其中，所述数据处理单元还被构造为前向扩展第二区域以得到与第一区域交叠的扩展的第二区域，计算被扩展的第二区域覆盖的数据的压缩误差，根据计算的压缩误差重新设置第一区域，并对被重新设置的第一区域覆盖的数据进行压缩。
  2. 如权利要求1所述的数据处理系统，其特征在于，数据处理单元被构造为根据旋转门SDT算法的规则来设置多个区域、计算压缩误差和进行压缩。
  3. 如权利要求1所述的数据处理系统，其特征在于，数据处理单元被构造为：将被第一区域与扩展的第二区域彼此交叠的交叠区域覆盖的数据作为重新设置的第一区域的结束边界。
  4. 如权利要求1所述的数据处理系统，其特征在于，所述多个数据包括在二维空间表示中顺序排列的数据y1、y2、……、yn-1、yn、yn+1、……、yn+m、……，其中，n和m为大于2的整数，
    数据处理单元被构造为将第一区域设置为以数据y1和数据yn为边界并覆盖数据y1、y2、……、yn-1、yn，
    数据处理单元被构造为将第二区域设置为以数据yn和数据yn+m为边界并覆盖数据yn、yn+1、……、yn+m-1、yn+m，
    其中，数据处理单元被构造为：在二维空间表示中沿与数据的排列所沿的方向相对的方向扩展第二区域以与第一区域交叠，从而得到覆盖了被交叠的区域覆盖的数据yi、yi+1、……、yn-1，其中，i为整数且1＜i≤n-1，和被第二区域覆盖的数据yn、yn+1、……、yn+m-1、yn+m的扩展的第二区域。
  5. 如权利要求4所述的数据处理系统，其特征在于，数据处理单元被构造为：
    计算被以数据y1和数据yi为边界的缩小的第一区域覆盖的数据y1、y2、……、yi的压缩误差CE1_i和被以数据yi和数据yn+m为边界的扩展的第二区域覆盖的数据yi、yi+1、……、yn-1、yn、yn+1、……、yn+m-1、yn+m的压缩误差CEi_n+m，并确定压缩误差CE1_i是否小于或等于第一区域的压缩误差CE1、且压缩误差CEi_n+m是否小于或 等于第二区域的压缩误差CE2，
    其中，数据处理单元被构造为：当确定压缩误差CE1_i小于或等于第一区域的压缩误差CE1、且压缩误差CEi_n+m小于或等于第二区域的压缩误差CE2时，将以数据y1和数据yi为边界且覆盖数据y1、y2、……、yi的缩小的第一区域作为重新设置的第一区域。
  6. 如权利要求5所述的数据处理系统，其特征在于，数据处理单元被构造为：
    当确定压缩误差CE1_i+j-1大于第一区域的压缩误差CE1、或压缩误差CEi+j-1_n+m大于第二区域的压缩误差CE2时，计算被以数据y1和数据yi+j为边界的缩小的第一区域覆盖的数据y1、y2、……、yi+j的压缩误差CE1_i+j和被以数据yi+j和数据yn+m为边界的扩展的第二区域覆盖的数据yi+j、……、yn-1、yn、yn+1、……、yn+m-1、yn+m的压缩误差CEi+j_n+m，其中，j为整数且1≤j＜n-i，
    其中，数据处理单元被构造为：当确定压缩误差CE1_i+j小于或等于第一区域的压缩误差CE1且压缩误差CEi+j_n+m小于或等于第二区域的压缩误差CE2时，将以数据y1和数据yi+j为边界的缩小的第一区域作为重新设置的第一区域。
  7. 如权利要求1所述的数据处理系统，其特征在于，数据处理单元被构造为将所述多个区域设置为包括第N区域和第N+1区域，其中，第N区域与第N-1区域彼此相邻并覆盖多个数据，第N+1区域与第N区域彼此相邻并覆盖多个数据，其中，N为整数且N＞1，
    其中，数据处理单元被构造为：
    当重新设置第N-1区域时，将第N区域前向扩展为以作为重新设置的第N-1区域的结束边界的数据作为开始边界的扩展的第N区域；
    前向扩展第N+1区域以得到与扩展的第N区域交叠的扩展的第N+1区域，计算被扩展的第N+1区域覆盖的数据的压缩误差，并根据计算的压缩误差重新设置扩展的第N区域；
    对重新设置的扩展的第N区域进行压缩。
  8. 一种数据处理方法，其特征在于，所述方法包括：
    根据压缩允许误差值在待压缩的多个数据的二维空间表示中设置彼此相邻的多个区域，其中，所述多个区域包括分别覆盖多个数据的彼此相邻的第一区域和第二区域；
    前向扩展第二区域以得到与第一区域交叠的扩展的第二区域，计算被扩展的第二区域覆盖的数据的压缩误差，并根据计算的压缩误差重新设置第一区域；
    对被重新设置的第一区域覆盖的数据进行压缩。
  9. 如权利要求8所述的方法，其特征在于，根据旋转门SDT算法的规则来执行设置多个区域的步骤、计算压缩误差的步骤、和进行压缩的步骤。
  10. 如权利要求8所述的方法，其特征在于，重新设置第一区域的步骤包括：
    将被第一区域与扩展的第二区域彼此交叠的交叠区域覆盖的数据作为重新设置的第一区域的结束边界。
  11. 如权利要求8所述的数据处理方法，其特征在于，
    所述多个数据包括在二维空间表示中顺序排列的数据y1、y2、……、yn-1、yn、yn+1、……、yn+m、……，其中，n和m为大于2的整数，
    第一区域以数据y1和数据yn为边界并覆盖数据y1、y2、……、yn-1、yn，
    第二区域以数据yn和数据yn+m为边界并覆盖数据yn、yn+1、……、yn+m-1、yn+m，
    前向扩展第二区域的步骤包括：
    在二维空间表示中沿与数据的排列所沿的方向相对的方向扩展第二区域以与第一区域交叠，从而得到覆盖了被交叠的区域覆盖的数据yi、yi+1、……、yn-1，其中，i为整数且1＜i≤n-1，和被第二区域覆盖的数据yn、yn+1、……、yn+m-1、yn+m的扩展的第二区域。
  12. 如权利要求11所述的数据处理方法，其特征在于，
    计算被扩展的第二区域覆盖的数据的压缩误差的步骤包括：
    计算被以数据y1和数据yi为边界的缩小的第一区域覆盖的数据y1、y2、……、yi的压缩误差CE1_i和被以数据yi和数据yn+m为边界的扩展的第二区域覆盖的数据yi、yi+1、……、yn-1、yn、yn+1、……、yn+m-1、yn+m的压缩误差CEi_n+m，并确定压缩误差CE1_i是否小于或等于第一区域的压缩误差CE1、且压缩误差CEi_n+m是否小于或等于第二区域的压缩误差CE2，
    根据计算的压缩误差重新设置第一区域的步骤包括：
    当确定压缩误差CE1_i小于或等于第一区域的压缩误差CE1、且压缩误差CEi_n+m小于或等于第二区域的压缩误差CE2时，将以数据y1和数据yi为边界且覆盖数据y1、y2、……、yi的缩小的第一区域作为重新设置的第一区域。
  13. 如权利要求12所述的数据处理方法，其特征在于，
    计算与扩展的第二区域对应的压缩误差的步骤包括：
    当确定压缩误差CE1_i+j-1大于第一区域的压缩误差CE1、或压缩误差CEi+j-1_n+m大于第二区域的压缩误差CE2时，计算被以数据y1和数据yi+j为边界的缩小的第一区域覆盖的数据y1、y2、……、yi+j的压缩误差CE1_i+j和被以数据yi+j和数据yn+m为边界的扩展的第二区域覆盖的数据yi+j、……、yn-1、yn、yn+1、……、yn+m-1、yn+m的压缩误差CEi+j_n+m，其中，j为整数且1≤j＜n-i，
    根据计算的压缩误差重新设置第一区域的步骤包括：当确定压缩误差CE1_i+j小于或等于第一区域的压缩误差CE1且压缩误差CEi+j_n+m小于或等于第二区域的压缩误差CE2时，将以数据y1和数据yi+j为边界的缩小的第一区域作为重新设置的第一区域。
  14. 如权利要求8所述的数据处理方法，其特征在于，所述多个区域包括第N区域和第N+1区域，其中，第N区域与第N-1区域彼此相邻并覆盖多个数据，第N+1区域与第N区域彼此相邻并覆盖多个数据，其中，N为整数且N＞1，
    所述方法还包括：
    当重新设置第N-1区域时，将第N区域前向扩展为以作为重新设置的第N-1区域的结束边界的数据作为开始边界的扩展的第N区域；
    前向扩展第N+1区域以得到与扩展的第N区域交叠的扩展的第N+1区域，计算被扩展的第N+1区域覆盖的数据的压缩误差，并根据计算的压缩误差重新设置扩展的第N区域；
    对重新设置的扩展的第N区域进行压缩。
  15. 电子设备，其特征在于，所述电子设备包括：
    至少一个处理器；以及
    与所述至少一个处理器连接的存储器，所述存储器具有存储于其中的指令，所述指令在被所述至少一个处理器执行时使所述电子设备执行如权利要求8到14中任一所述的方法。
  16. 非暂时性机器可读介质，其特征在于，所述非暂时性机器可读介质上存储有计算机可执行指令，所述计算机可执行指令在被执行时使至少一个处理器执行根据权利要求8至14中任一项所述的方法。
  17. 计算机程序产品，其特征在于，所述计算机程序产品包括计算机可执行指令，所述计算机可执行指令在被执行时使至少一个处理器执行根据权利要求8至14中任一项所述的方法。
PCT/CN2019/083874 2019-04-23 2019-04-23 数据处理系统和方法 WO2020215203A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP19926607.3A EP3917014A4 (en) 2019-04-23 2019-04-23 DATA PROCESSING SYSTEM AND PROCEDURES
PCT/CN2019/083874 WO2020215203A1 (zh) 2019-04-23 2019-04-23 数据处理系统和方法
CN201980091284.0A CN113383496A (zh) 2019-04-23 2019-04-23 数据处理系统和方法
US17/432,971 US11381251B2 (en) 2019-04-23 2019-04-23 Data processing system and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/083874 WO2020215203A1 (zh) 2019-04-23 2019-04-23 数据处理系统和方法

Publications (1)

Publication Number Publication Date
WO2020215203A1 true WO2020215203A1 (zh) 2020-10-29

Family

ID=72940670

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/083874 WO2020215203A1 (zh) 2019-04-23 2019-04-23 数据处理系统和方法

Country Status (4)

Country Link
US (1) US11381251B2 (zh)
EP (1) EP3917014A4 (zh)
CN (1) CN113383496A (zh)
WO (1) WO2020215203A1 (zh)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102664635A (zh) * 2012-03-06 2012-09-12 华中科技大学 一种精度可控的自适应数据压缩方法
EP2698679A1 (en) * 2012-08-16 2014-02-19 Siemens Aktiengesellschaft System and method for compressing production data stream and filtering compressed data with different criteria.
CN105117504A (zh) * 2015-07-27 2015-12-02 卡斯柯信号有限公司 一种用于信号系统的改进型旋转门数据采集方法

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9058422B2 (en) * 2013-09-28 2015-06-16 Intel Corporation Polling determination
US9779491B2 (en) * 2014-08-15 2017-10-03 Nikon Corporation Algorithm and device for image processing
US10685306B2 (en) * 2015-12-07 2020-06-16 Sap Se Advisor generating multi-representations of time series data
WO2018213617A1 (en) * 2017-05-18 2018-11-22 Uptake Technologies, Inc. Computing system and method for approximating predictive models and time-series values
WO2019199729A1 (en) * 2018-04-09 2019-10-17 The Government Of The United States Of America, As Represented By The Secretary Of The Navy Reducing error in data compression
US10756756B2 (en) * 2018-09-14 2020-08-25 Hewlett Packard Enterprise Development Lp Floating point data set compression
JP7345262B2 (ja) * 2019-03-11 2023-09-15 キヤノン株式会社 データ圧縮装置、データ処理装置、データ圧縮方法、プログラム、及び学習済みモデル
CN110996380B (zh) * 2019-11-29 2023-08-04 京东方科技集团股份有限公司 终端的唤醒周期调整方法、装置、存储介质及电子设备
US11405622B2 (en) * 2020-04-22 2022-08-02 Apple Inc. Lossless compression techniques
WO2022018466A1 (en) * 2020-07-22 2022-01-27 Citrix Systems, Inc. Determining server utilization using upper bound values

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102664635A (zh) * 2012-03-06 2012-09-12 华中科技大学 一种精度可控的自适应数据压缩方法
EP2698679A1 (en) * 2012-08-16 2014-02-19 Siemens Aktiengesellschaft System and method for compressing production data stream and filtering compressed data with different criteria.
CN105117504A (zh) * 2015-07-27 2015-12-02 卡斯柯信号有限公司 一种用于信号系统的改进型旋转门数据采集方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3917014A4 *

Also Published As

Publication number Publication date
EP3917014A1 (en) 2021-12-01
CN113383496A (zh) 2021-09-10
EP3917014A4 (en) 2022-11-16
US11381251B2 (en) 2022-07-05
US20220103184A1 (en) 2022-03-31

Similar Documents

Publication Publication Date Title
CN112214727A (zh) 运算加速器
JP5742275B2 (ja) 幾何学変換を行うためのシステム、方法、コンピュータプログラム及び記憶媒体
JP2002245098A (ja) 六面体メッシュ生成方法および装置
WO2020014893A1 (zh) 反卷积实现方法及相关产品
CN105678831A (zh) 一种图像渲染方法和装置
CN114995782A (zh) 数据处理方法、装置、设备和可读存储介质
EP3575947B1 (en) Systems and methods for interactive image caricaturing by an electronic device
WO2020215203A1 (zh) 数据处理系统和方法
CN108230454B (zh) 一种全景图片的切图方法、装置及存储介质
CN110678862B (zh) 用于硬连线乘法器电路的数据打包技术
US20160357878A1 (en) Modelling method and system
Heath-Brown Counting rational points on algebraic varieties
KR102611430B1 (ko) 반도체 설계 시뮬레이션을 위한 명령들을 포함하는 저장 매체, 반도체 설계 시스템, 그리고 반도체 설계 시뮬레이션의 방법
TW201704988A (zh) 用於蒙哥馬利域的計算方法、計算裝置和電腦軟體產品
CN110414663A (zh) 神经网络的卷积实现方法及相关产品
CN110048844B (zh) 实现sm2算法的多倍点计算方法、装置
CN106502775A (zh) 一种分时调度dsp算法的方法和系统
CN111179175A (zh) 基于卷积神经网络的图像处理方法、装置及存储介质
CN105354150A (zh) 一种内容匹配方法和装置
JP5349931B2 (ja) 情報処理装置、情報処理方法及びプログラム
CN113722746B (zh) 电缆施工图纸的混沌加密方法及系统
CN113595730B (zh) 一种用于工程中生成ecc曲线的处理方法和装置
CN110009559B (zh) 图像处理方法及装置
JP5595070B2 (ja) Ｃａｄデータ送信方法、ｃａｄデータ受信方法、ｃａｄデータ送信装置、ｃａｄデータ受信装置、ｃａｄデータ送信プログラム、およびｃａｄデータ受信プログラム、ならびにデータ構造
JP5997480B2 (ja) 画像処理装置、画像処理方法およびプログラム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19926607

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2019926607

Country of ref document: EP

Effective date: 20210824

NENP Non-entry into the national phase

Ref country code: DE