A Robust Smoothing Filtering Method, System, Device and Storage Medium

Abstract

The present application relates to a robust smoothing filtering method, system, device and storage medium. The method includes performing sliding window filtering on target data to obtain several segments of data to be detected; detecting the coincidence situation of all adjacent segments of data to be detected to mark several coincidence regions to be detected with outliers; performing iterative filtering detection on all the coincidence regions to be detected, and determining the outlier region and non-outlier region of the target data; adjusting the weights of the outlier region and non-outlier region of the target data, performing sliding filtering on the adjusted target data, and obtaining a filtering result through window weighting calculation. The present application effectively reduces the influence of outliers, improves the filtering effect, can effectively reduce the amount of calculation, improves the filtering efficiency, and is preferably applicable to scenarios with limited computing power.

Description

Technical Field

[0001] This application relates to the technical field of signal filtering, and in particular, to a robust smoothing filtering method, system, device, and storage medium. Background Art

[0002] Currently, signals such as data and images are in complex environments during the processes of acquisition, obtaining, transmission, and conversion. Affected by variables such as light and electromagnetism, all signals are disturbed by visible or invisible noise to varying degrees. Therefore, it is necessary to perform filtering processing on the signals.

[0003] When performing sliding filtering on signals, it is easily affected by outliers in the signals, making the processed signals less smooth than ideal. Existing smoothing filtering techniques use local weighted regression on multiple points around the target point of the input signal, and detect the outliers multiple times for the target point to adjust its weight. Since such calculations are performed for each point, the overall computational amount is large and not suitable for scenarios with limited computing power. Therefore, the inventor believes that existing smoothing filtering techniques need to be further improved. Summary of the Invention

[0004] In view of this, this application provides a robust smoothing filtering method, system, device, and storage medium to solve the technical problem that existing smoothing filtering techniques need to perform multiple local regression calculations for each point, resulting in a large computational amount.

[0005] To solve the above problems, in a first aspect, this application provides a robust smoothing filtering method, and the method includes:

[0006] Perform sliding window filtering on target data to obtain several segments of data to be detected;

[0007] Detect the coincidence situation of all adjacent segments of data to be detected to mark several overlapping regions to be detected with outliers;

[0008] Perform iterative filtering detection on all the overlapping regions to be detected, and determine the outlier region and non-outlier region of the target data;

[0009] Adjust the weights of the outlier region and non-outlier region of the target data, perform sliding filtering on the adjusted target data, and obtain the filtering result through window weighted calculation.

[0010] Optionally, the performing sliding window filtering on target data to obtain several segments of data to be detected includes:

[0011] Divide the target data into several segments of initial data with a preset first sliding window;

[0012] Perform average filtering on each segment of the initial data with a preset second sliding window, and use the initial data after the average filtering as the data to be detected; wherein the length of the second sliding window is less than the length of the first sliding window.

[0013] Optionally, detecting the coincidence situation of all adjacent segments of the data to be detected to mark a number of coincidence regions with outliers, including:

[0014] Calculate the difference of the overlapping part of the currently adjacent segments of the data to be detected by difference, wherein the length of the overlapping part is the difference between the lengths of the first sliding window and the second sliding window;

[0015] Determine whether the difference of the overlapping part of the currently adjacent segments of the data to be detected exceeds a preset threshold. If so, determine that there are outliers in the overlapping part of the currently adjacent segments of the data to be detected and mark it as a coincidence region to be detected;

[0016] After detecting the coincidence situation of all adjacent segments of the data to be detected, mark a number of coincidence regions with outliers.

[0017] Optionally, performing iterative filtering detection on all the coincidence regions to be detected to determine the outlier region and non-outlier region of the target data, including:

[0018] Perform sliding window filtering on each of the coincidence regions to be detected, and several segments of target detection data are obtained corresponding to each of the coincidence regions to be detected;

[0019] Detect the coincidence situation of adjacent segments of the target detection data in each of the coincidence regions to be detected, and mark all the outlier regions of the target data, and mark the data outside the outlier region of the target data as the non-outlier region.

[0020] Optionally, performing sliding window filtering on each of the coincidence regions to be detected, and several segments of target detection data are obtained corresponding to each of the coincidence regions to be detected, including:

[0021] Divide the currently detected coincidence region into several segments of initial detection data with a preset third sliding window, wherein the step size of the third sliding window is less than the length, and the length is less than the difference between the lengths of the first sliding window and the second sliding window;

[0022] Perform average filtering on each segment of the initial detection data with a preset fourth sliding window, and use the initial detection data after the average filtering as the target detection data; wherein the length of the fourth sliding window is less than the length of the third sliding window.

[0023] Optionally, detecting the coincidence of adjacent segment target detection data in each to-be-detected coincidence region, marking all outlier regions of the target data, and marking the data outside the outlier regions in the target data as non-outlier regions, including:

[0024] Calculating the difference of the overlapping part of the current adjacent segment target detection data by difference, where the length of the overlapping part is the length difference between the third sliding window and the fourth sliding window;

[0025] Judging whether the difference of the overlapping part of the current adjacent segment target detection data exceeds a preset threshold. If so, determining that there are outliers in the overlapping part of the adjacent segment target detection data and marking it as an outlier region;

[0026] After detecting the coincidence of all adjacent segment target detection data in all to-be-detected coincidence regions, marking all outlier regions of the target data.

[0027] Optionally, adjusting the weights of the outlier region and the non-outlier region of the target data, performing sliding filtering on the adjusted target data, and obtaining a filtering result through window weighting calculation, including:

[0028] Setting the weight of the points in all outlier regions of the target data to a preset weight, where the preset weight is less than 1;

[0029] Setting the weight of the points in all non-outlier regions of the target data to 1;

[0030] Filtering the target data with a preset fifth sliding window, calculating the mean value of the sum of the products of the weights of each point in each fifth sliding window during sliding, and taking it as the filtering result.

[0031] In a second aspect, a robust smoothing filtering system provided by the present application, the system includes:

[0032] A sliding filtering module, configured to perform sliding window filtering on target data to obtain several segments of to-be-detected data;

[0033] A detection module, configured to detect the coincidence of all adjacent segment to-be-detected data to mark several to-be-detected coincidence regions with outliers;

[0034] A determination module, configured to perform iterative filtering detection on all the to-be-detected coincidence regions and determine the outlier region and the non-outlier region of the target data;

[0035] A filtering calculation module, configured to adjust the weights of the outlier region and the non-outlier region of the target data, perform sliding filtering on the adjusted target data, and obtain a filtering result through window weighting calculation.

[0036] In a third aspect, a computer device provided by the present application adopts the following technical solution:

[0037] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the robust smoothing filtering method are implemented.

[0038] In a fourth aspect, a computer-readable storage medium provided by the present application adopts the following technical solution:

[0039] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the steps of the robust smoothing filtering method are implemented.

[0040] The beneficial effects of adopting the above embodiments are as follows: By sliding window filtering, the target data is first divided into several data to be detected, which is convenient for subsequent segmented detection; by detecting the coincidence situation of all adjacent segments of data to be detected, that is, the coincidence situation of the heads and tails of adjacent segments of data to be detected, it is convenient to mark several overlapping regions with outliers; then, by performing iterative filtering detection on all overlapping regions to be detected, the outlier regions are refined, so as to determine all the outlier regions and non-outlier regions of the target data, and the weights of the outlier regions and non-outlier regions can be automatically adjusted. Sliding filtering is performed according to the adjusted target data, and the filtering result is obtained through window weighted calculation, thereby effectively reducing the influence of outliers, improving the filtering effect, and being able to effectively reduce the calculation amount and improve the filtering efficiency, which is better applicable to scenarios with limited computing power. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] FIG. 1 is a schematic diagram of an application scenario of a robust smoothing filtering system provided by the present application;

[0042] Figure 2 is a flowchart of a method of an embodiment of the robust smoothing filtering method provided by the present application;

[0043] Figure 3 is a flowchart of a method of an embodiment of step S201 of the robust smoothing filtering method provided by the present application;

[0044] Figure 4 is a flowchart of a method of an embodiment of step S202 of the robust smoothing filtering method provided by the present application;

[0045] Figure 5 is a flowchart of a method of an embodiment of step S203 of the robust smoothing filtering method provided by the present application;

[0046] Figure 6 is a flowchart of a method of an embodiment of step S501 of the robust smoothing filtering method provided by the present application;

[0047] Figure 7 This is a flowchart of a method for an embodiment of step S502 of the robust smoothing filtering method provided by this application;

[0048] Figure 8 This is a flowchart of a method for an embodiment of step S204 of the robust smoothing filtering method provided by this application;

[0049] Figure 9 This is a schematic block diagram of the principle of an embodiment of the robust smoothing filtering system provided by this application;

[0050] Figure 10 This is a schematic block diagram of the principle of an embodiment of the computer device provided by this application. Detailed implementation manners

[0051] Next, the preferred embodiments of this application will be specifically described in conjunction with the accompanying drawings. Among them, the accompanying drawings form a part of this application and are used together with the embodiments of this application to explain the principle of this application, rather than to limit the scope of this application.

[0052] In the description of this application, "a plurality of" means two or more, unless otherwise specifically and clearly defined.

[0053] Referring to "embodiment" herein means that a specific feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of this application. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. Those skilled in the art will explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0054] This application provides a robust smoothing filtering method, system, device, and storage medium, which will be described in detail below respectively.

[0055] Figure 1 This is a schematic diagram of the scenario of the robust smoothing filtering system provided by the embodiments of this application. The system may include a server 100, and the robust smoothing filtering system is integrated in the server 100, such as Figure 1 the server in

[0056] In the embodiments of this application, the server 100 is mainly used for:

[0057] Performing sliding window filtering on target data to obtain several segments of data to be detected;

[0058] Detecting the coincidence situation of all adjacent segments of data to be detected to mark several overlapping regions to be detected with outliers;

[0059] Iteratively filter and detect all overlapping regions to be detected, and determine the outlier region and non-outlier region of the target data;

[0060] Adjust the weights of the outlier region and non-outlier region of the target data, perform sliding filtering on the adjusted target data, and obtain the filtering result through window weighting calculation.

[0061] In the embodiment of the present application, the server 100 may be an independent server or a server network or server cluster composed of servers. For example, the server 100 described in the embodiment of the present application includes, but is not limited to, a computer, a network host, a single network server, a set of multiple network servers, or a cloud server composed of multiple servers. Among them, the cloud server is composed of a large number of computers or network servers based on cloud computing.

[0062] It can be understood that the terminal 200 used in the embodiment of the present application may be a device that includes both receiving and transmitting hardware, that is, a device having receiving and transmitting hardware capable of performing two-way communication on a two-way communication link. Such a device may include: a cellular or other communication device, which has a single-line display or a multi-line display or a cellular or other communication device without a multi-line display. Specifically, the terminal 200 may be a desktop computer, a portable computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, an embedded device, etc. The type of the terminal 200 is not limited in this embodiment.

[0063] Those skilled in the art can understand that Figure 1 the application environment shown in Figure 1 is only an application scenario of the solution of the present application, and does not constitute a limitation on the application scenario of the solution of the present application. Other application environments may also include more or fewer terminals than Figure 1 shown in

[0064] For example, only 2 terminals are shown in Figure 1 . It can be understood that the robust smoothing filtering system may also include one or more other terminals, which are not specifically limited here.

[0065] It should be noted that Figure 1The schematic diagram of the scenario of the robust smoothing filter system shown is merely an example. The robust smoothing filter system and scenarios described in the embodiments of the present application are for more clearly explaining the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art know that with the evolution of the robust smoothing filter system and the emergence of new service scenarios, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems.

[0066] Referring to Figure 2 FIG. is a flowchart of a method according to an embodiment of the robust smoothing filter method provided by the present application. The robust smoothing filter method includes the following steps:

[0067] S201. Perform sliding window filtering on the target data to obtain several segments of data to be detected;

[0068] S202. Detect the coincidence situation of all adjacent segments of data to be detected to mark several coincidence regions to be detected with outliers;

[0069] S203. Perform iterative filtering detection on all coincidence regions to be detected, and determine the outlier region and non-outlier region of the target data;

[0070] S204. Adjust the weights of the outlier region and non-outlier region of the target data, perform sliding filtering on the adjusted target data, and obtain the filtering result through window weighting calculation.

[0071] In this embodiment, the target data refers to the sampled data of the input signal. Considering that when using smoothing filtering, the influence of outliers on the smoothing filtering result at the tail and head of the data is different from that at the middle of the data to detect the outliers of the data, and by adjusting the weights of the data at different positions, the influence of outliers is reduced.

[0072] In this embodiment, the target data is first divided into several segments of data to be detected through sliding window filtering, which is convenient for subsequent segmented detection; by detecting the coincidence situation of all adjacent segments of data to be detected, that is, the coincidence situation of the head and tail of the adjacent segments of data to be detected, it is convenient to mark several coincidence regions to be detected with outliers; then, through iterative filtering detection of all coincidence regions to be detected, the outlier region is refined, so as to determine all the outlier regions and non-outlier regions of the target data, and the weights of the outlier region and non-outlier region can be automatically adjusted. Perform sliding filtering according to the adjusted target data, and obtain the filtering result through window weighting calculation, so as to effectively reduce the influence of outliers, improve the filtering effect, and can effectively reduce the calculation amount and improve the filtering efficiency, which is better applicable to scenarios with limited computing power.

[0073] Referring to Figure 3The flowchart of a method for an embodiment of step S201 of the robust smoothing filtering method provided in this application. Step S201 includes the following steps:

[0074] S301. Divide the target data into several segments of initial data with a preset first sliding window;

[0075] S302. Perform average filtering on each segment of initial data with a preset second sliding window, and use the initial data after average filtering as the data to be detected; wherein the length of the second sliding window is less than the length of the first sliding window.

[0076] In this embodiment, the target data is represented as DATA1; the length of the first sliding window is L1, the step size is L2, and L2 < L1; the length of the second sliding window is L3, and L3 < L1.

[0077] Specifically, divide the target data DATA1 into N1 segments of initial data with the first sliding window, where the nth segment of initial data is DATA1_n; further, perform sliding window average filtering on the nth segment of initial data DATA1_n with the second sliding window to obtain the corresponding data to be detected DATA2_n; further, perform average filtering of the second sliding window on all N1 segments of initial data to obtain all N1 segments of data to be detected, facilitating subsequent detection of the coincidence of the tails and heads of adjacent segments of data to be detected.

[0078] Refer to Figure 4 The flowchart of a method for an embodiment of step S202 of the robust smoothing filtering method provided in this application. Step S202 includes the following steps:

[0079] S401. Calculate the difference of the overlapping part of the currently adjacent segments of data to be detected, where the length of the overlapping part is the difference between the lengths of the first sliding window and the second sliding window;

[0080] S402. Determine whether the difference of the overlapping part of the currently adjacent segments of data to be detected exceeds a preset threshold. If so, determine that there are outliers in the overlapping part of the currently adjacent segments of data to be detected, and mark it as the area to be detected for overlap;

[0081] S403. After detecting the coincidence of all adjacent segments of data to be detected, mark several areas to be detected for overlap where outliers exist.

[0082] In this embodiment, the area to be detected for overlap represents the data area with anomalies; after all the data to be detected are converted by digital-to-analog conversion, the corresponding voltage amplitudes are obtained.

[0083] Specifically, calculate the difference Dn between the overlapping parts of the nth piece of data to be detected DATA2_n and the (n - 1)th piece of data to be detected DATA2_(n - 1). D1 = DATA2_n - DATA2_(n - 1), and the length of the overlapping part is L1 - L2. If the difference Dn exceeds the preset threshold T1, it is determined that there are outliers in the current overlapping part, and the current overlapping part is marked as the overlapping area to be detected, denoted as OUTLIER_n. If the difference Dn does not exceed the preset threshold T1, it is determined that there are no outliers in the current overlapping part. In this embodiment, the threshold is an empirical value, specifically 0.2 mV. In other embodiments, the preset threshold can be adaptively modified according to actual needs, and no further limitation is provided here.

[0084] In this embodiment, after detecting the overlapping situations of all adjacent segments of data to be detected, N2 overlapping areas to be detected with outliers are obtained, thereby effectively detecting potential outlier areas and helping to improve the filtering efficiency.

[0085] Refer to Figure 5 FIG. is a flowchart of an embodiment of method step S203 of the robust smoothing filtering method provided by the present application. The step S203 includes the following steps:

[0086] S501. Perform sliding window filtering on each overlapping area to be detected, and several segments of target detection data are obtained corresponding to each overlapping area to be detected.

[0087] S502. Detect the overlapping situations of adjacent segments of target detection data in each overlapping area to be detected, mark all outlier areas of the target data, and mark the data outside the outlier areas of the target data as non-outlier areas.

[0088] In this embodiment, the target detection data refers to the segmented data that may contain outliers. The outlier area refers to the data area where outliers are determined to exist. The non-outlier area refers to the data area where outliers are determined not to exist.

[0089] In this embodiment, by performing sliding window filtering and outlier detection on each overlapping area to be detected, the area where outliers exist is more accurately determined, and the outlier area and non-outlier area of the target data are determined, which helps to accurately adjust the data weight and improve the filtering effect.

[0090] Refer to Figure 6 FIG. is a flowchart of an embodiment of method step S501 of the robust smoothing filtering method provided by the present application. The step S501 includes the following steps:

[0091] S601. Divide the current overlapping area to be detected into several segments of initial detection data with a preset third sliding window, where the step size of the third sliding window is less than the length, and the length is less than the length difference between the first sliding window and the second sliding window.

[0092] S602. Perform mean filtering on each segment of the initial detection data with a preset fourth sliding window, and use the initial detection data after mean filtering as the target detection data; wherein the length of the fourth sliding window is less than the length of the third sliding window.

[0093] In this embodiment, the length of the third sliding window is L4, the step size is L5, L5 < L4 < (L1 - L2); the length of the fourth sliding window is L6, L6 < L4.

[0094] Specifically, divide the current overlapping region to be detected OUTLIER_n into N3 segments of initial detection data with the third sliding window, where the nth segment of initial detection data is DATA3_n; further, perform sliding window mean filtering on the nth segment of initial detection data DATA3_n with the fourth sliding window to obtain the corresponding target detection data; further, perform mean filtering of the fourth sliding window on all N3 segments of initial detection data to obtain all N3 segments of target detection data, which is convenient for subsequent segmented detection of the overlapping conditions of the tails and heads of adjacent segments of target detection data.

[0095] Refer to Figure 7 FIG. S502 of the robust smoothing filtering method provided by the present application is a flowchart of an embodiment of the method, and this step S502 includes the following steps:

[0096] S701. Calculate the difference of the overlapping part of the current adjacent segments of target detection data by difference, where the length of the overlapping part is the difference between the lengths of the third sliding window and the fourth sliding window;

[0097] S702. Determine whether the difference of the overlapping part of the current adjacent segments of target detection data exceeds a preset threshold. If so, determine that there are outliers in the overlapping part of the adjacent segments of target detection data and mark it as an outlier region;

[0098] S703. After detecting the overlapping conditions of all adjacent segments of target detection data in all overlapping regions to be detected, mark all outlier regions of the target data.

[0099] Specifically, calculate the difference Qn of the overlapping part between the nth segment of target detection data DATA3_n and the (n - 1)th segment of target detection data DATA3_(n - 1) by difference, Qn = DATA3_n - DATA3_(n - 1), the length of the overlapping part is L4 - L5. If the difference Qn exceeds the preset threshold T1, determine that there are outliers in the overlapping part of the current adjacent segments of target detection data and mark it as an outlier region RANGE_n; after detecting the overlapping conditions of all adjacent segments of target detection data, obtain all outlier regions of the target data, and mark the data of the target data except the outlier regions as non - outlier regions.

[0100] Reference Figure 8 FIG. 227 is a flowchart of a method according to an embodiment of step S204 of the robust smoothing filtering method provided in the present application. The step S204 includes the following steps:

[0101] S801. Set the weights of the points in all outlier regions of the target data to a preset weight, where the preset weight is less than 1;

[0102] S802. Set the weights of the points in all non-outlier regions of the target data to 1;

[0103] S803. Filter the target data with a preset fifth sliding window, calculate the mean value of the sum of the products of the weights of each point in each fifth sliding window during the sliding, and use it as the filtering result.

[0104] Specifically, set the weights of the points in all outlier regions of the target data to a preset weight, where the preset weight is 1 / N4. In this embodiment, N4 takes an empirical value, specifically 10. In other embodiments, the value of N4 can be adaptively adjusted according to actual needs, and no further limitation is made here.

[0105] Further, the length of the fifth sliding window is L8, and the step size is 1. Filter the target data with the fifth sliding window. Exemplarily, for the i-th window, multiply the points in the window by their respective weights and add them, and then divide by L8 to obtain the filtered value of the points in the window; further, obtain the filtered values of all windows, which is the filtering result of the target data.

[0106] Compared with the prior art, performing multiple local regression calculations on each point of the filtered data has a large amount of calculation. In the embodiment of the present application, the target data is first divided into several pieces of data to be detected through sliding window filtering, which is convenient for subsequent segmented detection; by detecting the coincidence of all adjacent segments of the data to be detected, that is, the coincidence of the heads and tails of the adjacent segments of the data to be detected, it is convenient to mark several overlapping regions with outliers; then, through iterative filtering detection of all overlapping regions to be detected, the outlier regions are refined, so as to determine all outlier regions and non-outlier regions of the target data, and then the weights of the outlier regions and non-outlier regions can be automatically adjusted, and sliding filtering is performed according to the adjusted target data, and the filtering result is obtained through window weighted calculation, thereby effectively reducing the influence of outliers, improving the filtering effect, and being able to effectively reduce the amount of calculation and improve the filtering efficiency, and is preferably applicable to scenarios with limited computing power.

[0107] It should be understood that the magnitudes of the sequence numbers of the above steps do not mean the order of execution. The order of execution of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0108] This embodiment also provides a robust smoothing filtering system, which corresponds one-to-one with the robust smoothing filtering method in the above embodiment. As Figure 9 shown, the robust smoothing filtering system includes a sliding filtering module 901, a detection module 902, a determination module 903, and a filtering calculation module 904. The detailed description of each functional module is as follows:

[0109] The sliding filtering module 901 is used to perform sliding window filtering on the target data to obtain several segments of data to be detected;

[0110] The detection module 902 is used to detect the coincidence situation of all adjacent segments of data to be detected, so as to mark several coincidence regions to be detected with outliers;

[0111] The determination module 903 is used to perform iterative filtering detection on all coincidence regions to be detected, and determine the outlier region and non-outlier region of the target data;

[0112] The filtering calculation module 904 is used to adjust the weights of the outlier region and non-outlier region of the target data, perform sliding filtering on the adjusted target data, and obtain the filtering result through window weighted calculation.

[0113] For the specific limitations of each module of the robust smoothing filtering system, reference can be made to the limitations of the robust smoothing filtering method in the above text, which will not be elaborated here. Each module in the above robust smoothing filtering system can be implemented in whole or in part by software, hardware, and their combinations. The above modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0114] Referring to Figure 10 , this application also correspondingly provides a computer device, which can be a computing device such as a mobile terminal, a desktop computer, a notebook, a palm computer, and a server. The computer device includes a processor 10, a memory 20, and a display 30. Figure 10 Only some components of the computer device are shown, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively.

[0115] The memory 20 may be an internal storage unit of a computer device in some embodiments, such as the hard disk or memory of the computer device. The memory 20 may also be an external storage device of the computer device in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the computer device. Further, the memory 20 may also include both an internal storage unit and an external storage device of the computer device. The memory 20 is used to store application software installed on the computer device and various types of data, such as program codes installed on the computer device. The memory 20 may also be used to temporarily store data that has been output or will be output. In one embodiment, a robust smoothing filter program 40 is stored on the memory 20, and the robust smoothing filter program 40 can be executed by the processor 10, so as to implement the robust smoothing filter method of each embodiment of the present application.

[0116] The processor 10 may be a central processing unit (CPU), a microprocessor or other data processing chips in some embodiments, and is used to run the program codes stored in the memory 20 or process data, such as executing the robust smoothing filter method, etc.

[0117] The display 30 may be an LED display, a liquid crystal display, a touch liquid crystal display, an OLED (Organic Light-Emitting Diode) toucher, etc. in some embodiments. The display 30 is used to display information on the computer device and to display a visual user interface. The components 10 - 30 of the computer device communicate with each other through a system bus.

[0118] In one embodiment, when the processor 10 executes the robust smoothing filter program 40 in the memory 20, the following steps are implemented:

[0119] Perform sliding window filtering on the target data to obtain several segments of data to be detected;

[0120] Detect the coincidence situation of all adjacent segments of data to be detected, so as to mark several overlapping regions to be detected with outliers;

[0121] Perform iterative filtering detection on all overlapping regions to be detected, and determine the outlier region and non-outlier region of the target data;

[0122] Adjust the weights of the outlier region and non-outlier region of the target data, perform sliding filtering on the adjusted target data, and obtain a filtering result through window weighting calculation.

[0123] This embodiment also provides a computer-readable storage medium, on which a robust smoothing filter program is stored. When the robust smoothing filter program is executed by a processor, the following steps are implemented:

[0124] Perform sliding window filtering on the target data to obtain several segments of data to be detected;

[0125] Detect the coincidence situation of all adjacent segments of data to be detected to mark several coincidence regions to be detected with outliers;

[0126] Perform iterative filtering detection on all coincidence regions to be detected, and determine the outlier region and non-outlier region of the target data;

[0127] Adjust the weights of the outlier region and non-outlier region of the target data, perform sliding filtering on the adjusted target data, and obtain the filtering result through window weighting calculation.

[0128] Those of ordinary skill in the art can understand that all or part of the processes in the above method embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above method embodiments.

[0129] Any reference to a memory, storage, database, or other medium used in the embodiments provided in this application may include non-volatile and / or volatile memories. Non-volatile memories may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0130] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed in the present application should be covered by the protection scope of the present application.

Claims

1. A robust smoothing filtering method, characterized in that, The method includes: Performing sliding window filtering on the target data to obtain several segments of data to be detected; Detecting the coincidence situation of all adjacent segments of data to be detected to mark several coincidence regions to be detected with outliers; Performing iterative filtering detection on all the coincidence regions to be detected and determining the outlier region and non-outlier region of the target data; Adjusting the weights of the outlier region and non-outlier region of the target data, performing sliding filtering on the adjusted target data, and obtaining a filtering result through window weighting calculation; Among them, the performing iterative filtering detection on all the coincidence regions to be detected and determining the outlier region and non-outlier region of the target data includes: Performing sliding window filtering on each coincidence region to be detected, and several segments of target detection data are correspondingly obtained for each coincidence region to be detected; Detecting the coincidence situation of adjacent segments of target detection data in each coincidence region to be detected, marking all outlier regions of the target data, and marking the data outside the outlier region of the target data as non-outlier regions; The performing sliding window filtering on each coincidence region to be detected, and several segments of target detection data are correspondingly obtained for each coincidence region to be detected includes: Dividing the current coincidence region to be detected into several segments of initial detection data with a preset third sliding window, where the step size of the third sliding window is less than the length, and the length is less than the length difference between the first sliding window and the second sliding window; Performing average filtering on each segment of the initial detection data with a preset fourth sliding window, and taking the averaged initial detection data as the target detection data; where the length of the fourth sliding window is less than the length of the third sliding window; The detecting the coincidence situation of adjacent segments of target detection data in each coincidence region to be detected, marking all outlier regions of the target data, and marking the data outside the outlier region of the target data as non-outlier regions includes: Calculating the difference of the overlapping part of the current adjacent segments of target detection data by difference, where the length of the overlapping part is the length difference between the third sliding window and the fourth sliding window; Judging whether the difference of the overlapping part of the current adjacent segments of target detection data exceeds a preset threshold. If so, determining that there are outliers in the overlapping part of the adjacent segments of target detection data and marking it as an outlier region; After detecting the coincidence situation of all adjacent segments of target detection data in all the coincidence regions to be detected, marking all outlier regions of the target data.

2. The robust smoothing filtering method according to claim 1, wherein The performing sliding window filtering on the target data to obtain several segments of data to be detected includes: Dividing the target data into several segments of initial data with a preset first sliding window; Performing average filtering on each segment of the initial data with a preset second sliding window, and taking the averaged initial data as the data to be detected; where the length of the second sliding window is less than the length of the first sliding window.

3. The robust smoothing filtering method according to claim 2, wherein, The detecting the coincidence situation of all adjacent segments of data to be detected to mark several coincidence regions to be detected with outliers includes: Calculating the difference of the overlapping part of the current adjacent segments of data to be detected by difference, where the length of the overlapping part is the length difference between the first sliding window and the second sliding window; Determine whether the difference in the overlapping part of the data to be detected in the current adjacent segment exceeds a preset threshold. If so, determine that there are outliers in the overlapping part of the data to be detected in the current adjacent segment, and mark it as the area to be detected for overlap; After detecting the overlapping situations of all the data to be detected in adjacent segments, mark a number of areas to be detected for overlap with outliers.

4. The robust smoothing filtering method according to claim 1, wherein Adjust the weights of the outlier area and the non-outlier area of the target data, perform sliding filtering on the adjusted target data, and obtain the filtering result through window weighting calculation, including: Set the weights of the points in all outlier areas of the target data to a preset weight, and the preset weight is less than 1; Set the weights of the points in all non-outlier areas of the target data to 1; Perform filtering on the target data with a preset fifth sliding window, calculate the mean value of the sum of the products of the weights of each point in each fifth sliding window during the sliding, and use it as the filtering result.

5. A robust smoothing filter system, characterized in that, The system includes: A sliding filtering module, which is used to perform sliding window filtering on the target data to obtain several segments of data to be detected; A detection module, which is used to detect the overlapping situations of all the data to be detected in adjacent segments to mark a number of areas to be detected for overlap with outliers; A determination module, which is used to perform iterative filtering detection on all the areas to be detected for overlap and determine the outlier area and the non-outlier area of the target data; A filtering calculation module, which is used to adjust the weights of the outlier area and the non-outlier area of the target data, perform sliding filtering on the adjusted target data, and obtain the filtering result through window weighting calculation; Among them, performing sliding window filtering on each area to be detected for overlap, and each area to be detected for overlap correspondingly obtains several segments of target detection data, including: Divide the current area to be detected for overlap into several segments of initial detection data with a preset third sliding window, where the step size of the third sliding window is less than the length, and the length is less than the length difference between the first sliding window and the second sliding window; Perform average filtering on each segment of the initial detection data with a preset fourth sliding window, and use the averaged initial detection data as the target detection data; where the length of the fourth sliding window is less than the length of the third sliding window; Detect the overlapping situations of adjacent segments of target detection data in each area to be detected for overlap, and mark all the outlier areas of the target data, and mark the data outside the outlier areas in the target data as non-outlier areas, including: Calculate the difference in the overlapping part of the adjacent segments of target detection data by difference, where the length of the overlapping part is the length difference between the third sliding window and the fourth sliding window; Determine whether the difference in the overlapping part of the adjacent segments of target detection data exceeds a preset threshold. If so, determine that there are outliers in the overlapping part of the adjacent segments of target detection data, and mark it as the outlier area; After detecting the overlapping situations of all adjacent segments of target detection data in all the areas to be detected for overlap, mark all the outlier areas of the target data.

6. A computer device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the robust smoothing filtering method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the steps of the robust smoothing filtering method according to any one of claims 1 to 4.

Images