A real-time noise reduction method and device based on inter-frame filtering

Abstract

The application discloses a real-time noise reduction method and device based on interframe filtering, and the method comprises the following steps: receiving original data of a target; detecting the target by using the original data, calculating coordinate data of the target, and determining a target area; performing interframe filtering on the original data, and determining filtered data; determining a motion scene of the target according to the filtered data and the target area; and performing format integration on the filtered data, and determining noise reduction data. The application performs real-time interframe filtering on high-frame-frequency high-resolution video, applies a sobel algorithm to the interframe filtering algorithm, guarantees image processing effect, avoids the problem of image blurring in a motion scene in traditional interframe filtering noise reduction of an image, and takes into account excellent noise reduction performance of a static image, can perform adaptive filtering, and better adapts to a complex scene.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent road network, and particularly relates to a real-time noise reduction method and device based on inter-frame filtering. BACKGROUND

[0002] With the continuous development of video acquisition equipment, there are more and more scenes requiring image processing, and in order to ensure the resolution of the image, the resolution of the image is getting higher and higher. The traditional inter-frame filtering based on ARM architecture is more and more difficult to adapt to high-resolution and high-frame-rate image processing. The traditional inter-frame filtering noise reduction of the image is blurred in the motion scene, and cannot adapt to real-time scene changes, and is not suitable for multiple motion targets or small scenes and the like. SUMMARY

[0003] Therefore, the present application provides a real-time noise reduction method and device based on inter-frame filtering, which solves the problem that the prior art cannot adapt to high-resolution and high-frame-rate images and is not suitable for multiple motion targets or small scenes and the like.

[0004] According to a first aspect, the present application provides a real-time noise reduction method based on inter-frame filtering, comprising:

[0005] receiving original data of a target;

[0006] detecting the target by using the original data, calculating coordinate data of the target to determine a target region;

[0007] performing inter-frame filtering on the original data to determine filtered data;

[0008] determining a motion scene of the target according to the filtered data and the target region;

[0009] performing format integration on the filtered data to determine noise reduction data.

[0010] The real-time noise reduction method based on inter-frame filtering provided by the present application can ensure the image processing effect by performing real-time inter-frame filtering on high-frame-rate and high-resolution video, applying the sobel algorithm to the inter-frame filtering algorithm, avoiding the problem of image blurring in the motion scene of the traditional inter-frame filtering noise reduction of the image, and taking into account the excellent noise reduction performance of the static image, so that adaptive filtering can be performed and the method is better adapted to complex scenes.

[0011] In a first implementation manner of the first aspect, the receiving of the original data of the target comprises that the original data comprises video data and image data,

[0012] receiving the original data collected by an infrared detector;

[0013] The video data is formatted, the image data is intra-filtered, and processed data is determined.

[0014] In a second implementation form of the first aspect, in combination with the first implementation form of the first aspect, the target is detected using the original data, and coordinate data of the target is calculated, including:

[0015] The processed data is received through a line valid interface, a field valid interface, a data interface, and a clock interface;

[0016] A data matrix is formed according to different interface data;

[0017] The data matrix is calculated using a sobel operator, and coordinate data of the target is determined.

[0018] In a third implementation form of the first aspect, in combination with the second implementation form of the first aspect, the data matrix is formed according to different interface data, including:

[0019] When the line valid interface is in a valid state and is in a high level state, corresponding data is buffered to a local buffer when the clock interface is in a rising edge;

[0020] First line data of the processed data is received and written into a first buffer;

[0021] When second line data is buffered to the local buffer, first line data in the first buffer is read out, the read-out first line data is written into a second memory, and the second line data is written into the first buffer;

[0022] When third line data is buffered to the local buffer, second line data in the first buffer is read out, the read-out second line data is written into the second memory, and the third line data is written into the first buffer;

[0023] Data read out by each memory is used to determine a data matrix, and data in each buffer is simultaneously stored into a dynamic random access memory.

[0024] In a fourth implementation form of the first aspect, in combination with the third implementation form of the first aspect, the target is inter-filtered, and filtered data is determined, including:

[0025] First frame data of the target is stored into the dynamic random access memory;

[0026] When second frame data of the target is stored into the dynamic random access memory, data of a current frame after inter-filtering is calculated and stored into the dynamic random access memory;

[0027] determining a next frame of the current frame as the current frame, when the current frame data of the target is stored into the dynamic random memory, calculating the data of the current frame after inter-frame filtering, and storing the data into the dynamic random memory, until all frame data of the target is completed to be stored.

[0028] In a fifth implementation form of the first aspect, in combination with the fourth implementation form of the first aspect, the calculating the data of the current frame after inter-frame filtering comprises:

[0029] y(n)=x(n)*(1-k)+y(n-1)*k

[0030] wherein y(n) is the data of the current frame after inter-frame filtering, x(n) is the data of the current frame stored into the dynamic random memory, y(n-1) is the data of the last frame of the current frame after inter-frame filtering, and k is a preset filtering coefficient.

[0031] In a sixth implementation form of the first aspect, in combination with the first implementation form of the first aspect, the determining the motion scene of the target according to the filtered data and the target region comprises:

[0032] excluding the target region from the processed data to determine calculation data;

[0033] determining the current frame data of the calculation data and the data of the last frame after inter-frame filtering;

[0034] determining the motion scene by subtracting the current frame data from the data of the last frame after inter-frame filtering.

[0035] The real-time noise reduction method based on inter-frame filtering provided by the embodiment of the application can ensure the image processing effect by performing real-time inter-frame filtering on high-frame-frequency high-resolution video and applying the sobel algorithm to the inter-frame filtering algorithm, avoid the problem of image blur in a motion scene in the traditional inter-frame filtering noise reduction of images, and take into account the excellent noise reduction performance of static images, so that adaptive filtering can be performed and the method can better adapt to complex scenes.

[0036] According to the second aspect, the embodiment of the application provides a real-time noise reduction device based on inter-frame filtering, which comprises:

[0037] a data receiving module configured to receive original data of a target;

[0038] a motion target detection module configured to detect the target by using the original data and calculate coordinate data of the target to determine a target region;

[0039] an inter-frame filtering calculation module configured to perform inter-frame filtering on the original data and determine filtered data;

[0040] A scene determining module is configured to determine a motion scene of the target according to the filtered data and the target region.

[0041] An output module is configured to format and integrate the filtered data to determine noise reduction data.

[0042] The real-time noise reduction device based on inter-frame filtering provided by the embodiment can ensure the image processing effect by performing real-time inter-frame filtering on high-frame-frequency high-resolution video and applying the Sobel algorithm to the inter-frame filtering algorithm. The device can avoid the problem of image blur in a motion scene in traditional inter-frame filtering noise reduction of images, and can also take into account the excellent noise reduction performance of static images, and can perform adaptive filtering and better adapt to complex scenes.

[0043] According to a third aspect, an electronic device is provided, including a memory and a processor, which are in communication connection with each other, and the memory stores computer instructions, and the processor executes the computer instructions to perform the real-time noise reduction method based on inter-frame filtering in the first aspect or any one of the embodiments of the first aspect.

[0044] According to a fourth aspect, a computer readable storage medium is provided, which stores computer instructions for causing the computer to perform the real-time noise reduction method based on inter-frame filtering in the first aspect or any one of the embodiments of the first aspect. BRIEF DESCRIPTION OF DRAWINGS

[0045] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed in the specific embodiments or prior art description will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0046] Figure 1 is a flowchart of the real-time noise reduction method based on inter-frame filtering according to the embodiment of the present application;

[0047] Figure 2 is a schematic diagram of the Sobel operator convolution factor according to the preferred embodiment of the present application;

[0048] Figure 3 is a judgment range schematic diagram of coordinate data according to the preferred embodiment of the present application;

[0049] Figure 4 is a specific time flowchart of inter-frame filtering according to the preferred embodiment of the present application;

[0050] Figure 5 is a functional module schematic diagram of the inter-frame filtering based real-time noise reduction device according to the embodiment of the present application;

[0051] Figure 6 is a data flow schematic diagram of the inter-frame filtering based real-time noise reduction device according to the embodiment of the present application;

[0052] Figure 7 is a hardware structure schematic diagram of the electronic device provided by the embodiment of the present application. DETAILED DESCRIPTION

[0053] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below in connection with the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0054] In the embodiment, an inter-frame filtering based real-time noise reduction method is provided, which can be used in electronic devices such as computers, mobile phones, tablet computers and the like. Figure 1 is a flow chart of the inter-frame filtering based real-time noise reduction method according to the embodiment of the present application. As shown in the figure, the flow includes the following steps: Figure 1

[0055] S11, receiving original data of a target. In the embodiment, original data of a target (target object) is received, wherein the original data is obtained by using existing acquisition devices and acquisition technologies to collect information, and as long as the data acquisition can be realized, the embodiment is not limited thereto. The detailed information will be described in the subsequent steps, and the embodiment will not be described herein.

[0056] S12, detecting the target by using the original data and calculating coordinate data of the target to determine a target region. In the embodiment, the coordinate data information of the target is calculated by using the original data through a sobel algorithm, the top point coordinates of a rectangular frame are marked according to the coordinate position information, and the rectangular frame is determined as the target region. The detailed information will be described in the subsequent steps, and the embodiment will not be described herein.

[0057] S13, performing inter-frame filtering on the original data to determine filtered data. After the original data is received, the original data needs to be filtered, and the filtered data is determined through iterative data calculation, so as to determine the motion scene of the target. The detailed information will be described in the subsequent steps, and the embodiment will not be described herein.

[0058] S14, determining the motion scene of the target according to the filtered data and the target region.​

[0059] In this embodiment, after the object overall motion region of the target object is removed from the boundary region (i.e. the target region calculated above is removed), the difference between the two frames of images is used to determine the motion scene, and then a suitable filter coefficient is selected to determine the final motion scene of the target. Details will be described in subsequent steps, and this embodiment will not be described again.

[0060] S15, the filtered data is integrated in format to determine the noise reduction data. Among them, the video data in the original data after completing the inter-frame filtering is integrated in format to output an analog video; the image data is intra-frame filtered to output processed image data; and the final noise reduction data is determined.

[0061] The real-time noise reduction method based on inter-frame filtering provided in this embodiment uses the sobel algorithm in the inter-frame filtering algorithm to ensure the image processing effect by performing real-time inter-frame filtering on high-frame-rate high-resolution video. The problem of image blur in the motion scene in the traditional inter-frame filtering noise reduction of images is avoided, and the excellent noise reduction performance of static images is taken into account. The method can perform adaptive filtering and better adapt to complex scenes.

[0062] In another embodiment, a real-time noise reduction method based on inter-frame filtering is also provided. Another flow of the real-time noise reduction method based on inter-frame filtering according to the embodiment of the present application includes the following steps:

[0063] S21, receiving original data of a target; wherein the original data includes video data and image data.

[0064] In this embodiment, the above step S21 specifically further includes the following steps:

[0065] S211, receiving original data collected by an infrared detector. The original data collected by an insb640 infrared detector is received. It should be noted that the insb640 infrared detector is only used as an example in this embodiment, and as long as the data can be collected in actual application, this embodiment is not limited thereto.

[0066] S212, adjusting the format of the video data and performing intra-frame filtering on the image data to determine processed data.

[0067] S22, detecting the target by using the original data to calculate coordinate data of the target to determine a target region.

[0068] Specifically, the above step S22 specifically further includes the following steps:

[0069] S221, receiving the processed data through a line valid interface, a field valid interface, a data interface, and a clock interface.

[0070] S222, forming a data matrix according to different interface data;

[0071] Specifically, the above step S222 further comprises:

[0072] (1) when the line valid interface is in a valid state and is in a high level state, the corresponding data is buffered to the local buffer when the clock interface is in a rising edge;

[0073] (2) receiving the first line of processed data and writing it into the first buffer;

[0074] (3) when the second line of data is buffered to the local buffer, the first line of data in the first buffer is read out, the read-out first line of data is written into the second memory, and the second line of data is written into the first buffer;

[0075] (4) when the third line of data is buffered to the local buffer, the second line of data in the first buffer is read out, the read-out second line of data is written into the second memory, and the third line of data is written into the first buffer;

[0076] (5) using the read-out data of each memory, determining a data matrix, and simultaneously storing the data in each buffer to a dynamic random access memory.

[0077] In actual application, video data and image data are received through line valid, field valid, data, clock and other interfaces. When the line valid and field valid are both in a high level, the data is buffered to the local fifo (local buffer) at the rising edge of the clock. This module has two-stage fifos for pipelining, the first line of received data is written into fifo1 (first buffer), when the second line of video data reaches the module, the second line of data is written into fifo1, and the first line of data in fifo1 is read out and written into the second-stage fifo, i.e. fifo2 (second buffer); when the third line of data reaches the module, the third line of data is written into fifo1, and the second line of data in fifo1 is read out and written into fifo2. At the same time, the first line of data in fifo2 is read out. In this way, when the third line of data is received in the module, the second line and the first line of data are read out from fifo1 and fifo2 respectively, thereby forming a 3*3 matrix, and the sobel value is obtained according to the following formula according to the following convolution factor, and it is determined whether it is a boundary point.

[0078] It should be noted that the read-out data in fifo2 should be buffered to the original frame memory for sobel calculation, and then written into ddr (dynamic random access memory).

[0079] S223, calculating the data matrix using a sobel operator to determine the coordinate data of the target.

[0080] In this embodiment, the key of the sobel operator is to calculate the horizontal (Gx) and vertical (Yx) boundary values by the following two convolution factors, as shown in the following formula: Figure 2 Then, the two boundary values are taken the absolute value and summed. By judging whether the value exceeds a threshold value, it is determined whether it is a boundary point, and the coordinate data of the target is determined, and the specific judgment range is as shown in the following formula: Figure 3

[0081] Gx = (a3-a1) + (b3-b1) * 2 + (c3-c1),

[0082] Gy = (a1-c1) + (a2-c2) * 2 + (a3-c3),

[0083]

[0084] S23, inter-frame filtering is performed on the original data to determine the filtered data. For details, refer to step S13, and the embodiment will not be repeated here.

[0085] S24, the motion scene of the target is determined according to the filtered data and the target region. For details, refer to step S14, and the embodiment will not be repeated here.

[0086] S25, the format of the filtered data is integrated to determine the noise reduction data. For details, refer to step S15, and the embodiment will not be repeated here.

[0087] In another embodiment, a real-time noise reduction method based on inter-frame filtering is also provided. According to the real-time noise reduction method based on inter-frame filtering of the embodiment of the present application, the flow includes the following steps:

[0088] S31, receiving original data of a target.

[0089] For details, refer to step S21, and the embodiment will not be repeated here.

[0090] S32, detecting the target by using the original data to calculate the coordinate data of the target to determine the target region.

[0091] For details, refer to step S22, and the embodiment will not be repeated here.

[0092] S33, inter-frame filtering is performed on the original data to determine the filtered data.

[0093] In this embodiment, the above step S33 further includes the following steps:

[0094] S331, storing the first frame data of the target into a dynamic random access memory;

[0095] ​S332, when the second frame data of the target is stored in the dynamic random access memory, the data of the current frame after inter-frame filtering is calculated and stored in the dynamic random access memory;

[0096] S333, the next frame of the current frame is determined as the current frame, when the current frame data of the target is stored in the dynamic random access memory, the data of the current frame after inter-frame filtering is calculated and stored in the dynamic random access memory, until all the frame data of the target is completed.

[0097] In actual application, the first frame image received by the FPGA after power-on is not subjected to inter-frame filtering, and the first frame data of the original frame is stored in the ddr. After that, the first frame data of the original frame is read out from the ddr by the current frame rdfifo (read memory) and written into the video output module.

[0098] When the second frame data of the original frame is stored in the ddr by the motion target detection module and the filtering coefficient is updated, the mth row data of the second frame of the original frame is read out from the ddr by the current frame rdfifo at this time. At the same time, the mth row of the first frame is read out from the filtering frame rdfifo. The mth row data is calculated by the following formula: the filtered data of the mth row is iteratively written into the ddr by the filtering frame wrfifo (write memory). At the same time, the data is also sent to the next module for video output, i.e. the video output module.

[0099] When the nth frame data (n>2) of the original frame is stored in the ddr by the motion target detection module and the filtering coefficient is updated, the mth row data of the nth frame of the original frame is read out from the ddr by the current frame rdfifo at this time. At the same time, the mth row of the n-1th frame is read out from the filtering frame rdfifo. The mth row data is calculated by the above formula in a loop and stored in the ddr. The specific time flow is shown in Figure 4 .

[0100] The data of the current frame after inter-frame filtering is calculated by the following formula,

[0101] y(n)=x(n)*(1-k)+y(n-1)*k

[0102] wherein y(n) is the data of the current frame after inter-frame filtering, x(n) is the data of the current frame stored in the dynamic random access memory, y(n-1) is the data of the previous frame of the current frame after inter-frame filtering, and k is a preset filtering coefficient.

[0103] S34, determining the motion scene of the target according to the filtered data and the target region.

[0104] In the embodiment, the step S34 further includes the following steps:

[0105] S341, the processed data is removed from the target region, and the calculation data is determined;

[0106] S342, the current frame data of the calculation data is determined and the inter-frame filtered data of the last frame;

[0107] S343, the current frame data is subtracted from the inter-frame filtered data of the last frame, and the motion scene is determined by using the difference value.

[0108] In the embodiment, the difference value between the current frame image and the inter-frame filtered image of the last frame is calculated by removing the target region to judge the motion scene, and then the appropriate filter coefficient k is selected. When the difference value is larger, the filter parameter is smaller. The specific method of removing the target region: the target detection module at the end of each frame outputs the boundary coordinates of all targets in the scene to the module. After receiving the coordinates, the module is latched. The removal area is the sum of the rectangular area of the current frame and the rectangular area of the last frame.

[0109] S35, the filtered data is integrated in format, and the noise reduction data is determined.

[0110] For details, see step S25, which will not be described here.

[0111] The real-time noise reduction method based on inter-frame filtering provided by the embodiment of the application can ensure the image processing effect by applying the sobel algorithm to the inter-frame filtering algorithm, avoid the problem of image blur in the motion scene in the traditional inter-frame filtering noise reduction of image, and take into account the excellent noise reduction performance of the static image, so that adaptive filtering can be performed and better adapt to complex scenes.

[0112] The embodiment provides a real-time noise reduction device based on inter-frame filtering. As used in the following, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiment is preferably implemented in software, hardware or a combination of software and hardware is also possible and is contemplated.

[0113] The application discloses a real-time noise reduction device based on inter-frame filtering, as shown in the following figure, comprising: Figure 5

[0114] The data receiving module 01 is used for receiving the original data of the target.

[0115] The motion target detection module 02 is used for detecting the target by using the original data, calculating the coordinate data of the target, and determining the target region.

[0116] The inter-frame filtering calculation module 03 is used for inter-frame filtering of the original data to determine the filtered data.

[0117] ​The scene determining module 04 is used for determining the motion scene of the target according to the filtered data and the target region.

[0118] The output module 05 is used for performing format integration on the filtered data to determine the noise reduction data.

[0119] The data flow diagram of the real-time noise reduction device based on inter-frame filtering is shown in detail as follows: Figure 6 As shown in the figure, the original data is collected by an infrared detector, the data receiving module (video data receiving and image preprocessing module) receives the original data and sends the processed data to the moving target detection module, the original data is temporarily stored in the original frame storage wrfifo (original frame memory), and the data is stored in the ddr through the ddr controller. The filter frame reading rdfifo reads out the data in the ddr together with the target rectangular coordinates and the preprocessed data to calculate the filter coefficient by using the filter coefficient calculation module.

[0120] The current frame reading rdfifo and the filter frame reading rdfifo read out the data in the ddr, the filtered data is determined by the inter-frame filtering module, and the filtered data is stored in the filter frame storage wrfifo, and the data is output to the video output module (i.e. the output module).

[0121] The real-time noise reduction device based on inter-frame filtering provided by the embodiment of the application, by performing real-time inter-frame filtering on high-frame-frequency high-resolution video, applying the sobel algorithm to the inter-frame filtering algorithm, ensures the image processing effect; avoids the problem of image blur in the motion scene of the traditional inter-frame filtering noise reduction of the image, and takes into account the excellent noise reduction performance of the static image, can perform adaptive filtering, and better adapts to complex scenes.

[0122] The embodiment of the application further provides an electronic device, please refer to Figure 7 , Figure 7 is a structural schematic diagram of an electronic device provided by the optional embodiment of the application, as shown in Figure 7As shown, the electronic device can include at least one processor 601, such as a CPU (Central Processing Unit), at least one communication interface 603, a memory 604, and at least one communication bus 602. The communication bus 602 is used to realize the connection and communication between the components. The communication interface 603 can include a display, a keyboard, and can also include a standard wired interface and a wireless interface. The memory 604 can be a high-speed RAM memory, and can also be a non-volatile memory, such as at least one disk memory. The memory 604 can also be at least one storage device located away from the aforementioned processor 601. The processor 601 can be combined with Figure 7 The described device, the memory 604 stores an application program, and the processor 601 invokes the program code stored in the memory 604 to execute any of the above method steps.

[0123] The communication bus 602 can be a PCI bus or an EISA bus, etc. The communication bus 602 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, ​ Only one thick line is used in the figure, but it does not mean that there is only one bus or only one type of bus.

[0124] The memory 604 can include volatile memory, such as random access memory (RAM); the memory can also include non-volatile memory, such as flash memory, a hard disk (HDD) or a solid state disk (SSD); the memory 604 can also include a combination of the above types of memory.

[0125] The processor 601 can be a central processing unit (CPU), a network processor (NP), or a combination of CPU and NP.

[0126] The processor 601 can further include a hardware chip. The hardware chip can be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0127] Optionally, the memory 604 is further configured to store program instructions. The processor 601 can invoke the program instructions to implement the real-time noise reduction method based on inter-frame filtering as shown in the embodiments of the present application.

[0128] The embodiments of the present application further provide a non-transitory computer storage medium, and the computer storage medium stores computer executable instructions. The computer executable instructions can execute the real-time noise reduction method based on inter-frame filtering in any method embodiment described above. The storage medium can be a disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD), a solid-state drive (SSD), or the like. The storage medium can also include a combination of the above-mentioned storage devices.

[0129] Although the embodiments of the present application are described in conjunction with the accompanying drawings, various modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes are intended to fall within the scope of the appended claims.

Claims

1. A real-time noise reduction method based on inter-frame filtering, characterized in that, include: Receive the target's raw data; The target is detected using the raw data, and the coordinate data of the target is calculated to determine the target area; Perform inter-frame filtering on the original data to determine the filtered data; The motion scene of the target is determined based on the filtered data and the target region; Select filtering coefficients based on the motion scene of the target, and perform inter-frame filtering on the original data again based on the filtering coefficients; The data after the second inter-frame filtering is format-integrated to determine the noise-reduced data; The raw data received from the target includes: the raw data includes video data and image data. Receive the raw data collected by the infrared detector; The video data is formatted and the image data is intra-frame filtered to determine the processed data. The step of detecting the target using the original data and calculating the target's coordinate data includes: The processed data is received through the row active interface, field active interface, data interface, and clock interface. A data matrix is ​​formed based on data from different interfaces; The Sobel operator is used to calculate the coordinate data of the target by calculating the data matrix.

2. The method according to claim 1, characterized in that, The process of forming a data matrix based on data from different interfaces includes: When the row active interface is active and high, the corresponding data is buffered into the local buffer when the clock interface is rising. Receive the first line of processed data and write it into the first buffer; When the second row of data is cached in the local cache, the first row of data in the first cache is read out and written into the second memory, while the second row of data is written into the first cache. When the third row of data is cached in the local cache, the second row of data in the first cache is read out and written into the second memory, while the third row of data is written into the first cache. The data matrix is ​​determined using the data read from each memory, and the data in each cache is simultaneously stored in the dynamic random access memory.

3. The method according to claim 2, characterized in that, The step of performing inter-frame filtering on the target and determining the filtered data includes: The first frame of data of the target is stored in the dynamic random access memory; When the second frame data of the target is stored in the dynamic random access memory, the data of the current frame after inter-frame filtering is calculated and stored in the dynamic random access memory; The next frame of the current frame is determined as the current frame. When the current frame data of the target is stored in the dynamic random access memory, the data of the current frame after inter-frame filtering is calculated and stored in the dynamic random access memory, until all frame data of the target has been stored.

4. The method according to claim 3, characterized in that, The data of the current frame after inter-frame filtering is calculated using the following formula. y(n) = x(n)*(1-k) + y(n-1)*k Where y(n) is the data of the current frame after inter-frame filtering, x(n) is the data of the current frame stored in dynamic random access memory, y(n-1) is the data of the previous frame after inter-frame filtering, and k is the preset filtering coefficient.

5. The method according to claim 1, characterized in that, Determining the motion scene of the target based on the filtered data and the target region includes: The processed data is removed from the target region to determine the calculation data; Determine the current frame data and the inter-frame filtered data of the previous frame in the calculation data; The difference between the current frame data and the filtered data of the previous frame is used to determine the motion scene.

6. A real-time noise reduction device based on inter-frame filtering, characterized in that, include: The data receiving module is used to receive the target's raw data; The moving target detection module is used to detect the target using the raw data, calculate the coordinate data of the target, and determine the target area; The inter-frame filtering calculation module is used to perform inter-frame filtering on the original data and determine the filtered data. A scene determination module is used to determine the motion scene of the target based on the filtered data and the target region. The inter-frame filtering calculation module is further configured to select filtering coefficients based on the motion scene of the target, and perform inter-frame filtering on the original data again based on the filtering coefficients. The output module is used to integrate the format of the data after the second inter-frame filtering to determine the noise-reduced data; The data receiving module is specifically used to receive the raw data collected by the infrared detector, the raw data including video data and image data; to adjust the format of the video data, to perform intra-frame filtering on the image data, and to determine the processed data; The moving target detection module is specifically used to receive processed data through a row valid interface, a field valid interface, a data interface, and a clock interface; form a data matrix based on the data from different interfaces; and use the Sobel operator to calculate the coordinate data of the target.

7. An electronic device, characterized in that, include: The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes the computer instructions to perform the real-time noise reduction method based on inter-frame filtering as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the real-time noise reduction method based on inter-frame filtering as described in any one of claims 1-5.

Images