CN114529483A - Data processing method, device, terminal and readable storage medium - Google Patents

Abstract

The application provides a data processing method, which comprises the steps of obtaining the preset filtering radiuses of a plurality of filtering layers; determining a filtering area of data to be processed according to the filtering radiuses of the plurality of filtering layers; and acquiring the image data of the filtering area. The data processing method, the data processing device, the terminal and the nonvolatile computer readable storage medium can acquire all required filtering areas when data to be processed is filtered by a plurality of filtering layers, and compared with the data which are acquired in the filtering areas of each filtering layer respectively when each filtering layer is filtered, the data which are acquired in the filtering areas of different filtering layers have repeated parts, the data transmission quantity and the data transmission occupy more time, so that the power consumption of hardware is increased, the data processing efficiency is lower, the time occupied by the data transmission quantity and the data transmission is obviously reduced, the power consumption of the hardware is reduced, and the data processing efficiency is improved.

Description

Data processing method, device, terminal and readable storage medium

technical field

The present application relates to the field of image technology, and in particular, to a data processing method, a data processing device, a terminal, and a non-volatile computer-readable storage medium.

Background technique

At present, when processing data, data needs to be transmitted back and forth between memory and storage, and the data that different data processing processes need to obtain are different. Therefore, each data processing process needs to perform data transfer between memory and storage. Transmission, resulting in a large amount of data transmission and time occupied by data transmission, resulting in increased hardware power consumption and low data processing efficiency.

SUMMARY OF THE INVENTION

Embodiments of the present application provide a data processing method, a data processing apparatus, a terminal, and a non-volatile computer-readable storage medium.

The data processing method according to the embodiment of the present application includes acquiring the filter radii of a plurality of preset filter layers; determining a filter area of the data to be processed according to the filter radii of the multiple filter layers; and acquiring an image of the filter area data.

The data processing apparatus of the embodiment of the present application includes a first acquisition module, a first determination module, and a second acquisition module. The first obtaining module is used to obtain the preset filter radii of multiple filter layers; the first determination module is used to determine the filter area of the data to be processed according to the filter radii of the multiple filter layers; The second acquiring module is used for acquiring image data of the filtering area.

The terminal according to the embodiment of the present application includes a processor, the processor is configured to obtain the filter radius of a plurality of preset filter layers; determine the filter area of the data to be processed according to the filter radius of the plurality of filter layers; and obtain image data of the filtered region.

A non-volatile computer-readable storage medium containing a computer program of the present application, when the computer program is executed by one or more processors, causes the processors to perform a data processing method. The data processing method includes acquiring filter radii of a plurality of preset filter layers; determining a filter region of data to be processed according to the filter radii of the filter layers; and acquiring image data of the filter region.

In the data processing method, data processing device, terminal, and non-volatile computer-readable storage medium of the embodiments of the present application, by acquiring the filtering radii of multiple filtering layers of the data to be processed, according to the filtering radii of the multiple filtering layers, determine And when the data to be processed is filtered for multiple filter layers, all the required filter areas are compared to when each filter layer is filtered, and the data of the filter area of each filter layer is obtained separately. Different filter layers There are duplicate parts in the filtering area of the data transmission, and the amount of data transmission and the time occupied by data transmission are large, resulting in increased power consumption of the hardware and low data processing efficiency. The power consumption is reduced and the data processing efficiency is improved.

Additional aspects and advantages of the present application will be set forth, in part, from the following description, and in part will become apparent from the following description, or may be learned by practice of the present application.

Description of drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only These are some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without any creative effort.

1 is a schematic flowchart of a data processing method according to some embodiments of the present application;

2 is a schematic block diagram of a data processing apparatus according to some embodiments of the present application;

3 is a schematic plan view of a terminal of some embodiments of the present application;

4 to 6 are schematic schematic diagrams of some embodiments of the present application;

7 is a schematic flowchart of a data processing method according to some embodiments of the present application;

Fig. 8 is the principle schematic diagram of some embodiments of the present application;

9 is a schematic flowchart of a data processing method according to some embodiments of the present application;

10 is a schematic diagram of some embodiments of the present application;

11 and 12 are schematic flowcharts of data processing methods according to some embodiments of the present application;

FIG. 13 is a schematic diagram of connection between a processor and a computer-readable storage medium according to some embodiments of the present application.

Detailed ways

The embodiments of the present application will be further described below with reference to the accompanying drawings. The same or similar reference numbers refer to the same or similar elements or elements having the same or similar functions throughout the drawings. In addition, the embodiments of the present application described below in conjunction with the accompanying drawings are exemplary, only used to explain the embodiments of the present application, and should not be construed as limitations on the present application.

Please refer to FIG. 1 to FIG. 3, the data processing method of the embodiment of the present application includes the following steps:

011: Obtain the filter radius of the preset multiple filter layers;

012: Determine the filtering region of the data to be processed according to the filtering radii of the plurality of filtering layers; and

013: Acquire image data of the filtered area.

The data processing apparatus 10 in the embodiment of the present application includes a first acquisition module 11 , a first determination module 12 and a second acquisition module 13 . The first obtaining module 11 , the first determining module 12 and the second obtaining module 13 are used to execute step 011 , step 012 and step 013 respectively. That is, the first acquisition module 11 is used to acquire the preset filter radii of multiple filter layers. The first determination module 12 is used to determine the filter area of the data to be processed according to the filter radii of the multiple filter layers; the second acquisition module 13 uses to obtain image data of the filtered area.

The terminal 100 according to the embodiment of the present application includes the processor 30 . The processor 30 is configured to acquire the preset filter radii of the multiple filter layers; to determine the filter area of the data to be processed according to the filter radii of the multiple filter layers; and to acquire the image data of the filter area. That is, steps 011 , 012 and 013 may be implemented by the processor 30 .

Specifically, the terminal 100 further includes a housing 40 . The terminal 100 may be a mobile phone, a tablet computer, a display device, a notebook computer, an ATM, a gate, a smart watch, a head-mounted display device, a game console, and the like. As shown in FIG. 3 , the embodiment of the present application is described by taking the terminal 100 as a mobile phone as an example. It can be understood that the specific form of the terminal 100 is not limited to the mobile phone. The housing 40 can also be used to install functional modules such as a display device, an imaging device, a power supply device, and a communication device of the terminal 100 , so that the housing 40 provides protection against dust, drop, and water for the functional modules.

The data to be processed may be an image captured by the camera 20 of the terminal 100 or an image downloaded from the Internet, which is not limited herein. The data to be processed may also be a part of the image captured by the camera 20 ; or, the data to be processed may be multi-frame gesture data, acceleration data, etc. acquired by the terminal 100 . The embodiments of the present application are described by taking the data to be processed as an image to be processed as an example.

When filtering the image to be processed, it is generally performed based on a preset filtering algorithm. The preset filtering algorithm determines the preset size of the image to be processed, the preset number of filter layers, and the preset filter radius of each filter layer.

During filtering, if the size of the captured image is larger than the preset size, the captured image can be divided into a plurality of images to be processed by the preset size, so that the processor 30 can directly obtain the position information of each image to be processed. The vertex coordinates of the processed image (take the image to be processed as a rectangle as an example). At the same time, the processor 30 can also directly obtain the filter radius of each filter layer.

Then, the processor 30 can determine the filtering area of each filtering layer according to the position information of the image to be processed and the filtering radius of each filtering layer. For example, the filter radius is the distance from the edge of the image to be processed, and the location information includes the vertex coordinates of the image to be processed. According to the vertex coordinates of the image to be processed, the width and height of the image to be processed can be quickly determined. The processor 30 can determine the vertex coordinates of the filtering area according to the vertex coordinates and the filtering radius of the image to be processed

In an example, referring to FIG. 4 , the image coordinate system is established with the upper left corner of the image A1 to be processed as the origin, the width of the image A1 to be processed is 8 (the number of pixels along the W direction), and the height is 8 (the number of pixels along the H direction) number of pixels), the vertex coordinates of the image A1 to be processed are (0,0), (8,0), (0,8) and (8,8) respectively. If the filter radius is 2 pixels, the Two rows of pixels are added to the upper and lower edges of A1, and two columns of pixels are added to the left and right edges, respectively. The vertex coordinates of the image S1 to be processed are (-2,-2), (10,2), (- 2,10) and (10,10). In this way, according to the position information of the image A1 to be processed and the filtering radius of each filtering layer, the filtering area S1 of each filtering layer can be quickly determined.

In other embodiments, the filter radius includes a first filter radius (eg, corresponding to the upper side of the image), a second filter radius (eg, corresponding to the right side of the image), a third filter radius (eg, corresponding to the lower side of the image), and a fourth filter radius ( As shown in the left side of the corresponding image), the first filter radius, the second filter radius, the third filter radius and the fourth filter radius are all the same (the example shown in Figure 4); or the first filter radius and the third filter radius are the same, The second filtering radius and the fourth filtering radius are the same; or, the first filtering radius, the second filtering radius, the third filtering radius and the fourth filtering radius are different from each other.

In another example, referring to FIG. 5 , the image coordinate system is established with the upper left corner of the image A1 to be processed as the origin, and the coordinates of the vertices of the image A1 to be processed are (0,0), (8,0), (0 ,8) and (8,8), if the first filter radius and the third filter radius are the same and both are 2 pixels, the second filter radius and the fourth filter radius are the same and both are 1 pixel, then the image to be processed Two rows of pixels are added to the upper and lower edges of A1, and one column of pixels is added to the left and right edges, respectively. The vertex coordinates of the image S1 to be processed are (-1,-2), (9,-2), (- 1,10) and (9,10). In this way, according to the position information of the image A1 to be processed and the filtering radius of each filtering layer, the filtering area S1 of each filtering layer can be quickly determined.

In yet another example, please refer to FIG. 6 , the image coordinate system is established with the upper left corner of the image A1 to be processed as the origin, then the coordinates of the vertices of the image A1 to be processed are (0,0), (8,0), (0 ,8) and (8,8), if the first filter radius, the second filter radius, the third filter radius and the fourth filter radius are 4 pixels, 3 pixels, 2 pixels, and 1 pixel respectively, then in The upper and lower edges of the image A1 to be processed are respectively increased by 4 rows of pixels and 2 rows of pixels, and the left and right edges are respectively increased by 1 column of pixels and 3 columns of pixels, then the vertex coordinates of the to-be-processed image A1 are (-1,- 4), (11,-4), (-1,10) and (11,10). In this way, according to the position information of the image A1 to be processed and the filtering radius of each filtering layer, the filtering area S1 of each filtering layer can be quickly determined.

It can be understood that during the filtering process, two adjacent filter layers are generally dependent on each other. For example, the output image of the upper filter layer will be used as the input image of the next filter layer, and the input image of the next filter layer will be used as the input image of the next filter layer. Determined for the output image of the next filter layer and the filter radius of the next filter layer.

Therefore, the filtering area of the last filtering layer can be determined according to the output image of the last filtering layer and the filtering radius of the last filtering layer, so as to determine the output image of the penultimate filtering layer (that is, the filtering layer of the last filtering layer) The image data of the filter region of the layer, and also the input image of the last filter layer). Then, according to the output image of the second-to-last filter layer and the filter radius of the second-to-last filter layer, the filtering area of the second-to-last filter layer is determined, thereby determining the output image of the third-to-last filter layer (that is, the second-to-last filter layer). The image data of the filter region of the layer filter layer is also the input image of the second-to-last filter layer). In this way, by sequentially determining from the last layer to the first layer, the filtering area of the first filtering layer in all the filtering layers can be determined, so as to determine the filtering area of the image to be processed (that is, the filtering area of the first filtering layer The image data of the filtering area is also the input image of the first filtering layer).

It can be understood that the number of layers of the filtering layer may be 1 layer, 2 layers, 3 layers, 4 layers, 5 layers, etc. The above-described embodiment has been described with three or more filter layers. If there are two filter layers, it is only necessary to determine the filter area of the second-to-last filter layer in the above embodiment, and the filter area is the filter area of the image to be processed. If the filter layer is one layer, it is only necessary to determine the filter area of the first layer filter layer according to the output image of the first layer filter layer and the filter radius of the first layer filter layer, and the filter area is the filter area of the image to be processed. .

Of course, it is also possible to obtain only the filter radii of any adjacent 2 layers, adjacent 3 layers, 4 layers, etc. in the multi-layer filter layers, so as to obtain any adjacent 2 layers, adjacent 3 layers, 4 layers in sequence, etc. The filtering area corresponding to the layer, etc., after the data in the filtering area is read once, it can be used by any adjacent layer 2, layer 3, layer 4, etc., which are adjacent to each other, thereby reducing the amount of data read and write during filtering.

For example, taking the filter layer as an example of 4 layers, after the first layer is filtered, the filter area corresponding to the input image of the second layer can be calculated according to the filter radius of the second layer to the fourth layer. , the implementation of calculating the filtering region of the first layer will not be repeated here. In this way, it is only necessary to read the image data once for the successively adjacent layers 2 to 4, thereby reducing the amount of data reading and writing during filtering.

In addition, the image size of the output of each filtering layer may be fixed, for example, the size of the image to be processed is the same. For example, if the captured image is 100*100 and the size of the image to be processed is 50*50, the captured image is divided into 4 images to be processed. After filtering the image to be processed, the size of the final output filtered image is also 50*50.

Finally, the processor 30 directly obtains the image data in the filtering region of the image to be processed, so as to obtain all the image data required for subsequent filtering processing of the image to be processed. Among them, the image data that exists in the filtering area can be obtained directly (such as the part where the filtering area overlaps with the captured image), while for the part where there is no image data in the filtering area (the part where the filtering area does not overlap the captured image), you need to use the filtering area according to the filter area. Fill in the existing image data in the filter area, so as to obtain the image data of each pixel in the filter area.

The data processing method, the data processing device 10 and the terminal 100 according to the embodiments of the present application determine and obtain the data to be processed for multiple During the filtering of the filter layer, all the required filter areas are compared with the data of the filter area of each filter layer obtained separately when the filter layer is filtered. The filter areas of different filter layers have overlapping parts. The amount of transmission and the time occupied by data transmission are both, resulting in increased hardware power consumption and low data processing efficiency. In terms of data transmission amount and time occupied by data transmission, the power consumption of hardware is reduced and the data processing efficiency is improved.

Please refer to FIG. 2, FIG. 3 and FIG. 7. In some embodiments, the filter layer includes M layers, where M is a positive integer, and the filter area of the Mth layer is based on the filter radius of the Mth layer and the image area where the image to be processed is located. OK, step 012 further includes:

0121: Determine the filtering area of the N-1 layer according to the filtering radius of the N-1 layer and the filtering area of the N-th layer, where N is a positive integer less than or equal to M;

0122: After determining that the filter area of the N-1th layer is obtained, reduce the value of N by 1, and perform step 0121 again: Determine the N-1th layer according to the filter radius of the N-1th layer and the filter area of the Nth layer The filtering area of the layer is determined until the filtering area of the first layer is determined as the filtering area of the image to be processed.

In some embodiments, the first determination module 12 is further configured to perform steps 0121 and 0122. The first determining module 12 is further configured to determine the filtering region of the N-1th layer according to the filtering radius of the N-1th layer and the filtering region of the Nth layer, where N is a positive integer less than or equal to M; After the filtering area of the -1 layer, the value of N is reduced by 1, and the steps of determining the filtering area of the N-1 layer according to the filtering radius of the N-1 layer and the filtering area of the N layer are performed again until the first layer is determined. The filtering area of layer 1 is used as the filtering area of the image to be processed.

In some embodiments, the processor 30 is further configured to determine the filtering region of the N-1 th layer according to the filtering radius of the N-1 th layer and the filtering region of the N-th layer, where N is a positive integer less than or equal to M; After it is determined that the filtering area of the N-1th layer is obtained, the value of N is reduced by 1, and the filter area of the N-1th layer is determined again according to the filtering radius of the N-1th layer and the filtering area of the Nth layer. Steps until the filtering area of the first layer is determined as the filtering area of the image to be processed. That is, steps 0121 and 0122 may be implemented by the processor 30 .

Specifically, the filtering layer includes M layers, for example, M is 1, 2, 3, 4, 5, and so on. The M-th layer is the last layer for filtering processing. After the image data of the filtering area of the image to be processed is sequentially processed by the M filtering layers, filtering can be completed to generate a filtered image.

When determining the filtering area of the image to be processed, the processor 30 may first determine the filtering area of the M-th layer according to the filtering radius of the M-th layer and the image area where the image to be processed is located.

The size of the filtered image may be the same as that of the image to be processed. After filtering the filtering region of the Mth layer, only the image data in the image region where the image to be processed is located is retained to generate the filtering region. Therefore, the filter region of the Mth layer can be determined according to the image region where the image to be processed is located and the filter radius of the Mth layer. For example, if the filter radius is 1 pixel, then add one row of pixels to the upper and lower edges of the image area where the image to be processed is located, and add a column of pixels to the left and right edges to serve as the filtering area of the Mth layer. It can be understood that after each layer of filtering layer is filtered, the image data of the image region where the image to be processed is located may be changed.

Then, the processor 30 determines the filter area of the N-1th layer according to the filter radius of the N-1th layer and the filter area of the Nth layer; please refer to FIG. 1st floor L1 to 4th floor L4. Initially, the value of N is equal to the value of M, that is, the processor 30 determines the filtering area of the third layer according to the filtering radius of the third layer and the filtering area of the fourth layer; after determining the filtering area of the third layer, the value of N is The value is reduced by 1 to become 3, and the processor 30 determines the filter area of the second layer according to the filter radius of the second layer and the filter area of the third layer; after determining the filter area of the second layer, the value of N is reduced by 1 again to change. If it is 2, the processor 30 determines the filter area of the first layer according to the filter radius of the first layer and the filter area of the second layer, and after determining the filter area of the first layer, it can be used as the filter area of the image to be processed.

Please refer to FIG. 2, FIG. 3 and FIG. 9 again, in some embodiments, the filter layer includes a filter operator, and the filter area of the filter operator of the Mth layer is based on the filter radius of the filter operator of the Mth layer and the to-be-processed The image area where the image is located is determined, and step 0121 further includes:

01211: Determine the filter region of each filter operator of the N-1th layer according to the filter radius of the filter operator of the N-1th layer and the filter region of the N-th layer of filter operators.

In some embodiments, the first determination module 12 is further configured to perform step 01211. That is, the first determining module 12 is further configured to determine the filtering region of each filtering operator of the N-1 th layer according to the filtering radius of the filtering operator of the N-1 th layer and the filtering region of the filtering operator of the N-th layer.

In some embodiments, the processor 30 is further configured to determine each filter operator of the N-1th layer according to the filter radius of the filter operator of the N-1th layer and the filter region of the N-th layer of filter operators filter area. That is, step 01211 may be implemented by the processor 30 .

Specifically, each filter layer may include one or more filter operators, and each filter operator may independently filter the output image input by the previous filter layer to generate an intermediate filtered image. Please refer to Fig. 8, the filter layer is 4 layers, the first layer L1 includes 1 filter operator (ie, OP1), the second layer L2 includes 3 filter operators (ie OP2, OP3 and OP4), the third layer L3 It includes 2 filter operators (ie, OP5 and OP6), and the fourth layer L4 includes 1 filter operator (ie, OP7). 109*109, 107*107

After the filtering operator OP1 of the first layer performs filtering processing on the image data of the filtering area of the image to be processed (ie, the image P0 in FIG. 8 ), the first intermediate filtering image P1 is output; the first intermediate filtering image P1 is respectively input to the second Layers OP2, OP3 and OP4, OP2, OP3 and OP4 can process the first intermediate filter image P1 respectively, so as to output three second intermediate filter images P2; Layer OP5 (processing the second intermediate filtering image P2 of OP2 and OP3) and OP6 (processing the second intermediate filtering image P2 of OP4), OP5 and OP6 can process the second intermediate filtering image P2 respectively, and can output 2 images The third intermediate filtering image P3; the two third intermediate filtering images P3 are both input to the OP7 of the fourth layer, and the OP7 processes the two third intermediate filtering images P3, and the fourth intermediate filtering image P4 can be output. Four intermediate filtered images P4 are output as final filtered images.

As shown in Figure 8, there are associations between different filter operators, OP1 and OP2, OP1 and OP3, OP1 and OP4, OP2 and OP5, OP3 and OP5, OP4 and OP6, OP5 and OP7, OP6 and OP7 are associated, The filter operator refers to the output of the filter operator of the previous layer as the input of the filter operator of the next layer.

When calculating the filter area of each filter operator of the N-1th layer, first determine the filter area of the filter operator of the Mth layer according to the filter radius of the filter operator of the Mth layer and the image area where the image to be processed is located ( That is, the filter area of OP7). Then, after determining the filter operator of the Nth layer associated with each filter operator of the N-1th layer, according to the filter radius of the filter operator of the N-1th layer and the filtering of the filter operator of the Nth layer area, the filtering area of each filtering operator of the N-1th layer can be determined.

For example, if the value of N is equal to the value of M, which is 4, the filtering area of OP5 is determined according to the filtering area of OP7 and the filtering radius of OP5; the filtering area of OP6 is determined according to the filtering area of OP7 and the filtering radius of OP6. In this way, the filtering area of each filtering operator in each filtering layer is calculated layer by layer until the filtering area of the filtering operator OP1 of the first layer is calculated, which is the filtering area of the image to be processed.

It can be understood that one filter operator of the N-1th layer may be associated with multiple filter operators of the Nth layer. The first filter operator of the N-1th layer is associated with the second filter operator of the Nth layer. The output data of the filter operator of the N-1th layer is the input data of the filter operator of the Nth layer. The first filter The operator is any one of the filter operators of the N-1th layer, and the second filter operator is any one of the filter operators of the Nth layer.

For example, as shown in FIG. 8, the first filter operator (ie, OP1) of the N-1th layer (ie, the first layer as an example) and a plurality of second filter operators of the Nth layer (ie, the second layer) (ie, OP2, OP3, and OP4) are associated. The first intermediate filtered image P1 output by OP1 is the input image of OP2, OP3 and OP4. In order to ensure that the first intermediate filtered image P1 contains all the image data required by OP2, OP3 and OP4, after determining OP2, OP3 and OP4 After the filtering regions are determined, the intermediate filtering regions may be determined according to the filtering regions of each second filtering operator of the Nth layer (ie, the filtering regions of OP2, OP3, and OP4).

As shown in FIG. 10 , the union of the filtering regions of all the filtering operators of the current filtering layer (eg, the Nth layer) can be taken as the intermediate filtering region S2, such as the filtering regions of OP2, OP3 and OP4 (ie, OP2, OP3 and OP4). The union of the second intermediate filtering images P2) respectively output by OP4 is used as the intermediate filtering area S2, and since the filtering area is generally rectangular, the rectangular area corresponding to the union of the filtering areas of OP2, OP3 and OP4 can be used as the intermediate filtering area. Area S2.

Finally, the filter region of the first filter operator is determined according to the intermediate filter region and the filter radius of the first filter operator. Therefore, the image data outputted after the first filtering operator performs filtering processing includes all the image data required by OP2, OP3 and OP4. In this way, the image data of all filter operators in the current filter layer can be read at one time.

Alternatively, the union of the filtering regions of multiple filtering operators (at least two) of the current filtering layer may be used as the intermediate filtering region S2, for example, the union of the filtering regions of OP2 and OP3, and the filtering regions of OP2 and OP4 may be used. The union, or the union of the filter regions of OP3 and OP4, serves as the intermediate filter region S2.

Then, the filter region of the first filter operator is determined according to the intermediate filter region and the filter radius of the first filter operator. Therefore, the image data output by the first filter operator after filtering processing includes all the image data required by the multiple filter operators, so that the image data of the multiple filter operators can be obtained through one reading.

For the case where there are multiple first filter operators (ie, OP5 and OP6) of the N-1th layer (take the third layer as an example) and are associated with OP7 of the Nth layer (ie, the fourth layer), then The two third intermediate filtered images P3 generated by multiple OP5 and OP6 can be input into OP7, and OP7 can filter the two third intermediate filtered images P3 to obtain two fourth intermediate filtered images P4, and The two fourth intermediate filtered images P4 are fused, for example, by weighted fusion of the pixels in the same positions in the two fourth intermediate filtered images P4, and finally a fused fourth intermediate filtered image P4 is output.

Please refer to Fig. 2, Fig. 3 and Fig. 11, in some embodiments, the data processing method further comprises the following steps:

014: Determine the coordinate offset of each second filter operator according to the vertex coordinates of the filter region of each second filter operator and the vertex coordinates of the intermediate filter region;

015: When performing the filtering process on the Nth layer, according to the coordinate offset of each second filter operator and the vertex coordinates of the output image of the N-1th layer, determine the filter area of each second filter operator. vertex coordinates;

016: From the output image, acquire and process image data corresponding to the filtering region of each second filtering operator.

In some embodiments, the data processing apparatus 10 further includes a second determination module 14 , a third determination module 15 and a first processing module 16 . The second determination module 14 , the third determination module 15 and the first processing module 16 are used to execute step 014 , step 015 and step 016 respectively. That is, the second determination module 14 is configured to determine the coordinate offset of each second filter operator according to the vertex coordinates of the filter region of each second filter operator and the vertex coordinates of the intermediate filter region; the third determination module 15 uses When performing the filtering process on the Nth layer, according to the coordinate offset of each second filter operator and the vertex coordinates of the output image of the N-1th layer, determine the vertex of the filter area of each second filter operator Coordinates; the first processing module 16 is used for acquiring and processing image data corresponding to the filtering region of each second filtering operator from the output image.

In some embodiments, the processor 30 is further configured to determine the coordinate offset of each second filter operator according to the vertex coordinates of the filter region of each second filter operator and the vertex coordinates of the intermediate filter region; When the Nth layer performs filtering, according to the coordinate offset of each second filter operator and the vertex coordinates of the output image of the N-1th layer, determine the vertex coordinates of the filtering area of each second filter operator; In the output image, image data corresponding to the filter region of each second filter operator is acquired and processed. That is, steps 014 , 015 and 016 may be implemented by the processor 30 .

Specifically, since one first filter operator may be associated with multiple second filter operators, and the filter region of the first filter operator is based on the filter regions of multiple second filter operators (for example, the filter region of multiple second filter operators The union of the filtering regions, namely the intermediate filtering region) and the filtering radius of the first filtering operator are determined.

In order to ensure that the second filter operator can accurately obtain the required image data from the first intermediate filter image P1 (corresponding to the intermediate filter region) output by the first filter operator, it is necessary to determine that the filter region of the second filter operator is in the middle The location of the filter area.

Therefore, the processor 30 can determine the coordinate offset of each second filter operator according to the vertex coordinates of the filter region of each second filter operator and the vertex coordinates of the intermediate filter region; When the first filter operator of the first layer outputs the first intermediate filter image P1 (ie, the intermediate filter region), the second filter operator can obtain image data required for filtering from the first intermediate filter image P1.

For example, when filtering is performed at the Nth layer (eg, the second layer), the second filter operator can calculate the value of the filter region of the second filter operator according to its corresponding coordinate offset and the vertex coordinates of the first intermediate filter image P1. vertex coordinates, so as to obtain image data corresponding to the filtering region of the second filtering operator in the first intermediate filtering image P1, and perform filtering processing to ensure the accuracy of data acquisition and processing.

Please refer to Fig. 2, Fig. 3 and Fig. 12, in some embodiments, the filtering radius includes a fifth filtering radius and a sixth filtering radius, and the data processing method further comprises the following steps:

017: Determine the first difference between the fifth filter radius of the filter area of each second filter operator and the fifth filter radius of the intermediate filter area, and the sixth filter radius of the filter area of each second filter operator and the second difference of the sixth filter radius of the intermediate filter region;

018: When performing the filtering process on the Nth layer, determine the width of the filtering region of each second filter operator according to the first difference value of each second filter operator and the width of the output image of the N-1th layer , and according to the second difference value of each second filter operator and the height of the output image of the N-1th layer, determine the height of the filter area of each second filter operator;

019: determining the filter region of each second filter operator according to the width and height of each second filter operator; and

020: From the output image, acquire and process image data corresponding to the filtering region of each second filtering operator.

In some embodiments, the data processing apparatus 10 further includes a fourth determination module 17 , a fifth determination module 18 , a sixth determination module 19 and a second processing module 20 . The fourth determination module 17 , the fifth determination module 18 , the sixth determination module 19 and the second processing module 20 are used to execute step 017 , step 018 , step 019 and step 020 respectively. That is, the fourth determination module 17 is configured to determine the first difference between the fifth filter radius of the filter region of each second filter operator and the fifth filter radius of the intermediate filter region, and the filter value of each second filter operator The second difference between the sixth filtering radius of the region and the sixth filtering radius of the intermediate filtering region; the fifth determining module 18 is configured to, when performing filtering processing on the Nth layer, according to the first difference of each second filtering operator value and the width of the output image of the N-1th layer, determine the width of the filtering area of each second filter operator, and based on the second difference value of each second filter operator and the output image of the N-1th layer The height of each second filter operator determines the height of the filter region; the sixth determination module 19 is used to determine the filter region of each second filter operator according to the width and height of each second filter operator; the second The processing module 20 is used for acquiring and processing image data corresponding to the filtering region of each second filtering operator from the output image.

In some embodiments, the processor 30 is further configured to determine the first difference between the fifth filter radius of the filter region of each second filter operator and the fifth filter radius of the intermediate filter region, and each second filter The second difference between the sixth filter radius of the filter region of the operator and the sixth filter radius of the intermediate filter region; when performing the filtering process on the Nth layer, according to the first difference of each second filter operator and the sixth filter radius The width of the output image of the N-1 layer determines the width of the filtering area of each second filter operator, and according to the second difference value of each second filter operator and the height of the output image of the N-1th layer, Determine the height of the filter region of each second filter operator; determine the filter region of each second filter operator according to the width and height of each second filter operator; and from the output image, obtain and process and each second filter operator. Image data corresponding to the filtering region of the second filtering operator. That is, step 017 , step 018 , step 019 and step 020 may be implemented by the processor 30 .

Specifically, when the filtering radius of the filtering operator only includes the fifth filtering radius and the sixth filtering radius, that is, the filtering radius on the upper and lower sides of the filtering area is the same, both are the fifth filtering radius, and the filtering radius on the left and right are the same, are the sixth filter radius.

The processor 30 may determine the first difference between the fifth filter radius of the filter region of each second filter operator and the fifth filter radius of the intermediate filter region, and the sixth filter of the filter region of each second filter operator The second difference between the radius and the sixth filter radius of the intermediate filter region; thereby determining the offset of the filter radius of the filter region of each second filter operator relative to the filter radius of the intermediate filter region, thereby determining the offset of the filter radius of the filter region of each second filter operator ( For example, when the output image of the first filter operator of the first layer (that is, the first intermediate filter image P1, corresponding to the intermediate filter area), the second filter operator can obtain the filtering required from the first intermediate filter image P1. image data.

For example, when filtering is performed at the Nth layer (eg, the second layer), the second filter operator can be used for the height and width of the first intermediate filtered image P1 according to its corresponding first difference and second difference, respectively. difference to obtain the height and width of each second filter operator. Then, the processor 30 determines the filter region of each second filter operator according to the height and width of the second filter operator, for example, the first intermediate filter image P1 is 106*106, the first difference value and the second difference value are The values are 2 and 1 respectively, then the width and height of the second filter operator are 104*102, namely

Therefore, after determining the width and height of the second filter operator, the filter region of each second filter operator can be quickly determined. Finally, the processor 30 acquires image data corresponding to the filtering region of the second filtering operator from the first intermediate filtering image P1, and performs filtering processing to ensure the accuracy of data acquisition and processing.

In another embodiment, the filtering radii of the four sides of the filtering area are all the same, then only the difference between the filtering radius of the filtering area of each second filter operator and the filtering radius of the intermediate filtering area needs to be calculated. Therefore, during subsequent filtering processing, each filtering operator can quickly obtain all the image data required for filtering from the output image output by the previous layer.

In yet another embodiment, the filtering radii of the four sides of the filtering area are all different, that is, the filtering area includes the aforementioned first filtering radius to the fourth filtering radius. When calculating the difference of the filtering radius, it is necessary to calculate each Four differences between the first filter radius to the fourth filter radius of the second filter operator and the first filter radius to the fourth filter radius of the intermediate filter region, so as to facilitate the subsequent filter processing, each filter operator can quickly From the output image output by the previous layer, obtain all the image data required for filtering.

In an example, please refer to Fig. 8 again, when the data processing algorithm is used for data processing, the input data and output data of each filter operator are not down-sampled or up-sampled; the filter radius of each filter operator is is 5 pixels, and in the general method, the planned output of each filter operator is also planned as [w, h], and it is assumed that w=128, h=64.

For the existing filtering algorithm, each filtering layer independently reads and writes data from the memory 50 (such as dynamic random access memory, DRAM) of the terminal 100, the read and write of the first layer to the fourth layer memory 50 The quantities are as follows in Table 1:

Table 1

DRAM read DRAM write L1 (128+2*5)*(64+2*5)=10212 128*64=8192 L2 10212 128*64*3=8192*3 L3 10212*3 128*64*2=8192*2 L4 10212*2 128*64=8192 total 71484 57344

When using the data processing method of the present application for data processing, the read/write volume of the memory 50 is as follows in Table 2:

Table 2

DRAM read DRAM write total (128+2*5*4)*(64+2*5*4)=17472 w*h=8192

It can be seen that by using this method, the amount of data read by DRAM is reduced from 71484 to 17472, the amount of data written by DRAM is reduced from 57344 to 8192, and the amount of reading and writing required by the algorithm is greatly reduced, so the algorithm efficiency and power consumption performance can also be significantly improved. .

Referring to FIG. 13 , a non-volatile computer-readable storage medium 300 storing a computer program 302 according to an embodiment of the present application, when the computer program 302 is executed by one or more processors 30 , makes the processor 30 executable The data processing method of any one of the above embodiments.

For example, referring to FIG. 1, when the computer program 302 is executed by one or more processors 30, the processor 30 is caused to perform the following steps:

011: Obtain the filter radius of the preset multiple filter layers;

012: Determine the filtering region of the data to be processed according to the filtering radii of the plurality of filtering layers; and

013: Acquire image data of the filtered area.

For another example, referring to FIG. 7 , when the computer program 302 is executed by one or more processors 30, the processors 30 may further perform the following steps:

0121: Determine the filtering area of the N-1 layer according to the filtering radius of the N-1 layer and the filtering area of the N-th layer, where N is a positive integer less than or equal to M;

0122: After determining that the filter area of the N-1th layer is obtained, reduce the value of N by 1, and perform step 0121 again: Determine the N-1th layer according to the filter radius of the N-1th layer and the filter area of the Nth layer The filtering area of the layer is determined until the filtering area of the first layer is determined as the filtering area of the image to be processed.

In the description of this specification, reference to the terms "one embodiment," "some embodiments," "exemplary embodiment," "example," "specific example," or "some examples," or the like, is meant to combine embodiments or The particular feature, structure, material or characteristic described by example is included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, those skilled in the art may combine and combine the different embodiments or examples described in this specification, as well as the features of the different embodiments or examples, without conflicting each other.

Any description of a process or method in the flowcharts or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more steps of a program for implementing a specified logical function or process, and The scope of the preferred embodiments of the present application includes alternative implementations in which the functions may be performed out of the order shown or discussed, including performing the functions in a substantially simultaneous manner or in the reverse order depending upon the functions involved, which should be Those skilled in the art to which the embodiments of the application pertain are understood.

Although the embodiments of the present application have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limitations to the present application. Embodiments are subject to variations, modifications, substitutions and alterations.

Claims (13)

1. A method of data processing, comprising:

acquiring preset filtering radiuses of a plurality of filtering layers;

determining a filtering area of data to be processed according to the filtering radiuses of the plurality of filtering layers; and

and acquiring the image data of the filtering area.

2. The data processing method of claim 1, wherein the determining a filtering region of the data to be processed according to the filtering radii of a plurality of filtering layers comprises:

and determining the filtering area according to the filtering radiuses of the plurality of filtering layers and the position information of the data to be processed.

3. The data processing method according to claim 1, wherein the filter layer includes M layers, M is a positive integer, a filter region of the M layer is determined according to a filter radius of the M layer and an image region where the data to be processed is located, and the determining the filter region of the data to be processed according to the filter radius and the position information includes:

determining a filtering area of an N-1 layer according to the filtering radius of the N-1 layer and the filtering area of the Nth layer, wherein N is a positive integer less than or equal to M;

and after determining to obtain the filtering area of the (N-1) th layer, reducing the value of N by 1, and executing the step of determining the filtering area of the (N-1) th layer again according to the filtering radius of the (N-1) th layer and the filtering area of the (N) th layer until determining the filtering area of the (1) th layer as the filtering area of the data to be processed.

4. The data processing method of claim 3, wherein the filter radius comprises a first filter radius, a second filter radius, a third filter radius, and a fourth filter radius, and wherein the first filter radius, the second filter radius, the third filter radius, and the fourth filter radius are the same; or the first filtering radius is the same as the third filtering radius, and the second filtering radius is the same as the fourth filtering radius; or the first filtering radius, the second filtering radius, the third filtering radius and the fourth filtering radius are different from each other.

5. The data processing method according to claim 3, wherein the filter layer includes a filter operator, a filter region of the filter operator of an M-th layer is determined according to a filter radius of the filter operator of the M-th layer and an image region where the data to be processed is located, and the determining of the filter region of an N-1-th layer according to the filter radius of the N-1-th layer and the filter region of the N-1-th layer includes:

determining the filtering region of each filtering operator of the N-1 th layer according to the filtering radius of the filtering operator of the N-1 th layer and the filtering region of the filtering operator of the N-1 th layer.

6. The data processing method according to claim 5, wherein the first filter operator of the N-1 th layer is associated with the second filter operator of the N-th layer, and the association of the first filter operator of the N-1 th layer with the second filter operator of the N-1 th layer comprises the output data of the filter operator of the N-1 th layer being the input data of the filter operator of the N-1 th layer, the first filter operator being any one of the filter operators of the N-1 th layer, and the second filter operator being any one of the filter operators of the N-1 th layer;

said determining said filter region of each said filter operator of layer N-1 from said filter radius of said filter operator of layer N-1 and a filter region of said filter operator of layer N, comprising:

determining a middle filtering region according to the filtering region of each second filtering operator of the Nth layer; and

and determining the filtering area of the first filtering operator according to the intermediate filtering area and the filtering radius of the first filtering operator.

7. The data processing method according to claim 6, wherein the second filter operator associated with the first filter operator is plural, and the determining an intermediate filter region from the filter regions of each second filter operator of the Nth layer comprises:

and determining the intermediate filtering region according to a union set of the filtering regions of the plurality of second filtering operators.

8. The data processing method of claim 6, further comprising:

determining the coordinate offset of each second filter operator according to the vertex coordinates of the filter region of each second filter operator and the vertex coordinates of the middle filter region;

when the Nth layer is subjected to filtering processing, determining the vertex coordinates of the filtering area of each second filtering operator according to the coordinate offset of each second filtering operator and the vertex coordinates of the output image of the Nth-1 layer;

and acquiring and processing image data corresponding to the filtering region of each second filtering operator from the output image.

9. The data processing method of claim 6, wherein the filter radius comprises a fifth filter radius and a sixth filter radius; the data processing method further comprises:

determining a first difference of the fifth filter radius of the filter region of each of the second filter operators and the fifth filter radius of the intermediate filter region, and a second difference of the sixth filter radius of the filter region of each of the second filter operators and the sixth filter radius of the intermediate filter region;

when the Nth layer is subjected to filtering processing, determining the height of a filtering region of each second filtering operator according to the first difference value of each second filtering operator and the width of an output image of the Nth-1 layer, and determining the width of the filtering region of each second filtering operator according to the second difference value of each second filtering operator and the height of the output image of the Nth-1 layer;

determining a filtering area of each second filtering operator according to the width and the height of each second filtering operator; and

and acquiring and processing image data corresponding to the filtering region of each second filtering operator from the output image.

10. The data processing method of claim 1, further comprising:

and carrying out filtering processing on the image data of the filtering area to generate a filtering image.

11. A data processing apparatus, characterized by comprising:

the first obtaining module is used for obtaining the preset filtering radiuses of a plurality of filtering layers;

the first determining module is used for determining a filtering area of data to be processed according to the filtering radiuses of the plurality of filtering layers; and

and the second acquisition module is used for acquiring the image data of the filtering area.

12. The terminal is characterized by comprising a processor, a first filter layer and a second filter layer, wherein the processor is used for obtaining the filter radius of the preset plurality of filter layers; determining a filtering area of data to be processed according to the filtering radiuses of the plurality of filtering layers; and acquiring the image data of the filtering area.

13. A non-transitory computer-readable storage medium comprising a computer program which, when executed by a processor, causes the processor to perform the data processing method of any one of claims 1-10.

Images