CN111445408A - Method, device and storage medium for performing differentiation processing on image - Google Patents

Method, device and storage medium for performing differentiation processing on image Download PDF

Info

Publication number
CN111445408A
CN111445408A CN202010217463.9A CN202010217463A CN111445408A CN 111445408 A CN111445408 A CN 111445408A CN 202010217463 A CN202010217463 A CN 202010217463A CN 111445408 A CN111445408 A CN 111445408A
Authority
CN
China
Prior art keywords
image
region
original image
transition
blurred
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202010217463.9A
Other languages
Chinese (zh)
Inventor
王杰
尹浪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang Dahua Technology Co Ltd
Original Assignee
Zhejiang Dahua Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang Dahua Technology Co Ltd filed Critical Zhejiang Dahua Technology Co Ltd
Priority to CN202010217463.9A priority Critical patent/CN111445408A/en
Publication of CN111445408A publication Critical patent/CN111445408A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T9/00—Image coding
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/167—Position within a video image, e.g. region of interest [ROI]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20212—Image combination
    • G06T2207/20221—Image fusion; Image merging

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)
  • Image Processing (AREA)

Abstract

The invention discloses a method, a device and a storage medium for performing differentiation processing on an image, which are used for solving the technical problem that the display effect of an image cannot be improved while the image size is controlled in the prior art. The method comprises the following steps: acquiring an interested area corresponding to a target object from the first original image, and acquiring a second original image corresponding to the interested area; the first original image and the second original image have a first resolution; blurring the first original image to obtain a first blurred image; the first blurred image and the first original image are the same in size, and the second resolution of the first blurred image is smaller than the first resolution; superposing the second original image to an interested area corresponding to the target object in the first blurred image to obtain a superposed image; and circularly encoding the superposed image by adopting the estimated quality factor until the size of the encoded JPEG file reaches the preset size, and obtaining the target image.

Description

Method, device and storage medium for performing differentiation processing on image
Technical Field
The present invention relates to the field of image processing, and in particular, to a method, an apparatus, and a storage medium for performing differentiation processing on an image.
Background
With the implementation of the intelligent scheme on the external camera equipment such as the snapshot camera, higher requirements are put forward on the resolution and detail effect of the object in the snapshot process, for example, people expect to see images with clearer details and richer detail features.
The appearance of camera equipment with 4K + resolution meets the requirement on image definition, but after the original data collected by the camera equipment is coded by Motion Joint Photographic Experts Group, MJPEG)/(Joint Photographic Experts Group, JPEG), the size of the obtained MJPEG/JPEG picture is larger.
In practical applications, since a client often has a requirement on the size of a finally generated image (i.e., a target image), the size of the target image obtained after encoding raw data of an image with high resolution exceeds the actually required size.
In the related art, in order to control the image file size, it is a common practice to adjust the size of the image file by adjusting the quality factor after the image file is generated or to adjust the size of the image file by adjusting the quality factor before the image file is generated. Since both schemes encode the whole image, when they are used to control the size of the high-resolution image, the display effect of the whole target image will be large if the display effect is maintained, and the display effect of the region of interest in the target image will be reduced if the size is within the size required by the client.
Therefore, how to improve the display effect of the region of interest in the image while controlling the size of the image becomes an urgent technical problem to be solved.
Disclosure of Invention
The invention provides a method, a device and a storage medium for performing differentiation processing on an image, which are used for solving the technical problem that the display effect of an image cannot be improved while the image size is controlled in the prior art.
In a first aspect, to solve the above technical problem, a technical solution of a method for performing differentiation processing on an image according to an embodiment of the present invention is as follows:
acquiring an interesting region corresponding to a target object from a first original image by adopting a preset model, and acquiring a second original image corresponding to the interesting region from the first original image; the first original image is an image formed by YUV original data acquired by image acquisition equipment, the preset model is a model obtained after identification training of the target object, and the first original image and the second original image have first resolution;
blurring the first original image to obtain a first blurred image; wherein the first blurred image is the same size as the first original image, the first blurred image has a second resolution, and the second resolution is less than the first resolution;
superposing the second original image to a region of interest corresponding to the target object in the first blurred image to obtain a superposed image;
circularly encoding the superposed image according to a preset image format by adopting the estimated quality factor until the size of the encoded file reaches a preset size, and obtaining a target image; the estimated quality factor is determined according to the preset size and corresponds to the first resolution.
The image quality of the region of interest is improved by properly reducing the resolution of the image of the non-region of interest in one image and maintaining the high resolution of the region of interest, so that the same JPEG image size has a higher-quality coding effect on the image of the region of interest, and the image quality of the region of interest is improved under the condition that the control target image reaches the preset size, and the display effect of the whole image is improved.
Optionally, superimposing the second original image onto the region of interest corresponding to the target object in the first blurred image, to obtain a superimposed image, including:
determining a corresponding position region in the first blurred image based on the relative position of the region of interest in the first original image;
and covering the image of the part corresponding to the position area in the first blurred image by using the second original image to obtain the superposed image.
Optionally, before performing cyclic encoding on the superimposed image according to a preset picture format by using the pre-estimated quality factor, the method further includes:
acquiring the estimated quality factor corresponding to the preset size from a quality factor average value table; the quality factor average value table is a corresponding relation between different sizes of files with the preset picture format and used quality factors generated aiming at the same original image, the shooting resolution ratio of the image acquisition equipment adopted by the same original image and the first original image is the same, and the sizes of the shot original images are the same.
Optionally, after obtaining the superimposed image, before performing cyclic encoding on the superimposed image by using the estimated quality factor, the method further includes:
setting a transition region between the edge of the region of interest and the edge of the first original image; wherein an inner edge of the transition region is an edge of the region of interest, and an outer edge of the transition region is located between the edge of the region of interest and the edge of the first original image;
dividing the transition region into a plurality of transition sub-regions; wherein the distance between the inner and outer edges of each transition subregion is the same;
blurring each transition sub-region to obtain a blurred transition region, and overlaying the blurred transition region to the periphery of a position region in the superimposed image to obtain a new superimposed image; in the direction from the transition subarea along the outermost edge to the transition subarea along the innermost edge, the blurring factor used in the blurring processing of the transition subareas is increased by a set step.
Optionally, the blurring process is performed on each transition sub-region, including:
if the figure formed by the edges of the interested region is a polygon, dividing the transition sub-region into a plurality of transition micro-regions; the two adjacent transition micro regions are complementary, a boundary between the two adjacent micro regions is formed by extending an inner edge line of one micro region to an outer edge line of the other micro region, and the inner edge line and the outer edge line are straight lines;
and respectively carrying out the same fuzzification processing on each transition micro-region.
In a second aspect, an embodiment of the present invention provides an apparatus for performing differentiation processing on an image, including:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring an interesting region corresponding to a target object from a first original image by adopting a preset model and acquiring a second original image corresponding to the interesting region from the first original image; the first original image is an image formed by YUV original data acquired by image acquisition equipment, the preset model is a model obtained after identification training of the target object, and the first original image and the second original image have first resolution;
the blurring processing unit is used for performing blurring processing on the first original image to obtain a first blurred image; wherein the first blurred image is the same size as the first original image, the first blurred image has a second resolution, and the second resolution is less than the first resolution;
the superposition unit is used for superposing the second original image to an interested area corresponding to the target object in the first blurred image to obtain a superposed image;
the encoding unit is used for circularly encoding the superposed image according to a preset image format by adopting the estimated quality factor until the size of the encoded file reaches a preset size, so as to obtain a target image; the estimated quality factor is determined according to the preset size and corresponds to the first resolution.
Optionally, the superimposing unit is specifically configured to:
determining a corresponding position region in the first blurred image based on the relative position of the region of interest in the first original image;
and covering the image of the part corresponding to the position area in the first blurred image by using the second original image to obtain the superposed image.
Optionally, the encoding unit is further configured to:
acquiring the estimated quality factor corresponding to the preset size from a quality factor average value table; the quality factor average value table is a corresponding relation between different sizes of files with the preset picture format and used quality factors generated aiming at the same original image, the shooting resolution ratio of the image acquisition equipment adopted by the same original image and the first original image is the same, and the sizes of the shot original images are the same.
Optionally, the blur processing unit is further configured to:
setting a transition region between the edge of the region of interest and the edge of the first original image; wherein an inner edge of the transition region is an edge of the region of interest, and an outer edge of the transition region is located between the edge of the region of interest and the edge of the first original image;
dividing the transition region into a plurality of transition sub-regions; wherein the distance between the inner and outer edges of each transition subregion is the same;
blurring each transition sub-region to obtain a blurred transition region, and overlaying the blurred transition region to the periphery of a position region in the superimposed image to obtain a new superimposed image; in the direction from the transition subarea along the outermost edge to the transition subarea along the innermost edge, the blurring factor used in the blurring processing of the transition subareas is increased by a set step.
Optionally, the blur processing unit is further configured to:
if the figure formed by the edges of the interested region is a polygon, dividing the transition sub-region into a plurality of transition micro-regions; the two adjacent transition micro regions are complementary, a boundary between the two adjacent micro regions is formed by extending an inner edge line of one micro region to an outer edge line of the other micro region, and the inner edge line and the outer edge line are straight lines;
and respectively carrying out the same fuzzification processing on each transition micro-region.
In a third aspect, an embodiment of the present invention further provides an apparatus for performing differentiation processing on an image, where the apparatus includes:
at least one processor, and
a memory coupled to the at least one processor;
wherein the memory stores instructions executable by the at least one processor, the at least one processor implementing the method of the first aspect by executing the instructions stored by the memory.
In a fourth aspect, an embodiment of the present invention further provides a readable storage medium, including:
a memory for storing instructions that, when executed by the processor, cause an apparatus comprising the readable storage medium to perform the method of the first aspect.
Through the technical solutions in one or more of the above embodiments of the present invention, the embodiments of the present invention have at least the following technical effects:
in the embodiment provided by the invention, the region of interest corresponding to the target object is acquired from the first original image by adopting a preset model, and the second original image corresponding to the region of interest is acquired from the first original image; then, the first original image is fuzzified to obtain a first blurred image with the same size as the first original image, and the resolution of the first blurred image is smaller than that of the first original image; then, a second original image with the same resolution as the first original image is superposed to a region of interest corresponding to the target object in the first blurred image, and a superposed image is obtained; circularly encoding the superposed image according to a preset image format by adopting the estimated quality factor until the size of the encoded file reaches a preset size, and obtaining a target image; the system comprises a first original image, a preset model and a second original image, wherein the first original image is an image formed by YUV original data acquired by image acquisition equipment, the preset model is a model obtained after identification training of a target object, and the first original image and the second original image have a first resolution; the estimated quality factor is determined according to a preset size and corresponds to the first resolution. Therefore, the image quality of the region of interest is improved by properly reducing the resolution of the image of the non-region of interest in one image and maintaining the high resolution of the region of interest, so that the same image size with the preset picture format has a coding effect with higher quality for the image of the region of interest, and the image quality of the region of interest is improved under the condition that the control target image reaches the preset size, so that the display effect of the whole image is improved.
Drawings
FIG. 1 is a flow chart of adjusting the size of a picture after JPEG encoding;
FIG. 2 is a flow chart of resizing a picture before JPEG encoding;
fig. 3 is a flowchart of a method for performing differentiation processing on an image according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of acquiring a first original image according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of acquiring a first original image according to an embodiment of the present invention;
FIG. 6 is a diagram illustrating a first blurred image according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of obtaining an overlay image according to an embodiment of the present invention;
FIG. 8 is a diagram illustrating a transition region in a first original image according to an embodiment of the present invention;
FIG. 9 is a diagram of a transition sub-field provided by an embodiment of the present invention;
FIG. 10 is a diagram illustrating a transition sub-region being blurred according to an embodiment of the present invention;
fig. 11 is a schematic structural diagram of an apparatus for performing image differentiation processing according to an embodiment of the present invention.
Detailed Description
The embodiment of the invention provides a method, a device and a storage medium for performing differentiation processing on an image, and aims to solve the technical problem that the display effect of an image cannot be improved while the image size is controlled in the prior art.
In order to solve the technical problems, the general idea of the embodiment of the present application is as follows:
provided is a method for performing differentiation processing on an image, comprising the following steps: acquiring an interesting region corresponding to a target object from a first original image by adopting a preset model, and acquiring a second original image corresponding to the interesting region from the first original image; the system comprises a first original image, a preset model and a second original image, wherein the first original image is an image formed by YUV original data acquired by image acquisition equipment, the preset model is a model obtained after identification training of a target object, and the first original image and the second original image have a first resolution; blurring the first original image to obtain a first blurred image; the first blurred image and the first original image are the same in size, the first blurred image has a second resolution, and the second resolution is smaller than the first resolution; superposing the second original image to an interested area corresponding to the target object in the first blurred image to obtain a superposed image; circularly encoding the superposed image according to a preset image format by adopting the estimated quality factor until the size of the encoded file reaches a preset size, and obtaining a target image; the estimated quality factor is determined according to a preset size and corresponds to the first resolution.
In the scheme, the region of interest corresponding to the target object is acquired from the first original image by adopting the preset model, and the second original image corresponding to the region of interest is acquired from the first original image; then, the first original image is fuzzified to obtain a first blurred image with the same size as the first original image, and the resolution of the first blurred image is smaller than that of the first original image; then, a second original image with the same resolution as the first original image is superposed to a region of interest corresponding to the target object in the first blurred image, and a superposed image is obtained; circularly encoding the superposed image according to a preset image format by adopting the estimated quality factor until the size of the encoded file reaches a preset size, and obtaining a target image; the system comprises a first original image, a preset model and a second original image, wherein the first original image is an image formed by YUV original data acquired by image acquisition equipment, the preset model is a model obtained after identification training of a target object, and the first original image and the second original image have a first resolution; the estimated quality factor is determined according to a preset size and corresponds to the first resolution. Therefore, the image quality of the region of interest is improved by properly reducing the resolution of the image of the non-region of interest in one image and maintaining the high resolution of the region of interest, so that the same image size with the preset picture format has a coding effect with higher quality for the image of the region of interest, and the image quality of the region of interest is improved under the condition that the control target image reaches the preset size, so that the display effect of the whole image is improved.
In order to better understand the technical solutions of the present invention, the following detailed descriptions of the technical solutions of the present invention are provided with the accompanying drawings and the specific embodiments, and it should be understood that the specific features in the embodiments and the examples of the present invention are the detailed descriptions of the technical solutions of the present invention, and are not limitations of the technical solutions of the present invention, and the technical features in the embodiments and the examples of the present invention may be combined with each other without conflict.
Before introducing the scheme of the present application, two commonly used methods for adjusting the size of a picture are introduced, specifically as follows:
first, adjusting after file generation: after JPEG encoding is performed on the raw YUV data and a target file is generated, the size of the whole picture of the next encoding is adjusted by adjusting the quality factor of the picture encoding, please refer to fig. 1, which is a flowchart of adjusting the size of the picture after JPEG encoding.
Step 101: and determining a first quality factor corresponding to the estimated size of the target file according to the empirical values of the quality factors corresponding to different sizes of the target file.
Step 102: and JPEG coding is carried out on the original file corresponding to the target file according to the first quality factor to obtain a JPEG file, wherein the JPEG file is stored in the buffer area.
Step 103: and adjusting the value of the quality factor according to the difference value between the size of the JPEG file and the estimated size to obtain a new quality factor.
Step 104: and coding the JPEG file according to the new quality factor.
The technical scheme has the defects that for post adjustment, the difference between the size of the possibly generated picture file and the estimated size of the target file is large, and the accuracy is not high.
Second, adjusting before generating the file: if the difference between the size of the raw YUV data and the size of the target file is large, such as exceeding 10% of the size of the target file, or being less than 90% of the size of the target file, the raw YUV data is encoded for multiple times before generating the JPEG file, and the encoding quality factor is adjusted, please refer to fig. 2, which is a flowchart for adjusting the size of the picture before JPEG encoding.
Step 201: and determining a first quality factor corresponding to the estimated size of the target file according to the empirical values of the quality factors corresponding to different sizes of the target file.
Step 202: and judging whether the current coding frequency of the current JPEG coding reaches the maximum coding frequency, wherein the current coding frequency is 1 when a second quality factor used for JPEG coding is a first quality factor.
If the current encoding frequency does not reach the maximum encoding frequency, executing step 203; if the current encoding frequency reaches the maximum encoding frequency, go to step 206.
Step 203: JPEG coding is carried out on the intermediate file according to the second quality factor, and a coded file is obtained; and the current coding times are added by 1; when the second quality factor is the first quality factor, the intermediate file is an original file corresponding to the target file;
step 204: adjusting the value of the second quality factor according to a preset step length to obtain a new second quality factor;
step 205: and judging whether the difference value between the size of the encoded file and the estimated size is within a set range. For example, the set range is ± 10%.
If so, go to step 206, otherwise, go to step 202.
Step 206: reading the file from the buffer area for storing the encoded file to obtain a target file; and the target file is a JPEG file which is finally generated.
It should be noted that the JPEG file may also be an MJPEG file, and is not particularly limited.
According to the strategy, before the JPEG file is generated, the size of the JPEG file is adjusted by adjusting the quality factor for multiple times, and the target file which is in accordance with the expected estimated size is finally obtained, so that the size of the target file can be controlled more accurately.
The inventor finds that the above two schemes for controlling the size of the picture file use the same quality factor for the whole picture to encode, so that the realistic effect of the image is reduced under the condition of fixed picture size.
Referring to fig. 3, an embodiment of the invention provides a method for performing differentiation processing on an image, and the processing procedure of the method is as follows.
Step 301: acquiring an interesting region corresponding to a target object from a first original image by adopting a preset model, and acquiring a second original image corresponding to the interesting region from the first original image; the first original image is an image formed by YUV original data acquired by image acquisition equipment, the preset model is a model obtained after identification training of a target object, and the first original image and the second original image have first resolution.
The capturing device is generally arranged at an entrance and an exit of a highway, an airport, a subway and other public channels and used for monitoring vehicles and personnel, for example, the captured vehicles are subjected to information identification, fake-licensed vehicles can be found in time, criminals can be found in time through identifying the personnel, when the vehicles and the personnel are identified, the higher the resolution of the first original images of the captured vehicles and the personnel is, the more the details of target objects (the vehicles and the personnel) can be displayed, and therefore the identification accuracy can be improved.
The resolution of the first original image is usually determined by an image capturing device that captures the first original image, for example, the image capturing device that is used is a 4k high-definition device, and the captured first image is an ultra-high-definition device, such as 4096 × 2160 pixel, which has the disadvantages that even after JPEG encoding is performed on the first original image (i.e., YUV image), the storage space occupied by the obtained picture is very large, and a client usually has a limitation on the storage size of the picture, or a storage server that stores the pictures has a limitation on the size of the picture file, or in order to ensure that the file transfer rate is within an appropriate range, the size of the picture file is limited, etc., in order to control the storage size of the image after encoding the first original image according to a preset picture format without reducing the resolution of the target object, the present application makes the target object in the first original image maintain the original resolution, and reduces the resolution of other parts of the first original image except the target object, thereby solving the above technical problems, specifically, the scheme is that:
after a first original image (including a target object vehicle and a background) of a target object (such as a vehicle) is acquired through an image acquisition device, a region of interest corresponding to the target object is acquired from the first original image by using a preset model, and a second original image corresponding to the region of interest is acquired from the first original image, wherein the second original image has the same resolution as the first original image, and can be temporarily stored in a designated storage position and be taken out from the storage position when needed subsequently.
For example, please refer to fig. 4 and 5, fig. 4 is a schematic diagram of acquiring a first original image according to an embodiment of the present invention, and fig. 5 is a schematic diagram of acquiring a first original image according to an embodiment of the present invention.
In fig. 4, the target object is a vehicle, the background of the target object includes the road surface of the road and green plants on both sides of the road, and the lines of the road surface are clearly visible in the first original image, a preset model obtained after the vehicle (target object) is recognized and trained in advance is used, the origin coordinates of the region of interest in the first original image are (X, Y), the length is L, and the width is W, which are obtained from the region of interest (indicated by the area surrounded by the dotted line in fig. 2) corresponding to the target object in the first original image, and the second original image corresponding to the region of interest is obtained from the first original image (see fig. 2).
In fig. 4 and 5, the objects on the back side are illustrated by solid lines for convenience of description. The origin of the region of interest is the vertex of the region of interest closest to the origin of the first original image (assumed to be the upper left corner of fig. 4), and the vehicles used for training the preset model include various types of vehicles, not just one type of vehicle.
After the second original image is obtained, steps 302-303 can be performed.
Step 302: blurring the first original image to obtain a first blurred image; the first blurred image and the first original image are the same in size, the first blurred image has a second resolution, and the second resolution is smaller than the first resolution.
Step 303: and superposing the second original image to the region of interest corresponding to the target object in the first blurred image to obtain a superposed image.
Continuing to take the first original image as an example in fig. 4, after the first original image is blurred, a first blurred image is obtained, please refer to fig. 6, and fig. 6 is a schematic diagram (shown by a dotted line) of the first blurred image provided by the embodiment of the present invention. After the first original image is subjected to blurring processing, the second resolution efficiency of the obtained first blurred image is smaller than the first resolution of the first original image, and the size of the first blurred image is the same as that of the first original image. This can effectively reduce the storage space occupied by the first blurred image, and since the size of the first blurred image is the same as the size of the first original image, the second original image can be accurately superimposed on the first blurred image in step 303 according to the fact that the coordinates of the region of interest in the first blurred image are the same as the coordinates of the region of interest in the first original image.
And superposing the second original image to the region of interest corresponding to the first blurred image to obtain a superposed image, and specifically adopting the following processes:
firstly, determining a corresponding position area in a first blurred image based on the relative position of a region of interest in the first original image; and covering the image of the part corresponding to the position area in the first blurred image by using the second original image to obtain a superposed image.
For example, please refer to fig. 7, fig. 7 is a schematic diagram of obtaining an overlay image according to an embodiment of the present invention.
Continuing with fig. 4-6 as an example, in step 301, the origin coordinates of the region of interest obtained by the preset model are (X, Y), the length and the width are L, W, that is, the relative position of the region of interest in the first original image is determined, and since the size of the first original image is the same as that of the first blurred image, the position of the target object in the first blurred image can be determined by determining the corresponding position area in the first blurred image according to the relative position of the region of interest in the first original image, and then the image of the portion corresponding to the position area in the first blurred image is covered by the second original image with the first resolution, so that the superimposed image can be obtained.
It should be noted that, for convenience of illustration, in the embodiment provided by the present invention, the first blurred image is illustrated by a dotted line, and in practical applications, an object in the first blurred image is not a dotted line, and here, the object is only to form a difference with the first original image for convenience of illustration.
After the overlay image is obtained, step 304 may be performed.
Step 304: circularly encoding the superposed image according to a preset image format by adopting the estimated quality factor until the size of the encoded file reaches a preset size, and obtaining a target image; the estimated quality factor is determined according to a preset size and corresponds to the first resolution.
After obtaining the superimposed image, before performing cyclic coding on the superimposed image by adopting the pre-estimated quality factor, the method further comprises the following steps:
firstly, setting a transition region between the edge of a region of interest and the edge of a first original image; wherein, the inner edge of the transition region is the edge of the region of interest, and the outer edge of the transition region is located between the edge of the region of interest and the edge of the first original image.
Secondly, dividing the transition area into a plurality of transition sub-areas; wherein the distance between the inner and outer edges of each transition subregion is the same;
finally, blurring each transition sub-region to obtain a blurred transition region, and overlaying the blurred transition region to the periphery of a position region in the superimposed image to obtain a new superimposed image; in the direction from the transition subarea along the outermost edge to the transition subarea along the innermost edge, the blurring factor used in the blurring processing of the transition subareas is increased by a set step.
Please refer to fig. 8, which is a schematic diagram of a transition region in a first original image according to an embodiment of the present invention.
In a of fig. 8, the whole image is the first original image, the middle white region is the region of interest in the first original image where the second original image is located, and a transition region (the first loop region formed around the white region in a of fig. 8) is disposed between the edge of the region of interest and the edge of the first original image.
Dividing the transition region in a of fig. 8 into a plurality of transition sub-regions, wherein a closed figure formed by the outer edge and the inner edge of each transition sub-region is similar to a closed figure formed by the region of interest. Please refer to b of fig. 8, take the division into 2 transition sub-regions, i.e. transition sub-region 1 and transition sub-region 2 as an example. When the transition region is divided into a plurality of transition subregions, the transition subregions can be divided at equal distances, and when a figure formed by the outer edges of the interested region is a polygon, the distances d between two adjacent parallel edges on the same side are equal; when the graph formed by the outer edge of the region of interest is a closed curve, the distance d between the outer edges of two adjacent transition sub-regions, and so on, please refer to fig. 9, which is a schematic diagram of the transition sub-regions provided by the embodiment of the present invention.
Then, performing fuzzy processing on each transition sub-region to obtain a blurred transition region, please refer to c in fig. 8, when processing the transition sub-regions, if the transition sub-region 1 adopts a fuzzy factor as the fuzzy factor 1, the fuzzy factor 2 adopted by the transition sub-region superposes the fuzzy factor 1 to set a step length. If the transition area is divided into more transition sub-areas, the transition factor of the transition sub-area at the outermost edge is minimum, the transition factor of the transition sub-area at the innermost edge is maximum, and the transition factors of the transition sub-areas between the transition sub-areas are gradually increased according to the set step length. And finally, overlapping the blurred transition region to the periphery of the position region in the superimposed image to obtain a new superimposed image.
For example, the transition region is divided into n transition sub-regions, the nth transition sub-region is the region of interest, the 1 st transition sub-region is farthest from the region of interest, and the blurring factor corresponding to the nth transition sub-region is:
dim _ factor (an) ═ Dim _ factor _ a-adjust Dim × (n-1) formula 1;
wherein Dim _ factor _ a is a blurring factor of the first blurred image, Dim _ factor (an) is a blurring factor of the nth transition sub-region, and adjust Dim is a set step length.
The smaller the value of Dim _ factor (an), the lower the degree of blurring.
By adding the transition region around the target position (i.e. the target object) of the superimposed image and performing the blurring processing on the transition region by using the formula 1, the blurring degree of the region of interest (i.e. the region corresponding to the target object) in the transition to the region of no interest can be gradually reduced, so that the quality difference between the target object and the surrounding region thereof can be effectively reduced, and the display effect is improved.
Specifically, the blurring process is performed on each transition sub-region, and the following method may be adopted:
referring to fig. 10, which is a schematic diagram of performing a blurring process on a transition sub-region according to an embodiment of the present invention, if a graph formed by edges of a position region is a polygon, the transition sub-region is divided into a plurality of transition micro regions, and in fig. 10, a transition sub-region 1 is divided into 4 transition micro regions, and the dividing manners of other transition sub-regions are similar and are not repeated; wherein, two adjacent transition micro regions are complementary, and a boundary line (indicated by a dotted line in fig. 9) between two adjacent micro regions is formed by extending an inner edge line of one micro region to an outer edge line of the other micro region, and the inner edge line and the outer edge line are both straight lines.
And respectively carrying out the same fuzzification treatment on each transition micro-region, namely carrying out the fuzzification treatment on each transition micro-region contained in each transition sub-region by adopting the same fuzzification factor.
By dividing the transition sub-region into a plurality of transition micro-regions, fuzzification processing is conveniently carried out on each transition micro-region.
Before the overlapped image is circularly coded by adopting the estimated quality factor, the method further comprises the following steps:
acquiring the estimated quality factor corresponding to a preset size from a quality factor average value table; the quality factor average value table is a corresponding relation between different sizes of image files with preset image formats generated aiming at the same original image and used quality factors, the shooting resolution ratio of the image acquisition equipment adopted by the same original image and the first original image is the same, and the sizes of the shot original images are the same.
For example, please refer to table 1, which is a quality factor average value table, wherein data is obtained by performing multiple encoding experiments using an encoder under a specific application scenario according to a specific device and at a specific width and height (e.g. the width and height of a picture corresponding to YUV raw data is 4096 × 2160).
TABLE 1
Figure BDA0002424864720000151
Assuming that the length and width of the first original image is 4096 × 2160 and the type of photographing apparatus used is the same as that used in the experiment in table 1, if a JPEG image with a preset size of 1M (the preset picture format is JPEG) is to be reached by the target image, the estimated quality factor is set to 65 if the quality factor corresponding to 1M is found from table 1 to be 65.
After the quality factor value is determined, the pre-estimated quality factor 65 may be used to cyclically encode the superimposed image according to a preset picture format (i.e., JPEG format) (or cyclically encode a new superimposed image if a transition region is set), until the size of the encoded JPEG file reaches a preset size, thereby obtaining the target image. Specifically, the method for encoding the image may be referred to in the related art described earlier in fig. 1 of the present application, and will not be described herein again.
Optionally, when performing cyclic coding, the current quality factor used by the current coding may be determined in a manner of increasing or decreasing the prediction step size based on the prediction quality factor.
For example, the estimated quality factor is 65, the estimated step size is 5, if the current quality factor is determined in a descending manner, when the superimposed image is coded for the first time, the current quality factor used is 65, the image after the first coding is obtained, when the image is coded for the second time (that is, the image after the first coding is coded), the current quality factor used is 65-5-60, the image after the second coding is obtained, and so on, until the size of the coded image reaches the preset size, the target image is obtained.
Hereinafter, the method of determining the estimated step size will be described.
The ratio of the first original image to the target image in the storage space is determined and recorded as a ratio.
ratio ═ currjegsize/targetjegsize equation 2;
wherein, currjpegsize is the size of the storage space occupied by the first original image, and targetJpegSize is the size of the storage space occupied by the target image (i.e. the preset size).
If the first resolution is 4096 × 2160, the preset size is targetJpegSize is 1024 bytes, and the current encoded file size currjpegsize is 800k, then:
ration＝curJpegSize/targetJpegSize＝0.8。
if the estimated quality factor (marked as base _ QP _ factor) is 65, and the currently used quality factor is marked as qpValue when JPEG encoding is performed, then:
qpValue base _ QP _ factor + qpStep formula 3;
wherein qpStep is the estimated step size.
According to different ratios of the storage space occupied by the first original image and the target image, an encoding test is performed to obtain estimated step sizes qpStep corresponding to different ratios, specifically referring to table 2.
TABLE 2
Figure BDA0002424864720000161
After the estimated quality factor and the estimated step length are determined, the superposed image or the new superposed image can be coded according to a coding method of a preset picture format until a target image with a preset size is obtained.
In the embodiment provided by the present invention, the preset picture format may be a JPEG format, an MJPEG format, a Portable Network Graphics (PNG) format, a Bitmap image (BMP) format, and other picture formats, and the specific preset picture format is any picture format, which is not limited herein.
The resolution of the region of interest in the first original image is kept unchanged, the resolution of other regions except the region of interest is reduced, and then picture coding is performed, so that the display quality of the region of interest can be improved, the storage space occupied by the whole target picture can be reduced, and the actual effect of the target image can be effectively improved.
Based on the same inventive concept, an embodiment of the present invention provides an apparatus for performing differentiation processing on an image, where specific implementation of a processing method of the apparatus may refer to descriptions in the method embodiment, and repeated descriptions are omitted, please refer to fig. 11, and the apparatus includes:
an obtaining unit 1101, configured to obtain, by using a preset model, a region of interest corresponding to a target object from a first original image, and obtain, from the first original image, a second original image corresponding to the region of interest; the first original image is an image formed by YUV original data acquired by image acquisition equipment, the preset model is a model obtained after identification training of the target object, and the first original image and the second original image have first resolution;
a blurring processing unit 1102, configured to perform blurring processing on the first original image to obtain a first blurred image; wherein the first blurred image is the same size as the first original image, the first blurred image has a second resolution, and the second resolution is less than the first resolution;
a superimposing unit 1103, configured to superimpose the second original image onto the region of interest corresponding to the target object in the first blurred image, so as to obtain a superimposed image;
the encoding unit 1104 is configured to perform cyclic encoding on the superimposed image according to a preset picture format by using the estimated quality factor until the size of the encoded file reaches a preset size, so as to obtain a target image; the estimated quality factor is determined according to the preset size and corresponds to the first resolution.
Optionally, the superimposing unit 1103 is specifically configured to:
determining a corresponding position region in the first blurred image based on the relative position of the region of interest in the first original image;
and covering the image of the part corresponding to the position area in the first blurred image by using the second original image to obtain the superposed image.
Optionally, the encoding unit 1104 is further configured to:
acquiring the estimated quality factor corresponding to the preset size from a quality factor average value table; the quality factor average value table is a corresponding relation between different sizes of files with preset picture formats generated aiming at the same original image and used quality factors, the shooting resolution ratio of the image acquisition equipment adopted by the same original image and the first original image is the same, and the sizes of the shot original images are the same.
Optionally, the blur processing unit 1102 is further configured to:
setting a transition region between the edge of the region of interest and the edge of the first original image; wherein an inner edge of the transition region is an edge of the region of interest, and an outer edge of the transition region is located between the edge of the region of interest and the edge of the first original image;
dividing the transition region into a plurality of transition sub-regions; wherein the distance between the inner and outer edges of each transition subregion is the same;
blurring each transition sub-region to obtain a blurred transition region, and overlaying the blurred transition region to the periphery of a position region in the superimposed image to obtain a new superimposed image; in the direction from the transition subarea along the outermost edge to the transition subarea along the innermost edge, the blurring factor used in the blurring processing of the transition subareas is increased by a set step.
Optionally, the blur processing unit 1102 is further configured to:
if the figure formed by the edges of the interested region is a polygon, dividing the transition sub-region into a plurality of transition micro-regions; the two adjacent transition micro regions are complementary, a boundary between the two adjacent micro regions is formed by extending an inner edge line of one micro region to an outer edge line of the other micro region, and the inner edge line and the outer edge line are straight lines;
and respectively carrying out the same fuzzification processing on each transition micro-region.
Based on the same inventive concept, an embodiment of the present invention further provides an apparatus for performing differentiation processing on an image, where the apparatus includes:
at least one processor, and
a memory coupled to the at least one processor;
wherein the memory stores instructions executable by the at least one processor, and the at least one processor performs the method for processing image difference as described above by executing the instructions stored in the memory.
Based on the same inventive concept, an embodiment of the present invention further provides a readable storage medium, including:
a memory for storing instructions that, when executed by the processor, cause an apparatus comprising the readable storage medium to perform a method of differencing an image as described above.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (10)

1. A method for performing differentiation processing on an image, comprising:
acquiring an interesting region corresponding to a target object from a first original image by adopting a preset model, and acquiring a second original image corresponding to the interesting region from the first original image; the first original image is an image formed by YUV original data acquired by image acquisition equipment, the preset model is a model obtained after identification training of the target object, and the first original image and the second original image have first resolution;
blurring the first original image to obtain a first blurred image; wherein the first blurred image is the same size as the first original image, the first blurred image has a second resolution, and the second resolution is less than the first resolution;
superposing the second original image to a region of interest corresponding to the target object in the first blurred image to obtain a superposed image;
circularly encoding the superposed image according to a preset image format by adopting the estimated quality factor until the size of the encoded file reaches a preset size, and obtaining a target image; the estimated quality factor is determined according to the preset size and corresponds to the first resolution.
2. The method of claim 1, wherein superimposing the second original image onto the region of interest corresponding to the target object in the first blurred image, obtaining a superimposed image, comprises:
determining a corresponding position region in the first blurred image based on the relative position of the region of interest in the first original image;
and covering the image of the part corresponding to the position area in the first blurred image by using the second original image to obtain the superposed image.
3. The method of claim 1, wherein prior to cyclically encoding the overlay image in the predetermined picture format using the estimated quality factor, further comprising:
acquiring the estimated quality factor corresponding to the preset size from a quality factor average value table; the quality factor average value table is a corresponding relation between different sizes of files with the preset picture format and used quality factors generated aiming at the same original image, the shooting resolution ratio of the image acquisition equipment adopted by the same original image and the first original image is the same, and the sizes of the shot original images are the same.
4. The method of claim 2, wherein after obtaining the overlay image and before performing the loop coding on the overlay image using the estimated quality factor, further comprising:
setting a transition region between the edge of the region of interest and the edge of the first original image; wherein an inner edge of the transition region is an edge of the region of interest, and an outer edge of the transition region is located between the edge of the region of interest and the edge of the first original image;
dividing the transition region into a plurality of transition sub-regions; wherein the distance between the inner and outer edges of each transition subregion is the same;
blurring each transition sub-region to obtain a blurred transition region, and overlaying the blurred transition region to the periphery of a position region in the superimposed image to obtain a new superimposed image; in the direction from the transition subarea along the outermost edge to the transition subarea along the innermost edge, the blurring factor used in the blurring processing of the transition subareas is increased by a set step.
5. The method of claim 4, wherein blurring each transition sub-region comprises:
if the figure formed by the edges of the interested region is a polygon, dividing the transition sub-region into a plurality of transition micro-regions; the two adjacent transition micro regions are complementary, a boundary between the two adjacent micro regions is formed by extending an inner edge line of one micro region to an outer edge line of the other micro region, and the inner edge line and the outer edge line are straight lines;
and respectively carrying out the same fuzzification processing on each transition micro-region.
6. An apparatus for performing a differentiation process on an image, comprising:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring an interesting region corresponding to a target object from a first original image by adopting a preset model and acquiring a second original image corresponding to the interesting region from the first original image; the first original image is an image formed by YUV original data acquired by image acquisition equipment, the preset model is a model obtained after identification training of the target object, and the first original image and the second original image have first resolution;
the blurring processing unit is used for performing blurring processing on the first original image to obtain a first blurred image; wherein the first blurred image is the same size as the first original image, the first blurred image has a second resolution, and the second resolution is less than the first resolution;
the superposition unit is used for superposing the second original image to an interested area corresponding to the target object in the first blurred image to obtain a superposed image;
the encoding unit is used for circularly encoding the superposed image according to a preset image format by adopting the estimated quality factor until the size of the encoded file reaches a preset size, so as to obtain a target image; the estimated quality factor is determined according to the preset size and corresponds to the first resolution.
7. The apparatus of claim 6, wherein the superimposing unit is specifically configured to:
determining a corresponding position region in the first blurred image based on the relative position of the region of interest in the first original image;
and covering the image of the part corresponding to the position area in the first blurred image by using the second original image to obtain the superposed image.
8. The apparatus of claim 7, wherein the obfuscation unit is further to:
setting a transition region between the edge of the region of interest and the edge of the first original image; wherein an inner edge of the transition region is an edge of the region of interest, and an outer edge of the transition region is located between the edge of the region of interest and the edge of the first original image;
dividing the transition region into a plurality of transition sub-regions; wherein the distance between the inner and outer edges of each transition subregion is the same;
blurring each transition sub-region to obtain a blurred transition region, and overlaying the blurred transition region to the periphery of a position region in the superimposed image to obtain a new superimposed image; in the direction from the transition subarea along the outermost edge to the transition subarea along the innermost edge, the blurring factor used in the blurring processing of the transition subareas is increased by a set step.
9. An apparatus for performing a differentiation process on an image, comprising:
at least one processor, and
a memory coupled to the at least one processor;
wherein the memory stores instructions executable by the at least one processor, the at least one processor implementing the method of any one of claims 1-5 by executing the instructions stored by the memory.
10. A readable storage medium, comprising a memory,
the memory is for storing instructions that, when executed by the processor, cause an apparatus comprising the readable storage medium to perform the method of any of claims 1-5.
CN202010217463.9A 2020-03-25 2020-03-25 Method, device and storage medium for performing differentiation processing on image Pending CN111445408A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010217463.9A CN111445408A (en) 2020-03-25 2020-03-25 Method, device and storage medium for performing differentiation processing on image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010217463.9A CN111445408A (en) 2020-03-25 2020-03-25 Method, device and storage medium for performing differentiation processing on image

Publications (1)

Publication Number Publication Date
CN111445408A true CN111445408A (en) 2020-07-24

Family

ID=71652448

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010217463.9A Pending CN111445408A (en) 2020-03-25 2020-03-25 Method, device and storage medium for performing differentiation processing on image

Country Status (1)

Country Link
CN (1) CN111445408A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111818311A (en) * 2020-08-25 2020-10-23 北京中联合超高清协同技术中心有限公司 Ultra-high-definition video monitor with variable vision field
CN112102320A (en) * 2020-11-18 2020-12-18 粒子文化科技集团(杭州)股份有限公司 Image compression method, image compression device, electronic device, and storage medium

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20010024528A1 (en) * 1997-06-04 2001-09-27 Nikon Corporation Image compression apparatus, method and recording medium storing an image compression program
CN105023252A (en) * 2015-07-14 2015-11-04 厦门美图网科技有限公司 Method and system for enhancement processing of beautified image and shooting terminal
CN108848367A (en) * 2018-07-26 2018-11-20 宁波视睿迪光电有限公司 A kind of method, device and mobile terminal of image procossing
CN109559285A (en) * 2018-10-26 2019-04-02 北京东软医疗设备有限公司 A kind of image enhancement display methods and relevant apparatus
CN109995962A (en) * 2017-12-29 2019-07-09 浙江宇视科技有限公司 The control method and system of coding of graphics upper dimension bound
CN110390657A (en) * 2018-04-20 2019-10-29 北京中科晶上超媒体信息技术有限公司 A kind of image interfusion method
CN110536138A (en) * 2018-05-25 2019-12-03 杭州海康威视数字技术股份有限公司 A kind of lossy compression coding method, device and system grade chip
CN110572579A (en) * 2019-09-30 2019-12-13 联想(北京)有限公司 image processing method and device and electronic equipment
CN110889352A (en) * 2019-11-18 2020-03-17 珠海全志科技股份有限公司 Image blurring processing method, computer device, and computer-readable storage medium

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20010024528A1 (en) * 1997-06-04 2001-09-27 Nikon Corporation Image compression apparatus, method and recording medium storing an image compression program
CN105023252A (en) * 2015-07-14 2015-11-04 厦门美图网科技有限公司 Method and system for enhancement processing of beautified image and shooting terminal
CN109995962A (en) * 2017-12-29 2019-07-09 浙江宇视科技有限公司 The control method and system of coding of graphics upper dimension bound
CN110390657A (en) * 2018-04-20 2019-10-29 北京中科晶上超媒体信息技术有限公司 A kind of image interfusion method
CN110536138A (en) * 2018-05-25 2019-12-03 杭州海康威视数字技术股份有限公司 A kind of lossy compression coding method, device and system grade chip
CN108848367A (en) * 2018-07-26 2018-11-20 宁波视睿迪光电有限公司 A kind of method, device and mobile terminal of image procossing
CN109559285A (en) * 2018-10-26 2019-04-02 北京东软医疗设备有限公司 A kind of image enhancement display methods and relevant apparatus
CN110572579A (en) * 2019-09-30 2019-12-13 联想(北京)有限公司 image processing method and device and electronic equipment
CN110889352A (en) * 2019-11-18 2020-03-17 珠海全志科技股份有限公司 Image blurring processing method, computer device, and computer-readable storage medium

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111818311A (en) * 2020-08-25 2020-10-23 北京中联合超高清协同技术中心有限公司 Ultra-high-definition video monitor with variable vision field
CN111818311B (en) * 2020-08-25 2021-06-04 北京中联合超高清协同技术中心有限公司 Ultra-high-definition video monitor with variable vision field
CN112102320A (en) * 2020-11-18 2020-12-18 粒子文化科技集团(杭州)股份有限公司 Image compression method, image compression device, electronic device, and storage medium

Similar Documents

Publication Publication Date Title
CN109871760B (en) Face positioning method and device, terminal equipment and storage medium
CN112218107B (en) Live broadcast rendering method and device, electronic equipment and storage medium
US20110058049A1 (en) Real-Time Image Stabilization
CN111445408A (en) Method, device and storage medium for performing differentiation processing on image
US6959117B2 (en) Method and apparatus for deblurring and re-blurring image segments
CN112543327B (en) Method, encoder system and storage medium for providing privacy masking for video streams
JP7010122B2 (en) Video generator, video generation method, and program
US20200151888A1 (en) Intelligent method for viewing surveillance videos with improved efficiency
EP4116462A3 (en) Method and apparatus of processing image, electronic device, storage medium and program product
CN107977989B (en) Method and equipment for determining image offset
CN109934793B (en) Real-time image fusion method based on integer DCT
IL180154A (en) Real-time stabilization
EP3568832A1 (en) Adding motion effects to digital still images
BR102020027013A2 (en) METHOD TO GENERATE AN ADAPTIVE MULTIPLANE IMAGE FROM A SINGLE HIGH RESOLUTION IMAGE
CN112330541A (en) Live video processing method and device, electronic equipment and storage medium
CN109688413A (en) To support the method and encoder of the video code model encoded video stream of ancillary frame
JP2006509292A (en) Apparatus and method for removing false edges from a segmented image
CN111447444A (en) Image processing method and device
Katramados et al. Real-time visual saliency by division of gaussians
US8866970B1 (en) Method for real-time processing of a video sequence on mobile terminals
US11044399B2 (en) Video surveillance system
CN111932594A (en) Billion pixel video alignment method and device based on optical flow and medium
CN113111770B (en) Video processing method, device, terminal and storage medium
JP6615818B2 (en) VIDEO GENERATION DEVICE, VIDEO GENERATION METHOD, AND PROGRAM
JP2020194532A (en) Method, system, and device for detecting object in distorted image

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20200724