CN114567797A - Image processing method and device and vehicle - Google Patents
Image processing method and device and vehicle Download PDFInfo
- Publication number
- CN114567797A CN114567797A CN202110310965.0A CN202110310965A CN114567797A CN 114567797 A CN114567797 A CN 114567797A CN 202110310965 A CN202110310965 A CN 202110310965A CN 114567797 A CN114567797 A CN 114567797A
- Authority
- CN
- China
- Prior art keywords
- image
- video
- frame
- area
- target object
- 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
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 32
- 238000000034 method Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims description 36
- 238000001514 detection method Methods 0.000 claims description 20
- 238000009499 grossing Methods 0.000 claims description 20
- 238000012549 training Methods 0.000 claims description 12
- 238000012790 confirmation Methods 0.000 claims description 8
- 230000000694 effects Effects 0.000 abstract description 36
- 230000033001 locomotion Effects 0.000 description 9
- 238000010200 validation analysis Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/234—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs
- H04N21/23424—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs involving splicing one content stream with another content stream, e.g. for inserting or substituting an advertisement
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs
- H04N21/44016—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving splicing one content stream with another content stream, e.g. for substituting a video clip
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
- H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
- H04N21/835—Generation of protective data, e.g. certificates
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Business, Economics & Management (AREA)
- Marketing (AREA)
- Computer Security & Cryptography (AREA)
- Image Processing (AREA)
Abstract
The application provides an image processing method and device, and belongs to the technical field of automobiles. The method comprises the following steps: obtaining an original video stream to be processed; determining an image area where a target object in each frame of video image in the original video stream is located; determining boundary points of the image area; replacing the image area with a camouflage image corresponding to the target object according to the boundary point of the image area; and obtaining a target video stream obtained by replacing the image area in each frame of video image with the camouflage image. By using the image processing method and device provided by the application, the privacy-related image in the original image is replaced by the camouflage image, and the problems that the privacy of others is invaded by the content shared by the video, the content is blurred, and the watching effect and the attractiveness of the video are influenced by the mosaic-processed privacy content are solved.
Description
Technical Field
The embodiment of the application relates to the technical field of automobiles, in particular to an image processing method and device and a vehicle.
Background
With the explosive growth of video sharing applications, the number of cases that the shared videos contain other people's private contents is increased gradually, and if the videos are directly shared, the privacy of other people is violated.
The existing image processing technology is to fuzzify the content of privacy or make mosaic for shielding, but the processed video can shift the attention focus of a viewer to the processed privacy content, and the viewing effect and the attractiveness of the video content are affected, so that how to enable the content shared by the video not to infringe the privacy of other people and not affect the viewing effect and the attractiveness of the video is a technical problem to be solved urgently.
Disclosure of Invention
The embodiment of the application aims to provide an image processing method, and aims to solve the problems that the content shared by a video violates the privacy of others and is fuzzy, and the mosaic processed privacy content influences the viewing effect and the attractiveness of the video.
A first aspect of an embodiment of the present application provides an image processing method, including:
obtaining an original video stream to be processed;
determining an image area where a target object in each frame of video image in an original video stream is located;
determining boundary points of an image area;
replacing the image area with a camouflage image corresponding to the target object according to the boundary point of the image area;
and obtaining a target video stream obtained by replacing the image area in each frame of video image with a camouflage image.
Optionally, replacing the image area with a camouflage image corresponding to the target object according to the boundary point of the image area, including:
determining the area boundary of the image area according to the boundary point of the image area;
a camouflaged image corresponding to the target object is overlaid to the image area of the frame of video image based on the area boundary of the image area of the frame of video image.
Optionally, determining a region boundary of an image region of the frame of video image includes:
performing correlation matching on the positions of four endpoints of an image area of the frame of video image to obtain a moving track coordinate curve of each endpoint in the image area;
smoothing the moving track coordinate curve of each endpoint to obtain the positions of four endpoints of the image area of the frame of video image after smoothing;
and determining the area boundary of the image area of the frame video image according to the positions of the four end points of the image area of the frame video image after the smoothing processing.
Optionally, the disguised image corresponding to the target object is generated by:
obtaining a template image;
and adjusting the color and the style of the template image according to the color and the style of each frame of video image in the original video stream to obtain a camouflage image corresponding to the target object in each frame of video image.
Optionally, obtaining a template image comprises:
randomly generating a plurality of first character contents, and arranging the first character contents according to a preset rule;
and generating a template image according to the arranged first character contents.
Optionally, obtaining a template image comprises:
acquiring a plurality of second character contents input by a user, and arranging the plurality of second character contents according to a preset rule;
and generating a template image according to the arranged second character contents.
Optionally, determining an image area where a target object in each frame of video image in the original video stream is located includes:
inputting each frame of video image in the original video stream into an object detection model to obtain an image area where a target object in each frame of video image output by the object detection model is located;
the object detection model is obtained by training a preset model by taking a plurality of sample images as training samples, wherein each sample image in the plurality of sample images comprises an image area of a target object.
A second aspect of the present application provides an image processing apparatus comprising:
a first obtaining module: for obtaining an original video stream to be processed;
a first confirmation module: the method comprises the steps of determining an image area where a target object in each frame of video image in an original video stream is located;
a second confirmation module: for determining boundary points of the image area;
a first control module: the device comprises a camera, a camera module, a display module and a control module, wherein the camera is used for acquiring a target object;
a second obtaining module: and the target video stream is obtained by replacing the image area in each frame of video image with the camouflage image.
Optionally, the first control module comprises:
a first confirmation unit: the boundary of the image area of the frame video image is determined according to the boundary point of the image area;
a first control unit: and based on the area boundary of the image area of the frame video image, the method is used for covering the disguised image corresponding to the target object to the image area of the frame video image.
A third aspect of the present application provides an automobile provided with an image processing apparatus as provided in the second aspect of the present application, the image processing apparatus being configured to perform the image processing method as provided in the first aspect of the present application.
Has the advantages that:
the image processing method comprises the steps of firstly, accurately positioning each frame of video image in an original video stream to obtain the position, shape and size of a target object image, and then replacing an image area in each frame of video image in the original video stream by a camouflage image with the same shape and size as the target object image at the same position of the target object image to obtain a target video stream; therefore, on one hand, the individual privacy is protected, on the other hand, the camouflage effect is natural and is not easy to be perceived by a viewer, the focus of the viewer on the video content is kept, the technical effects that the privacy of other people is not invaded and the ornamental effect and the attractiveness of the video content are not influenced are achieved, and the problems that the privacy of other people is invaded and the content shared by the video is fuzzy, and the ornamental effect and the attractiveness of the video are influenced by mosaic processing the privacy content are solved.
Secondly, the positions of four endpoints of an image area of each frame of video image in each frame of original video stream are associated and matched, so that a moving track coordinate curve of each endpoint in the whole original video stream is obtained, then the moving track coordinate curve of each endpoint is smoothed, the positions of the four endpoints of the image area of the frame of video image after smoothing are obtained, a camouflage image corresponding to a target object covers the image area of the frame of video image according to a determined boundary, the positions of the endpoints of the target object in the video are smoothed, the problem that due to the positioning precision, the positioning of the same endpoint in each frame of image has slight deviation, human vision is very sensitive to change and jitter, the position change caused by the deviation can influence the ornamental performance of the video, and the aim of eliminating the position jitter of the target object is achieved, the camouflage effect is more natural.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required to be used in the description of the embodiments of the present application will be briefly described below, and it is obvious that the drawings in the description below are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings may be obtained according to these drawings without inventive labor.
Fig. 1 is a flowchart illustrating steps of an image processing method according to an embodiment of the present application;
fig. 2 is a flowchart illustrating sub-steps of step S104 of an image processing method according to an embodiment of the present application;
fig. 3 is a flowchart illustrating sub-steps of step S1041 of an image processing method according to an embodiment of the application;
fig. 4 is a block diagram of an image processing apparatus according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some, but not all, embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The existing image processing technology is to blur or mosaic the private content, but the processed video will focus the attention of the viewer to the processed private content, which affects the viewing effect and the aesthetic property of the video content.
In view of this, the image processing method provided by the present application, first, accurately positioning each frame of video image in an original video stream, where the frame of video image relates to the privacy of others, to obtain the position, shape and size of a target object image, and then replacing an image area in each frame of video image in the original video stream with a camouflage image having the same shape and size as the target object image at the same position of the target object image to obtain a target video stream; therefore, on one hand, the individual privacy is protected, on the other hand, the camouflage effect is natural and is not easy to be perceived by a viewer, the focus of the viewer on the video content is kept, the technical effects that the privacy of other people is not invaded and the ornamental effect and the attractiveness of the video content are not influenced are achieved, and the problems that the privacy of other people is invaded and the content shared by the video is fuzzy, and the ornamental effect and the attractiveness of the video are influenced by mosaic processing the privacy content are solved.
Example one
Referring to fig. 1, a flowchart illustrating steps of an image processing method according to the present application is shown, where as shown in fig. 1, the image processing method includes:
step S101: an original video stream to be processed is obtained.
In this embodiment, the original video stream may be a video shot by a mobile phone or a video shot by a vehicle-mounted camera, which is not limited herein, and the original video stream may include contents of privacy of others, for example: a license plate number; if the content containing the privacy of others is directly subjected to video sharing without being processed, the privacy of others is infringed, and therefore the original video stream needs to be processed in advance.
Step S102: and determining an image area where the target object in each frame of video image in the original video stream is located.
In this embodiment, the original video stream is formed by combining each frame of video image, and the target object in each frame of video image is processed, so as to achieve the purpose of processing the target object in the original video stream; the target object can be a license plate, a name, a face and other contents containing privacy of others, and the image area where the target object is located is accurately positioned to obtain an image of the target object; for example: the original video comprises vehicles of other people, the part of the vehicle related to privacy is a license plate, the target object is the license plate of the vehicle, and the image area where the license plate is located is positioned in each frame of video image so as to obtain the image of the license plate.
Step S103: the boundary points of the image area are determined.
In this embodiment, the object containing the privacy content in the frame of video image is first located, the privacy content is then located in the located object area, and then four endpoints of the located privacy content are located to obtain the boundary points of the image area of the frame of video image, so as to achieve the purpose of accurately locating the privacy content.
Step S104: and replacing the image area with a camouflage image corresponding to the target object according to the boundary point of the image area.
In the embodiment, the area range to be replaced is accurately positioned according to the boundary point of the image area, and then the original image related to the privacy of other people in the image area is replaced by the camouflage image corresponding to the target object, so that the problem that the original image infringes the privacy of other people is solved.
Step S105: and obtaining a target video stream obtained by replacing the image area in each frame of video image with a camouflage image.
In this embodiment, an image area in each frame of video image in the original video stream, that is, an image related to the privacy of another person in each frame of video image in the original video stream is replaced by a disguised image with a corresponding size and shape to obtain a replaced target video stream, and thus the obtained target video stream does not affect the viewing effect and the aesthetic property of the video content, and does not infringe the privacy of another person.
The image processing method includes the steps that firstly, each frame of video image in an original video stream is accurately positioned according to the image related to privacy of others, the position, the shape and the size of a target object image are obtained, then a camouflage image with the same shape and the same size as the target object image is used for replacing an image area in each frame of video image in the original video stream at the same position of the target object image, and the target video stream is obtained; therefore, on one hand, the individual privacy is protected, on the other hand, the camouflage effect is natural and is not easy to be perceived by a viewer, the focus of the viewer on the video content is kept, the technical effects that the privacy of other people is not invaded and the ornamental effect and the attractiveness of the video content are not influenced are achieved, and the problems that the privacy of other people is invaded and the content shared by the video is fuzzy, and the ornamental effect and the attractiveness of the video are influenced by mosaic processing the privacy content are solved.
Based on the image processing method, the following specific examples are provided, and the examples can be arbitrarily combined to form another image processing method without mutual conflict.
Referring to fig. 2, a flowchart illustrating a sub-step of step S104 of an image processing method according to the present application, as shown in fig. 2, for replacing an image region with a disguised image corresponding to a target object according to a boundary point of the image region, includes:
step S1041: and determining the area boundary of the image area according to the boundary point of the image area.
In this embodiment, on the image area where the determined target object is located, the boundary of the image area is determined by determining the boundary point of the image area, so as to achieve the purpose of accurate positioning.
Step S1042: a camouflaged image corresponding to the target object is overlaid on an image area of the frame of video image based on an area boundary of the image area of the frame of video image.
And accurately positioning the target object based on the area boundary of the image area of the frame of video image, and then covering a camouflage image corresponding to the target object on the image area of the frame of video image to replace the content of the privacy part contained in the frame of video image.
By accurately positioning the target object and then covering the target object by using the corresponding camouflage image based on the boundary of the positioned image area, the aim of accurate replacement is fulfilled, and the problem that the coverage is incomplete due to inaccurate positioning or exceeds the original image area, so that other contents in the frame of video image are covered to influence the ornamental performance of the video is avoided.
In a possible implementation manner, referring to fig. 3, a flowchart illustrating a sub-step of step S1041 of an image processing method of the present application is shown, as shown in fig. 3, where determining a region boundary of an image region of the frame of video image includes:
step S10411: and performing correlation matching on the positions of four endpoints of the image area of each frame of video image to obtain a movement track coordinate curve of each endpoint in the image area.
And performing correlation matching on the positions of four end points of the image area of each frame of video image in the original video stream, thereby obtaining a moving track coordinate curve of each end point in the whole original video stream.
Step S10412: and smoothing the moving track coordinate curve of each endpoint to obtain the positions of four endpoints of the image area of the frame of video image after smoothing.
In a video, the moving coordinate curve of the same point in the video should be a smooth curve, but due to the problem of positioning accuracy, the positioning of the same endpoint in each frame of image has slight deviation, human vision is very sensitive to change and jitter, and the position change caused by the deviation can affect the ornamental value of the video.
Step S10413: and determining the area boundary of the image area of the frame video image according to the positions of the four end points of the image area of the frame video image after the smoothing processing.
And determining the boundaries of the image areas of the frame video images according to the positions of the four end points of the image areas of the frame video images after the smoothing processing, wherein the camouflage image corresponding to the target object covers the image areas of the frame video images according to the determined area boundaries.
The positions of four endpoints of an image area of each frame of video image in each frame of original video stream are associated and matched, so that a moving track coordinate curve of each endpoint in the whole original video stream is obtained, then the moving track coordinate curve of each endpoint is subjected to smoothing processing, the positions of the four endpoints of the image area of the frame of video image after the smoothing processing are obtained, a camouflage image corresponding to a target object covers the image area of the frame of video image according to a determined boundary, the positions of the endpoints of the target object in the video are subjected to smoothing processing, the problem that due to the positioning precision, the positioning of the same endpoint in each frame of image has slight deviation, human vision is very sensitive to change and jitter, the problem that the position change caused by the deviation affects the ornamental performance of the video is solved, and the aim of eliminating the position jitter of the target object is achieved, the camouflage effect is more natural.
In one possible embodiment, the target object corresponds to a camouflage image generated by:
step S201: a template image is obtained.
The template image is an image of the object, which may be provided by a database or stored by a computer, and is not limited herein.
Step S202: and adjusting the color and the style of the template image according to the color and the style of each frame of video image in the original video stream to obtain a camouflage image corresponding to the target object in each frame of video image.
The method has the advantages that the color and the style of the template image are adjusted according to the color and the style of each frame of video image in the original video stream, the camouflage image which is consistent with the color and the style of the target object in each frame of video image is obtained, the problem that the camouflage image can cause human visual attention due to the fact that the color and the style of the camouflage image are inconsistent with the color and the style of each frame of video image in the original video stream is solved, and a better camouflage effect is achieved.
In this embodiment, obtaining a template image includes:
step S20111: and randomly generating a plurality of first character contents, and arranging the first character contents according to a preset rule.
The contents related to privacy in the image all contain specific character contents, and the character contents contained in different objects all have their own arrangement rules, for example: the license plate number, the license plate structure and the character arrangement all have corresponding rules specified by the corresponding national standard, if the target object is the license plate, the license plate number for disguising the license plate is automatically generated, and the characters in the license plate number are arranged according to the rules in the national standard for arranging the characters of the license plate number.
Step S20112: and generating a template image according to the arranged first character contents.
And generating a corresponding template image according to the plurality of first character contents arranged according to a preset rule.
According to the difference of the target objects, the corresponding rules for arranging the character contents of the target objects are preset, and the template images obtained after arrangement according to the preset rules can highly restore the target images of the objects, so that a better camouflage effect is achieved.
In one possible embodiment, obtaining a template image includes:
s20121: acquiring a plurality of second character contents input by a user, and arranging the plurality of second character contents according to a preset rule;
the user can input the character content of the target object according to the preference of the user, the input character content is arranged according to the preset rule, the user inputs a plurality of second character contents by himself, the interestingness of the video is enhanced, the participation of the user can be improved, meanwhile, the character content is arranged according to the preset rule, the content cannot be obtrusive, and the ornamental performance of the video cannot be influenced.
S20122: and generating a template image according to the arranged second character contents.
And generating a template image according to the arranged second character contents, wherein the template image can highly restore the target image of the object, thereby achieving a better camouflage effect.
In one possible embodiment, determining an image area where a target object in each frame of video image in an original video stream is located includes:
s1021: inputting each frame of video image in the original video stream into an object detection model to obtain an image area where a target object in each frame of video image output by the object detection model is located;
the object detection model is obtained by training a preset model by taking a plurality of sample images as training samples, wherein each sample image in the plurality of sample images comprises an image area of a target object.
The object detection model takes a plurality of sample images as training samples, and is obtained by training a preset model, so that the detection accuracy of the object detection model is improved, each frame of video image of an original video stream is input into the object detection model, and therefore the image area where the target object in each frame of video image is located.
Example two
Based on the same inventive concept, another embodiment of the present application provides an image processing apparatus for performing the image processing method as provided in the first embodiment of the present application; referring to fig. 4, a block diagram of the image processing apparatus according to the present invention is shown, and as shown in fig. 4, the image processing apparatus includes:
the first acquisition module 11: for obtaining an original video stream to be processed;
the first validation module 12: the method comprises the steps of determining an image area where a target object in each frame of video image in an original video stream is located;
the second confirmation module 13: for determining boundary points of the image area;
the first control module 14: the image area is replaced by a camouflage image corresponding to the target object according to the boundary point of the image area;
the second acquisition module 15: and the target video stream is obtained by replacing the image area in each frame of video image with the camouflage image.
The image processing method includes the steps that firstly, each frame of video image in an original video stream is accurately positioned according to the image related to privacy of others, the position, the shape and the size of a target object image are obtained, then a camouflage image with the same shape and the same size as the target object image is used for replacing an image area in each frame of video image in the original video stream at the same position of the target object image, and the target video stream is obtained; therefore, on one hand, the individual privacy is protected, on the other hand, the camouflage effect is natural and is not easy to be perceived by a viewer, the focus of the viewer on the video content is kept, the technical effects that the privacy of other people is not invaded and the ornamental effect and the attractiveness of the video content are not influenced are achieved, and the problems that the privacy of other people is invaded and the content shared by the video is fuzzy, and the ornamental effect and the attractiveness of the video are influenced by mosaic processing the privacy content are solved.
In one possible embodiment, the first control module 14 includes:
first confirmation unit 141: the image area boundary determining unit is used for determining the area boundary of the image area of the frame of video image according to the boundary point of the image area;
the first control unit 142: and based on the area boundary of the image area of the frame video image, the method is used for covering the disguised image corresponding to the target object to the image area of the frame video image.
The target object is accurately positioned through the first confirming unit 141, and then the first control unit 142 covers the corresponding disguised image based on the boundary of the positioned image area, so that the purpose of accurate replacement is achieved, and the problem that the coverage is incomplete due to inaccurate positioning or exceeds the original image area, and other contents in the frame of video image are covered to influence the enjoyment of the video is avoided.
In a possible embodiment, the first validation unit 141 includes:
the first acquisition sub-unit 1411: the system comprises a frame of video image, a motion track coordinate curve acquisition module and a motion track coordinate curve acquisition module, wherein the motion track coordinate curve acquisition module is used for carrying out correlation matching on the positions of four endpoints of an image area of the frame of video image to obtain a motion track coordinate curve of each endpoint in the image area;
the second acquisition subunit 1412: the system comprises a frame video image acquisition unit, a motion track coordinate curve generation unit and a motion track coordinate curve generation unit, wherein the motion track coordinate curve generation unit is used for generating a motion track coordinate curve of each endpoint;
the first control subunit 1413: and the boundary of the image area of the frame video image is determined according to the positions of the four end points of the image area of the frame video image after the smoothing processing.
The positions of four endpoints of an image area of each frame of video image in each frame of original video stream are associated and matched through a first acquiring subunit 1411, so that a moving track coordinate curve of each endpoint in the whole original video stream is acquired, then a second acquiring subunit 1412 performs smoothing processing on the moving track coordinate curve of each endpoint to acquire the positions of the four endpoints of the image area of the frame of video image after smoothing processing, and a first control subunit 1413 covers a camouflage image corresponding to a target object on the image area of the frame of video image according to a determined boundary; the method has the advantages that the positions of the end points of the target object in the video are smoothed, the problem that due to the problem of positioning accuracy, the same end point has slight deviation in positioning of each frame of image, human vision is very sensitive to change and shake, and the position change caused by the deviation can influence the ornamental performance of the video is solved, and the technical effects of eliminating the shake of the position of the target object and enabling the camouflage effect to be more natural are achieved.
In one possible embodiment, the target object corresponds to a camouflage image generated by means comprising:
the third obtaining module 21: obtaining a template image;
the second control module 22: and adjusting the color and the style of the template image according to the color and the style of each frame of video image in the original video stream to obtain a camouflage image corresponding to the target object in each frame of video image.
The third obtaining module 21 obtains the template image, and the second control module 22 adjusts the color and style of the template image according to the color and style of each frame of video image in the original video stream to obtain a camouflage image which is consistent with the color and style of the target object in each frame of video image, so that the problem that the camouflage image can cause human visual attention due to the fact that the color and style of the camouflage image are inconsistent with the color and style of each frame of video image in the original video stream is solved, and a better camouflage effect is achieved.
In a possible implementation, the third obtaining module 21 includes:
the second control subunit 2111: randomly generating a plurality of first character contents, and arranging the plurality of first character contents according to a preset rule;
the third control subunit 2112: and generating a template image according to the arranged first character contents.
According to the difference of the target objects, the corresponding rules for arranging the character contents of the target objects are preset, and the template images obtained after arrangement according to the preset rules can highly restore the target images of the objects, so that a better camouflage effect is achieved.
In a possible implementation, the third obtaining module 21 includes:
the fourth control subunit 2121: acquiring a plurality of second character contents input by a user, and arranging the plurality of second character contents according to a preset rule;
the fifth control subunit 2122: and generating a template image according to the arranged second character contents.
The user can input the character content of the target object according to the preference of the user, the input character content is arranged according to the preset rule, the user inputs a plurality of second character contents by himself, the interestingness of the video is enhanced, the participation of the user can be improved, meanwhile, the character content is arranged according to the preset rule, the content cannot be obtrusive, and the ornamental performance of the video cannot be influenced.
In a possible embodiment, the first validation module 12 comprises:
the second acquisition unit 121: inputting each frame of video image in the original video stream into an object detection model to obtain an image area where a target object in each frame of video image output by the object detection model is located;
the object detection model is obtained by training a preset model by taking a plurality of sample images as training samples, wherein each sample image in the plurality of sample images comprises an image area of the target object.
The object detection model takes a plurality of sample images as training samples, and is obtained by training a preset model, so that the detection accuracy of the object detection model is improved, each frame of video image of an original video stream is input into the object detection model, and therefore the image area where the target object in each frame of video image is located.
For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
EXAMPLE III
The embodiment of the present application further provides an automobile, which may include an image processing apparatus as provided in the second embodiment of the present application, where the image processing apparatus is configured to execute the image processing method as provided in the first embodiment of the present application.
According to the automobile provided by the application, firstly, each frame of video image in an original video stream is accurately positioned to obtain the position, the shape and the size of a target object image, and then a camouflage image with the same shape and the same size as the target object image is used for replacing an image area in each frame of video image in the original video stream at the same position of the target object image to obtain a target video stream; therefore, on one hand, the individual privacy is protected, on the other hand, the camouflage effect is natural and is not easy to be perceived by a viewer, the focus of the viewer on the video content is kept, the technical effects that the privacy of other people is not invaded and the ornamental effect and the attractiveness of the video content are not influenced are achieved, and the problems that the privacy of other people is invaded and the content shared by the video is fuzzy, and the ornamental effect and the attractiveness of the video are influenced by mosaic processing the privacy content are solved.
The positions of four endpoints of an image area of each frame of video image in each frame of original video stream are associated and matched, so that a moving track coordinate curve of each endpoint in the whole original video stream is obtained, then the moving track coordinate curve of each endpoint is subjected to smoothing processing, the positions of the four endpoints of the image area of the frame of video image after the smoothing processing are obtained, a camouflage image corresponding to a target object covers the image area of the frame of video image according to a determined boundary, the positions of the endpoints of the target object in the video are subjected to smoothing processing, the problem that due to the positioning precision, the positioning of the same endpoint in each frame of image has slight deviation, human vision is very sensitive to change and jitter, the problem that the position change caused by the deviation affects the ornamental performance of the video is solved, and the aim of eliminating the position jitter of the target object is achieved, the camouflage effect is more natural.
It should be understood that while the present specification has described preferred embodiments of the present application, additional variations and modifications of those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the true scope of the embodiments of the application.
The image processing method, the image processing device and the automobile provided by the application are introduced in detail, specific examples are applied in the description to explain the principle and the implementation of the application, and the description of the embodiments is only used for helping to understand the method and the core idea of the application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Claims (10)
1. An image processing method, characterized in that the method comprises:
obtaining an original video stream to be processed;
determining an image area where a target object in each frame of video image in the original video stream is located;
determining boundary points of the image area;
replacing the image area with a camouflage image corresponding to the target object according to the boundary point of the image area;
and obtaining a target video stream obtained by replacing the image area in each frame of video image with the camouflage image.
2. The image processing method according to claim 1, wherein replacing the image area with a camouflaged image corresponding to the target object based on the boundary point of the image area comprises:
determining the region boundary of the image region according to the boundary point of the image region;
and overlaying a camouflage image corresponding to the target object to the image area of the frame of video image based on the area boundary of the image area.
3. The image processing method of claim 2, wherein determining the region boundary of the image region comprises:
performing correlation matching on the positions of four end points of the image area of the frame of video image to obtain a moving track coordinate curve of each end point in the image area;
smoothing the moving track coordinate curve of each endpoint to obtain the positions of four endpoints of the image area of the frame of video image after smoothing;
and determining the area boundary of the image area of the frame of video image according to the positions of the four end points of the image area of the frame of video image after the smoothing processing.
4. The image processing method of claim 1, wherein the target object's corresponding camouflage image is generated by:
obtaining a template image;
and adjusting the color and the style of the template image according to the color and the style of each frame of video image in the original video stream to obtain a camouflage image corresponding to the target object in each frame of video image.
5. The image processing method of claim 4, wherein obtaining the template image comprises:
randomly generating a plurality of first character contents, and arranging the first character contents according to a preset rule;
and generating the template image according to the arranged first character contents.
6. The image processing method of claim 4, wherein obtaining the template image comprises:
acquiring a plurality of second character contents input by a user, and arranging the plurality of second character contents according to a preset rule;
and generating the template image according to the arranged second character contents.
7. The image processing method of any one of claims 1 to 6, wherein determining an image area in which a target object in each frame of video image in the original video stream is located comprises:
inputting each frame of video image in the original video stream into an object detection model to obtain an image area where a target object in each frame of video image output by the object detection model is located;
the object detection model is obtained by training a preset model by taking a plurality of sample images as training samples, wherein each sample image in the plurality of sample images comprises an image area of the target object.
8. An image processing apparatus, characterized in that the method comprises:
a first obtaining module: for obtaining an original video stream to be processed;
a first confirmation module: the method comprises the steps of determining an image area where a target object in each frame of video image in the original video stream is located;
a second confirmation module: for determining boundary points of the image region;
a first control module: the image area is replaced by a camouflage image corresponding to the target object according to the boundary point of the image area;
a second obtaining module: and the target video stream is obtained by replacing the image area in each frame of video image with the camouflage image.
9. The image processing apparatus according to claim 8, wherein the first control module comprises:
a first confirmation unit: the image processing device is used for determining the area boundary of the image area according to the boundary point of the image area;
a first control unit: and based on the area boundary of the image area of the frame of video image, covering a camouflage image corresponding to the target object to the image area of the frame of video image.
10. An automobile, characterized in that it is provided with an image processing apparatus according to any one of claims 8-9, said image processing apparatus being adapted to perform the image processing method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110310965.0A CN114567797A (en) | 2021-03-23 | 2021-03-23 | Image processing method and device and vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110310965.0A CN114567797A (en) | 2021-03-23 | 2021-03-23 | Image processing method and device and vehicle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114567797A true CN114567797A (en) | 2022-05-31 |
Family
ID=81712538
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110310965.0A Pending CN114567797A (en) | 2021-03-23 | 2021-03-23 | Image processing method and device and vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114567797A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117061581A (en) * | 2023-10-08 | 2023-11-14 | 腾讯科技（深圳）有限公司 | Remote control method, remote control device, electronic apparatus, storage medium, and program product |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103810725A (en) * | 2014-03-12 | 2014-05-21 | 北京理工大学 | Video stabilizing method based on global optimization |
| CN104219304A (en) * | 2014-09-01 | 2014-12-17 | 北京优特捷信息技术有限公司 | Device and method for protecting privacy information of users |
| CN105243668A (en) * | 2015-10-13 | 2016-01-13 | 中山大学 | License plate image privacy removal method |
| CN105469379A (en) * | 2014-09-04 | 2016-04-06 | 广东中星电子有限公司 | Video target area shielding method and device |
| CN108564554A (en) * | 2018-05-09 | 2018-09-21 | 上海大学 | A kind of video stabilizing method based on movement locus optimization |
| CN108965982A (en) * | 2018-08-28 | 2018-12-07 | 百度在线网络技术（北京）有限公司 | Video recording method, device, electronic equipment and readable storage medium storing program for executing |
| CN109284684A (en) * | 2018-08-21 | 2019-01-29 | Oppo广东移动通信有限公司 | A kind of information processing method, device and computer storage medium |
| CN109977937A (en) * | 2019-03-26 | 2019-07-05 | 北京字节跳动网络技术有限公司 | Image processing method, device and equipment |
| CN110620891A (en) * | 2019-09-27 | 2019-12-27 | 上海依图网络科技有限公司 | Imaging system and video processing method |
| CN111429330A (en) * | 2020-03-31 | 2020-07-17 | 上海兑观信息科技技术有限公司 | Method and system for protecting privacy of monitoring video |
| CN112040145A (en) * | 2020-08-28 | 2020-12-04 | 维沃移动通信有限公司 | Image processing method and device and electronic equipment |
| CN112528318A (en) * | 2020-11-27 | 2021-03-19 | 国家电网有限公司大数据中心 | Image desensitization method and device and electronic equipment |
-
2021
- 2021-03-23 CN CN202110310965.0A patent/CN114567797A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103810725A (en) * | 2014-03-12 | 2014-05-21 | 北京理工大学 | Video stabilizing method based on global optimization |
| CN104219304A (en) * | 2014-09-01 | 2014-12-17 | 北京优特捷信息技术有限公司 | Device and method for protecting privacy information of users |
| CN105469379A (en) * | 2014-09-04 | 2016-04-06 | 广东中星电子有限公司 | Video target area shielding method and device |
| CN105243668A (en) * | 2015-10-13 | 2016-01-13 | 中山大学 | License plate image privacy removal method |
| CN108564554A (en) * | 2018-05-09 | 2018-09-21 | 上海大学 | A kind of video stabilizing method based on movement locus optimization |
| CN109284684A (en) * | 2018-08-21 | 2019-01-29 | Oppo广东移动通信有限公司 | A kind of information processing method, device and computer storage medium |
| CN108965982A (en) * | 2018-08-28 | 2018-12-07 | 百度在线网络技术（北京）有限公司 | Video recording method, device, electronic equipment and readable storage medium storing program for executing |
| CN109977937A (en) * | 2019-03-26 | 2019-07-05 | 北京字节跳动网络技术有限公司 | Image processing method, device and equipment |
| CN110620891A (en) * | 2019-09-27 | 2019-12-27 | 上海依图网络科技有限公司 | Imaging system and video processing method |
| CN111429330A (en) * | 2020-03-31 | 2020-07-17 | 上海兑观信息科技技术有限公司 | Method and system for protecting privacy of monitoring video |
| CN112040145A (en) * | 2020-08-28 | 2020-12-04 | 维沃移动通信有限公司 | Image processing method and device and electronic equipment |
| CN112528318A (en) * | 2020-11-27 | 2021-03-19 | 国家电网有限公司大数据中心 | Image desensitization method and device and electronic equipment |
Non-Patent Citations (1)
| Title |
|---|
| 王敬东;薛重飞;魏雪迎;刘云霄;: "三焦点张量重投影视频稳像算法", 中国图象图形学报, no. 07, 16 July 2017 (2017-07-16), pages 77 - 87 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117061581A (en) * | 2023-10-08 | 2023-11-14 | 腾讯科技（深圳）有限公司 | Remote control method, remote control device, electronic apparatus, storage medium, and program product |
| CN117061581B (en) * | 2023-10-08 | 2024-04-19 | 腾讯科技（深圳）有限公司 | Remote control method, remote control device, electronic apparatus, storage medium, and program product |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8849013B2 (en) | System and method of image processing | |
| US20150279121A1 (en) | Active Point Cloud Modeling | |
| EP3276951A1 (en) | Image processing system, image processing method, and program | |
| WO2019128137A1 (en) | Online three-dimensional editing systems and methods | |
| US9392248B2 (en) | Dynamic POV composite 3D video system | |
| CN107870961B (en) | Method and system for searching and sorting space objects and computer readable storage device | |
| DE102021118479A1 (en) | EMBEDDING COMPLEX 3D OBJECTS INTO AN AUGMENTED REALITY SCENE USING IMAGE SEGMENTATION | |
| CN114003190B (en) | Augmented reality method and device suitable for multiple scenes and multiple devices | |
| US20220343520A1 (en) | Image Processing Method and Image Processing Apparatus, and Electronic Device Using Same | |
| CN111835982B (en) | Image acquisition method, image acquisition device, electronic device, and storage medium | |
| DE102021119882A1 (en) | VIDEO BACKGROUND ESTIMATION USING SPATIAL-TEMPORAL MODELS | |
| CN111246196B (en) | Video processing method and device, electronic equipment and computer readable storage medium | |
| CN107564085B (en) | Image warping processing method and device, computing equipment and computer storage medium | |
| CN114567797A (en) | Image processing method and device and vehicle | |
| JP2024041895A (en) | Modular image interpolation method | |
| CN103377469A (en) | Terminal and image processing method | |
| CN113870213A (en) | Image display method, image display device, storage medium, and electronic apparatus | |
| CN109785439A (en) | Human face sketch image generating method and Related product | |
| CN106470337A (en) | For the method for the personalized omnirange video depth of field, device and computer program | |
| CN112165629A (en) | Intelligent live broadcast method, wearable device and intelligent live broadcast system | |
| CN110047126B (en) | Method, apparatus, electronic device, and computer-readable storage medium for rendering image | |
| Peng et al. | Mpib: An mpi-based bokeh rendering framework for realistic partial occlusion effects | |
| KR101426785B1 (en) | A user-friendly image obfuscation method for the sensitive area in smart devices by reflecting art sensitivity | |
| CN107787507A (en) | The apparatus and method for obtaining the registration error figure for the acutance rank for representing image | |
| CN116977392A (en) | Image generation method, device, electronic equipment and storage medium |
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 |