CN115294220A - Intelligent image compression method and system - Google Patents

Abstract

The invention provides an intelligent image compression method and system, and relates to the technical field of image processing. In the invention, a to-be-processed image set acquired by each image acquisition terminal device is respectively acquired, and a plurality of to-be-processed image sets corresponding to a plurality of image acquisition terminal devices are obtained; performing duplicate removal screening on the image sets to be processed aiming at each image set to be processed in a plurality of image sets to be processed to obtain a target image set corresponding to the image set to be processed, wherein the number of frames of the images to be processed included in the target image set is less than or equal to the number of frames of the images to be processed included in the corresponding image set to be processed; and for each target image set, compressing the images to be processed included in the target image set to obtain a compressed image set corresponding to the target image set. Based on the method, the problem that image compression resources are easily wasted in the prior art can be solved.

Description

Intelligent image compression method and system

Technical Field

The invention relates to the technical field of image processing, in particular to an intelligent image compression method and system.

Background

With the continuous development of image processing technology, the application range of image processing technology is increasing, for example, application scenarios such as video production and image monitoring. Among them, in various applications of image processing, it is generally necessary to compress an image to reduce the amount of image data. However, in the prior art, all the acquired images are generally directly compressed, and thus, the resource for compressing the images may be wasted.

Disclosure of Invention

In view of this, the present invention provides an intelligent image compression method and system to solve the problem in the prior art that image compression resources are easily wasted.

In order to achieve the above purpose, the embodiment of the invention adopts the following technical scheme:

an intelligent image compression method is applied to an image processing server, the image processing server is in communication connection with a plurality of image acquisition terminal devices, and the method comprises the following steps:

respectively acquiring a to-be-processed image set acquired by each image acquisition terminal device in the plurality of image acquisition terminal devices to obtain a plurality of to-be-processed image sets corresponding to the plurality of image acquisition terminal devices, wherein each to-be-processed image set comprises a plurality of frames of to-be-processed images, and the plurality of frames of to-be-processed images are sorted in the corresponding to-be-processed image sets according to acquisition time;

for each image set to be processed in the plurality of image sets to be processed, performing duplicate removal screening on the image set to be processed to obtain a target image set corresponding to the image set to be processed, wherein the number of frames of images to be processed included in the target image set is less than or equal to the number of frames of images to be processed included in the corresponding image set to be processed;

and for each target image set, compressing the images to be processed included in the target image set to obtain a compressed image set corresponding to the target image set, wherein the compressed images included in the compressed image set are obtained by compressing the corresponding images to be processed.

In some preferred embodiments, in the above method for intelligently compressing an image, the step of respectively obtaining a set of images to be processed acquired by each of the plurality of image acquisition terminal devices to obtain a plurality of sets of images to be processed corresponding to the plurality of image acquisition terminal devices includes:

determining whether image compression processing is needed, generating corresponding image acquisition request information when the image compression processing is needed, and sending the image acquisition request information to each image acquisition terminal device in the plurality of image acquisition terminal devices, wherein each image acquisition terminal device is used for sending a currently acquired and stored image set to be processed to the image processing server after receiving the image acquisition request information, and deleting the image set to be processed;

and respectively acquiring a to-be-processed image set sent by each image acquisition terminal device of the plurality of image acquisition terminal devices based on the image acquisition request information to obtain a plurality of to-be-processed image sets corresponding to the plurality of image acquisition terminal devices.

In some preferred embodiments, in the above intelligent image compression method, the step of determining whether image compression processing is required, and when image compression processing is required, generating corresponding image acquisition request information, and sending the image acquisition request information to each of the plurality of image acquisition terminal devices includes:

acquiring the time of image compression processing last time in history to obtain corresponding history time information, and calculating a history time interval between the history time information and current time information;

acquiring a preset time interval reference value, acquiring a data volume of a historical image to be processed corresponding to image compression processing performed last time in history to obtain a corresponding historical image data volume, calculating a ratio between the historical image data volume and the preset image data volume reference value to obtain a corresponding data volume ratio, and calculating a quotient between the time interval reference value and the data volume ratio to obtain a corresponding time interval threshold;

determining a relative magnitude relationship between the historical time interval and the time interval threshold;

if the historical time interval is greater than or equal to the time interval threshold, determining that image compression processing is required, if the historical time interval is smaller than the time interval threshold, determining that image compression processing is not required, generating corresponding image acquisition request information when image compression processing is required, and sending the image acquisition request information to each image acquisition terminal device in the plurality of image acquisition terminal devices.

In some preferred embodiments, in the above method for intelligently compressing an image, the step of performing duplicate removal and screening on each to-be-processed image set in the multiple to-be-processed image sets to obtain a target image set corresponding to the to-be-processed image set includes:

for each to-be-processed image set in the to-be-processed image sets, performing similarity calculation on to-be-processed images included in the to-be-processed image sets to obtain image similarity between every two frames of to-be-processed images included in the to-be-processed image sets;

and aiming at each image set to be processed in the plurality of image sets to be processed, based on the image similarity between every two frames of images to be processed included in the image set to be processed, performing duplicate removal screening on the image set to be processed to obtain a target image set corresponding to the image set to be processed.

In some preferred embodiments, in the above method for intelligently compressing an image, the step of performing, for each to-be-processed image set in the multiple to-be-processed image sets, duplicate removal and screening on the to-be-processed image set based on an image similarity between every two frames of to-be-processed images included in the to-be-processed image set to obtain a target image set corresponding to the to-be-processed image set includes:

for each two frames of images to be processed included in each image set to be processed in the multiple image sets to be processed, determining a sorting interval length between the two frames of images to be processed, and determining a similarity weighting coefficient corresponding to the two frames of images to be processed based on the sorting interval length, wherein the similarity weighting coefficient and the sorting interval length have a positive correlation;

for each two frames of images to be processed in each image set to be processed in the plurality of image sets to be processed, updating the image similarity corresponding to the two frames of images to be processed based on the similarity weighting coefficients corresponding to the two frames of images to be processed to obtain corresponding image similarity updated values, wherein the image similarity updated values are greater than or equal to the corresponding image similarities;

determining a relative size relationship between image similarity update values corresponding to two frames of images to be processed and a pre-configured image similarity threshold value aiming at every two frames of images to be processed in each image set to be processed in the plurality of image sets to be processed, and determining the two frames of images to be processed as a repeated image combination when the image similarity update value is greater than or equal to the image similarity threshold value;

and respectively performing duplicate removal screening on each repeated image combination corresponding to the image set to be processed aiming at each image set to be processed in the plurality of image sets to be processed to obtain a target image set corresponding to the image set to be processed, wherein one frame of image to be processed in one corresponding repeated image combination is screened out during duplicate removal screening.

In some preferred embodiments, in the above intelligent image compression method, the step of, for each target image set, performing compression processing on the image to be processed included in the target image set to obtain a compressed image set corresponding to the target image set includes:

for every two target image sets, calculating to obtain the set similarity between the two target image sets based on the images to be processed included in the two target image sets;

for each target image set, performing fusion processing on the set similarity between the target image set and each other target image set to obtain a set similarity fusion value corresponding to the target image set, and determining a corresponding image compression coefficient based on the set similarity fusion value, wherein the image compression coefficient and the set similarity fusion value have a positive correlation;

and for each target image set, compressing each frame of image to be processed included in the target image set based on an image compression coefficient corresponding to the target image set to obtain a compressed image set corresponding to the target image set, wherein the ratio of the data volume of the image to be processed to the data volume of the corresponding compressed image obtained by compression is equal to the image compression coefficient.

In some preferred embodiments, in the above intelligent image compression method, the step of calculating, for each two target image sets, a set similarity between the two target image sets based on the images to be processed included in the two target image sets includes:

for each to-be-processed image set in the to-be-processed image sets, performing similarity calculation on to-be-processed images included in the to-be-processed image sets to obtain image similarity between every two frames of to-be-processed images included in the to-be-processed image sets;

calculating the average value of the image similarity between the image to be processed and other images to be processed of each frame in the image set to be processed aiming at each image to be processed in each image set to be processed in the plurality of image sets to be processed to obtain the similarity coefficient of the image to be processed;

calculating the image similarity between every two frames of images to be processed between every two target image sets, obtaining the coefficient difference value corresponding to the two frames of images to be processed based on the difference value between the similarity coefficients corresponding to the two frames of images to be processed, and determining the image fusion coefficient corresponding to the two frames of images to be processed based on the coefficient difference value, wherein the image fusion coefficient and the coefficient difference value have a negative correlation relationship;

aiming at each two frames of images to be processed between each two target image sets, calculating the product of the image similarity between the two frames of images to be processed and the image fusion coefficient corresponding to the two frames of images to be processed to obtain the image fusion similarity corresponding to the two frames of images to be processed;

and aiming at every two target image sets, performing mean value calculation based on image fusion similarity corresponding to every two frames of images to be processed between the two target image sets to obtain set similarity between the two target image sets.

The embodiment of the invention also provides an intelligent image compression system, which is applied to an image processing server, wherein the image processing server is in communication connection with a plurality of image acquisition terminal devices, and the system comprises:

an image set obtaining module, configured to obtain a to-be-processed image set collected by each of the multiple image collection terminal devices, respectively, to obtain multiple to-be-processed image sets corresponding to the multiple image collection terminal devices, where each of the to-be-processed image sets includes multiple frames of to-be-processed images, and the multiple frames of to-be-processed images are sorted according to collection time in the corresponding to-be-processed image set;

an image set duplicate removal module, configured to perform duplicate removal screening on each to-be-processed image set in the multiple to-be-processed image sets to obtain a target image set corresponding to the to-be-processed image set, where a frame number of the to-be-processed images included in the target image set is less than or equal to a frame number of the to-be-processed images included in the corresponding to-be-processed image set;

and the image set compression module is used for compressing the images to be processed included in the target image set aiming at each target image set to obtain a compressed image set corresponding to the target image set, wherein the compressed images included in the compressed image set are obtained by compressing the corresponding images to be processed.

In some preferred embodiments, in the above intelligent image compression system, the image set deduplication module is specifically configured to:

for each to-be-processed image set in the to-be-processed image sets, performing similarity calculation on to-be-processed images included in the to-be-processed image sets to obtain image similarity between every two frames of to-be-processed images included in the to-be-processed image sets;

and aiming at each image set to be processed in the plurality of image sets to be processed, based on the image similarity between every two frames of images to be processed included in the image set to be processed, performing duplicate removal screening on the image set to be processed to obtain a target image set corresponding to the image set to be processed.

In some preferred embodiments, in the above intelligent image compression system, the image set compression module is specifically configured to:

for every two target image sets, calculating to obtain set similarity between the two target image sets based on images to be processed included in the two target image sets;

for each target image set, performing fusion processing on the set similarity between the target image set and each other target image set to obtain a set similarity fusion value corresponding to the target image set, and determining a corresponding image compression coefficient based on the set similarity fusion value, wherein the image compression coefficient and the set similarity fusion value have a positive correlation;

and for each target image set, compressing each frame of image to be processed included in the target image set based on an image compression coefficient corresponding to the target image set to obtain a compressed image set corresponding to the target image set, wherein the ratio of the data volume of the image to be processed to the data volume of the corresponding compressed image obtained by compression is equal to the image compression coefficient.

The method and the system for intelligently compressing the image, provided by the embodiment of the invention, can firstly respectively obtain the to-be-processed image sets acquired by each image acquisition terminal device to obtain a plurality of to-be-processed image sets corresponding to the plurality of image acquisition terminal devices, and then, aiming at each to-be-processed image set in the plurality of to-be-processed image sets, the to-be-processed image sets are subjected to de-duplication screening to obtain the target image sets corresponding to the to-be-processed image sets, so that the to-be-processed images included in the target image sets can be compressed aiming at each target image set to obtain the compressed image sets corresponding to the target image sets.

In order to make the aforementioned and other objects, features and advantages of the present invention comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

Fig. 1 is a block diagram of an image processing server according to an embodiment of the present invention.

Fig. 2 is a schematic diagram of steps included in an intelligent image compression method according to an embodiment of the present invention.

Fig. 3 is a schematic diagram of modules included in an intelligent image compression system according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.

Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be obtained by a person skilled in the art without inventive step based on the embodiments of the present invention, are within the scope of protection of the present invention.

As shown in fig. 1, an embodiment of the present invention provides an image processing server. Wherein the image processing server may include a memory and a processor.

In detail, the memory and the processor are electrically connected directly or indirectly to realize data transmission or interaction. For example, they may be electrically connected to each other via one or more communication buses or signal lines. The memory can have stored therein at least one software function (computer program) which can be present in the form of software or firmware. The processor may be configured to execute the executable computer program stored in the memory, so as to implement the image intelligent compression method provided by the embodiment of the present invention (as described later).

Alternatively, in one possible embodiment, the Memory may be, but is not limited to, random Access Memory (RAM), read Only Memory (ROM), programmable Read-Only Memory (PROM), erasable Read-Only Memory (EPROM), electrically Erasable Read-Only Memory (EEPROM), and the like.

Alternatively, in one possible implementation, the Processor may be a general-purpose Processor including a Central Processing Unit (CPU), a Network Processor (NP), a System on Chip (SoC), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components.

Moreover, the structure shown in fig. 1 is only an illustration, and the image processing server may further include more or fewer components than those shown in fig. 1, or have a different configuration from that shown in fig. 1, for example, may include a communication unit for information interaction with other devices (such as an image capturing terminal device). Also, the image processing server may be coupled via the bus to a display, such as a liquid crystal display, or an active matrix display, for displaying information to a user. An input device, such as a keyboard including alphanumeric and other keys, may also be coupled to the bus for communicating information and command selections to the processor. In an embodiment, the input device has a touch screen display. The input device may include a cursor control (such as a mouse, a trackball, or cursor direction keys) for communicating direction information and command selections to the processor and for controlling cursor movement on the display.

With reference to fig. 2, an embodiment of the present invention further provides an intelligent image compression method, which is applicable to the image processing server. The method steps defined by the flow related to the intelligent image compression method can be realized by the image processing server. And the image processing server is in communication connection with a plurality of image acquisition terminal devices. The specific flow shown in fig. 2 will be explained below.

Step S110, respectively obtaining a to-be-processed image set acquired by each image acquisition terminal device of the plurality of image acquisition terminal devices, and obtaining a plurality of to-be-processed image sets corresponding to the plurality of image acquisition terminal devices.

In the embodiment of the present invention, the image processing server may respectively obtain a to-be-processed image set acquired by each of the plurality of image acquisition terminal devices, so as to obtain a plurality of to-be-processed image sets corresponding to the plurality of image acquisition terminal devices. Each to-be-processed image set comprises a plurality of frames of to-be-processed images, and the plurality of frames of to-be-processed images are sequenced in the corresponding to-be-processed image sets according to the acquisition time.

Step S120, aiming at each image set to be processed in the plurality of image sets to be processed, carrying out duplicate removal screening on the image set to be processed to obtain a target image set corresponding to the image set to be processed.

In the embodiment of the present invention, the image processing server may perform duplicate removal and screening on each to-be-processed image set in the plurality of to-be-processed image sets to obtain a target image set corresponding to the to-be-processed image set. And the frame number of the images to be processed included in the target image set is less than or equal to the frame number of the images to be processed included in the corresponding image set to be processed.

Step S130, for each target image set, performing compression processing on the to-be-processed image included in the target image set to obtain a compressed image set corresponding to the target image set.

In this embodiment of the present invention, the image processing server may perform compression processing on the to-be-processed image included in each target image set to obtain a compressed image set corresponding to the target image set. And the compressed images included in the compressed image set are obtained by compressing the corresponding images to be processed.

Based on the steps included in the above image intelligent compression method, the to-be-processed image set acquired by each image acquisition terminal device may be first obtained, a plurality of to-be-processed image sets corresponding to the plurality of image acquisition terminal devices are obtained, and then, for each to-be-processed image set in the plurality of to-be-processed image sets, the to-be-processed image set is subjected to deduplication screening, so as to obtain a target image set corresponding to the to-be-processed image set, so that the to-be-processed images included in the target image set may be compressed for each target image set, and the compressed image set corresponding to the target image set is obtained.

Alternatively, in one possible implementation, step S110 may include the following:

firstly, determining whether image compression processing is needed, generating corresponding image acquisition request information when the image compression processing is needed, and sending the image acquisition request information to each image acquisition terminal device in the plurality of image acquisition terminal devices, wherein each image acquisition terminal device is used for sending a currently acquired and stored image set to be processed to the image processing server and deleting the image set to be processed after receiving the image acquisition request information;

secondly, a to-be-processed image set sent by each image acquisition terminal device of the plurality of image acquisition terminal devices based on the image acquisition request information is acquired respectively, and a plurality of to-be-processed image sets corresponding to the plurality of image acquisition terminal devices are obtained.

Alternatively, in a possible implementation, the step of determining whether image compression processing is required, and when image compression processing is required, generating corresponding image acquisition request information, and sending the image acquisition request information to each of the plurality of image acquisition terminal devices may include the following steps:

firstly, acquiring the time of image compression processing which is performed last time in history to obtain corresponding historical time information, and calculating the historical time interval between the historical time information and the current time information;

secondly, acquiring a pre-configured time interval reference value, acquiring the data volume of a historical image to be processed corresponding to the latest image compression processing in history to obtain a corresponding historical image data volume, calculating the ratio of the historical image data volume to the pre-configured image data volume reference value to obtain a corresponding data volume ratio, and calculating the quotient of the time interval reference value and the data volume ratio to obtain a corresponding time interval threshold value;

then, determining a relative magnitude relationship between the historical time interval and the time interval threshold (i.e., determining whether the historical time interval is greater than or equal to the time interval threshold);

and finally, if the historical time interval is greater than or equal to the time interval threshold, determining that image compression processing is required, if the historical time interval is smaller than the time interval threshold, determining that image compression processing is not required, generating corresponding image acquisition request information when image compression processing is required, and sending the image acquisition request information to each image acquisition terminal device in the plurality of image acquisition terminal devices.

Alternatively, in one possible implementation, step S120 may include the following:

firstly, aiming at each image set to be processed in the plurality of image sets to be processed, carrying out similarity calculation on the images to be processed included in the image sets to be processed to obtain the image similarity between every two frames of images to be processed included in the image sets to be processed;

secondly, aiming at each image set to be processed in the plurality of image sets to be processed, based on the image similarity between every two frames of images to be processed included in the image set to be processed, the image set to be processed is subjected to duplicate removal screening, and a target image set corresponding to the image set to be processed is obtained.

Optionally, in a possible implementation manner, the step of performing similarity calculation on the to-be-processed images included in the to-be-processed image set to obtain the image similarity between every two frames of to-be-processed images included in the to-be-processed image set for each to-be-processed image set in the plurality of to-be-processed image sets may include the following steps:

firstly, determining two frames of images to be processed included in the image set to be processed as a first image to be processed and a second image to be processed respectively, segmenting the first image to be processed to obtain a first segmented image set corresponding to the first image to be processed, and segmenting the second image to be processed to obtain a second segmented image set corresponding to the second image to be processed, wherein multiple frames of first segmented images included in the first segmented image set are spliced to form the first image to be processed, multiple frames of second segmented images included in the second segmented image set are prominently spliced to form the second image to be processed, and the number of the first segmented images is the same as that of the second segmented images;

secondly, acquiring multi-frame historical images to be processed which are acquired historically by image acquisition terminal equipment corresponding to the image set to be processed, clustering the multi-frame historical images to be processed to obtain corresponding clustering results, and determining at least one frame of clustering center image corresponding to the multi-frame historical images to be processed based on the clustering results;

then, for each frame of first divided image in the first divided image set, respectively calculating pixel similarity between the first divided image and each frame of the cluster center image (for example, ordering pixel values of each pixel position to obtain a corresponding pixel value sequence, then calculating sequence similarity between two pixel value sequences to obtain the pixel similarity), and determining the pixel similarity having the maximum value as a first target pixel similarity corresponding to the first divided image in the pixel similarity between the first divided image and each frame of the cluster center image, and determining the cluster center image corresponding to the first target pixel similarity as a target cluster center image corresponding to the first divided image;

then, for each frame of second segmentation image in the second segmentation image set, respectively calculating pixel similarity between the second segmentation image and each frame of the clustering center image, determining the pixel similarity with the maximum value as a second target pixel similarity corresponding to the second segmentation image in the pixel similarity between the second segmentation image and each frame of the clustering center image, and determining the clustering center image corresponding to the second target pixel similarity as a target clustering center image corresponding to the second segmentation image;

further, based on whether the corresponding target clustering center images are the same or not, combining the first segmentation images and the second segmentation images, determining the number of formed segmentation image combinations to obtain the number of the corresponding segmentation image combinations, and based on the number ratio between the number of the segmentation image combinations and the number of the first segmentation images, determining corresponding first similarity, wherein the target clustering center image corresponding to the first segmentation image and the target clustering center image corresponding to the second segmentation image included in each segmentation image combination are the same, and the first similarity and the number ratio have a positive correlation;

further, for each of the segmented image combinations, fusing (for example, fusing in a manner of calculating a product or calculating a weighted sum value, or the like) the target pixel similarity corresponding to the first segmented image and the target pixel similarity corresponding to the second segmented image included in the segmented image combination to obtain a corresponding similarity fusion value, and obtaining a corresponding second similarity based on an average value of the similarity fusion values corresponding to each of the segmented images;

and finally, obtaining the image similarity between the two to-be-processed images based on the first similarity and the second similarity (such as calculating a weighted sum value and the like).

Optionally, in a possible implementation manner, the step of, for each to-be-processed image set in the multiple to-be-processed image sets, performing de-duplication screening on the to-be-processed image set based on an image similarity between every two frames of to-be-processed images included in the to-be-processed image set to obtain a target image set corresponding to the to-be-processed image set may include the following steps:

firstly, aiming at each two frames of images to be processed in each image set to be processed in the plurality of image sets to be processed, determining the length of a sequencing interval between the two frames of images to be processed, and determining a similarity weighting coefficient corresponding to the two frames of images to be processed based on the length of the sequencing interval, wherein the similarity weighting coefficient and the sequencing interval have positive correlation;

secondly, for each two frames of images to be processed included in each image set to be processed in the plurality of image sets to be processed, updating the image similarity corresponding to the two frames of images to be processed based on the similarity weighting coefficients corresponding to the two frames of images to be processed to obtain a corresponding image similarity updating value, wherein the image similarity updating value is greater than or equal to the corresponding image similarity;

then, for each two frames of images to be processed included in each image set to be processed in the plurality of image sets to be processed, determining a relative size relationship between an image similarity update value corresponding to the two frames of images to be processed and a pre-configured image similarity threshold, and determining the two frames of images to be processed as a repeated image combination when the image similarity update value is greater than or equal to the image similarity threshold;

and finally, for each to-be-processed image set in the to-be-processed image sets, respectively performing duplicate removal screening on each repeated image combination corresponding to the to-be-processed image set to obtain a target image set corresponding to the to-be-processed image set, wherein when the duplicate removal screening is performed, one frame of to-be-processed image in one corresponding repeated image combination is screened out.

Alternatively, in one possible implementation, step S130 may include the following:

firstly, aiming at every two target image sets, calculating to obtain the set similarity between the two target image sets based on the images to be processed included in the two target image sets;

secondly, performing fusion processing (such as average value calculation) on the set similarity between the target image set and each other target image set aiming at each target image set to obtain a set similarity fusion value corresponding to the target image set, and determining a corresponding image compression coefficient based on the set similarity fusion value, wherein the image compression coefficient and the set similarity fusion value have positive correlation;

then, for each target image set, performing compression processing on each frame of image to be processed included in the target image set based on an image compression coefficient corresponding to the target image set to obtain a compressed image set corresponding to the target image set, where a ratio of a data amount of the image to be processed to a data amount of a corresponding compressed image obtained by performing compression processing is equal to the image compression coefficient (that is, the image compression coefficient represents a corresponding compression degree).

Alternatively, in a possible embodiment, the step of calculating, for each two target image sets, a set similarity between the two target image sets based on the images to be processed included in the two target image sets may include the following steps:

firstly, aiming at each image set to be processed in the plurality of image sets to be processed, carrying out similarity calculation on the images to be processed included in the image sets to be processed to obtain the image similarity between every two frames of images to be processed included in the image sets to be processed;

secondly, calculating an average value of image similarity between the image to be processed and other images to be processed of each frame in the image set to be processed aiming at each image to be processed in each image set to be processed in the plurality of image sets to be processed, and obtaining a similarity coefficient of the image to be processed (namely taking the obtained average value as the similarity coefficient);

then, for each two frames of images to be processed between every two target image sets, calculating the image similarity between the two frames of images to be processed, obtaining a coefficient difference value corresponding to the two frames of images to be processed based on a difference value between similarity coefficients corresponding to the two frames of images to be processed, and determining an image fusion coefficient corresponding to the two frames of images to be processed based on the coefficient difference value, wherein the image fusion coefficient and the coefficient difference value have a negative correlation relationship;

then, aiming at each two frames of images to be processed between each two target image sets, calculating the product of the image similarity between the two frames of images to be processed and the image fusion coefficient corresponding to the two frames of images to be processed to obtain the image fusion similarity corresponding to the two frames of images to be processed;

and finally, performing mean value calculation on every two target image sets based on image fusion similarity corresponding to every two frames of images to be processed between the two target image sets to obtain set similarity between the two target image sets.

With reference to fig. 3, an embodiment of the present invention further provides an intelligent image compression system, which is applicable to the image processing server. Wherein, the image intelligent compression system can comprise the following modules:

the image collection acquisition module is used for respectively acquiring to-be-processed image collections acquired by each image acquisition terminal device in the image acquisition terminal devices to obtain a plurality of to-be-processed image collections corresponding to the image acquisition terminal devices, wherein each to-be-processed image collection comprises a plurality of frames of to-be-processed images which are sequenced in the corresponding to-be-processed image collections according to acquisition time;

an image set duplicate removal module, configured to perform duplicate removal screening on each to-be-processed image set in the multiple to-be-processed image sets to obtain a target image set corresponding to the to-be-processed image set, where a frame number of the to-be-processed images included in the target image set is less than or equal to a frame number of the to-be-processed images included in the corresponding to-be-processed image set;

and the image set compression module is used for compressing the images to be processed included in the target image set aiming at each target image set to obtain a compressed image set corresponding to the target image set, wherein the compressed images included in the compressed image set are obtained by compressing based on the corresponding images to be processed.

Optionally, in a possible implementation, the image set deduplication module is specifically configured to:

for each to-be-processed image set in the to-be-processed image sets, performing similarity calculation on to-be-processed images included in the to-be-processed image sets to obtain image similarity between every two frames of to-be-processed images included in the to-be-processed image sets;

and aiming at each image set to be processed in the plurality of image sets to be processed, based on the image similarity between every two frames of images to be processed included in the image set to be processed, performing duplicate removal screening on the image set to be processed to obtain a target image set corresponding to the image set to be processed.

Optionally, in a possible implementation, the image set compression module is specifically configured to:

for every two target image sets, calculating to obtain the set similarity between the two target image sets based on the images to be processed included in the two target image sets;

for each target image set, carrying out fusion processing on the set similarity between the target image set and each other target image set to obtain a set similarity fusion value corresponding to the target image set, and determining a corresponding image compression coefficient based on the set similarity fusion value, wherein the image compression coefficient and the set similarity fusion value have a positive correlation;

and for each target image set, compressing each frame of image to be processed included in the target image set based on an image compression coefficient corresponding to the target image set to obtain a compressed image set corresponding to the target image set, wherein the ratio of the data volume of the image to be processed to the data volume of the corresponding compressed image obtained by compression is equal to the image compression coefficient.

In summary, according to the method and system for intelligently compressing images provided by the present invention, a to-be-processed image set acquired by each image acquisition terminal device may be first obtained, a plurality of to-be-processed image sets corresponding to the plurality of image acquisition terminal devices are obtained, and then, for each to-be-processed image set in the plurality of to-be-processed image sets, the to-be-processed image set is subjected to deduplication screening, so as to obtain a target image set corresponding to the to-be-processed image set, so that, for each target image set, a to-be-processed image included in the target image set may be compressed, and a compressed image set corresponding to the target image set is obtained.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (10)

1. An intelligent image compression method is applied to an image processing server, the image processing server is in communication connection with a plurality of image acquisition terminal devices, and the method comprises the following steps:

respectively acquiring a to-be-processed image set acquired by each image acquisition terminal device in the plurality of image acquisition terminal devices to obtain a plurality of to-be-processed image sets corresponding to the plurality of image acquisition terminal devices, wherein each to-be-processed image set comprises a plurality of frames of to-be-processed images, and the plurality of frames of to-be-processed images are sorted in the corresponding to-be-processed image sets according to acquisition time;

for each to-be-processed image set in the to-be-processed image sets, performing duplicate removal screening on the to-be-processed image set to obtain a target image set corresponding to the to-be-processed image set, wherein the number of frames of the to-be-processed images included in the target image set is less than or equal to the number of frames of the to-be-processed images included in the corresponding to-be-processed image set;

and for each target image set, compressing the images to be processed included in the target image set to obtain a compressed image set corresponding to the target image set, wherein the compressed images included in the compressed image set are obtained by compressing the corresponding images to be processed.

2. The method according to claim 1, wherein the step of obtaining the to-be-processed image sets acquired by each of the image acquisition terminal devices to obtain a plurality of to-be-processed image sets corresponding to the image acquisition terminal devices comprises:

determining whether image compression processing is needed, generating corresponding image acquisition request information when the image compression processing is needed, and sending the image acquisition request information to each image acquisition terminal device in the plurality of image acquisition terminal devices, wherein each image acquisition terminal device is used for sending a currently acquired and stored image set to be processed to the image processing server after receiving the image acquisition request information, and deleting the image set to be processed;

and respectively acquiring a to-be-processed image set sent by each image acquisition terminal device of the plurality of image acquisition terminal devices based on the image acquisition request information to obtain a plurality of to-be-processed image sets corresponding to the plurality of image acquisition terminal devices.

3. The intelligent image compression method according to claim 2, wherein the step of determining whether image compression processing is required, and when image compression processing is required, generating corresponding image acquisition request information and sending the image acquisition request information to each of the plurality of image acquisition terminal devices comprises:

acquiring the time of image compression processing which is performed last time in history to obtain corresponding historical time information, and calculating the historical time interval between the historical time information and the current time information;

acquiring a preset time interval reference value, acquiring a data volume of a historical image to be processed corresponding to image compression processing performed last time in history to obtain a corresponding historical image data volume, calculating a ratio between the historical image data volume and the preset image data volume reference value to obtain a corresponding data volume ratio, and calculating a quotient between the time interval reference value and the data volume ratio to obtain a corresponding time interval threshold;

determining a relative magnitude relationship between the historical time interval and the time interval threshold;

if the historical time interval is greater than or equal to the time interval threshold, determining that image compression processing is required, if the historical time interval is smaller than the time interval threshold, determining that image compression processing is not required, generating corresponding image acquisition request information when image compression processing is required, and sending the image acquisition request information to each image acquisition terminal device in the plurality of image acquisition terminal devices.

4. The method of claim 1, wherein the step of performing de-duplication screening on each to-be-processed image set in the to-be-processed image sets to obtain a target image set corresponding to the to-be-processed image set comprises:

for each to-be-processed image set in the to-be-processed image sets, performing similarity calculation on to-be-processed images included in the to-be-processed image sets to obtain image similarity between every two frames of to-be-processed images included in the to-be-processed image sets;

and aiming at each image set to be processed in the plurality of image sets to be processed, based on the image similarity between every two frames of images to be processed included in the image set to be processed, performing duplicate removal screening on the image set to be processed to obtain a target image set corresponding to the image set to be processed.

5. The method of claim 4, wherein the step of, for each to-be-processed image set in the plurality of to-be-processed image sets, performing de-duplication screening on the to-be-processed image set based on image similarity between every two frames of to-be-processed images included in the to-be-processed image set to obtain a target image set corresponding to the to-be-processed image set includes:

for each two frames of images to be processed included in each image set to be processed in the multiple image sets to be processed, determining a sorting interval length between the two frames of images to be processed, and determining a similarity weighting coefficient corresponding to the two frames of images to be processed based on the sorting interval length, wherein the similarity weighting coefficient and the sorting interval length have a positive correlation;

for each two frames of images to be processed included in each image set to be processed in the plurality of image sets to be processed, updating the image similarity corresponding to the two frames of images to be processed based on the similarity weighting coefficients corresponding to the two frames of images to be processed, and obtaining corresponding image similarity update values, wherein the image similarity update values are greater than or equal to the corresponding image similarity;

for each two frames of images to be processed in each image set to be processed in the plurality of image sets to be processed, determining a relative size relationship between an image similarity update value corresponding to the two frames of images to be processed and a pre-configured image similarity threshold, and determining the two frames of images to be processed as a repeated image combination when the image similarity update value is greater than or equal to the image similarity threshold;

and respectively carrying out duplicate removal screening on each repeated image combination corresponding to the image set to be processed aiming at each image set to be processed in the plurality of image sets to be processed to obtain a target image set corresponding to the image set to be processed, wherein one frame of image to be processed in one corresponding repeated image combination is screened out when the duplicate removal screening is carried out.

6. The method according to any one of claims 1 to 5, wherein the step of compressing the image to be processed included in each target image set to obtain the compressed image set corresponding to the target image set comprises:

for every two target image sets, calculating to obtain set similarity between the two target image sets based on images to be processed included in the two target image sets;

for each target image set, carrying out fusion processing on the set similarity between the target image set and each other target image set to obtain a set similarity fusion value corresponding to the target image set, and determining a corresponding image compression coefficient based on the set similarity fusion value, wherein the image compression coefficient and the set similarity fusion value have a positive correlation;

and for each target image set, compressing each frame of image to be processed included in the target image set based on an image compression coefficient corresponding to the target image set to obtain a compressed image set corresponding to the target image set, wherein the ratio of the data volume of the image to be processed to the data volume of the corresponding compressed image obtained by compression is equal to the image compression coefficient.

7. The method of claim 6, wherein the step of calculating the set similarity between two target image sets based on the images to be processed included in the two target image sets for each two target image sets comprises:

for each to-be-processed image set in the to-be-processed image sets, performing similarity calculation on to-be-processed images included in the to-be-processed image sets to obtain image similarity between every two frames of to-be-processed images included in the to-be-processed image sets;

calculating the average value of the image similarity between the image to be processed and other images to be processed of each frame in the image set to be processed aiming at each image to be processed in each image set to be processed in the plurality of image sets to be processed to obtain the similarity coefficient of the image to be processed;

calculating the image similarity between two frames of images to be processed aiming at every two frames of images to be processed between every two target image sets, obtaining the coefficient difference value corresponding to the two frames of images to be processed based on the difference value between the similarity coefficients corresponding to the two frames of images to be processed, and determining the image fusion coefficient corresponding to the two frames of images to be processed based on the coefficient difference value, wherein the image fusion coefficient and the coefficient difference value have a negative correlation relationship;

aiming at each two frames of images to be processed between each two target image sets, calculating the product of the image similarity between the two frames of images to be processed and the image fusion coefficient corresponding to the two frames of images to be processed to obtain the image fusion similarity corresponding to the two frames of images to be processed;

and aiming at every two target image sets, performing mean value calculation based on image fusion similarity corresponding to every two frames of images to be processed between the two target image sets to obtain set similarity between the two target image sets.

8. The utility model provides an image intelligence compression system which characterized in that is applied to image processing server, image processing server communication connection has a plurality of image acquisition terminal equipment, the system includes:

an image set obtaining module, configured to obtain a to-be-processed image set collected by each of the multiple image collection terminal devices, respectively, to obtain multiple to-be-processed image sets corresponding to the multiple image collection terminal devices, where each of the to-be-processed image sets includes multiple frames of to-be-processed images, and the multiple frames of to-be-processed images are sorted according to collection time in the corresponding to-be-processed image set;

the image set duplicate removal module is used for performing duplicate removal screening on the image sets to be processed aiming at each image set to be processed in the image sets to be processed to obtain a target image set corresponding to the image set to be processed, wherein the number of frames of the images to be processed included in the target image set is less than or equal to the number of frames of the images to be processed included in the corresponding image set to be processed;

and the image set compression module is used for compressing the images to be processed included in the target image set aiming at each target image set to obtain a compressed image set corresponding to the target image set, wherein the compressed images included in the compressed image set are obtained by compressing based on the corresponding images to be processed.

9. The intelligent image compression system of claim 8, wherein the image set deduplication module is specifically configured to:

for each to-be-processed image set in the to-be-processed image sets, performing similarity calculation on to-be-processed images included in the to-be-processed image sets to obtain image similarity between every two frames of to-be-processed images included in the to-be-processed image sets;

and aiming at each image set to be processed in the plurality of image sets to be processed, based on the image similarity between every two frames of images to be processed included in the image set to be processed, performing duplicate removal screening on the image set to be processed to obtain a target image set corresponding to the image set to be processed.

10. The intelligent image compression system of claim 8, wherein the image set compression module is specifically configured to:

for every two target image sets, calculating to obtain the set similarity between the two target image sets based on the images to be processed included in the two target image sets;

for each target image set, carrying out fusion processing on the set similarity between the target image set and each other target image set to obtain a set similarity fusion value corresponding to the target image set, and determining a corresponding image compression coefficient based on the set similarity fusion value, wherein the image compression coefficient and the set similarity fusion value have a positive correlation;

and for each target image set, compressing each frame of image to be processed included in the target image set based on an image compression coefficient corresponding to the target image set to obtain a compressed image set corresponding to the target image set, wherein the ratio of the data volume of the image to be processed to the data volume of the corresponding compressed image obtained by compression is equal to the image compression coefficient.

Images