An image processing method, an electronic device, and a storage medium

Abstract

The application discloses an image processing method, an electronic device and a storage medium, and the method comprises the following steps: receiving at least one region of interest image and basic information corresponding to each region of interest image sent by a target processing end; the region of interest image is obtained by dividing a target image by using preset configuration information by the target processing end; in response to successful reception of the target region of interest image and the corresponding basic information, encoding the target region of interest image and the corresponding basic information received successfully, and transmitting the encoded code stream to a second processing end; the target region of interest image is one of the at least one region of interest image; that is, the application divides the target image into at least one region of interest image, obtains basic information of each region of interest image, encodes, transmits and processes each region of interest image respectively, reduces the processing time of each link, and further reduces the delay, thereby improving the efficiency of image processing.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to an image processing method, electronic device and storage medium. Background Technology

[0002] With the explosive growth of computer applications, image processing is constantly evolving towards intelligence and humanization, and various intelligent scenarios are being developed and applied. In the development of image processing, image acquisition and processing are typically achieved through the characteristics of devices to realize the desired image processing objectives.

[0003] In current image processing, due to the characteristics of the devices, the processing time will cause a delay of at least one frame. In addition, there is the time for network transmission and decoding, which makes the delay of image acquisition and processing even longer. Current solutions mostly use the method of increasing the frame rate to reduce the delay, but this will correspondingly increase the dependence on system performance and data transmission bandwidth, affecting the efficiency of image processing and restricting the development of image processing. Summary of the Invention

[0004] The main technical problem addressed by this application is to provide an image processing method, electronic device, and storage medium that divides a target image into at least one region of interest (ROI) image, obtains basic information of each ROI image, and then encodes and transmits each ROI image separately, thereby reducing the processing time of each step, reducing latency, and improving the efficiency of image processing.

[0005] To address the aforementioned technical problems, this application provides an image processing method applied to a first processing end, comprising: receiving at least one region of interest (ROI) image and basic information corresponding to each ROI image sent by a target processing end; wherein the at least one ROI image is obtained by the target processing end dividing a target image using preset configuration information; in response to successful reception of the target ROI image and the corresponding basic information, encoding the successfully received target ROI image and the corresponding basic information, and transmitting the encoded bitstream to a second processing end; wherein the target ROI image is one of the at least one ROI images.

[0006] In some embodiments, receiving at least one region of interest image and basic information corresponding to each region of interest image sent by the target processing terminal includes: receiving at least two region of interest images output by the image acquisition module in the target processing terminal, wherein the at least two region of interest images are obtained by the target processing terminal dividing the current frame target image using preset configuration information; and receiving the basic information corresponding to each region of interest image output by the parameter configuration module in the target processing terminal, wherein the basic information represents the division parameters of each region of interest in the target image.

[0007] In some embodiments, at least two region-of-interest (ROI) images sent by a target processing terminal are received; the step of encoding the successfully received ROI image and the corresponding basic information in response to successful reception of the ROI image and the corresponding basic information, and transmitting the encoded bitstream to a second processing terminal, includes: binding the ROI image and the corresponding basic information in response to successful reception of any ROI image and the corresponding basic information; encoding the bound ROI image and the basic information, and transmitting the encoded bitstream to the second processing terminal.

[0008] In some embodiments, before encoding, the method further includes: responding to the current region of interest image being the last region of interest image in the current frame target image, determining a first change in the exposure parameters of different frame target images using external influence information, and setting the exposure parameters corresponding to the next frame target image based on the first change; determining a second change in the basic information of the region of interest images in different frame target images using preset configuration information, and setting the basic information corresponding to the region of interest images in the next frame target image based on the second change; and feeding back the exposure parameters corresponding to the next frame target image and the basic parameters corresponding to the region of interest images in the next frame target image as feedback information to the target processing terminal, so that the target processing terminal controls the image acquisition module to acquire multiple region of interest images and corresponding basic information of the next frame target image.

[0009] To address the aforementioned technical problems, another technical solution adopted in this application is: providing an image processing method applied to a second processing end, comprising: receiving and decoding an encoded bitstream sent by a first processing end to obtain multiple region-of-interest (ROI) images, basic information corresponding to each RIO image, and a binding relationship between the RIO images and the corresponding basic information; using the binding relationship and preset configuration information, analyzing and fusing at least two RIO images to determine the analysis result and each frame of target image obtained by fusion.

[0010] In some embodiments, the step of analyzing and fusing at least two region-of-interest (ROI) images using the binding relationship and preset configuration information to determine the analysis result and each frame of target image obtained by fusion includes: using the binding relationship and preset configuration information to determine a target region image with at least two associated ROI images, wherein the at least two associated ROI images are at least two ROI images among multiple ROI images divided into a target image frame; in response to analyzing the target region image, determining the analysis result of the target region image and displaying the analysis result; and using the binding relationship and preset configuration information to fuse multiple ROI image data to determine a target image frame.

[0011] In some embodiments, fusing multiple region-of-interest (ROI) image data using the binding relationship and preset configuration information to determine a target image frame includes: using the binding relationship and preset configuration information to obtain difference information between the basic information corresponding to the multiple ROI images; optimizing the difference information; and then stitching the multiple ROI images into a complete target image frame.

[0012] In some embodiments, the method further includes: in response to the current region of interest image not being the last region of interest image in the target image of the current frame, querying whether an analysis result exists; in response to the existence of the analysis result, outputting the analysis result; or in response to the absence of the analysis result, continuing to receive the encoded bitstream and performing subsequent processing.

[0013] To solve the above-mentioned technical problems, another technical solution adopted in this application is: to provide an image processing method applied to a target processing end, comprising: acquiring a target image and receiving preset configuration information sent by a first processing end; dividing the target image into at least one region of interest image using the preset configuration information, and obtaining basic information corresponding to each region of interest image; in response to the successful division of the target image, transmitting at least one region of interest image and the corresponding basic information to the first processing end.

[0014] In some embodiments, the method further includes: receiving feedback information sent by a first processing terminal, and using the feedback information to adjust the segmentation of the target image in the next frame.

[0015] In some embodiments, the method of dividing the target image includes at least: uniform division, non-uniform division, path division, and complex division.

[0016] To solve the above-mentioned technical problems, another technical solution adopted in this application is: to provide an electronic device, the electronic device including a memory and a processor coupled to the memory, the memory storing at least one computer program, and when the at least one computer program is loaded and executed by the processor, it is used to implement the image processing method as described above.

[0017] To solve the above-mentioned technical problems, another technical solution adopted in this application is to provide a computer-readable storage medium having at least one program, which, when loaded and executed by a processor, is used to implement the image processing method described above.

[0018] Unlike current technologies, the image processing method provided in this application includes: receiving at least one region of interest (ROI) image and basic information corresponding to each ROI image sent by a target processing end; wherein the at least one ROI image is obtained by the target processing end dividing the target image using preset configuration information; in response to successful reception of the target ROI image and corresponding basic information, encoding the successfully received target ROI image and corresponding basic information, and transmitting the encoded bitstream to a second processing end; wherein the target ROI image is one of the at least one ROI image. That is, in this application, by dividing the target image into at least one ROI image and obtaining the basic information of each ROI image, and then encoding and transmitting each ROI image separately, the processing time of each step is reduced, thereby reducing latency and improving the efficiency of image processing. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0020] Figure 1 This is a flowchart illustrating an embodiment of the image processing method applied to the first processing end in this application;

[0021] Figure 2 This is a flowchart illustrating an embodiment of the image processing method applied to the second processing end in this application;

[0022] Figure 3 This is a flowchart illustrating an embodiment of the image processing method applied to the target processing end in this application;

[0023] Figure 4This is a schematic diagram of the structure of an embodiment of an image processing system according to this application;

[0024] Figure 5 This is a schematic diagram of the structure of an embodiment of the electronic device in this application;

[0025] Figure 6 This is a schematic diagram of an embodiment of a computer-readable storage medium in this application. Detailed Implementation

[0026] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be particularly noted that the following embodiments are for illustrative purposes only and do not limit the scope of the invention. Similarly, the following embodiments are only some, not all, embodiments of the present invention, and all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0028] Traditional image processing methods, due to the characteristics of the devices involved, generally introduce at least one frame of latency. Adding the time required for network transmission and decoding, this latency in image acquisition and processing becomes a bottleneck restricting technological development. Current solutions often reduce latency by increasing the frame rate, but this correspondingly increases the dependence on system performance and data transmission bandwidth, impacting image processing efficiency and hindering its further development.

[0029] Therefore, an image processing method is provided, which divides the target image into at least one region of interest (ROI) image, obtains the basic information of each ROI image, and then encodes and transmits each ROI image separately, thereby reducing the processing time of each step, reducing latency, and improving the efficiency of image processing.

[0030] Please see Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the image processing method applied to the first processing end in this application; it should be noted that, if there is a substantial result, the method of this application does not rely on... Figure 1 The sequence of processes shown is limited.

[0031] like Figure 1 As shown, the image processing method applied to the first processing end of this application may include the following steps:

[0032] S10. Receive at least one region of interest image and basic information corresponding to each region of interest image sent by the target processing terminal; wherein, the at least one region of interest image is obtained by the target processing terminal dividing the target image using preset configuration information.

[0033] The target processing end is a processing end used to divide the target image into at least one region of interest (ROI) image; the region of interest image refers to the ROI region, which is divided using the ROI acquisition characteristics of the target processing end and preset configuration information; the basic information refers to the parameter information that characterizes the region of interest.

[0034] Specifically, after the target image is divided into at least one region of interest image using preset configuration information at the target processing end, the first processing end receives the region of interest image sent by the target processing end, as well as the basic information corresponding to each region of interest image.

[0035] S20. In response to the successful reception of the target region of interest image and the corresponding basic information, the successfully received target region of interest image and the corresponding basic information are encoded, and the encoded bitstream is transmitted to the second processing end; wherein, the target region of interest image is one of at least one region of interest image.

[0036] Here, the target region of interest image refers to one of the at least one region of interest images currently accepted; encoding refers to compressing the target region of interest image to obtain the compressed encoded bitstream; and the second processing end refers to the processing end used for decoding and subsequent processing of the encoded bitstream.

[0037] Specifically, during the process of receiving the region of interest image, if a region of interest image and its corresponding basic information are successfully received, it is set as the target region of interest image. In response to the successful reception of the target region of interest image and its corresponding basic information, the first processing end can encode the successfully received target region of interest image and its corresponding basic information, that is, perform image compression on the target region of interest image and its corresponding basic information to obtain the corresponding encoded bitstream, and then transmit the obtained encoded bitstream to the second processing end.

[0038] In some embodiments, there may be multiple target region of interest images, and these multiple target region of interest images can be processed in parallel for subsequent encoding and transmission.

[0039] In this embodiment, the target image is divided into at least one region of interest (ROI) image, and the basic information of each ROI image is obtained. Then, each ROI image is encoded and transmitted separately, reducing the processing time of each step, thereby reducing latency and improving the efficiency of image processing.

[0040] In some embodiments, step S10, receiving at least one region of interest image and basic information corresponding to each region of interest image sent by the target processing terminal, may include the following operations.

[0041] The receiver receives at least two region-of-interest (ROI) images output by the image acquisition module in the target processing unit. The at least two ROI images are obtained by the target processing unit dividing the target image of the current frame using preset configuration information.

[0042] The target processing end may include an image acquisition module for acquiring and outputting target images; the region of interest images output by the target processing end may be two or more.

[0043] Specifically, after the image acquisition module in the target processing terminal acquires the target image and divides the current frame target image into at least two regions of interest images using preset configuration information, the first processing terminal receives the at least two regions of interest images output by the image acquisition module in the target processing terminal.

[0044] It also receives basic information corresponding to each region of interest image output by the parameter configuration module in the target processing end, wherein the basic information represents the segmentation parameters of each region of interest in the target image.

[0045] The target processing end may also include a parameter configuration module, which provides preset configuration information and obtains the basic information corresponding to each region of interest image.

[0046] Specifically, after the parameter configuration module provides preset configuration information to the image acquisition module, the image acquisition module uses the preset configuration information to divide the current frame target image into at least two regions of interest images. Then, the first processing end receives the basic information corresponding to each region of interest image output by the parameter configuration module in the target processing end.

[0047] In some embodiments, the preset configuration information includes at least the following: number of region of interest (ROI) images, size (height and width) of ROI images, starting position, switching / moving mode, etc.; the basic information includes at least the following: starting position, size (height and width) of ROI images, sequence number of each ROI image, and exposure order; wherein, the sequence number may be the division sequence number of the ROI image in the same frame of the target image, and the exposure order may be the exposure order of the ROI images in the same frame of the target image.

[0048] In this embodiment, the first processing end receives at least one region of interest image and corresponding basic information of the target image, effectively dividing each frame of the target image, thereby providing a basis for subsequent encoding, decoding and other processing work to be performed earlier, thus reducing latency and improving the efficiency of image processing.

[0049] In some embodiments, step S20, in response to the successful reception of the target region of interest image and the corresponding basic information, encodes the successfully received target region of interest image and the corresponding basic information, and transmits the encoded bitstream to the second processing end, may include the following operation steps.

[0050] Upon successful reception of any target region of interest image and corresponding basic information, the target region of interest image and corresponding basic information are bound together.

[0051] In this context, binding refers to linking the target region of interest image with its related basic information, so that the corresponding target region of interest images can be distinguished based on the basic information in subsequent processing.

[0052] Specifically, during the process of receiving the region of interest image at the first processing end, in response to the successful reception of any target region of interest and its corresponding basic information, the successfully received target region of interest image and its corresponding basic information are bound together so that they can be processed together in subsequent processing.

[0053] The bound target region of interest image and basic information are encoded, and the encoded bitstream is transmitted to the second processing end.

[0054] Specifically, once a target region of interest image and its corresponding basic information are bound together, subsequent processing can begin, namely, encoding the bound target region of interest image and basic information to obtain the encoded bitstream corresponding to the target region of interest image and basic information, and then the first processing end transmits the obtained encoded bitstream to the second processing end.

[0055] In some embodiments, if there are multiple target region of interest images, the multiple region of interest images and their corresponding basic information can be bound in parallel, and the encoding and transmission processing of the multiple target region of interest images can be performed in parallel.

[0056] In this embodiment, once any target region of interest image and its corresponding basic information are successfully received, the subsequent processing can begin. In the case of multiple target regions of interest images, binding, encoding, and transmission can be performed in parallel, which reduces the processing time for each step, thereby reducing latency and improving the efficiency of image processing.

[0057] Furthermore, the following operations can be included before encoding.

[0058] In response to the fact that the current region of interest image is the last region of interest image in the current frame target image, the first change of the exposure parameters of the target images in different frames is determined by using external influence information, and the exposure parameters corresponding to the target image in the next frame are set according to the first change.

[0059] The determination of whether the current region of interest image is the last region of interest image of the target image in the current frame is to determine whether the segmentation and exposure of the target image in different frames have changed, that is, whether the target image in different frames has been processed differently; external influence information refers to the changes in the external environment that cause changes in the quality of the target image; the first change situation refers to the change between the exposure parameters of the target image in the current frame and the exposure parameters of the target image in the next frame.

[0060] Specifically, in the process of binding the region of interest (ROI) image and its corresponding basic information, it is necessary to determine the sequence number of the current ROI image to ascertain whether it is the last ROI image in the target image of the current frame. When it is determined that the current ROI image is the last ROI image in the target image of the current frame, the first processing end responds to the fact that the current ROI image is the last ROI image in the target image of the current frame by using external influence information to determine the first change of the exposure parameters of the target images of different frames, that is, to determine the first change of the exposure parameters of the target image of the current frame and the target image of the next frame, and then set the exposure information of the ROI image divided by the target image of the next frame.

[0061] For example, if the external environment corresponding to external influence information becomes darker, then it is necessary to increase the exposure time or increase the gain.

[0062] The second change of basic information of the region of interest image in different frame target images is determined by using preset configuration information, and the basic information corresponding to the region of interest image in the next frame target image is set according to the second change.

[0063] The second change refers to the change between the region of interest segmentation information in the current frame target image and the region of interest segmentation information in the next frame target image.

[0064] Specifically, in the process of binding the region of interest image and the corresponding basic information, the second change of the basic information of the region of interest image in different frame target images is determined by using preset configuration information. The basic information corresponding to the region of interest image in the next frame target image is set according to the second change. That is, the second change of the basic information of the current frame target image and the basic information of the next frame target image is determined, and then the basic information of the next frame target image is set, that is, the segmentation information.

[0065] The exposure parameters corresponding to the target image in the next frame and the basic parameters corresponding to the region of interest image in the target image in the next frame are used as feedback information and fed back to the target processing end, so that the target processing end controls the image acquisition module to acquire multiple region of interest images and corresponding basic information of the target image in the next frame.

[0066] The feedback information refers to the exposure parameters and basic parameters corresponding to the next frame target image obtained by the first processing end.

[0067] Specifically, after the first processing end obtains the exposure parameters and basic information corresponding to the next frame target image, it feeds them back to the target processing end as feedback information, so that the target processing end can use the exposure parameters and basic information corresponding to the next frame to control the image acquisition module to acquire multiple regions of interest images of the next frame target image, thereby enabling the first processing end to receive multiple regions of interest images and corresponding basic information of the next frame target image.

[0068] In this embodiment, applied to the first processing end, the target image is divided into at least one region of interest image, and the basic information of each region of interest image is obtained. Then, each region of interest image is encoded, transmitted and processed separately, reducing the processing time of each step, thereby reducing latency and improving the efficiency of image processing.

[0069] This application also provides an image processing method applied to a second processing end.

[0070] See Figure 2 , Figure 2 This is a flowchart illustrating an embodiment of the image processing method applied to the second processing end in this application.

[0071] like Figure 2 As shown, the image processing method applied to the second processing end includes the following operations.

[0072] Y10: Receive and decode the encoded bitstream sent by the first processing end to obtain multiple region of interest (ROI) images, basic information corresponding to each ROI image, and the binding relationship between the ROI images and their corresponding basic information.

[0073] The encoded bitstream is obtained by the first processing end after encoding the region of interest image and its corresponding basic information; the binding relationship refers to the association relationship between the region of interest image and its corresponding basic information.

[0074] Specifically, after the first processing end obtains the encoded bitstream of the region of interest image and the corresponding basic information, the second processing end receives the encoded bitstream sent by the first processing end and decodes the encoded bitstream to obtain multiple region of interest images, the basic information corresponding to each region of interest image, and the binding relationship between the region of interest images and the corresponding basic information.

[0075] Y20. Using the binding relationship and preset configuration information, analyze and fuse at least two region of interest images to determine the analysis results and the target image of each frame obtained by fusion.

[0076] Here, analysis refers to analyzing the information contained in the image of the region of interest; fusion refers to merging multiple region of interest images into a complete target image.

[0077] Specifically, the second processing end uses binding information and preset configuration information to analyze at least two region of interest images to obtain corresponding analysis results, such as the content contained in the region of interest images; and to fuse at least two region of interest images, that is, to stitch at least two region of interest images into a complete target image frame.

[0078] In this embodiment, the method is applied to the second processing end, namely the decoding end, to analyze and fuse multiple acquired region of interest images. That is, each region of interest is processed separately to improve the parallel processing degree, thereby reducing latency and improving the efficiency of image processing.

[0079] In some embodiments, step Y20 uses the binding relationship and preset configuration information to analyze and fuse at least two region of interest images, and determines the analysis results and the target image of each frame obtained by fusion, which may include the following operations.

[0080] Using binding relationships and preset configuration information, a target region image is determined from at least two associated region-of-interest (ROI) images, wherein the at least two associated ROI images are at least two of the multiple ROI images divided into a single frame of the target image.

[0081] Among them, the at least two associated region of interest images can be a target that can be determined by multiple region of interest images, such as a car. If a complete frame of target image is divided into multiple region of interest images, then the part corresponding to the target is at least a part of a frame of target image.

[0082] Specifically, the second processor uses the binding relationship and preset configuration information to determine the target region image corresponding to a target from at least two associated region of interest images.

[0083] In response to the analysis of the target region image, the analysis results of the target region image are determined and displayed.

[0084] The target region image can correspond to a complete target.

[0085] Specifically, the target area image is analyzed to determine the analysis results of the target area image for the complete target, and then the analysis results of the complete target are displayed, for example, by showing the result.

[0086] By using binding relationships and preset configuration information, image data from multiple regions of interest are fused to determine a single target image frame.

[0087] Fusion refers to the process of restoring multiple regions of interest (ROIs) within the same frame into a single, complete target image.

[0088] Specifically, the second processing end uses the binding relationship and preset configuration information to determine the multiple regions of interest (ROI) images divided into each frame of the target image, and then merges the corresponding multiple ROI images to obtain a complete frame of the target image.

[0089] Furthermore, by utilizing the binding relationship and preset configuration information, multiple regions of interest image data can be fused to determine a target image frame, which may also include the following operations.

[0090] By utilizing the binding relationship and preset configuration information, the difference information between the basic information corresponding to multiple regions of interest images is obtained.

[0091] Since the exposure start time of each region of interest image is different, there will be slight misalignment, distortion or brightness difference in each region of interest image, and it is necessary to obtain the corresponding difference information.

[0092] Specifically, before fusion, it is necessary to bind relationships and preset configuration information to obtain difference information between related regions of interest images.

[0093] The difference information is optimized, and then multiple region-of-interest images are stitched together into a complete target image.

[0094] The process involves optimizing the difference information, specifically addressing subtle misalignments, distortions, or brightness differences in each region of interest image. These optimized region of interest images are then stitched together to obtain a complete, error-free target image.

[0095] In some embodiments, the following operations may also be included.

[0096] In response to the current region of interest image not being the last region of interest image in the target image of the current frame, query whether there are analysis results.

[0097] The step of determining whether the current region of interest (ROI) image is the last ROI image in the current frame of the target image is to ensure that, in the absence of a complete frame of the target image, a partial ROI can identify a complete target in a target image.

[0098] If an analysis result exists, the analysis result will be output.

[0099] Specifically, if a complete objective exists, an analysis of the complete objective is performed to obtain the analysis results, and the analysis results are output and displayed.

[0100] Alternatively, if no analysis result is found, the system continues to receive the encoded bitstream and perform subsequent processing.

[0101] Specifically, if no analysis result is found, the encoded bitstream continues to be received. After waiting for multiple region of interest images corresponding to a complete frame of the target image, the complete frame of the target image is analyzed to obtain the analysis result of the complete frame of the target image.

[0102] In this embodiment, the second processing end, namely the decoding end, is applied to obtain multiple region of interest images based on the received encoded bitstream, and then perform analysis and fusion processing simultaneously. That is, multiple region of interest images can be processed in parallel at each stage, reducing latency and thus effectively improving image processing efficiency.

[0103] This application also provides an image processing method applied to the target processing end.

[0104] See Figure 3 , Figure 3 This is a flowchart illustrating an embodiment of the image processing method applied to the target processing end in this application.

[0105] like Figure 3 As shown, the image processing method applied to the target processing end may include the following operations.

[0106] Z10: Acquire the target image and receive the preset configuration information sent by the first processing terminal.

[0107] The target processing end includes an image acquisition module and a parameter configuration module. The image acquisition module is used to acquire and output images, while the parameter configuration module is used to provide configuration information so that the image acquisition module outputs according to the configuration information.

[0108] Specifically, the image acquisition module of the target processing end acquires the target image, while the parameter configuration module of the target processing end receives the preset configuration information or feedback information sent by the first processing end, and then adjusts the configuration information according to the feedback information.

[0109] Z20. Divide the target image into at least one region of interest image using the preset configuration information, and obtain the basic information corresponding to each region of interest image.

[0110] Specifically, after acquiring the target image, the image acquisition module uses preset configuration information to divide the target image into at least one region of interest image, and the parameter configuration module acquires the basic information corresponding to each region of interest image.

[0111] Z30, in response to the successful segmentation of the target image, transmits at least one region of interest image and corresponding basic information to the first processing end.

[0112] The first processing end is the encoding end.

[0113] Specifically, after successfully dividing a complete frame of target image, at least one region of interest image and its corresponding basic information are transmitted to the first processing end for encoding and other processing.

[0114] In this embodiment, a complete frame of target image is divided into at least one region of interest (ROI) image using preset configuration information. Then, the ROI image corresponding to the complete frame of target image is transmitted to subsequent processing. After being divided into multiple ROI images, each ROI image can be processed subsequently, reducing the processing time for each frame of target image, i.e., reducing latency, and thus improving image processing efficiency.

[0115] In some embodiments, the following operations may also be included.

[0116] It receives feedback information sent by the first processing end and uses the feedback information to adjust the segmentation of the target image in the next frame.

[0117] The feedback information is determined by the first processing unit based on changes in exposure parameters and basic information.

[0118] Specifically, after receiving the feedback information, the parameter configuration module of the target processing end uses the feedback information to adjust the preset configuration information, thereby enabling the image acquisition module to output the corresponding region of interest image using the adjusted configuration information.

[0119] That is, when the division of the region of interest required by the target image in the current frame and the target image in the next frame is different, the above adjustment operation is performed.

[0120] In some embodiments, the following operations may also be included.

[0121] The methods for dividing a target image include at least: uniform division, non-uniform division, path division, and complex division.

[0122] Uniform partitioning refers to dividing each region of interest (ROI) image in the same frame of the target image into images of uniform size (height and width), suitable for scenarios requiring low-latency full-scene analysis and processing; non-uniform partitioning, on the other hand, randomly divides a frame of the target image into multiple ROI images, each of which can have different sizes (height and width), suitable for scenarios where low-latency triggering of scene analysis of one or more fixed associated regions, such as ROI0 for intrusion analysis and ROI1 for departure analysis; path partitioning is based on the target's motion trajectory, with widths that can be the same and can be smaller than the width of the complete frame of the target image, suitable for scenarios where the target moves along a specified path; complex partitioning can be based on multiple different targets, with partial overlap and different heights and widths, suitable for complex scenarios with multiple analysis requirements simultaneously.

[0123] In this embodiment, the target image is divided into multiple regions of interest (ROI) images using preset configuration information. This provides less than one frame of ROI images for subsequent encoding, transmission, and decoding of each ROI image. Furthermore, multiple ROI images can be processed in parallel, reducing latency and improving image processing efficiency.

[0124] In some embodiments, an image processing system is also provided.

[0125] See Figure 4 , Figure 4 This is a schematic diagram of an embodiment of an image processing system according to this application.

[0126] like Figure 4 As shown, the image processing system 400 includes a target processing terminal 410, a first processing terminal 420, and a second processing terminal 430. The target processing terminal 410 includes an image acquisition module 411 and a parameter configuration module 412.

[0127] Specifically, the parameter configuration module 412 in the target processing end 410 provides preset configuration information to the image acquisition module 411, thereby enabling the image acquisition module 411 to divide the acquired target image into multiple regions of interest (ROI) images and output them according to the configuration information. The first processing end 420 receives the multiple ROI images and the basic information corresponding to each ROI image output by the target processing end 410, and then encodes each ROI and the corresponding basic information, and transmits the encoded bitstream to the second processing end 430. The second processing end 430 receives the encoded bitstream and decodes it to obtain multiple ROI images and the corresponding basic information. It can then analyze and fuse the ROI images to obtain the analysis results of some regions, fuse them to obtain a complete frame of target image and the analysis results of the target image, and display the analysis results.

[0128] This application also provides an electronic device.

[0129] Please see Figure 5 , Figure 5 This is a schematic diagram of an embodiment of the electronic device in this application. This electronic device can perform the steps of the image processing method described above.

[0130] The electronic device 500 includes a memory 520, a processor 510 coupled to the memory, and at least one computer program stored in the memory 520 and executable on the processor 510. When the processor 510 loads and executes the at least one computer program, it implements the image processing steps in the above method. For related details, please refer to the detailed description in the above method; further elaboration will not be repeated here.

[0131] This application also includes a computer-readable storage medium.

[0132] Please see Figure 6 , Figure 6 This is a schematic diagram of an embodiment of a computer-readable storage medium in this application.

[0133] The computer-readable storage medium 600 stores at least one program 610, which, when loaded and executed by a processor, is used to implement the image processing steps in the above method. For related details, please refer to the detailed description in the above method; it will not be repeated here.

[0134] The above scheme divides the target image into at least one region of interest (ROI) image, obtains the basic information of each ROI image, and then encodes and transmits each ROI image separately, reducing the processing time of each step, thereby reducing latency and improving the efficiency of image processing.

[0135] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection of apparatuses or units, and may be electrical, mechanical, or other forms.

[0136] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0137] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0138] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0139] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. An image processing method, characterized in that, Applied to the first processing end, including: The system receives at least one region of interest (ROI) image and basic information corresponding to each ROI image from a target processing terminal; wherein the at least one ROI image is obtained by the target processing terminal dividing a target image using preset configuration information, and the basic information includes at least the sequence number and exposure order of each ROI image; In response to successful reception of the target region of interest image and the corresponding basic information, the target region of interest image and the corresponding basic information are bound together, the bound target region of interest image and the corresponding basic information are encoded, and the encoded bitstream is transmitted to the second processing end; wherein, the target region of interest image is one of the at least one region of interest images.

2. The method according to claim 1, characterized in that, The receiving target processing terminal sends at least one region of interest image and basic information corresponding to each region of interest image, including: The target processing terminal receives at least two region of interest images output by the image acquisition module, wherein the at least two region of interest images are obtained by the target processing terminal dividing the current frame target image using preset configuration information; And receive the basic information corresponding to each region of interest image output by the parameter configuration module in the target processing terminal, wherein the basic information represents the segmentation parameters of each region of interest in the target image.

3. The method according to claim 1, characterized in that, Before encoding, the following is also included: In response to the fact that the current region of interest image is the last region of interest image in the current frame target image, the first change of the exposure parameters of the target images in different frames is determined by using external influence information, and the exposure parameters corresponding to the next frame target image are set according to the first change. The second change of the basic information of the region of interest image in different frame target images is determined by using preset configuration information, and the basic information corresponding to the region of interest image in the next frame target image is set according to the second change. The exposure parameters corresponding to the target image in the next frame and the basic parameters corresponding to the region of interest image in the target image in the next frame are fed back to the target processing end as feedback information, so that the target processing end controls the image acquisition module to acquire multiple region of interest images and corresponding basic information of the target image in the next frame.

4. An image processing method, characterized in that, Applied to the second processing end, including: The system receives and decodes the encoded bitstream sent by the first processing end to obtain multiple region of interest (ROI) images, basic information corresponding to each ROI image, and the binding relationship between the ROI images and the corresponding basic information. The basic information includes at least the sequence number and exposure order of each ROI image, and the binding relationship is obtained when the ROI image and the corresponding basic information are successfully received and bound. Using the binding relationship and preset configuration information, at least two region of interest (ROI) images are analyzed and fused to determine the analysis results and each frame of the target image obtained by fusion. The binding relationship and preset configuration information are used to obtain the difference information between the basic information corresponding to the multiple ROI images. The difference information includes misalignment, distortion, or brightness differences caused by the difference in the exposure start time of each ROI image. The difference information is optimized, and then the multiple ROI images are stitched together into a complete frame of the target image.

5. The method according to claim 4, characterized in that, The step of analyzing and fusing at least two region-of-interest images using the binding relationship and preset configuration information, and determining the analysis results and each frame of the target image obtained by fusion, includes: Using the binding relationship and preset configuration information, a target region image is determined by at least two associated region of interest images, wherein the at least two associated region of interest images are at least two region of interest images among multiple region of interest images divided in a frame of target image; In response to analyzing the target region image, determining the analysis result of the target region image, and displaying the analysis result; Using the binding relationship and preset configuration information, multiple image data of the region of interest are fused to determine a target image frame.

6. The method according to claim 5, characterized in that, Also includes: In response to the fact that the current region of interest image is not the last region of interest image in the target image of the current frame, query whether there is an analysis result; If the analysis result exists, then the analysis result is output; Alternatively, if the analysis result is not found, the system continues to receive the encoded bitstream and perform subsequent processing.

7. An image processing method, characterized in that, Applied to the target processing end, including: Acquire the target image and receive the preset configuration information sent by the first processing end; The target image is divided into at least one region of interest image using preset configuration information, and basic information corresponding to each region of interest image is obtained, wherein the basic information includes at least the sequence number and exposure order of each region of interest image; In response to the successful segmentation of the target image, at least one region of interest image and the corresponding basic information are transmitted to the first processing end. The target image is obtained by stitching together multiple regions of interest images after optimizing the difference information between the basic information corresponding to multiple regions of interest images using binding relationships and preset configuration information. The binding relationship is obtained by successfully receiving and binding the region of interest images and the corresponding basic information.

8. The method according to claim 7, characterized in that, Also includes: The system receives feedback information from the first processing unit and uses the feedback information to adjust the segmentation of the target image in the next frame.

9. The method according to claim 7, characterized in that, Also includes: The methods for dividing the target image include at least: uniform division, non-uniform division, path division, and complex division.

10. An electronic device, characterized in that, The electronic device includes a memory and a processor coupled to the memory, the memory storing at least one computer program, which, when loaded and executed by the processor, is used to implement the image processing method as described in any one of claims 1-9.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium has at least one program that, when loaded and executed by a processor, is used to implement the image processing method as described in any one of claims 1-9.

Images