A multi-frame image fusion method and fusion system based on frame-accurate technology

Abstract

The present invention belongs to the field of image processing technology, and proposes a multi-frame image fusion method and a fusion system based on frame precision technology. The main scheme is: when the basic first image acquisition unit senses the target to be monitored, multiple first image information of the target to be monitored is simultaneously acquired at the first set frame node through multiple first image acquisition units, and / or the second image acquisition unit is started by the edge computing gateway control, and multiple second image information of the target to be monitored is simultaneously acquired at the second set frame node through the second image acquisition unit; after the multiple first image information and / or the multiple second image information are aligned and preprocessed, image fusion is performed to obtain a fused image, and the fused image is sent to the server as the final image for storage. The present invention only needs to integrate the corresponding image acquisition unit and edge computing gateway on the basis of the existing hardware configuration, and does not need to directly perform image fusion on the service, which can reduce the processing burden of the server.

Description

Technical Field

[0001] The present invention relates to the field of fluid measurement technology, and in particular to a multi-frame image fusion method based on frame precision technology and a fusion system thereof. Background Art

[0002] Image fusion refers to the process of extracting key information from each channel through image processing technology from image data of the same target acquired by multiple source channels, and fusing them into the image required by the user to assist the user in observing the image.

[0003] However, most of the image data that need to be fused currently comes directly from the image acquisition device. When more image acquisition devices are set up for the same target, the traditional image fusion technology also needs to build a data set and train a neural network to achieve better fusion. In addition, the data set construction and neural network training process, and even the image fusion process, are all completed in the server. In many cases, a server will bear a large amount of command sending and receiving, data storage and data processing functions. Therefore, traditional image fusion methods will occupy a higher processing burden on the server, which may cause delays in the rest of the server functions or even cause the server to crash.

[0004] Therefore, there is an urgent need for a solution for multi-frame images that can reduce the processing burden of the server. Summary of the invention

[0005] The purpose of the present invention is to provide a multi-frame image fusion method and a fusion system based on frame-precision technology. It only needs to integrate the corresponding image acquisition unit and edge computing gateway on the basis of the existing hardware configuration. The start timing of the additional integrated image acquisition unit can be controlled in the edge computing gateway. After the image acquisition unit acquires the image to be fused of the same target, the image fusion can be performed in the edge computing gateway. There is no need to perform image fusion directly in the service, which can reduce the processing burden of the server.

[0006] The present invention solves the technical problem and adopts the following technical solution:

[0007] On the one hand, the present invention provides a multi-frame image fusion method based on frame accurate technology, comprising the following steps:

[0008] Acquire position information and quantity information of a plurality of first image acquisition units of a target to be monitored in a scene to be monitored, wherein each first image acquisition unit is integrated with a corresponding second image acquisition unit;

[0009] Selecting a basic first image acquisition unit from a first image acquisition unit having the largest number of features to be extracted based on the target to be monitored;

[0010] An edge computing gateway is set based on the quantity information of all first image acquisition units and the position information of the basic first image acquisition unit, and the first image acquisition unit is controlled to be in a startup state and the second image acquisition unit is in a standby state through the edge computing gateway;

[0011] When the basic first image acquisition unit senses the target to be monitored, multiple first image information of the target to be monitored is simultaneously acquired at the first set frame node by multiple first image acquisition units, and / or the second image acquisition unit is started by the edge computing gateway control, and multiple second image information of the target to be monitored is simultaneously acquired at the second set frame node by the second image acquisition unit;

[0012] After registering and preprocessing the plurality of first image information and / or the plurality of second image information, image fusion is performed to obtain a fused image, and the fused image is sent as a final image to a server for storage.

[0013] As a further optimization, the resolution of the first image acquisition unit is different from the resolution of the second image acquisition unit integrated thereon.

[0014] As a further optimization, selecting a basic first image acquisition unit from a first image acquisition unit having the largest number of features to be extracted based on the target to be monitored includes the following steps:

[0015] Presetting the number of features to be extracted of the target to be monitored and setting the target to be monitored to arrive at a specified position of the scene to be monitored;

[0016] When the target to be monitored appears at a specified position, each first image information acquisition unit is used to acquire preset first image information;

[0017] Extract features from each piece of first image information respectively, and sort the extracted features from large to small;

[0018] A first image acquisition unit with the largest number of features is selected and used as a basic first image acquisition unit.

[0019] As a further optimization, the edge computing gateway is set based on the quantity information of all first image acquisition units and the location information of the basic first image acquisition unit, which means:

[0020] Obtaining the quantity information of all the first image acquisition units in the area to be monitored, and determining whether the quantity of all the first image acquisition units in the area to be monitored reaches a specified quantity, and if so, setting the edge computing gateway in the area to be monitored;

[0021] Obtain the location information of the basic first image acquisition unit. When the edge computing gateway is set in the area to be monitored, draw a straight line between the center of mass of the edge computing gateway and the center of mass of the basic first image acquisition unit. The straight line distance is lower than the first preset distance. When the edge computing gateway is set outside the area to be monitored, the straight line distance is lower than the second preset distance. The first distance is less than the second preset distance.

[0022] As a further optimization, after registering and preprocessing the plurality of first image information, performing image fusion to obtain a fused image, and sending the fused image as a final image to a server for storage, means:

[0023] A plurality of first image information of the target to be monitored is simultaneously acquired at a first set frame node by a plurality of first image acquisition units, and after registering and preprocessing the plurality of first image information, image fusion is performed to obtain a first fused image;

[0024] It is determined whether the number of features contained in the first fused image reaches a first preset number. If so, the first fused image is sent as a final image to a server for storage.

[0025] As a further optimization, after registering and preprocessing the plurality of second image information, performing image fusion to obtain a fused image, and sending the fused image as a final image to a server for storage, means:

[0026] When the number of features contained in the first fused image does not reach the first preset number, the second image acquisition unit is started through the edge computing gateway control, and the second image acquisition unit simultaneously acquires multiple second image information of the target to be monitored at the second set frame node, and after registering and preprocessing the multiple second image information, image fusion is performed to obtain a second fused image;

[0027] It is determined whether the number of features contained in the second fused image reaches a second preset number. If so, the second fused image is sent as a final image to the server for storage.

[0028] As a further optimization, after registering and preprocessing the plurality of first image information and the plurality of second image information, performing image fusion to obtain a fused image, and sending the fused image as a final image to the server for storage, means:

[0029] When the number of features contained in the second fused image does not reach the second preset number, the first image acquisition unit and the second image processing unit are started simultaneously, and the first fused image and the second fused image are fused to obtain a third fused image, and the third fused image is sent to the server as the final image for storage.

[0030] As a further optimization, the registering of the first image information means: taking the first image information acquired by the basic first image acquisition unit as a reference, matching the first image information acquired by other first image acquisition units to the reference first image information;

[0031] The registering of the second image information refers to: selecting a basic second image acquisition unit from a second image acquisition unit having the largest number of features to be extracted based on the target to be monitored, taking the second image information acquired by the basic second image acquisition unit as a reference, and matching the second image information acquired by other second image acquisition units to the reference second image information;

[0032] The registering of the first image information and the second image information refers to: selecting a basic image acquisition unit from an image acquisition unit with the largest number of features to be extracted based on the target to be monitored, and matching the image information acquired by other image acquisition units to the benchmark image information based on the image information acquired by the basic image acquisition unit.

[0033] As a further optimization, preprocessing the plurality of first image information and / or the plurality of second image information means:

[0034] Prior knowledge of the scene to be monitored is obtained, and the prior knowledge is added to the first image information and / or the plurality of second image information.

[0035] On the other hand, the present invention further provides a multi-frame image fusion system based on frame-accurate technology, which is applied to the multi-frame image fusion method based on frame-accurate technology, comprising:

[0036] An image acquisition unit acquisition module, used to acquire position information and quantity information of a plurality of first image acquisition units of a target to be monitored in a scene to be monitored, wherein each first image acquisition unit is integrated with a corresponding second image acquisition unit;

[0037] A basic first image acquisition unit acquisition module, used to select a basic first image acquisition unit from a first image acquisition unit having the largest number of features to be extracted based on the target to be monitored;

[0038] The edge gateway is used to set the edge computing gateway based on the quantity information of all the first image acquisition units and the location information of the basic first image acquisition units, and control the first image acquisition units to be in the startup state and the second image acquisition units to be in the standby state through the edge computing gateway;

[0039] An image information acquisition module, which is used to simultaneously acquire multiple first image information of the target to be monitored at a first set frame node through multiple first image acquisition units when the basic first image acquisition unit senses the target to be monitored, and / or to control the start of the second image acquisition unit through the edge computing gateway, and simultaneously acquire multiple second image information of the target to be monitored at a second set frame node through the second image acquisition unit;

[0040] The image fusion module is used to perform image fusion after registering and preprocessing the plurality of first image information and / or the plurality of second image information to obtain a fused image, and send the fused image as a final image to the server for storage.

[0041] The beneficial effects of the present invention are as follows: through the above-mentioned multi-frame image fusion method based on frame precision technology and its fusion system, firstly, the position information and quantity information of multiple first image acquisition units of the monitored target in the monitored scene are obtained, and each first image acquisition unit is integrated with a corresponding second image acquisition unit; secondly, a basic first image acquisition unit is selected from the first image acquisition unit with the largest number of features to be extracted of the monitored target, and an edge computing gateway is set based on the quantity information of all the first image acquisition units and the position information of the basic first image acquisition unit, and the first image acquisition unit is controlled to be in the start-up state and the second image acquisition unit is in the standby state through the edge computing gateway.

[0042] Then, when the basic first image acquisition unit senses the target to be monitored, multiple first image information of the target to be monitored is simultaneously acquired at the first set frame node through multiple first image acquisition units, and / or the second image acquisition unit is started through the edge computing gateway control, and multiple second image information of the target to be monitored is simultaneously acquired at the second set frame node through the second image acquisition unit. Finally, after the multiple first image information and / or multiple second image information are registered and preprocessed, image fusion is performed to obtain a fused image, and the fused image is sent to the server as the final image for storage.

[0043] Therefore, the present invention only needs to integrate the corresponding image acquisition unit and edge computing gateway on the basis of the existing hardware configuration, without the need to perform image fusion directly on the service, which can reduce the processing burden of the server. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flow chart of a multi-frame image fusion method based on frame-accurate technology in Example 1 of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, 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 in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0046] Example 1

[0047] This embodiment provides a multi-frame image fusion method based on frame-accurate technology, and its flow chart is shown in Figure 1 , wherein the method comprises the following steps:

[0048] S1, obtaining position information and quantity information of a plurality of first image acquisition units of a target to be monitored in a scene to be monitored, wherein each first image acquisition unit is integrated with a corresponding second image acquisition unit;

[0049] S2, selecting a basic first image acquisition unit from a first image acquisition unit having the largest number of features to be extracted based on the target to be monitored;

[0050] S3, setting an edge computing gateway based on the quantity information of all first image acquisition units and the location information of the basic first image acquisition units, and controlling the first image acquisition units to be in a startup state and the second image acquisition units to be in a standby state through the edge computing gateway;

[0051] S4, when the basic first image acquisition unit senses the target to be monitored, multiple first image information of the target to be monitored is simultaneously acquired at the first setting frame node by multiple first image acquisition units, and / or the second image acquisition unit is started by the edge computing gateway control, and multiple second image information of the target to be monitored is simultaneously acquired at the second setting frame node by the second image acquisition unit;

[0052] S5. After registering and preprocessing the plurality of first image information and / or the plurality of second image information, perform image fusion to obtain a fused image, and send the fused image as a final image to a server for storage.

[0053] In the above method, since the scene to be monitored is relatively flexible and the target to be monitored is set according to the nature of the scene to be monitored, different numbers of first image acquisition units can be installed for different targets to be monitored. In order to achieve image fusion, it is necessary to ensure that there are at least two images to be fused. In this embodiment, each first image acquisition unit will only acquire one image. Therefore, there are at least two first image acquisition units. In actual application, the first image acquisition unit is an image acquisition terminal containing an image sensor.

[0054] In the traditional solution, after the image to be fused is acquired by the image acquisition unit, the image to be fused will be immediately transmitted to the server for alignment and preprocessing operations, and then the image fusion will be completed in the server. After the image fusion, it will also be stored in the server for the client to call. Therefore, when the server is responsible for the image fusion of multiple monitored areas or multiple monitored targets, there will be a large data processing pressure. Therefore, in this embodiment, an edge computing gateway will be set up to alleviate the computing pressure of the server. The entire image fusion process is implemented in the edge computing gateway, and the server is only responsible for storing the fused image information.

[0055] It should be pointed out that, after the installation, the traditional first image acquisition unit will continue to acquire the image to be fused. However, with the development of electronic technology, the frequency of updating and replacing the first image acquisition unit (such as image sensors) is getting higher and higher. This results in that not all first image acquisition units have the same resolution. The long-term use of the first image acquisition unit will result in the acquisition of the image information to be fused. The image features contained therein will be relatively fewer and fewer, and some may even no longer be suitable for the monitored scene. Therefore, the frequency of equipment replacement of the first image acquisition unit will be accelerated, thereby increasing labor costs. Therefore, in this embodiment, the second image acquisition unit can be integrated in the installed first image acquisition unit. Since the second image acquisition unit is integrated in the first image acquisition unit, it can be easily replaced as the scene to be monitored differs and the target to be monitored changes.

[0056] At the same time, since the second image acquisition unit is added, it is adapted to the scene and target to be monitored during integration. Therefore, in general, the resolution of the first image acquisition unit is different from the resolution of the second image acquisition unit integrated thereon. The difference in resolution can also facilitate the acquisition of more feature information in the image to be fused during subsequent image fusion.

[0057] Generally speaking, due to the excellent computing power of the edge computing gateway, an edge computing gateway can be responsible for the data processing of multiple monitored scenes and multiple monitored targets at the same time. However, in this embodiment, after using the first image information to perform image fusion in the edge computing gateway, it is also necessary to control the opening of the second image acquisition unit, and it is also necessary to send the final fused image to the server after determining it. Therefore, the setting position of the edge gateway is as adaptable to the monitored scene as possible. In addition, since a basic first image acquisition unit will be selected when performing image fusion in this embodiment, after the basic first image acquisition unit obtains the image information to be fused, it is also necessary to align the image information to be fused obtained by other first image acquisition units based on this. Therefore, in this embodiment, selecting a basic first image acquisition unit from a first image acquisition unit with the largest number of features to be extracted based on the target to be monitored can include the following steps:

[0058] Presetting the number of features to be extracted of the target to be monitored and setting the target to be monitored to arrive at a specified position of the scene to be monitored;

[0059] When the target to be monitored appears at a specified position, each first image information acquisition unit is used to acquire preset first image information;

[0060] Extract features from each piece of first image information respectively, and sort the extracted features from large to small;

[0061] A first image acquisition unit with the largest number of features is selected and used as a basic first image acquisition unit.

[0062] Therefore, setting the edge computing gateway based on the quantity information of all first image acquisition units and the location information of the basic first image acquisition unit may refer to:

[0063] Obtaining the quantity information of all the first image acquisition units in the area to be monitored, and determining whether the quantity of all the first image acquisition units in the area to be monitored reaches a specified quantity, and if so, setting the edge computing gateway in the area to be monitored;

[0064] Obtain the location information of the basic first image acquisition unit. When the edge computing gateway is set in the area to be monitored, draw a straight line between the center of mass of the edge computing gateway and the center of mass of the basic first image acquisition unit. The straight line distance is lower than the first preset distance. When the edge computing gateway is set outside the area to be monitored, the straight line distance is lower than the second preset distance. The first distance is less than the second preset distance.

[0065] It should be pointed out that after obtaining the image information to be fused, it is necessary to perform registration to meet the consistency in time and space. Moreover, after registration, for some images with prior knowledge, the prior knowledge representation can be added to the image through preprocessing operations. In this way, the resulting images can be fused to obtain better results.

[0066] Therefore, in this embodiment, after registering and preprocessing the plurality of first image information, performing image fusion to obtain a fused image, and sending the fused image as a final image to a server for storage, may refer to:

[0067] A plurality of first image information of the target to be monitored is simultaneously acquired at a first set frame node by a plurality of first image acquisition units, and after registering and preprocessing the plurality of first image information, image fusion is performed to obtain a first fused image;

[0068] It is determined whether the number of features contained in the first fused image reaches a first preset number. If so, the first fused image is sent as a final image to a server for storage.

[0069] In addition, after registering and preprocessing the plurality of second image information, performing image fusion to obtain a fused image, and sending the fused image as a final image to a server for storage may refer to:

[0070] When the number of features contained in the first fused image does not reach the first preset number, the second image acquisition unit is started through the edge computing gateway control, and the second image acquisition unit simultaneously acquires multiple second image information of the target to be monitored at the second set frame node, and after registering and preprocessing the multiple second image information, image fusion is performed to obtain a second fused image;

[0071] It is determined whether the number of features contained in the second fused image reaches a second preset number. If so, the second fused image is sent as a final image to the server for storage.

[0072] In addition, after registering and preprocessing the plurality of first image information and the plurality of second image information, performing image fusion to obtain a fused image, and sending the fused image as a final image to a server for storage may refer to:

[0073] When the number of features contained in the second fused image does not reach the second preset number, the first image acquisition unit and the second image processing unit are started simultaneously, and the first fused image and the second fused image are fused to obtain a third fused image, and the third fused image is sent to the server as the final image for storage.

[0074] Specifically, in this embodiment, the registering of the first image information may refer to: taking the first image information acquired by the basic first image acquisition unit as a reference, matching the first image information acquired by other first image acquisition units to the reference first image information;

[0075] The registering of the second image information refers to: selecting a basic second image acquisition unit from a second image acquisition unit having the largest number of features to be extracted based on the target to be monitored, taking the second image information acquired by the basic second image acquisition unit as a reference, and matching the second image information acquired by other second image acquisition units to the reference second image information;

[0076] The registering of the first image information and the second image information refers to: selecting a basic image acquisition unit from an image acquisition unit with the largest number of features to be extracted based on the target to be monitored, and matching the image information acquired by other image acquisition units to the benchmark image information based on the image information acquired by the basic image acquisition unit.

[0077] After the image registration operation, a preprocessing operation needs to be performed to obtain a better image fusion result. Therefore, in this embodiment, preprocessing the plurality of first image information and / or the plurality of second image information may refer to:

[0078] Prior knowledge of the scene to be monitored is obtained, and the prior knowledge is added to the first image information and / or the plurality of second image information.

[0079] Example 2

[0080] Based on Example 1, this embodiment provides a multi-frame image fusion system based on frame-accurate technology, including:

[0081] An image acquisition unit acquisition module, used to acquire position information and quantity information of a plurality of first image acquisition units of a target to be monitored in a scene to be monitored, wherein each first image acquisition unit is integrated with a corresponding second image acquisition unit;

[0082] A basic first image acquisition unit acquisition module, used to select a basic first image acquisition unit from a first image acquisition unit having the largest number of features to be extracted based on the target to be monitored;

[0083] The edge gateway is used to set the edge computing gateway based on the quantity information of all the first image acquisition units and the location information of the basic first image acquisition units, and control the first image acquisition units to be in the startup state and the second image acquisition units to be in the standby state through the edge computing gateway;

[0084] An image information acquisition module, which is used to simultaneously acquire multiple first image information of the target to be monitored at a first set frame node through multiple first image acquisition units when the basic first image acquisition unit senses the target to be monitored, and / or to control the start of the second image acquisition unit through the edge computing gateway, and simultaneously acquire multiple second image information of the target to be monitored at a second set frame node through the second image acquisition unit;

[0085] The image fusion module is used to perform image fusion after registering and preprocessing the plurality of first image information and / or the plurality of second image information to obtain a fused image, and send the fused image as a final image to the server for storage.

[0086] Referring to Example 1, the application scenario and implementation principle of this embodiment are consistent with those of Example 1, and thus will not be described in detail.

[0087] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A multi-frame image fusion method based on frame-accurate technology, characterized in that: The steps include: Acquire position information and quantity information of a plurality of first image acquisition units of a target to be monitored in a scene to be monitored, each first image acquisition unit being integrated with a corresponding second image acquisition unit; Selecting a basic first image acquisition unit from a first image acquisition unit having the largest number of features to be extracted based on the target to be monitored; An edge computing gateway is set based on the quantity information of all first image acquisition units and the position information of the basic first image acquisition unit, and the first image acquisition unit is controlled to be in a startup state and the second image acquisition unit is in a standby state through the edge computing gateway; When the basic first image acquisition unit senses the target to be monitored, multiple first image information of the target to be monitored is simultaneously acquired at the first set frame node by multiple first image acquisition units, and / or the second image acquisition unit is started by the edge computing gateway control, and multiple second image information of the target to be monitored is simultaneously acquired at the second set frame node by the second image acquisition unit; After registering and preprocessing the plurality of first image information and / or the plurality of second image information, image fusion is performed to obtain a fused image, and the fused image is sent as a final image to a server for storage; The setting of the edge computing gateway based on the quantity information of all the first image acquisition units and the position information of the basic first image acquisition unit refers to: obtaining the quantity information of all the first image acquisition units in the area to be monitored, and judging whether the quantity of all the first image acquisition units in the area to be monitored reaches the prescribed quantity, and when it reaches the prescribed quantity, setting the edge computing gateway in the area to be monitored; obtaining the position information of the basic first image acquisition unit, and when the edge computing gateway is set in the area to be monitored, drawing a straight line between the centroid of the edge computing gateway and the centroid of the basic first image acquisition unit, the straight line distance is lower than the first preset distance, and when the edge computing gateway is set outside the area to be monitored, the straight line distance is lower than the second preset distance, and the first preset distance is smaller than the second preset distance.

2. The multi-frame image fusion method based on frame-accurate technology according to claim 1, characterized in that: The resolution of the first image acquisition unit is different from the resolution of the second image acquisition unit integrated thereon.

3. The multi-frame image fusion method based on frame-accurate technology according to claim 1, characterized in that: The step of selecting a basic first image acquisition unit from a first image acquisition unit having the largest number of features to be extracted based on the target to be monitored comprises the following steps: Presetting the number of features to be extracted of the target to be monitored and setting the target to be monitored to arrive at a specified position of the scene to be monitored; When the target to be monitored appears at a specified position, each first image information acquisition unit is used to acquire preset first image information; Extract features from each piece of first image information respectively, and sort the extracted features from large to small; A first image acquisition unit with the largest number of features is selected and used as a basic first image acquisition unit.

4. The multi-frame image fusion method based on frame-accurate technology according to claim 1, characterized in that: After registering and preprocessing the plurality of first image information, performing image fusion to obtain a fused image, and sending the fused image as a final image to a server for storage, means: A plurality of first image information of the target to be monitored is simultaneously acquired at a first set frame node by a plurality of first image acquisition units, and after registering and preprocessing the plurality of first image information, image fusion is performed to obtain a first fused image; It is determined whether the number of features contained in the first fused image reaches a first preset number. If so, the first fused image is sent as a final image to a server for storage.

5. The multi-frame image fusion method based on frame-accurate technology according to claim 4, characterized in that: After registering and preprocessing the plurality of second image information, performing image fusion to obtain a fused image, and sending the fused image as a final image to a server for storage, means: When the number of features contained in the first fused image does not reach the first preset number, the second image acquisition unit is started through the edge computing gateway control, and the second image acquisition unit simultaneously acquires multiple second image information of the target to be monitored at the second set frame node, and after registering and preprocessing the multiple second image information, image fusion is performed to obtain a second fused image; It is determined whether the number of features contained in the second fused image reaches a second preset number. If so, the second fused image is sent as a final image to the server for storage.

6. The multi-frame image fusion method based on frame-accurate technology according to claim 5, characterized in that: After registering and preprocessing the plurality of first image information and the plurality of second image information, performing image fusion to obtain a fused image, and sending the fused image as a final image to a server for storage, means: When the number of features contained in the second fused image does not reach the second preset number, the first image acquisition unit and the second image processing unit are started simultaneously, and the first fused image and the second fused image are fused to obtain a third fused image, and the third fused image is sent to the server as the final image for storage.

7. The multi-frame image fusion method based on frame-accurate technology according to claim 1, characterized in that: The registering of the first image information refers to: taking the first image information acquired by the basic first image acquisition unit as a reference, matching the first image information acquired by other first image acquisition units to the reference first image information; The registering of the second image information refers to: selecting a basic second image acquisition unit from a second image acquisition unit having the largest number of features to be extracted based on the target to be monitored, taking the second image information acquired by the basic second image acquisition unit as a reference, and matching the second image information acquired by other second image acquisition units to the reference second image information; The registering of the first image information and the second image information refers to: selecting a basic image acquisition unit from an image acquisition unit with the largest number of features to be extracted based on the target to be monitored, and matching the image information acquired by other image acquisition units to the benchmark image information based on the image information acquired by the basic image acquisition unit.

8. The multi-frame image fusion method based on frame-accurate technology according to claim 1, characterized in that: Preprocessing the plurality of first image information and / or the plurality of second image information refers to: Prior knowledge of the scene to be monitored is obtained, and the prior knowledge is added to the first image information and / or the plurality of second image information.

9. A multi-frame image fusion system based on frame-accurate technology, implementing a multi-frame image fusion method based on frame-accurate technology as claimed in any one of claims 1 to 8, characterized in that: include: An image acquisition unit acquisition module, used to acquire position information and quantity information of a plurality of first image acquisition units of a target to be monitored in a scene to be monitored, wherein each first image acquisition unit is integrated with a corresponding second image acquisition unit; A basic first image acquisition unit acquisition module, used to select a basic first image acquisition unit from a first image acquisition unit having the largest number of features to be extracted based on the target to be monitored; The edge gateway is used to set the edge computing gateway based on the quantity information of all the first image acquisition units and the location information of the basic first image acquisition units, and control the first image acquisition units to be in the startup state and the second image acquisition units to be in the standby state through the edge computing gateway; An image information acquisition module, which is used to simultaneously acquire multiple first image information of the target to be monitored at a first set frame node through multiple first image acquisition units when the basic first image acquisition unit senses the target to be monitored, and / or to control the start of the second image acquisition unit through the edge computing gateway, and simultaneously acquire multiple second image information of the target to be monitored at a second set frame node through the second image acquisition unit; The image fusion module is used to perform image fusion after registering and preprocessing the plurality of first image information and / or the plurality of second image information to obtain a fused image, and send the fused image as a final image to the server for storage.

Images