Image enhancement model rapid training method, system and medium for monitoring device

Abstract

The application belongs to the technical field of monitoring image processing, and relates to a kind of image enhancement model fast training method, system and medium for monitoring equipment, comprising: determining the original shooting device of the image to be processed;The image is projected, and the original shooting device is used for shooting, after shooting is completed, the image is switched to the next one, and shooting is continued until the shooting of all images is completed;Read the image shot by the original shooting device, combine with the original image to be processed, form a data set;According to the data set, the image enhancement model is trained, and the final image enhancement model is generated. The image enhancement model derived thereby has stronger adaptability to specific monitoring equipment, so that better image enhancement effect can be obtained.

Description

TECHNICAL FIELD

[0001] The application relates to a quick training method, system and medium for an image enhancement model of a monitoring device, and belongs to the technical field of monitoring image processing. BACKGROUND

[0002] With the progress of science and technology and the development of society, a large number of monitoring devices are arranged in every corner of the city to realize monitoring of each region. The video pictures recorded by the monitoring devices play a great role in the fields of public safety, traffic and natural disaster warning. However, in the actual use process, we often encounter the problem that the target is blurred or has low visibility in the monitoring picture due to low monitoring resolution, defects of the monitoring probe, long distance of the target and dim environment light, so that the target cannot be recognized or effective information cannot be extracted. At present, there are two ways to repair the images. One is to use image processing based on various algorithms to enhance the images. This method is highly dependent on the performance of the algorithm and the experience of the operator, and thus is low in efficiency and unstable in result. The other is to use an image enhancement model to automatically repair the images. This method is high in efficiency, but the performance of the model depends on the training data set, and in most cases, the training image data comes from other imaging devices, so the model is difficult to adapt to each monitoring probe. When the image to be processed comes from a monitoring probe with significant individual differences, such as having bad points or picture color distortion, the model may not achieve good results. SUMMARY

[0003] In view of the above problems, the application provides a quick training method, system and medium for an image enhancement model of a monitoring device. The image enhancement model obtained by the application has stronger adaptability to specific monitoring devices, so that better image enhancement effect can be obtained.

[0004] To achieve the above object, the application provides the following technical scheme: a quick training method for an image enhancement model of a monitoring device, comprising the following steps: determining an original shooting device of an image to be processed; projecting the image and shooting the image by using the original shooting device; after shooting is completed, switching the image to the next one and continuing shooting until shooting of all images is completed; reading the images shot by the original shooting device and combining the images with the original image to be processed to form a data set; and training an image enhancement model according to the data set to generate a final image enhancement model.

[0005] Further, the image to be processed is preprocessed before the image is projected.

[0006] Further, the preprocessing comprises reducing resolution or lowering brightness.

[0007] Furthermore, the image is projected through an image projection terminal, which is a display or a projector.

[0008] Furthermore, the original imaging device takes less than or equal to 1 second to capture an image.

[0009] Furthermore, the image enhancement model is the RetinexNet model.

[0010] This invention also discloses a rapid training system for image enhancement models of surveillance equipment, comprising: a device determination module for determining the original capturing device of the image to be processed; a capturing module for projecting the image and capturing it with the original capturing device, switching the image to the next image after capturing, and continuing to capture until all images are captured; a dataset forming module for reading the images captured by the original capturing device and combining them with the original image to be processed to form a dataset; and a model training module for training the image enhancement model based on the dataset to generate the final image enhancement model.

[0011] Furthermore, before projecting the image, the image to be processed is preprocessed, including reducing the resolution, adding blur, reducing the contrast, or lowering the brightness.

[0012] Furthermore, the image is projected through an image projection terminal, which is a display or a projector.

[0013] The present invention also discloses a computer-readable storage medium storing a computer program, which is executed by a processor to implement the rapid training method for image enhancement models of monitoring equipment as described in any of the preceding claims.

[0014] The present invention has the following advantages due to the adoption of the above technical solutions: the solution of the present invention can quickly obtain an image enhancement model, and the obtained image enhancement model has stronger adaptability to specific monitoring equipment, thereby achieving better image enhancement results; compared with traditional manual collection and annotation, this method can save time and greatly improve efficiency. Attached Figure Description

[0015] Figure 1 This is a flowchart of a rapid training method for an image enhancement model of a monitoring device according to an embodiment of the present invention;

[0016] Figure 2 This is a schematic diagram of a rapid training system for image enhancement models of monitoring equipment according to an embodiment of the present invention.

[0017] Instruction manual illustrations:

[0018] 1 - original shooting device; 2 - image projection end; 3 - control device. DETAILED DESCRIPTION

[0019] In order for those skilled in the art to better understand the technical solutions of the present application, the present application will be described in detail through specific embodiments. However, it should be understood that the specific embodiments are provided only for better understanding of the present application, and they should not be understood as limiting the present application. In the description of the present application, it should be understood that the terms used are only for the purpose of description, and cannot be understood as indicating or implying relative importance.

[0020] In order to solve the problems existing in the prior art, such as the general image enhancement model using data sets often derived from the Internet, other shooting devices or synthesized, the image data and the image to be processed are not derived from the same device, when the monitoring device or the shooting environment is special, the model trained using the data set is greatly different, and the model cannot achieve good results, etc. The present application proposes a fast training method, system and medium for an image enhancement model for a monitoring device, which can collect a large amount of image data in a short time for a specific monitoring device to generate a data set. After generating the data set based on the specific monitoring device, the model is trained to obtain a model with strong adaptability to the monitoring device. Based on the model, the image to be processed is enhanced to obtain good results. The image enhancement model obtained has stronger adaptability to the specific monitoring device, thereby being able to obtain better image enhancement effect. The scheme of the present application will be described in detail through embodiments in combination with the drawings.

[0021] Embodiment one

[0022] The present embodiment discloses a fast training method for an image enhancement model for a monitoring device, as shown in Figure 1 , Figure 2 , comprising the following steps:

[0023] S1 determines the original shooting device 1 of the image to be processed.

[0024] S2 projects the image and shoots it with the original shooting device 1. After shooting, switch the image to the next one and continue shooting until all images are shot.

[0025] Before projecting the image, the image to be processed is preprocessed. The preprocessing includes reducing the resolution or lowering the brightness. The image is projected through the image projection end 2, which is a display or a projector. The shooting time of the original shooting device 1 for one image is less than or equal to 1 second. The switching of the image is controlled by the control device 3, and the original shooting device 1 and the control device 3 are in communication connection. When it shoots one image, it sends a signal to the control device 3, and the control device 3 controls the image projection end 2 to switch the image.

[0026] S3 reads the image photographed by the original photographing device 1, combines the original image to be processed to form a data set;

[0027] S4 trains the image enhancement model according to the data set to generate a final image enhancement model.

[0028] The image enhancement model is a RetinexNet model.

[0029] Embodiment two

[0030] Based on the same inventive concept, the embodiment discloses a fast training system of an image enhancement model for a monitoring device, as shown in the figure, comprising: Figure 2

[0031] The device determination module is configured to determine the original photographing device 1 of the image to be processed.

[0032] The photographing module is configured to project the image and photograph it with the original photographing device 1. After photographing, the image is switched to the next one, and the photographing is continued until the photographing of all images is completed. Before the image is projected, the image to be processed is preprocessed, and the preprocessing includes reducing the resolution or lowering the brightness. The image is projected through the image projection end 2, and the image projection end 2 is a display or a projector. The switching of the image is controlled by the control device 3, and the original photographing device 1 is in communication connection with the control device 3. When it photographs an image, it sends a signal to the control device 3, and the control device 3 controls the image projection end 2 to switch the image.

[0033] The data set formation module is configured to read the image photographed by the original photographing device 1, combine the original image to be processed to form a data set;

[0034] The model training module is configured to train the image enhancement model according to the data set to generate a final image enhancement model.

[0035] Embodiment three

[0036] Based on the same inventive concept, the embodiment further discloses a computer readable storage medium, and the computer readable storage medium stores a computer program. The computer program is executed by a processor to implement the fast training method of the image enhancement model for the monitoring device according to any one of the above embodiments.

[0037] The scheme in the application can simulate the photographing environment of the image to be processed according to the photographing device of the image to be processed, obtain a large number of photographed pictures in a short time, and then train the model with these data. The image enhancement model with high adaptability to the image to be processed and the device can be obtained.

[0038] ​Those skilled in the art will appreciate that embodiments of the application can be readily used as a method, a system or a computer program product. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer readable program code.

[0039] The present application is described in reference to the flowchart illustrations and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams.

[0040] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams.

[0041] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams. Figure 1 one or more functions specified in the flowchart illustrations and / or block diagrams.

[0042] It should be noted that the above examples are only used to illustrate the technical solutions of the present application and not to limit it. Although the present application has been described in detail with reference to the above examples, those skilled in the art should understand that the specific embodiments of the present application can still be modified or replaced equivalently without departing from the spirit and scope of the present application, and any modification or equivalent replacement without departing from the spirit and scope of the present application should be covered within the protection scope of the claims of the present application. The above content is only a specific embodiment of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0043] It should be noted that the above examples are only used to illustrate the technical solutions of the present application and not to limit it. Although the present application has been described in detail with reference to the above examples, those skilled in the art should understand that the specific embodiments of the present application can still be modified or replaced equivalently without departing from the spirit and scope of the present application, and any modification or equivalent replacement without departing from the spirit and scope of the present application should be covered within the protection scope of the claims of the present application. The above content is only a specific embodiment of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for rapid training of image enhancement models for surveillance equipment, characterized in that, include: Determine the original shooting device of the image to be processed; The image is projected and captured using the original shooting device. After the capture is completed, the image is switched to the next one, and the shooting continues until all images are captured. The images captured by the original imaging device are read and combined with the original images to be processed to form a dataset; The image enhancement model is trained based on the dataset to generate the final image enhancement model; Before projecting the image, the image to be processed is preprocessed; The preprocessing includes reducing resolution, adding blur, reducing contrast, or lowering brightness.

2. The rapid training method for image enhancement models of monitoring equipment as described in claim 1, characterized in that, The image is projected through an image projection device, which is a display or a projector.

3. The rapid training method for image enhancement models of monitoring equipment as described in claim 1, characterized in that, The original imaging device takes less than or equal to 1 second to capture an image.

4. The rapid training method for image enhancement models of monitoring equipment as described in claim 1, characterized in that, The image enhancement model is the RetinexNet model.

5. A rapid training system for image enhancement models of surveillance equipment, characterized in that, include: The device determination module is used to determine the original capturing device of the image to be processed; The shooting module is used to project the image and take a picture with the original shooting device. After the picture is taken, the image is switched to the next one and the shooting continues until all images are taken. The dataset forming module is used to read the images captured by the original imaging device and combine them with the original images to be processed to form a dataset; The model training module is used to train the image enhancement model based on the dataset and generate the final image enhancement model. Before projecting the image, the image to be processed is preprocessed; The preprocessing includes reducing resolution, adding blur, reducing contrast, or lowering brightness.

6. The rapid training system for image enhancement models of monitoring equipment as described in claim 5, characterized in that, The image is projected through an image projection device, which is a display or a projector.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that is executed by a processor to implement the rapid training method for an image enhancement model of a monitoring device as described in any one of claims 1-4.

Images