Image task automatic deployment method and device, medium and program product

Abstract

The invention provides an image task automatic deployment method and device, a medium and a program product, and relates to the technical field of image processing.The method comprises the steps that a user configures target image content through a front-end interaction interface; sending the target image content to a task manager, and deploying an image task according to the target image content through a preset task base and a plurality of image processing units; running the image task to obtain a target image; wherein the task base comprises a conventional task base and a free task base; the conventional task base acquires video image frames according to a preset acquisition rule, processes the video image frames by adopting image processing algorithms corresponding to respective image tasks in sequence according to the sequence of the image processing units, and outputs a target image; and the free task base is used for controlling acquisition and processing of video image frames by the image processing unit and outputting a target image. The problem that the image task is difficult to automatically deploy according to the target image content configured by the user in the prior art is solved.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to an automatic image task deployment method, apparatus, medium, and program product. Background Technology

[0002] With the rapid development of computer vision technology, image processing tasks are increasingly being used in fields such as security monitoring, intelligent transportation, and industrial inspection, and users' demands for automation in image acquisition, processing, and target image acquisition are constantly increasing. However, existing image processing task deployment methods still have significant limitations.

[0003] Traditional image task deployment relies heavily on manual configuration, requiring technicians to manually set image acquisition device parameters, select image processing algorithms, and plan task execution processes according to specific scenarios. This not only demands high professional skills from operators but is also cumbersome, inefficient, and difficult to adapt to the personalized needs of different users for target image content. Summary of the Invention

[0004] The embodiments of the present invention provide an automatic image task deployment method, apparatus, medium and program product, which aims to solve the problem that the prior art relies on manual configuration and is difficult to automatically deploy image tasks according to the target image content configured by the user.

[0005] To achieve the above objectives, in a first aspect, the present invention provides an automatic image task deployment method, comprising the following steps: Users configure the target image content through the front-end interactive interface; The target image content is sent to the task manager, and a predetermined image task is deployed based on the target image content through a preset task base and multiple image processing units. Run the image task to acquire the target image; The task base includes a conventional task base and a free task base. The conventional task base acquires video image frames from one or more cameras according to preset acquisition rules, and performs predetermined conventional task processing on the video image frames in the order of multiple image processing units using the image processing algorithms corresponding to their respective image tasks, and outputs the target image. The free-task base provides the image processing unit with a collector and a streamer, and the image processing unit controls the acquisition and processing of video image frames according to the predetermined free task, as well as the output of the target image.

[0006] Furthermore, when the number of video image frames acquired exceeds a preset threshold, the image task is run using a task manager deployed on the edge device; when the number of video image frames acquired does not exceed the preset threshold, the image task is run using a task manager deployed on the server.

[0007] Furthermore, when deploying an image task, the task manager is selected to run the image task based on network bandwidth, hardware resources, and / or camera connectivity.

[0008] Furthermore, the acquisition of video image frames is performed in a non-blocking manner, and when a predetermined number of video frame images are acquired, the acquired images are processed together.

[0009] Furthermore, the task manager uses the FastAPI framework to provide an external HTTP interface; it uses a dictionary to establish the connection between image task IDs and image task instances; and it uses a queue to facilitate communication between the image task center and specific task processes.

[0010] Furthermore, the image task is run through a preset streamer, task unit, streamer, output module, log module, storage module, timer and / or message middleware communication module.

[0011] Furthermore, it also includes a custom unit, which allows users to customize the image processing algorithm according to the target image requirements, and to place it together with the image processing unit in the task base to form an image task for deployment.

[0012] Secondly, the present invention provides an automatic image task deployment apparatus, including a memory and a processor, wherein the memory stores at least one program, and the at least one program is executed by the processor to implement the automatic image task deployment method as described above.

[0013] Thirdly, the present invention provides a computer-readable storage medium storing at least one program that is executed by a processor to implement the image task automatic deployment method as described above.

[0014] Fourthly, the present invention provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the image task automatic deployment method as described above.

[0015] The above technical solution has the following technical effects: Users configure target image content through a front-end interactive interface; the target image content is sent to the task manager, and image tasks are deployed according to the target image content through a preset task base and multiple image processing units; the image tasks are run to acquire the target image; wherein, the task base includes a regular task base and a free task base; the regular task base acquires video image frames from one or more cameras according to preset acquisition rules, and processes the video image frames sequentially using the image processing algorithms corresponding to their respective image tasks according to the order of multiple image processing units, and the last image processing unit outputs the target image; the free task base provides the image processing units with a collector and a streamer, and the image processing units control the acquisition and processing of video image frames, as well as the output of the target image. This invention solves the problem in the prior art that it is difficult to automatically deploy image tasks according to the target image content configured by the user. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating an embodiment of the automatic deployment method for image tasks according to the present invention.

[0017] Figure 2 This is a schematic diagram of the structure of an automatic image task deployment method according to an embodiment of the present invention.

[0018] Figure 3 This is a diagram showing the relationship between the task manager and the backend in an embodiment of the present invention.

[0019] Figure 4 This is an internal structure diagram of a task manager according to an embodiment of the present invention.

[0020] Figure 5 This is a schematic diagram of the structure of an image task automatic deployment device according to an embodiment of the present invention. Detailed Implementation

[0021] To further illustrate the various embodiments, the present invention provides accompanying drawings. These drawings are part of the disclosure of the present invention, primarily used to illustrate the embodiments and to explain the operating principles of the embodiments in conjunction with the relevant descriptions in the specification. With reference to these drawings, those skilled in the art should be able to understand other possible implementations and the advantages of the present invention. Components in the drawings are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0022] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments.

[0023] Example 1: Figure 1 This is a flowchart illustrating an embodiment of the automatic deployment method for image tasks according to the present invention, as shown below. Figure 1 As shown, the method of this embodiment includes the following steps: Users configure the desired target image content through the front-end interactive interface; In one specific implementation, Figure 2 This is a schematic diagram of the structure of an automatic image task deployment method according to an embodiment of the present invention, as shown below. Figure 2 As shown, the structure consists of three parts: a front-end, a back-end, and a task manager. The front-end interacts with the user, providing an interface for configuring and managing tasks. The back-end is responsible for saving the user-configured target image content to the database and sending user commands such as starting a task, stopping a task, and querying task running information to the task manager. The task manager is responsible for receiving commands from the back-end and running the target task according to the task configuration.

[0024] In one specific implementation, the front-end uses the Vue technology stack, the back-end uses the Spring Cloud technology stack, and the task management part is written in Python. The front-end, back-end, and task management can be deployed using Docker, with both the front-end and back-end deployed in a distributed manner to improve the system's stability, reliability, and concurrency.

[0025] The command is sent to the task manager, and the predetermined image task is deployed according to the target image content through the preset task base and multiple image processing units. In one specific implementation, a task deployment model of task base and task unit is designed. A task consists of a base and one or more image processing units. The base determines how the task runs, and the unit determines the specific image processing logic. One specific implementation includes two types of bases: a conventional task base for regular image processing tasks, which acquires image frames from the acquisition unit according to certain rules, then sequentially transmits the image frames to each unit, simultaneously transmitting the output of the previous unit to the next, and finally pushing the output image frame of the last unit out for viewing. It can also be configured not to push the image frame to save overhead; the unit only focuses on how to process the image and the output of the previous unit. The other type is a free task base, which provides the acquisition unit and the pusher to the unit. The specific process of acquiring image frames from the acquisition unit, processing them, and outputting them is controlled by the unit. The unit is the core of image processing. For regular image processing tasks, it can be roughly divided into three steps: preprocessing, various image processing algorithms, and finally, analysis and output of the results. By providing some common and general image preprocessing units, image algorithm processing units, and post-processing units, users can complete the deployment of image tasks through some configuration. In one specific implementation, if the built-in units cannot meet the requirements, users can also customize units and place them together with the built-in units in the base to form a task for deployment.

[0026] In one specific implementation, an image processing task can simultaneously acquire images from multiple cameras. To accelerate image processing, multiple video frames are typically processed concurrently. Since different cameras can be configured with different frame rates, one implementation uses a non-blocking method to acquire a specified number of images from multiple cameras. Once enough video frames have been acquired, these images are processed together.

[0027] In one specific implementation, Figure 3 This is a diagram illustrating the relationship between the task manager and the backend in an embodiment of the present invention, as follows: Figure 3 As shown, the task manager can be deployed on multiple devices. Each task manager has its backend interface configured during deployment. After deployment, the task manager automatically connects to the backend and reports its information. The backend records this information in Redis, achieving data and program separation and facilitating easy horizontal scaling of the system, i.e., deploying multiple backend services. The task manager can be deployed on servers or edge devices. In one specific implementation, for large-scale real-time video processing, using edge devices can save significant bandwidth resources, while for small-scale real-time video processing, servers can be used directly, eliminating the need to purchase additional equipment. When configuring a task, the task manager can be configured to run the task based on network bandwidth, hardware resources, camera connectivity, etc. When the "Run" button is clicked, the backend sends task-related information to the corresponding task manager. The task manager retrieves the necessary unit code and AI model from the file service and then runs the task.

[0028] In one specific implementation, when the number of video image frames acquired exceeds a preset threshold, an image task is run using a task manager deployed on the edge device; when the number of video image frames acquired does not exceed the preset threshold, an image task is run using a task manager deployed on the server.

[0029] In one specific implementation, when deploying an image task, the task manager is selected to run the image task based on network bandwidth, hardware resources, and / or camera connectivity.

[0030] In one specific implementation, Figure 4 This is an internal structure diagram of a task manager according to an embodiment of the present invention, as shown below. Figure 4As shown, the Task Manager uses the FastAPI framework to provide an external HTTP interface; it uses a dictionary to establish the connection between task IDs and task instances; and it uses queues for communication between the task center and specific task processes. Internally, both regular and free tasks can have a puller, task unit, pusher, output module, logging module, storage module, timer, and message middleware communication module. These modules are not essential for every task but depend on the task configuration. For example, during the debugging phase, the processing screen needs to be displayed for easy viewing; once debugging is complete, the algorithm processing screen is no longer needed, so the pusher is no longer required. The output module is responsible for outputting the content required by the task through a database or message middleware; the logging module is responsible for recording logs during task execution; the puller pulls images from various cameras; the task unit executes specific image processing logic; the storage module is responsible for retrieving the unit script and files needed to run the task, as well as storing files such as images or videos that the task needs to save; the timer is used to run the task periodically (some tasks do not need to run continuously but run at intervals); and the pusher pushes the task output screen to the streaming server.

[0031] In one specific implementation, we will take two specific image-related tasks as examples to illustrate how to deploy the tasks.

[0032] Task 1: Identify the bag number of the iron bag.

[0033] Typically, there are many locations requiring packet identification, so multiple cameras need to be identified simultaneously. Since the same model is used to identify these packets, it's suitable to deploy this as a single task (rather than multiple tasks) to simultaneously identify images captured by multiple cameras. Furthermore, this task uses neural networks for target detection and identification, which involves significant computation. Identifying only one image at a time incurs considerable overhead, so it's usually necessary to identify multiple images simultaneously to reduce this overhead. The processing flow for this task is basically standard: first, image frames are acquired; then, preprocessing is performed; next, the model identifies the image; and finally, the results are output. The entire process is sequential, and the identification of each camera is independent. This means that each processing step doesn't consider which cameras' images were acquired or processed; it simply outputs the corresponding result. Therefore, this task is suitable for using a standard task platform with the appropriate processing units configured.

[0034] Task 2: Identifying the sequence of ladles on the molten iron line.

[0035] The task involves identifying the order of ladles on the molten iron production line. Since the ladle numbers painted on the ladles can be in any direction, cameras need to be placed on both sides of the line. Because the line is quite long, multiple cameras are needed on each side. These cameras simultaneously perform identification, and the results are matched to reconstruct the ladle positions on the line, outputting the final result. In this example, each identification requires processing images from multiple cameras simultaneously. However, conventional task frame-taking logic only considers the number of images captured, not the relationships between them. For example, in one processing session, camera A might capture three frames while camera B captures none. Therefore, this situation necessitates using a free task platform, writing free units, and implementing frame-taking logic within these units for subsequent processing.

[0036] Example 2: Figure 5 This is a schematic diagram of the structure of an automatic image task deployment device according to an embodiment of the present invention, as shown below. Figure 5 As shown, the device includes a processor 501, a memory 502, a bus 503, and a computer program stored in the memory 502 and executable on the processor 501. The processor 501 includes one or more processing cores. The memory 502 is connected to the processor 501 via the bus 503. The memory 502 is used to store program instructions. When the processor executes the computer program, it implements the steps in the above-described method embodiment of Embodiment 1 of the present invention.

[0037] Furthermore, as an executable solution, the automatic image task deployment device can be a computer unit, which can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer unit may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above-described structure of the computer unit is merely an example and does not constitute a limitation on the computer unit. It may include more or fewer components, or combine certain components, or use different components. For example, the computer unit may also include input / output devices, network access devices, buses, etc., and this embodiment of the invention does not limit this.

[0038] Furthermore, as an executable solution, the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. The processor is the control center of the computer unit, connecting various parts of the entire computer unit via various interfaces and lines.

[0039] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the computer unit by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a function; the data storage area may store data created based on the use of the mobile phone, etc. In addition, the memory may include high-speed random access memory and may also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital card (SD card), flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0040] Example 3: The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method described in the embodiments of the present invention.

[0041] If the modules / units integrated in the computer unit are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), and software distribution media, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction.

[0042] Example 4: The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the image task automatic deployment method as described above.

[0043] Although the invention has been specifically shown and described in conjunction with preferred embodiments, those skilled in the art should understand that various changes in form and detail may be made to the invention without departing from the spirit and scope of the invention as defined in the appended claims, all of which shall be within the scope of protection of the invention.

Claims

1. An image task automatic deployment method, characterized in that, The method comprises the following steps: A user configures target image content through a front-end interactive interface; The target image content is sent to a task manager, and a predetermined image task is deployed according to the target image content through a preset task base and multiple image processing units; The image task is run to obtain the target image; The task base comprises a regular task base and a free task base; The regular task base collects video image frames from one or more cameras according to a preset collection rule, and sequentially adopts an image processing algorithm corresponding to each image task to perform predetermined regular task processing on the video image frames according to the order of the multiple image processing units, and outputs the target image; The free task base provides a collector and a streamer for the image processing unit, and controls the video image frames to collect and process according to a predetermined free task, and outputs the target image.

2. The image task automatic deployment method of claim 1, wherein, When the number of collected video image frames exceeds a preset threshold, the task manager deployed on an edge device is used to run the image task; when the number of collected video image frames does not exceed the preset threshold, the task manager deployed on a server is used to run the image task.

3. The method of claim 1, wherein, When deploying the image task, the task manager is selected to run the image task according to network bandwidth, hardware resources, and / or camera connection conditions.

4. The method of claim 1, wherein, The collected video image frames are collected in a non-blocking manner, and when a predetermined number of video frame images are collected, the collected images are processed together.

5. The method of claim 1, wherein, The task manager uses a fastapi framework to provide an external http interface; a dictionary is used to establish a connection between an image task ID and an image task instance; a queue is used for communication between an image task center and a specific task process.

6. The method of claim 1, wherein, The image task is run through a preset pull streamer, task unit, push streamer, output module, log module, storage module, timer, and / or message middleware communication module.

7. The method of claim 1, wherein, A custom unit is further included for a user to customize an image processing algorithm according to target image requirements and place the image processing algorithm in the task base together with the image processing unit to form an image task for deployment.

8. An image task automatic deployment apparatus characterized by comprising: The storage medium stores at least one program, and the at least one program is executed by the processor to implement the image task automatic deployment method according to any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The storage medium stores at least one program, and the at least one program is executed by the processor to implement the image task automatic deployment method according to any one of claims 1-7.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the image task automatic deployment method according to any one of claims 1-7.

Images