Computer equipment

Abstract

The invention relates to computer equipment. The computer equipment comprises a memory, a processor and a monitoring computer algorithm operating system which is stored on the memory and can run on the processor, according to the computer equipment, the processing efficiency and performance can be improved.

Description

Technical Field

[0001] The present invention relates to the field of computer systems, and in particular to a computer device. Background Art

[0002] The characteristics of computer systems are that they can perform accurate and fast calculations and judgments, and they are versatile, easy to use, and can be connected to a network. The specific characteristics are as follows.

[0003] (1) Calculation: Almost all complex calculations can be performed by computers through arithmetic and logical operations.

[0004] (2) Judgment: Computers have the ability to distinguish different situations and choose different treatments, so they can be used in management, control, confrontation, decision-making, reasoning and other fields.

[0005] (3) Storage: Computers can store huge amounts of information.

[0006] (4) Accuracy: As long as the word length is sufficient, the calculation accuracy is theoretically unlimited.

[0007] (5) Fast: The time required for a computer operation is now measured in nanoseconds.

[0008] (6) General: Computers are programmable, and different programs can realize different applications.

[0009] (7) Ease of use: Rich high-performance software and intelligent human-computer interface greatly facilitate use.

[0010] (8) Networking: Multiple computer systems can transcend geographical boundaries and share remote information and software resources through communication networks.

[0011] CN119226182A provides a main control chip, a data processing method, a solid-state hard disk and a computer system, wherein the main control chip includes: a flash memory host interface controller, including an allocation address acquisition module and a command execution module; the allocation address acquisition module is used to obtain a first cache address and a second cache address corresponding to a data write command, the first cache address is a cache space address located inside the main control chip, and the second cache address is a cache space address located outside the main control chip; the command execution module is used to write write data to the first cache address and the second cache address; the main control chip performs a programming operation based on the write data in the first cache address, and releases the write data in the second cache address after the main control chip confirms the data consistency of the write data in the first cache address.

[0012] CN119226201A discloses a firmware processing method, a retiming card, a computer system, and a readable storage medium. The method includes: obtaining level information of two single-ended signals of a high-speed expansion slot by connecting a wire to the high-speed expansion slot; determining the target bandwidth of an external device based on the level information according to a mapping relationship between level and bandwidth; determining a target memory to be connected to the retiming component based on the target bandwidth according to a mapping relationship between bandwidth and firmware; wherein firmware of different bandwidths are stored in different memories; and sending a control signal to connect to the target memory to a selection switch so that the retiming component performs signal adjustment based on the target firmware. The technical effect of this application is that the firmware switching of the retiming card can be completed without turning on the computer, without relying on the baseboard management controller or bandwidth information stored in other devices, thus achieving seamless operation.

[0013] CN119201357A relates to a computer system, including a simulation virtual machine, which includes a translator and an instruction executor for translating the simulation virtual machine instructions into instructions that can be executed by a test platform. The simulation virtual machine uses shared memory to send instructions to the test platform; the instruction translator includes a CPU simulator, and the simulator receives and sends instructions to the BIOS and the OS. The instruction translator also includes multiple instruction grabbers for grabbing corresponding instructions from the simulator; the instruction executor includes: a request filtering module that filters and distributes various instructions received from the instruction translator, and sends the instructions required for the test platform to execute the test after being filtered to a request control module; the request control module sends the instructions required to execute the test after being filtered to a transmission module, and receives the test execution results from the test platform from the transmission module; the transmission module uses shared memory to send the instructions required by the test platform to the test platform. Summary of the Invention

[0014] In order to solve the technical problems in the prior art, the present invention provides a computer device, including a memory, a processor and a monitoring computer algorithm operating system stored in the memory and capable of running on the processor. The present invention can obtain a plurality of scene information of the work scene in a specific scene, including the internal space volume of the work scene inside the office building, the background board area, the distance from the background board to the fisheye camera mechanism, the number of employees and the number of desks, and also obtain various picture parameters of the work scene picture, including the horizontal resolution, vertical resolution, maximum pixel depth of field, minimum pixel depth of field, signal-to-noise ratio and picture content repetition of the work scene picture, so as to provide intelligent prediction of different encoding times for the work scene picture in a specific scene under different size encoding blocks. The measurement provides sufficient and comprehensive basic data, and adopts a coding prediction model to intelligently predict the predicted value of the coding time required to perform intra-frame coding on the working scene picture under the coding block size based on the coding block size used for intra-frame coding, various picture parameters of the working scene picture and multiple scene information of the working scene inside the office building, so as to obtain the predicted values ​​of the various coding times required to perform intra-frame coding on the working scene picture under various coding block sizes, and output the coding block size corresponding to the predicted value of the coding time with the smallest value as the optimal coding block size with coding rate priority, so as to effectively predict the coding time required before performing intra-frame coding on the working scene picture in a specific scenario, and provide an important basis for the selection of the size of the intra-frame coding block.

[0015] According to the present invention, a computer device is provided, comprising a memory, a processor, and a monitoring computer algorithm operating system stored in the memory and executable on the processor, the system comprising: A multiple learning device is used to perform multi-layer learning on the Hoffet neural network to obtain a coding prediction model output, wherein performing each layer of learning on the Hoffet neural network is performing each learning process on the Hoffet neural network; An image receiving device is connected to the fisheye camera mechanism and is used to obtain a work scene image, wherein the work scene image is an image obtained by the fisheye camera mechanism disposed inside the office building performing a wide-field shooting of the work scene inside the office building; a first detection device, connected to the picture receiving device, for detecting various picture parameters of the working scene picture, wherein the various picture parameters of the working scene picture are horizontal resolution, vertical resolution, maximum pixel depth of field, minimum pixel depth of field, signal-to-noise ratio, and picture content repetitiveness of the working scene picture; a second detection device for detecting multiple pieces of scene information of a work scene inside an office building, wherein the multiple pieces of scene information of the work scene inside the office building include an internal space volume of the work scene inside the office building, an area of ​​a background board, a distance from the background board to a fisheye camera mechanism, the number of employees, and the number of desks; a time prediction mechanism, connected to the multiple learning device, the first detection device, and the second detection device, respectively, for using a coding prediction model to intelligently predict a predicted value of the coding time required to perform intra-frame coding on the work scene picture under the coding block size based on the coding block size used for intra-frame coding, various picture parameters of the work scene picture, and multiple scene information of the work scene inside the office building; A block selection mechanism is connected to the time prediction mechanism and is used to obtain predicted values ​​of the encoding time required to perform intra-frame encoding on the working scene image at various encoding block sizes, and output the encoding block size corresponding to the predicted value of the encoding time with the smallest value as the optimal encoding block size with priority on encoding rate; The image receiving device is connected to the fisheye camera mechanism and is used to obtain a work scene image, wherein the work scene image is an image obtained by the fisheye camera mechanism disposed inside the office building performing wide-field shooting of the work scene inside the office building, and includes: the shooting field of view angle of the fisheye camera mechanism is greater than or equal to a set field of view angle threshold; Among them, the multiple learning device is used to perform multi-layer learning on the Hoffitt neural network to obtain the coding prediction model output, and performing each layer of learning on the Hoffitt neural network is performing each learning process on the Hoffitt neural network, including: the number of learning layers performed on the Hoffitt neural network is proportional to the total number of pixels in the working scene picture.

[0016] It can be seen that the present invention has at least the following three main inventive concepts: First, for a work scene image in a specific scenario, multiple pieces of scene information of the work scene are obtained, including the internal space volume of the work scene inside the office building, the area of ​​the background board, the distance from the background board to the fisheye camera mechanism, the number of employees, and the number of desks. Various picture parameters of the work scene image are also obtained, including the horizontal resolution, vertical resolution, maximum pixel depth of field, minimum pixel depth of field, signal-to-noise ratio, and picture content repetition of the work scene image, thereby providing sufficient and comprehensive basic data for intelligent prediction of different encoding times for different sizes of encoding blocks of the work scene image in a specific scenario. Second, the artificial intelligence model based on which the intelligent prediction of different encoding times of different-sized encoding blocks of a work scene image under a specific scenario is a coding prediction model with a customized structure is implemented. The customization of the structure is that a multi-layer learning of the Hoffitt neural network is performed to obtain the coding prediction model output. The execution of each layer of learning on the Hoffitt neural network is equivalent to executing each learning process on the Hoffitt neural network. The number of layers of learning performed on the Hoffitt neural network is proportional to the total number of pixels of the work scene image. Third: A coding prediction model is used to intelligently predict the coding time required to perform intra-frame coding on the working scene picture under the coding block size, various picture parameters of the working scene picture, and multiple scene information of the working scene inside the office building based on the coding block size used for intra-frame coding, the various picture parameters of the working scene picture, and the multiple scene information of the working scene inside the office building, so as to obtain the predicted values ​​of the various coding times required to perform intra-frame coding on the working scene picture under various coding block sizes, and output the coding block size corresponding to the predicted value of the coding time with the smallest value as the optimal coding block size with coding rate priority, so as to effectively predict the coding time required before performing intra-frame coding on the working scene picture in a specific scene, and provide an important basis for the selection of the size of the intra-frame coding block. DETAILED DESCRIPTION

[0017] In the existing technology, it is impossible to use the computer system used for monitoring to obtain multiple scene information and various picture parameters of the working scene picture in a specific scene, and thus it is impossible to provide sufficient and comprehensive basic data for the intelligent prediction of different encoding times of different coding blocks of different sizes of working scene pictures in a specific scene, and naturally it is impossible to parse the optimal coding block size with coding rate priority.

[0018] The present invention provides a computer device comprising a memory, a processor, and a monitoring computer algorithm operating system stored in the memory and operable on the processor. An embodiment of the monitoring computer algorithm operating system of the present invention will be described in detail below.

[0019] The monitoring computer algorithm operating system shown in the embodiment of the present invention includes: A multiple learning device is used to perform multi-layer learning on the Hoffet neural network to obtain a coding prediction model output, wherein performing each layer of learning on the Hoffet neural network is performing each learning process on the Hoffet neural network; Specifically, a multiple learning device is used to perform multi-layer learning on the Hoffet neural network to obtain a coding prediction model output, and performing each layer of learning on the Hoffet neural network is to perform each learning process on the Hoffet neural network, including: using a MATLAB toolbox to complete the simulation and testing of the multi-layer learning of the Hoffet neural network; An image receiving device is connected to the fisheye camera mechanism and is used to obtain a work scene image, wherein the work scene image is an image obtained by the fisheye camera mechanism disposed inside the office building performing a wide-field shooting of the work scene inside the office building; a first detection device, connected to the picture receiving device, for detecting various picture parameters of the working scene picture, wherein the various picture parameters of the working scene picture are horizontal resolution, vertical resolution, maximum pixel depth of field, minimum pixel depth of field, signal-to-noise ratio, and picture content repetitiveness of the working scene picture; a second detection device for detecting multiple pieces of scene information of a work scene inside an office building, wherein the multiple pieces of scene information of the work scene inside the office building include an internal space volume of the work scene inside the office building, an area of ​​a background board, a distance from the background board to a fisheye camera mechanism, the number of employees, and the number of desks; a time prediction mechanism, connected to the multiple learning device, the first detection device, and the second detection device, respectively, for using a coding prediction model to intelligently predict a predicted value of the coding time required to perform intra-frame coding on the work scene picture under the coding block size based on the coding block size used for intra-frame coding, various picture parameters of the work scene picture, and multiple scene information of the work scene inside the office building; A block selection mechanism is connected to the time prediction mechanism and is used to obtain predicted values ​​of the encoding time required to perform intra-frame encoding on the working scene image at various encoding block sizes, and output the encoding block size corresponding to the predicted value of the encoding time with the smallest value as the optimal encoding block size with priority on encoding rate; The image receiving device is connected to the fisheye camera mechanism and is used to obtain a work scene image, wherein the work scene image is an image obtained by the fisheye camera mechanism disposed inside the office building performing wide-field shooting of the work scene inside the office building, and includes: the shooting field of view angle of the fisheye camera mechanism is greater than or equal to a set field of view angle threshold; The multiple learning device is used to perform multi-layer learning on the Hoffet neural network to obtain the coding prediction model output, and performing each layer of learning on the Hoffet neural network is performing each learning process on the Hoffet neural network, including: the number of learning layers performed on the Hoffet neural network is proportional to the total number of pixels of the working scene screen; Among them, the first detection device is connected to the picture receiving device and is used to detect various picture parameters of the working scene picture, and the various picture parameters of the working scene picture are the horizontal resolution, vertical resolution, maximum pixel depth of field, minimum pixel depth of field, signal-to-noise ratio and picture content repetition of the working scene picture, including: the maximum pixel depth of field of the working scene picture is the maximum value of each depth of field value corresponding to each pixel in the working scene picture; wherein, a coding prediction model is used to intelligently predict the coding time required to perform intra-frame coding on the work scene picture at the coding block size based on the coding block size used for intra-frame coding, various picture parameters of the work scene picture, and multiple scene information of the work scene inside the office building; and input data of the predicted value is subjected to numerical normalization processing to obtain input data after the numerical normalization processing is performed; Among them, the method of using a coding prediction model to intelligently predict the coding time required for performing intra-frame coding on the working scene picture under the coding block size, various picture parameters of the working scene picture, and multiple scene information of the working scene inside the office building, and performing numerical normalization processing on the input data of the predicted value to obtain the input data after the numerical normalization processing is completed includes: using a coding prediction model to intelligently predict the coding time required for performing intra-frame coding on the working scene picture under the coding block size, various picture parameters of the working scene picture, and multiple scene information of the working scene inside the office building, and performing numerical normalization processing based on hexadecimal numerical conversion to obtain the input data after the numerical normalization processing is completed; And wherein, the input data of the predicted value of the coding time required for performing intra-frame coding on the work scene picture at the coding block size is intelligently predicted based on the coding block size used for intra-frame coding, the various picture parameters of the work scene picture, and multiple scene information of the work scene inside the office building is subjected to numerical normalization processing to obtain the input data after the numerical normalization processing is completed, including: the input data of the predicted value of the coding time required for performing intra-frame coding on the work scene picture at the coding block size is intelligently predicted based on the coding block size used for intra-frame coding, the various picture parameters of the work scene picture, and multiple scene information of the work scene inside the office building is subjected to numerical normalization processing based on octal value conversion to obtain the input data after the numerical normalization processing is completed; And wherein, the input data of the predicted value of the coding time required for performing intra-frame coding on the work scene picture at the coding block size is intelligently predicted based on the coding block size used for intra-frame coding, various picture parameters of the work scene picture, and multiple scene information of the work scene inside the office building is subjected to numerical normalization processing to obtain the input data after the numerical normalization processing is completed, including: the input data of the predicted value of the coding time required for performing intra-frame coding on the work scene picture at the coding block size is intelligently predicted based on the coding block size used for intra-frame coding, various picture parameters of the work scene picture, and multiple scene information of the work scene inside the office building is subjected to numerical normalization processing based on binary numerical value conversion to obtain the input data after the numerical normalization processing is completed; And wherein, the input data of the predicted value of the coding time required for performing intra-frame coding on the work scene picture at the coding block size is intelligently predicted based on the coding block size used for intra-frame coding, the various picture parameters of the work scene picture, and multiple scene information of the work scene inside the office building is subjected to numerical normalization processing to obtain the input data after the numerical normalization processing is completed, including: the input data of the predicted value of the coding time required for performing intra-frame coding on the work scene picture at the coding block size is intelligently predicted based on the coding block size used for intra-frame coding, the various picture parameters of the work scene picture, and multiple scene information of the work scene inside the office building is subjected to numerical normalization processing based on decimal value conversion to obtain the input data after the numerical normalization processing is completed; And wherein, the output data of the predicted value of the encoding time required to perform intra-frame encoding on the work scene picture at the encoding block size is intelligently predicted using the encoding prediction model based on the encoding block size used for intra-frame encoding, various picture parameters of the work scene picture, and multiple scene information of the work scene inside the office building, and the predicted value is in the form of a numerical representation after numerical normalization processing; And wherein, the output data of the predicted value of the coding time required to perform intra-frame coding on the working scene picture under the coding block size is intelligently predicted based on the coding prediction model used for intra-frame coding, the various picture parameters of the working scene picture, and multiple scene information of the working scene inside the office building is a numerical representation after numerical normalization processing, including: the output data of the predicted value of the coding time required to perform intra-frame coding on the working scene picture under the coding block size is intelligently predicted based on the coding prediction model used for intra-frame coding, the various picture parameters of the working scene picture, and multiple scene information of the working scene inside the office building is a numerical representation after numerical normalization processing based on hexadecimal numerical conversion, a numerical representation after numerical normalization processing based on octal numerical conversion, a numerical representation after numerical normalization processing based on binary numerical conversion, and a numerical representation after numerical normalization processing based on decimal numerical conversion.

[0020] In addition, in the monitoring computer algorithm operating system, a first detection device is connected to the picture receiving device and is used to detect various picture parameters of the working scene picture. The various picture parameters of the working scene picture are the horizontal resolution, vertical resolution, maximum pixel depth of field, minimum pixel depth of field, signal-to-noise ratio and picture content repetition of the working scene picture, and also include: the maximum pixel depth of field of the working scene picture is the minimum value of the depth of field values ​​corresponding to each pixel in the working scene picture.

[0021] The monitoring computer algorithm operating system of the present invention addresses the technical problem in the prior art of being unable to select the optimal coding block size for the working scene picture in a specific scenario due to the lack of coding rate priority. By obtaining multiple pieces of scene information of the working scene and various picture parameters of the working scene picture, the different coding times of the working scene picture in the specific scenario under different sizes of coding blocks are intelligently predicted, and the coding block size corresponding to the coding time with the smallest value is used as the optimal coding block size with coding rate priority, thereby improving the efficiency and performance of image processing and solving the above technical problems.

[0022] Having described the embodiments of the present invention and its advantages, it should be noted that various changes, substitutions and alterations can be made herein without departing from the spirit and scope of the invention as defined by the appended claims. Additionally, the use of terms such as "first" and "second" does not denote any order or importance, but rather the terms "first", "second", etc. are used to distinguish one element from another.

Claims

1. A computer device comprising a memory, a processor, and a monitoring computer algorithm operating system stored in the memory and operable on the processor, characterized in that: The system includes: A multiple learning device is used to perform multi-layer learning on the Hoffet neural network to obtain a coding prediction model output, wherein performing each layer of learning on the Hoffet neural network is performing each learning process on the Hoffet neural network, wherein the number of layers of learning performed on the Hoffet neural network is proportional to the total number of pixels of the working scene image; an image receiving device connected to the fisheye camera mechanism and configured to capture an image of a work scene, wherein the work scene image is an image captured by the fisheye camera mechanism disposed inside the office building through a wide field of view of the work scene inside the office building, wherein the field of view angle of the fisheye camera mechanism is greater than or equal to a set field of view angle threshold; A first detection device is connected to the image receiving device and is used to detect various image parameters of the working scene image, where the various image parameters of the working scene image are horizontal resolution, vertical resolution, maximum pixel depth of field, minimum pixel depth of field, signal-to-noise ratio, and image content repetitiveness; a second detection device for detecting multiple pieces of scene information of a work scene inside an office building, wherein the multiple pieces of scene information of the work scene inside the office building include an internal space volume of the work scene inside the office building, an area of ​​a background board, a distance from the background board to the fisheye camera mechanism, the number of employees, and the number of desks; a time prediction mechanism, connected to the multiple learning device, the first detection device, and the second detection device, respectively, for using a coding prediction model to intelligently predict a predicted value of the coding time required to perform intra-frame coding on the work scene picture at the coding block size based on the coding block size used for intra-frame coding, various picture parameters of the work scene picture, and multiple scene information of the work scene inside the office building; The block selection mechanism is connected to the time prediction mechanism and is used to obtain the predicted values ​​of the encoding time required to perform intra-frame encoding on the working scene picture under various encoding block sizes, and output the encoding block size corresponding to the predicted value of the encoding time with the smallest value as the optimal encoding block size with coding rate priority.

2. The computer device according to claim 1, wherein: The first detection device is connected to the picture receiving device and is used to detect various picture parameters of the working scene picture. The various picture parameters of the working scene picture are the horizontal resolution, vertical resolution, maximum pixel depth of field, minimum pixel depth of field, signal-to-noise ratio and picture content repetition of the working scene picture, including: the maximum pixel depth of field of the working scene picture is the maximum value of the depth of field values ​​corresponding to each pixel in the working scene picture.

3. The computer device according to claim 2, wherein: A coding prediction model is used to intelligently predict the coding time required to perform intra-frame coding on the work scene picture under the coding block size based on the coding block size used for intra-frame coding, various picture parameters of the work scene picture, and multiple scene information of the work scene inside the office building. The input data of the predicted value is numerically normalized to obtain the input data after the numerical normalization processing.

4. The computer device according to claim 3, wherein: A coding prediction model is used to intelligently predict the coding time required to perform intra-frame coding on the working scene picture under the coding block size, various picture parameters of the working scene picture, and multiple scene information of the working scene inside the office building. The input data of the predicted value is subjected to numerical normalization processing to obtain the input data after the numerical normalization processing is completed. The method includes: a coding prediction model is used to intelligently predict the coding block size used for intra-frame coding, various picture parameters of the working scene picture, and multiple scene information of the working scene inside the office building. The input data of the predicted value is subjected to numerical normalization processing based on hexadecimal numerical conversion to obtain the input data after the numerical normalization processing is completed.

5. The computer device according to claim 3, wherein: A coding prediction model is used to intelligently predict the coding time required to perform intra-frame coding on the working scene picture under the coding block size, various picture parameters of the working scene picture, and multiple scene information of the working scene inside the office building. The input data of the predicted value is subjected to numerical normalization processing to obtain the input data after the numerical normalization processing is completed. The method includes: a coding prediction model is used to intelligently predict the coding block size used for intra-frame coding, various picture parameters of the working scene picture, and multiple scene information of the working scene inside the office building. The input data of the predicted value is subjected to numerical normalization processing based on octal value conversion to obtain the input data after the numerical normalization processing is completed.

6. The computer device according to claim 3, wherein: A coding prediction model is used to intelligently predict the coding time required to perform intra-frame coding on the working scene picture under the coding block size, various picture parameters of the working scene picture, and multiple scene information of the working scene inside the office building. The input data of the predicted value is subjected to numerical normalization processing to obtain the input data after the numerical normalization processing is completed. The method includes: a coding prediction model is used to intelligently predict the coding block size used for intra-frame coding, various picture parameters of the working scene picture, and multiple scene information of the working scene inside the office building. The input data of the predicted value is subjected to numerical normalization processing based on binary numerical conversion to obtain the input data after the numerical normalization processing is completed.

7. The computer device according to claim 3, wherein: A coding prediction model is used to intelligently predict the coding time required to perform intra-frame coding on the working scene picture under the coding block size, various picture parameters of the working scene picture, and multiple scene information of the working scene inside the office building. The input data of the predicted value is subjected to numerical normalization processing to obtain the input data after the numerical normalization processing is completed. The method includes: a coding prediction model is used to intelligently predict the coding block size used for intra-frame coding, various picture parameters of the working scene picture, and multiple scene information of the working scene inside the office building. The input data of the predicted value is subjected to numerical normalization processing based on decimal value conversion to obtain the input data after the numerical normalization processing is completed.

8. The computer device according to claim 2, wherein: A coding prediction model is used to intelligently predict the coding time required to perform intra-frame coding on the work scene picture under the coding block size based on the coding block size used for intra-frame coding, various picture parameters of the work scene picture, and multiple scene information of the work scene inside the office building. The output data of the predicted value is a numerical representation after numerical normalization.

9. The computer device according to claim 8, wherein: The coding prediction model is used to intelligently predict the coding time required to perform intra-frame coding on the working scene picture under the coding block size based on the coding block size used for intra-frame coding, various picture parameters of the working scene picture, and multiple scene information of the working scene inside the office building. The output data of the predicted value is a numerical representation after numerical normalization processing, including: the coding prediction model is used to intelligently predict the coding block size used for intra-frame coding, various picture parameters of the working scene picture, and multiple scene information of the working scene inside the office building. The output data of the predicted value is a numerical representation after numerical normalization processing based on hexadecimal numerical conversion, a numerical representation after numerical normalization processing based on octal numerical conversion, a numerical representation after numerical normalization processing based on binary numerical conversion, and a numerical representation after numerical normalization processing based on decimal numerical conversion.