FPGA-based image processor prototype verification platform and image verification method

By running RTL code on FPGA and utilizing its parallel computing characteristics, the processing speed and energy efficiency problems of image processor module verification in the prior art are solved, and more efficient image processing and verification are achieved, reducing power consumption and development cycles are achieved. .

CN118918420APending Publication Date: 2024-11-08XI AN JIAOTONG UNIV
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411202873.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-29
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

Existing CPU or GPU-based image processor module verification methods have shortcomings in terms of processing speed, energy efficiency ratio, operational complexity and verification cycle, which limits the overall performance and efficiency of the system, especially when processing complex image data or high frame rate videos. Poor performance at the time.

Method used

Using the FPGA-based image processor prototype verification platform, combining the central processing unit, image processing module, asynchronous transceiver and synchronous dynamic random memory, the image data processing and verification are used to run RTL code on the FPGA, and the parallel computing characteristics of FPGA are used. Improve processing efficiency and debug and monitor with integrated logic analyzer.

Benefits of technology

Significantly improves the parallel processing capability and real-time performance of the image processor, reduces power consumption, shortens development cycle, improves the reliability and efficiency of the verification platform, reduces subsequent rework and adjustments, and achieves more efficient image processing and verification. .

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118918420A_ABST
    Figure CN118918420A_ABST
Patent Text Reader

Abstract

The invention belongs to the field of prototype verification of image processing devices, and discloses an image processor prototype verification platform based on an FPGA and an image verification method. The method comprises the following steps: a central processing unit is connected with a flash memory processor, reads image data in the flash memory processor, and transmits the image data to a second synchronous dynamic random access memory for verification; reading the processed image data in the second synchronous dynamic random access memory, and comparing and checking the processed image data; the to-be-verified image processing module processes the verified image data in the second synchronous dynamic random access memory on the FPGA through an RTL code, and stores the processed image data to the second synchronous dynamic random access memory; the asynchronous transceiving transmitter transmits the processed image data after successful verification to a computer; and the first synchronous dynamic random access memory stores the processed image data after successful verification. According to the invention, the problem of insufficient parallel processing capability when the prototype verification task of the complex image processor is executed in the prior art can be solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of prototype verification of image processing devices, and in particular relates to an image processor prototype verification platform and an image verification method based on FPGA. Background Art

[0002] Existing image processor module verification usually relies on the central processing unit (CPU) or graphics processing unit (GPU) for calculation. Although this method has high flexibility, it is often not ideal in terms of processing speed and energy efficiency. Specifically, although CPU and GPU have strong versatility and programming portability when processing complex image algorithms, they have certain limitations in parallel processing capabilities and real-time performance. Especially when it is necessary to process large amounts of image data or high-frame rate videos, the performance bottlenecks of CPU and GPU will significantly affect the overall efficiency of the system. In addition, when performing high-intensity image processing tasks, CPU and GPU usually generate high power consumption, which is particularly disadvantageous in battery-powered embedded systems or mobile devices.

[0003] When performing RTL (Register Transfer Level) level verification of hardware circuits, the UVM (Universal Verification Methodology) method is usually used for verification on a server with a built verification environment. UVM is a standard method widely used in integrated circuit design verification. It helps designers effectively verify complex hardware designs by providing a set of systematic verification components and mechanisms. However, the use of the UVM verification method and platform is relatively complex, involving multiple links and steps: First, the setting of configuration parameters: the verification environment needs to configure a large number of parameters to ensure that all functions and boundary conditions of the design can be covered. These parameters include the timing of the input signal, trigger conditions, verification coverage, etc. Second, the construction of the UVM platform environment: a complete UVM verification platform needs to be built on the server, including installing and configuring verification tools, building verification buses, creating verification components, etc. This process consumes a lot of time and resources. Third, test case development: in order to fully verify the functions of the hardware design, a large number of test cases need to be written. The test case needs to describe in detail various possible input situations and their expected outputs, and include tests for abnormal situations and boundary conditions. Fourth, simulation model writing: in addition to test cases, simulation models also need to be written to simulate the behavior of the hardware design in actual operation. These models need to be highly accurate to ensure the reliability of the verification results. Fifth, the operation process is cumbersome: the entire verification process requires multiple iterations, and a large number of simulations need to be rerun after each design modification to ensure that the modification does not introduce new problems. This process is very cumbersome and requires verification engineers to have deep expertise and rich experience. Sixth, the verification cycle is long: due to the complexity and comprehensiveness of the UVM method, the entire verification cycle is usually long. From the construction of the verification environment to the development of test cases, to simulation and debugging, each step requires a lot of time and effort. This may become a bottleneck in the project progress when the product development cycle is tight.

[0004] Therefore, through the above content, the existing image processor module verification method based on CPU or GPU, and the RTL level verification method based on UVM, although they have certain advantages in flexibility and comprehensiveness of verification, have obvious deficiencies in processing speed, energy efficiency ratio, operation complexity and verification cycle. These deficiencies limit the overall performance and efficiency of the system and increase the difficulty of development and verification. Therefore, there is an urgent need for a more efficient and simpler image processor verification platform to cope with the increasingly complex image processing tasks and the increasingly compact product development cycle. Summary of the invention

[0005] The present invention provides an image processor prototype verification platform and an image verification method based on FPGA, so as to solve the problem of insufficient parallel processing capability in the prior art when executing complex image processing tasks.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] In a first aspect, an FPGA-based image processor prototype verification platform includes: a central processing unit, an image processing module, an asynchronous receiver-transmitter, a first synchronous dynamic random access memory and a second synchronous dynamic random access memory;

[0008] The central processing unit is connected to the flash memory processor and is used to read the image data in the flash memory processor, transfer it to the second synchronous dynamic random access memory and perform verification; and at the same time, read the processed image data in the second synchronous dynamic random access memory and perform comparison verification;

[0009] The image processing module to be verified is used for, after the verification of the central processing unit is completed, processing the verified image data in the second synchronous dynamic random access memory through the RTL code on the FPGA, and then storing the processed image data in the second synchronous dynamic random access memory;

[0010] an asynchronous receiver-transmitter, connected to the central processing unit, for transmitting the processed image data after successful verification to a computer;

[0011] The first synchronous dynamic random access memory is connected to the central processing unit and is used to store the processed image data after successful verification.

[0012] In some embodiments, an integrated logic analyzer is also included, which is connected to the image processing module to be verified and is used to debug and monitor the signals and data flows in the image processing module to be verified; at the same time, the signals in the image processing module to be verified are captured and analyzed inside the FPGA.

[0013] In some embodiments, a switch control unit is further included, which is used to drive an LED to provide a prompt after the image data in the second synchronous dynamic random access memory is successfully verified.

[0014] In some implementations, the first synchronous dynamic random access memory and the second synchronous dynamic random access memory are double rate synchronous dynamic random access memory.

[0015] In a second aspect, an image verification method based on an FPGA-based image processor prototype verification platform includes the following steps:

[0016] Read the image data and verify it. After successful verification, transmit the verified image data to the image processing module to be verified;

[0017] The image processing module to be verified processes the verified image data by the RTL code transplanted on the FPGA to obtain processed image data;

[0018] The processed image data is compared and verified with the expected image data. If the comparison and verification is successful, the processed image data is output to complete the image verification; if the comparison and verification fails, the image processing module to be verified is returned to process the verified image data and then the comparison and verification is performed again.

[0019] In some embodiments, the step of reading the image data and verifying it, and after successful verification, transmitting the verified image data to the image processing module to be verified, specifically includes:

[0020] The central processing unit reads the image data and performs verification. If the verification is successful, the verified image data is transmitted to the image processing module to be verified; if the verification is unsuccessful, the image data is read again and verified until the verification is successful.

[0021] In some embodiments, the central processing unit reads the processed image data and compares and verifies it with the expected image data.

[0022] In some implementations, the verified image data is stored in a second synchronous dynamic random access memory, waiting to be read by the to-be-verified image processing module;

[0023] storing the processed image data before comparison and verification in the second synchronous dynamic random access memory, waiting for the central processing unit to read;

[0024] The processed image data after comparison and verification is stored in the first synchronous dynamic random access memory.

[0025] In a third aspect, a computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the image verification method of the FPGA-based image processor prototype verification platform are implemented.

[0026] In a fourth aspect, a computer program product is provided, the computer product comprising a computer program, and the computer program, when executed by a processor, implements the steps of the image verification method of the FPGA-based image processor prototype verification platform.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] The present invention provides an image processor prototype verification platform based on FPGA, including a central processing unit, an image processing module, an asynchronous transceiver, a first synchronous dynamic random access memory and a second synchronous dynamic random access memory. The present invention mainly processes the verified image data in the second synchronous dynamic random access memory on the FPGA through the RTL code through the image processing module to be verified, and then verifies it through the central processing unit. The present invention is different from the existing image processor prototype verification platform. The present invention is based on the FPGA prototype verification system platform, and the RTL code is transplanted to the FPGA, and the RTL-level hardware circuit can be verified close to the real use scenario, avoiding the high cost when directly using ASIC for design. The FPGA can flexibly adjust and optimize the design to reduce the development cost; at the same time, it can avoid the situation that the existing technology simply relies on the parallel processing capability and real-time performance of the CPU or GPU to process complex images, and solve the problem of insufficient parallel processing capability of the existing technology when performing complex image processing tasks. Compared with CPU and GPU, FPGA has lower power consumption when processing high-intensity computing tasks, thereby extending the battery life of the device and reducing the heat dissipation requirements and related hardware costs.

[0029] Furthermore, the present invention uses an integrated logic analyzer to connect the image data module to be verified, so as to solve the problem of RTL level design and implementation of the image data module to be verified in FPGA.

[0030] Furthermore, the present invention adopts a double-rate synchronous dynamic random access memory as a storage medium, which can greatly improve the data reading and writing speed and storage efficiency, and improve the image processing speed of the image processor prototype verification platform.

[0031] The present invention provides an image verification method for an image processor prototype verification platform based on FPGA, wherein the image processing module to be verified processes the verified image data in the RTL code transplanted on the FPGA to obtain the processed image data. By implementing the image processing algorithm as the RTL code on the FPGA, the parallel computing characteristics of the FPGA can be fully utilized, and the data processing throughput can be greatly improved, which has higher efficiency and shorter processing delay than the traditional CPU or GPU; at the same time, the present invention transplants the RTL code to the FPGA hardware system platform, and can be tested and optimized in an actual operating environment close to the final product, so that potential problems can be discovered and solved earlier, the reliability and performance of the image processor prototype verification platform can be improved, and rework and adjustment in the subsequent development stage can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 A schematic diagram of the structure of the FPGA-based image processor prototype verification platform provided in Example 1;

[0033] Figure 2 This is a flowchart of the PS part of the FPGA-based image processor prototype verification platform provided in Example 1;

[0034] Figure 3 This is a flowchart of the PL part of the FPGA-based image processor prototype verification platform provided in Example 1;

[0035] Figure 4 A flowchart of an image verification method for an FPGA-based image processor prototype verification platform provided in Example 2. DETAILED DESCRIPTION

[0036] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solution of the present invention will be further described in detail below in conjunction with the accompanying drawings, and the described content is intended to explain the present invention rather than to limit it.

[0037] It should be noted that the terms "including" and "having" and any variations thereof in the specification and claims of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or are inherent to these processes, methods, systems, products or apparatus.

[0038] Embodiment 1

[0039] This embodiment provides an image processor prototype verification platform based on FPGA (Field-Programmable Gate Array), which combines the high-performance parallel processor capability of FPGA and the flexible control capability of CPU, which is conducive to accelerating the execution of complex image processing algorithms while maintaining the programmability and adaptability of the system. The prototype verification platform of this embodiment realizes rapid prototype verification, shortens the development cycle, and allows tasks to be allocated between hardware and software to optimize overall performance. At the same time, this platform also supports comprehensive system-level debugging and performance analysis, which can provide more design and verification flexibility. The specific design ideas are as follows:

[0040] The design idea of ​​the RTL-level prototype verification system for the image processor module based on the FPGA platform includes the following key steps: First, the RTL code, IP core and constraints of the image processor are converted into a design suitable for the FPGA platform to ensure that the gated clock, latch and asynchronous delay structures that are not applicable in the FPGA are avoided; second, preprocessing instructions such as define and ifdef are used to manage specific code blocks, isolate ASIC-specific designs such as test insertion and low-power optimization; then, wrappers are used to make necessary interface and function adjustments to ensure the portability of the ASIC RTL code. When implemented on the FPGA, its high-performance parallel processing capabilities are used to optimize the execution speed of the image processing algorithm while maintaining low power consumption and high energy efficiency. Finally, the FPGA prototype verification system is used to perform testing and debugging close to the actual usage scenario, and potential problems are discovered and solved early, thereby improving the efficiency and reliability of the entire design process.

[0041] In summary, when porting ASICRTL code to the FPGA platform, the following principles were followed to ensure smooth porting:

[0042] (1) Avoid using latches: Latches may cause unstable timing behavior in FPGAs, so flip-flops are used instead.

[0043] (2) Avoid combinational logic loops: Combinatorial logic loops can cause feedback delays and instability, which can be avoided through proper logic design.

[0044] (3) Optimize RTL code: Avoid including ASIC optimization techniques such as clock gating, test insertion, and low power consumption in the original RTL code. These techniques may not be implemented in FPGA or may cause additional complexity.

[0045] (4) Use preprocessor directives: Include or remove prototype code by using preprocessor directives such as 'define and 'ifdef. These structures can be used to isolate specific code blocks such as DFT design, memory instantiation, etc., so that they can be flexibly switched between different platforms.

[0046] (5) Use wrappers: To avoid modifying the original ASIC RTL code, use wrappers to make necessary adjustments and replacements. This method can adapt to FPGA design requirements without changing the original file, ensuring code portability and maintainability.

[0047] Through the above ideas, the RTL code of ASIC design can be effectively transplanted to the FPGA platform for prototype verification. This not only ensures the functional correctness of the design, but also can discover and solve potential problems in the early stage, thereby improving the efficiency and reliability of the entire prototype verification platform process.

[0048] Therefore, the FPGA-based image processor prototype verification platform provided by this embodiment is built through the above content. Figure 1 As shown in the figure, it consists of an FPGA with a CPU processor, which is divided into a PS (Processing System) processing system part and a PL (Programmable Logic) programmable logic part. Each part has different functional divisions. The PS part is responsible for the function scheduling of the entire system platform, external input and output interfaces, result comparison, and state representation, and is developed and implemented using an embedded C program; the PL part is responsible for migrating the RTL code of the image module, internal signal debug viewing, and the external interface of the functional components to be verified.

[0049] The PS (Processing System) processing system consists of the following functional components:

[0050] (1) CPU (Central Processing Unit): Responsible for the overall control and processing of the system. It manages system operations, data flow, and peripheral interfaces by running embedded software.

[0051] exist Figure 1 In the system, the CPU is connected to each interface module (UIF, SDIF, DIF0, DIF1) and the control unit (CU), and is responsible for directing and coordinating their work.

[0052] (2) UIF (UART (Universal Asynchronous Receiver / Transmitter) Interface): Provides communication functions with external UART devices and PC (Personal Computer).

[0053] exist Figure 1 In the process, the UART interface communicates with the CPU in serial mode for debugging, data transmission, and printing verification result data to the PC.

[0054] (3) SDIF (SD Card Interface): allows the system to exchange data with the SD card for data storage and reading. The SD card is a flash memory processor;

[0055] exist Figure 1 In the process, SDIF is connected to the CPU, and data transmission is realized through SPI (Serial Peripheral Interface) or SD (Secure Digital) protocol to store image source data.

[0056] (4) DIF0 (Data Interface 0): responsible for communicating with PS_DDR and managing data storage and reading. Through this interface, the CPU can store the processed process data or execute programs in PS_DDR (Double Data Rate Synchronous Dynamic Random Access Memory), realizing high-bandwidth data interaction and large program execution.

[0057] (5) CU (Control Unit): manages the status of LEDs and switches and is controlled by the CPU. It is used to process user input (such as switches), system status indication (such as LEDs), external trigger control or external interrupt enable, etc., to achieve flexible and free external control of the system.

[0058] like Figure 2 As shown, the PS part of the workflow is as follows:

[0059] First, generate the image source data for testing through PC software. This data can be pre-recorded image files or dynamically generated image data. Load the generated data source into the SD card, so that the SD card acts as an external storage device to provide image source data for the system.

[0060] Next, start the CPU and put it into working state, ready to execute a series of scheduling control instructions to process image data. The CPU reads the image data from the SD card through the SD card interface (SDIF) and loads it into the system. After reading, the CPU writes the data to PL_DDR (dynamic random access memory of the programmable logic part). The high bandwidth and low latency characteristics of PL_DDR make it an ideal choice for image data storage, supporting subsequent high-speed image processing operations of the image processing module to be verified. After that, the CPU will read and verify the written data. If the verification is incorrect, the data reading and writing operations will be performed again. If the verification is correct, the CPU drives the LED through the control unit to indicate that the data transmission operation is successfully completed. This step not only provides visual feedback, but also helps developers confirm the completion of data transmission.

[0061] Subsequently, the CPU performs data processing and result comparison. This step involves executing the image processing function algorithm implemented by the module to be verified and comparing the processing results with the expected results to verify the accuracy and completeness of the data processing of the module to be verified.

[0062] After processing is complete, the CPU outputs the comparison results to the connected PC via the UART interface (UIF). The UART interface provides a standard serial communication method that facilitates data transfer to an external computer for further analysis and recording. Finally, the CPU ends the current task, the entire image processing and verification process is completed, and the system enters standby mode, waiting for the next operation.

[0063] Through the above process, the system can effectively utilize the high-performance parallel processing capabilities of FPGA to achieve efficient image processing and verification operations. This verification method provides an environment close to the actual usage scenario, significantly improving the development and verification efficiency of the image processor module.

[0064] The PL (Programmable Logic) programmable logic part consists of the following functional components:

[0065] (1) Image processing module to be verified (DUT, Device Under Test): The verified image processing unit in the system is responsible for executing the RTL migration of the image processing module. This module is implemented on the FPGA through RTL code, which can perform high-performance parallel processing, significantly improve the image processing speed, and effectively test potential problems of the image processing module.

[0066] (2) ILA (Integrated Logic Analyzer): used to debug and monitor signals and data flows in the PL. It can capture and analyze the signals of the image processing module inside the FPGA to solve problems in design and implementation.

[0067] (3) DIF1 (Data Interface 1): responsible for communicating with PL_DDR and managing data storage and reading. This interface enables the CPU to access the data memory of the PL part, which mainly stores the data and processing results generated during the working process of the image processing module, and provides data storage support for image processing task verification.

[0068] (4) JTAG (Joint Test Action Group, FPGA debugging interface): used to debug and program FPGA. Provides a standard interface for communicating, programming, and debugging FPGA.

[0069] It should be noted that PS_DDR corresponds to the aforementioned first synchronous dynamic random access memory, and PL_DDR corresponds to the aforementioned second synchronous dynamic random access memory.

[0070] like Figure 3 As shown, the PL part of the workflow is as follows:

[0071] First, the CPU triggers the DUT (image processing module to be verified) to start working. The DUT enters the data processing stage by reading the image source data in the PL_DDR (dynamic random access memory of the programmable logic part). After the DUT reads the image data from the PL_DDR, it starts to process the data and performs image processing operations according to the predetermined functional algorithm or logic.

[0072] Once the image data processing is completed, the DUT stores the processing results in PL_DDR. If the DUT does not store the results, it returns to the step where the DUT reads the image data in PL_DDR and re-processes and stores the data. This process ensures that the image data source in PL_DDR read by the DUT is correct.

[0073] If the DUT module stores the calculation result in PL_DDR, the CPU will compare and verify the result of DUT processing to confirm whether the processing result is as expected. If the comparison shows that the result data is correct, the CPU stores the result in PS_DDR to complete the final result data comparison. If the comparison shows that the result data is incorrect, it returns to DUT calculation and adds ILA for internal signal error correction.

[0074] After the result is stored in PL_DDR, the DUT ends the current task, the entire image processing and verification process is completed, and the system enters the standby state, waiting for the next operation to be triggered. This process is verified and verified multiple times to ensure the accuracy of the image processing data and the efficient operation of the system.

[0075] Overall, the flowchart shows the complete process of DUT prototype verification of the image processor module on the PL platform on FPGA through detailed steps. Through a series of operations such as CPU triggering, data reading, processing, verification and storage, the system can effectively verify the function and performance of the image processing module and ensure its reliability and stability in practical applications.

[0076] The FPGA-based image processor prototype verification platform provided in this embodiment realizes efficient image processing and system control functions through the collaborative work of PS and PL. The PS part is responsible for overall control scheduling and communication with external devices, while the PL part focuses on high-performance image processing verification tasks. Through the cooperation of various interface modules and memory, the system can flexibly expand functions and significantly improve data processing capabilities and efficiency.

[0077] Embodiment 2

[0078] Based on the image processor prototype verification platform provided in the first embodiment, this embodiment provides an image verification method based on the FPGA image processor prototype verification platform, such as Figure 4As shown in the figure, it is divided into data preparation and loading process, image processing and verification process and result output and indication process. The data preparation and loading process is mainly responsible for the preparation of image data source and data loading; the image processing and verification process mainly realizes the function realization and correctness verification of the image module to be verified; the result output and indication process mainly outputs the calculation results and comparison results to an external PC to facilitate the verification personnel to observe and correct errors.

[0079] The image verification method of the FPGA-based image processor prototype verification platform provided in this embodiment specifically includes the following steps:

[0080] (1) Data preparation and loading process

[0081] In the data preparation and loading process, the image source data needs to be generated first, which will be used for subsequent image processing and verification. The generated image data is stored in the SD card and then loaded into the system through the SD card interface. The CPU is started to start processing the data. The CPU reads the image data in the SD card and writes it to the PL_DDR (Programmable Logic Dynamic Random Access Memory). Next, the CPU reads the data written to the PL_DDR and verifies it to ensure the integrity and accuracy of the data. If the data verification is incorrect, the CPU will re-read the data in the SD card and write it to the PL_DDR until the data verification is correct. After the verification is correct, the CPU indicates the completion status of the data writing by driving the LED.

[0082] (2) Image processing and verification process

[0083] During the image processing and verification process, the CPU first triggers the DUT (Device Under Test, the image module to be verified) to start processing the image data. The DUT reads the pre-stored image data from the PL_DDR and processes the data according to the designed algorithm. After the processing is completed, the DUT stores the processing result data in the PL_DDR. If the storage result is unsuccessful, the DUT will reprocess the image data and store the result. The CPU then reads the result data processed by the DUT from the PL_DDR and performs a comparison check to ensure the correctness and consistency of the processing result. If the verification result is incorrect, the CPU will trigger the DUT to process the image data again until the result is verified to be correct.

[0084] (3) Result output and instruction process

[0085] During the result output and indication process, the CPU stores the final processing result in PS_DDR. Once the DUT completes the image processing task, the CPU outputs the processing result to the PC through the UART interface for recording and analysis. The correctness of the final processing result can be confirmed by the verification result compared by the CPU. After completing all operations, the CPU ends the task and the system returns to the initial state. Through the LED status indication and the data output of the UART interface, the user can intuitively understand the progress of the entire processing process and the final processing result, thereby ensuring the correctness and effectiveness of the image processing module verification.

[0086] The division of modules in the embodiments of the present invention is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional module in each embodiment of the present invention may be integrated into one processor, or may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of software functional modules.

[0087] This embodiment also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in a computer device and, of course, an extended storage medium supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by a processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of an image verification method of an image processor prototype verification platform based on FPGA in the above embodiment.

[0088] This embodiment further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the corresponding steps of the image verification method of the FPGA-based image processor prototype verification platform in the above embodiment are implemented.

[0089] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0090] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0091] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0092] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0093] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. An FPGA-based image processor prototype verification platform, characterized in that: include: A central processing unit, an image processing module, an asynchronous receiver-transmitter, a first synchronous dynamic random access memory and a second synchronous dynamic random access memory; The central processing unit is connected to the flash memory processor and is used to read the image data in the flash memory processor, transfer it to the second synchronous dynamic random access memory and perform verification; and at the same time, read the processed image data in the second synchronous dynamic random access memory and perform comparison verification; The image processing module to be verified is used for processing the verified image data in the second synchronous dynamic random access memory through the RTL code on the FPGA after the verification of the central processing unit is completed, and then storing the processed image data in the second synchronous dynamic random access memory; an asynchronous receiver-transmitter, connected to the central processing unit, for transmitting the processed image data after successful verification to a computer; The first synchronous dynamic random access memory is connected to the central processing unit and is used to store the processed image data after successful verification.

2. The FPGA-based image processor prototype verification platform according to claim 1, characterized in that: It also includes an integrated logic analyzer, which is connected to the image processing module to be verified and is used to debug and monitor the signals and data flows in the image processing module to be verified; at the same time, the signals in the image processing module to be verified are captured and analyzed inside the FPGA.

3. The FPGA-based image processor prototype verification platform according to claim 1, characterized in that: It also includes a switch control unit, which is used to drive the LED to give a prompt after the image data in the second synchronous dynamic random access memory is successfully verified.

4. The FPGA-based image processor prototype verification platform according to claim 1, characterized in that: The first synchronous dynamic random access memory and the second synchronous dynamic random access memory are double rate synchronous dynamic random access memory.

5. An image verification method based on an FPGA-based image processor prototype verification platform, characterized in that: The image processor prototype verification platform according to any one of claims 1 to 4 comprises the following steps: Read the image data and verify it. After successful verification, transmit the verified image data to the image processing module to be verified; The image processing module to be verified processes the verified image data through the RTL code transplanted on the FPGA to obtain processed image data; The processed image data is compared and verified with the expected image data. If the comparison and verification is successful, the processed image data is output to complete the image verification; if the comparison and verification fails, the image processing module to be verified is returned to process the verified image data and then the comparison and verification is performed again.

6. The image verification method of the FPGA-based image processor prototype verification platform according to claim 5, characterized in that: The step of reading the image data and verifying it, and after the verification is successful, transmitting the verified image data to the image processing module to be verified specifically includes: The central processing unit reads the image data and performs verification. If the verification is successful, the verified image data is transmitted to the image processing module to be verified; if the verification is unsuccessful, the image data is read again and verified until the verification is successful.

7. The image verification method of the FPGA-based image processor prototype verification platform according to claim 5, characterized in that: The central processing unit reads the processed image data and compares and verifies it with the expected image data.

8. The image verification method of the FPGA-based image processor prototype verification platform according to claim 7, characterized in that: The verified image data is stored in a second synchronous dynamic random access memory, waiting for the image processing module to be verified to read it; storing the processed image data before comparison and verification in the second synchronous dynamic random access memory, waiting for the central processing unit to read; The processed image data after comparison and verification is stored in the first synchronous dynamic random access memory.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the image verification method of the FPGA-based image processor prototype verification platform according to any one of claims 5 to 8 are implemented.

10. A computer program product, the computer product comprising a computer program, characterized in that: When the computer program is executed by a processor, the steps of the image verification method of an FPGA-based image processor prototype verification platform according to any one of claims 5 to 8 are implemented.

Citation Information

Patent Citations

  • SOC prototype verification method, system and device and medium

    CN111737933A

  • Prototype verification system, prototype verification method, processing unit, and apparatus

    CN113326670A

  • Resource evaluation method and device for chip prototype verification, equipment and medium

    CN118410751A

  • Verification platform and verification method, and computer storage medium

    WO2018218954A1