CN110400249A - A kind of image processing method based on CPU - Google Patents
A kind of image processing method based on CPU Download PDFInfo
- Publication number
- CN110400249A CN110400249A CN201910666233.8A CN201910666233A CN110400249A CN 110400249 A CN110400249 A CN 110400249A CN 201910666233 A CN201910666233 A CN 201910666233A CN 110400249 A CN110400249 A CN 110400249A
- Authority
- CN
- China
- Prior art keywords
- cpu
- fpga
- image
- hardware
- test
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T1/00—General purpose image data processing
- G06T1/20—Processor architectures; Processor configuration, e.g. pipelining
Abstract
The invention discloses a kind of image processing methods based on CPU, comprising the following steps: architecture testing CPU-FPGA processing module, basic algorithm is tested, be arranged operating value, framework operation CPU hardware and operation CPU-FPGA processing module, operation CPU-FPGA processing module to image data operation processing, operation CPU hardware to image data operation processing, operation CPU-FPGA processing result image with run CPU processing result image synthesize；The present invention, which handles hardware operating value by CPU hardware operating value and test CPU-FPGA, can choose out the CPU hardware of most suitable image operation processing to be processed, it can be improved the image procossing speed of service and effect, CPU-FPGA processing module is tested by framework and image processing speed can be improved in operation CPU-FPGA processing module, can provide the processing result image of a high quality for later period synthesis process.
Description
Technical field
The present invention relates to technical field of image processing more particularly to a kind of image processing methods based on CPU.
Background technique
As user constantly exceeds the requirement in terms of image procossing the computing capability of CPU, CPU processing capacity is also continuous
It is powerful, but after entering the 3D epoch, it has been found that huge 3D rendering processing data calculate so that CPU increasingly can't bear heavy burden,
And far beyond its computing capability.
CPU is the work such as the processing for being designed to processing common tasks, processing, operation and system core control, CPU's
Micro-architecture is to handle the work such as the little calculating class of data dependence, complicated cumbersome non-computational class for high efficiency and optimize.
As graphics calculations demand is increasing, the graphics calculations that the computing capability of CPU is also unable to satisfy rapid growth are needed, at present people
All image procossing, obtained processing result image quality and image procossing only are carried out simply by simple CPU hardware
Efficiency is not high enough, and therefore, the present invention proposes a kind of image processing method based on CPU, to solve deficiency in the prior art
Place.
Summary of the invention
In view of the above-mentioned problems, the present invention proposes a kind of image processing method based on CPU, by CPU hardware operating value and
Test CPU-FPGA processing hardware operating value can choose out the CPU hardware of most suitable image operation processing to be processed, Neng Gouti
Hi-vision handles the speed of service and effect, tests CPU-FPGA processing module by framework and operation CPU-FPGA processing module can
To improve image processing speed, the processing result image of a high quality can be provided for later period synthesis process, ensure that final
Processing result image quality it is high.
The present invention proposes a kind of image processing method based on CPU, comprising the following steps:
Step 1: CPU and FPGA are combined framework, form a new test CPU-FPGA processing module；
Step 2: utilize test CPU hardware and test CPU-FPGA processing module to the basic algorithm for image procossing
It is tested, obtains CPU hardware test result and test CPU-FPGA processing module test result；
Step 3: it is transported according to CPU hardware test result and test CPU-FPGA processing module test result setting CPU hardware
Row value and test CPU-FPGA handle hardware operating value；
Step 4: framework runs CPU hardware and operation CPU-FPGA processing module；
Step 5: image data to be processed is sent to operation CPU-FPGA processing module, and passes through operation CPU-
FPGA processing module carries out basic algorithm operation processing to image data, then obtains operation CPU-FPGA processing result image,
Obtained operation CPU-FPGA processing result image copy is sent to operation CPU hardware；
Step 6: image data to be processed is sent to operation CPU hardware, and by operation CPU hardware to picture number
According to basic algorithm operation processing is carried out, operation CPU processing result image is obtained；
Step 7: operation CPU-FPGA processing result image and operation CPU processing result image are synthesized, obtained most
Whole processing result image.
Further improvement lies in that: it is also embedded with driving unit in the test CPU-FPGA processing module in the step 1,
The driving unit is used to carry out image data to be processed to generate BitTorrent source file.
Further improvement lies in that: CPU hardware and test CPU-FPGA processing module are tested in the step 2 to for scheming
Before being tested as the basic algorithm of processing, need to carry out operation test to basic algorithm using a test image data.
Further improvement lies in that: the CPU hardware operating value and test CPU-FPGA processing hardware being arranged in the step 3
Operating value and be 1.
Further improvement lies in that: in the step 4 method particularly includes: according to the CPU hardware operating value set and test
CPU-FPGA handles hardware operating value and selects suitable CPU and FPGA, and runs CPU using selected CPU and FPGA architecture
Hardware and operation CPU-FPGA processing module.
Further improvement lies in that: in the step 4 during framework operation CPU-FPGA processing module, need to utilize drive
Moving cell generates BitTorrent source file and carries out framework to operation CPU-FPGA processing module.
Further improvement lies in that: the operation CPU-FPGA image procossing knot of operation CPU hardware is sent in the step 5
Fruit needs to be kept separately in operation CPU hardware and guarantees that calling can be read at any time.
Further improvement lies in that: in the step 7 method particularly includes: by being sent to, the carry out in operation CPU hardware is independent
It saves operation CPU-FPGA processing result image and is read out calling, with operation CPU processing result image in operation CPU hardware
It is synthesized, obtains final processing result image.
The invention has the benefit that by using test CPU hardware and test CPU-FPGA processing module to for scheming
The CPU hardware operating value and test CPU-FPGA processing hardware operating value tested and be provided as the basic algorithm of processing
The CPU hardware that can choose out most suitable image operation processing to be processed, can be improved the image procossing speed of service and effect, obtains
The processing result image quality arrived is high, and testing CPU-FPGA processing module and operation CPU-FPGA processing module by framework can be with
Image processing speed is improved, the processing result image of a high quality can be provided for later period synthesis process, ensure that final
Processing result image quality is high.
Specific embodiment
In order to deepen the understanding of the present invention, the present invention is further described below in conjunction with embodiment, the present embodiment
For explaining only the invention, it is not intended to limit the scope of the present invention..
A kind of image processing method based on CPU, comprising the following steps:
Step 1: being combined framework for CPU and FPGA, forms a new test CPU-FPGA processing module, test
Driving unit is also embedded in CPU-FPGA processing module, the driving unit is for generating image data to be processed
BitTorrent source file；
Step 2: carrying out operation test to basic algorithm using a test image data, then using test CPU hardware and
Test CPU-FPGA processing module tests the basic algorithm for image procossing, obtains CPU hardware test result and survey
Try CPU-FPGA processing module test result；
Step 3: it is transported according to CPU hardware test result and test CPU-FPGA processing module test result setting CPU hardware
Row value and test CPU-FPGA handle hardware operating value, CPU hardware operating value and test CPU-FPGA processing hardware operating value and
It is 1；
Step 4: suitable according to the CPU hardware operating value and test CPU-FPGA processing hardware operating value selection that set
CPU and FPGA, and using selected CPU and FPGA architecture operation CPU hardware and operation CPU-FPGA processing module, framework
During running CPU-FPGA processing module, need to generate BitTorrent source file using driving unit to operation CPU-FPGA
Processing module carries out framework；
Step 5: image data to be processed is sent to operation CPU-FPGA processing module, and passes through operation CPU-
FPGA processing module carries out basic algorithm operation processing to image data, then obtains operation CPU-FPGA processing result image,
Obtained operation CPU-FPGA processing result image copy is sent to operation CPU hardware, is sent to the operation of operation CPU hardware
CPU-FPGA processing result image needs to be kept separately in operation CPU hardware and guarantees that calling can be read at any time；
Step 6: image data to be processed is sent to operation CPU hardware, and by operation CPU hardware to picture number
According to basic algorithm operation processing is carried out, operation CPU processing result image is obtained；
Step 7: will be sent to operation CPU hardware in progress individually saves run CPU-FPGA processing result image into
Row, which is read, to be called, and is synthesized in operation CPU hardware with operation CPU processing result image, obtains final image procossing knot
Fruit.
By utilizing test CPU hardware and testing CPU-FPGA processing module to the basic algorithm progress for image procossing
The CPU hardware operating value and test CPU-FPGA processing hardware operating value tested and be provided can choose out most suitable wait locate
The CPU hardware for managing image operation processing, can be improved the image procossing speed of service and effect, obtained processing result image quality
Height tests CPU-FPGA processing module by framework and image processing speed can be improved in operation CPU-FPGA processing module, can
To provide the processing result image of a high quality for later period synthesis process, it ensure that final processing result image quality is high.
The basic principles, main features and advantages of the invention have been shown and described above.The technical staff of the industry should
Understand, the present invention is not limited to the above embodiments, and the above embodiments and description only describe originals of the invention
Reason, without departing from the spirit and scope of the present invention, various changes and improvements may be made to the invention, these changes and improvements
It all fall within the protetion scope of the claimed invention.The claimed scope of the invention is by appended claims and its equivalent circle
It is fixed.
Claims (8)
1. a kind of image processing method based on CPU, which comprises the following steps:
Step 1: CPU and FPGA are combined framework, form a new test CPU-FPGA processing module；
Step 2: utilizing test CPU hardware and tests CPU-FPGA processing module to the basic algorithm progress for image procossing
Test obtains CPU hardware test result and test CPU-FPGA processing module test result；
Step 3: CPU hardware operating value is arranged according to CPU hardware test result and test CPU-FPGA processing module test result
Hardware operating value is handled with test CPU-FPGA；
Step 4: framework runs CPU hardware and operation CPU-FPGA processing module；
Step 5: image data to be processed is sent to operation CPU-FPGA processing module, and by running at CPU-FPGA
It manages module and basic algorithm operation processing is carried out to image data, then obtain operation CPU-FPGA processing result image, will obtain
Operation CPU-FPGA processing result image copy be sent to operation CPU hardware；
Step 6: being sent to operation CPU hardware for image data to be processed, and by operation CPU hardware to image data into
Row basic algorithm operation processing obtains operation CPU processing result image；
Step 7: operation CPU-FPGA processing result image and operation CPU processing result image are synthesized, obtained final
Processing result image.
2. a kind of image processing method based on CPU according to claim 1, it is characterised in that: in the step 1
Driving unit is also embedded in test CPU-FPGA processing module, the driving unit is used to carry out image data to be processed
Generate BitTorrent source file.
3. a kind of image processing method based on CPU according to claim 1, it is characterised in that: surveyed in the step 2
Before examination CPU hardware and test CPU-FPGA processing module test the basic algorithm for image procossing, need to utilize one
Test image data carry out operation test to basic algorithm.
4. a kind of image processing method based on CPU according to claim 1, it is characterised in that: set in the step 3
The CPU hardware operating value and test CPU-FPGA set handle hardware operating value and are 1.
5. a kind of image processing method based on CPU according to claim 1, it is characterised in that: have in the step 4
Body method are as follows: according to the CPU hardware operating value that sets and test CPU-FPGA processing hardware operating value select suitable CPU and
FPGA, and utilize selected CPU and FPGA architecture operation CPU hardware and operation CPU-FPGA processing module.
6. a kind of image processing method based on CPU according to claim 2, it is characterised in that: frame in the step 4
During structure runs CPU-FPGA processing module, need to generate BitTorrent source file using driving unit to operation CPU-
FPGA processing module carries out framework.
7. a kind of image processing method based on CPU according to claim 1, it is characterised in that: sent out in the step 5
The operation CPU-FPGA processing result image to operation CPU hardware is sent to need to be kept separately and protected in operation CPU hardware
Card can be read calling at any time.
8. a kind of image processing method based on CPU according to claim 7, it is characterised in that: have in the step 7
Body method are as follows: the progress being sent in operation CPU hardware is individually saved into operation CPU-FPGA processing result image and is read out
It calls, is synthesized in operation CPU hardware with operation CPU processing result image, obtain final processing result image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910666233.8A CN110400249A (en) | 2019-07-23 | 2019-07-23 | A kind of image processing method based on CPU |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910666233.8A CN110400249A (en) | 2019-07-23 | 2019-07-23 | A kind of image processing method based on CPU |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110400249A true CN110400249A (en) | 2019-11-01 |
Family
ID=68325883
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910666233.8A Pending CN110400249A (en) | 2019-07-23 | 2019-07-23 | A kind of image processing method based on CPU |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110400249A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104331858A (en) * | 2014-11-24 | 2015-02-04 | 厦门美图之家科技有限公司 | Acceleration method for processing image by utilizing both CPU and GPU |
| CN107515860A (en) * | 2017-08-07 | 2017-12-26 | 中译语通科技（青岛）有限公司 | A kind of machine translation method based on neuron |
| CN108829530A (en) * | 2018-06-15 | 2018-11-16 | 郑州云海信息技术有限公司 | A kind of image processing method and device |
| CN109547367A (en) * | 2018-11-09 | 2019-03-29 | 中国航空无线电电子研究所 | Software Radio platform based on SCA |
| CN109583186A (en) * | 2018-12-04 | 2019-04-05 | 江西财经大学 | A kind of cloud computing formula image processing system |
-
2019
- 2019-07-23 CN CN201910666233.8A patent/CN110400249A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104331858A (en) * | 2014-11-24 | 2015-02-04 | 厦门美图之家科技有限公司 | Acceleration method for processing image by utilizing both CPU and GPU |
| CN107515860A (en) * | 2017-08-07 | 2017-12-26 | 中译语通科技（青岛）有限公司 | A kind of machine translation method based on neuron |
| CN108829530A (en) * | 2018-06-15 | 2018-11-16 | 郑州云海信息技术有限公司 | A kind of image processing method and device |
| CN109547367A (en) * | 2018-11-09 | 2019-03-29 | 中国航空无线电电子研究所 | Software Radio platform based on SCA |
| CN109583186A (en) * | 2018-12-04 | 2019-04-05 | 江西财经大学 | A kind of cloud computing formula image processing system |
Non-Patent Citations (3)
| Title |
|---|
| 包明 著: "《EDA技术与可编程器件的应用》", 31 October 2007 * |
| 日经BP社编: "《完全读懂AI应用前线》", 31 August 2018 * |
| 杨勇 等: "联合字典鲁棒稀疏表示的多聚焦图像融合", 《小型微型计算机系统》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI646502B (en) | Mapping multi-rate shading to monolithic programs | |
| US9454843B2 (en) | Modulated and blended anti-aliasing | |
| DE102019117585A1 (en) | Selectively pack patches for immersive video | |
| TW200844897A (en) | Processor with adaptive multi-shader | |
| DE102019117469A1 (en) | Video processing mechanism | |
| DE102019114970A1 (en) | ADVANCED IMMERSIVE MEDIA PIPELINE FOR CORRECTING ARTIFACTS AND CLARITY OF OBJECTS IN COMPUTING ENVIRONMENTS | |
| CN106339984B (en) | Distributed image ultra-resolution method based on K mean value driving convolutional neural networks | |
| TW200818054A (en) | Tile based precision rasterization in graphics pipeline | |
| EP1647935A3 (en) | Method and apparatus for enhancing image quality of a two-dimensional ultrasound image | |
| JP6379225B2 (en) | Method and apparatus for effective texture compression | |
| CN103888771A (en) | Parallel video image processing method based on GPGPU technology | |
| CN114223196A (en) | Systems and methods for foveal rendering | |
| TWI769138B (en) | Graphics processing method, pipeline circuit and computer program | |
| CN1282128C (en) | Image processing device and method | |
| JP2004073360A (en) | Method for preparing projection data, method for preparing pixel data, and multi-detector x-ray ct equipment | |
| JP2018502381A (en) | Reduced sorting in limited pipelines | |
| DE102019115130A1 (en) | Device and method for conservative morphological anti-aliasing with multiple scanning | |
| CN1329325A (en) | Image processing method and device, recording medium and imaging device | |
| US20150348222A1 (en) | Method and apparatus for parallel pixel shading | |
| DE102021127982A1 (en) | STREAMING A FIELD OF LIGHT WITH LOSSLESS OR LOSSY COMPRESSION | |
| Ignatov et al. | Pynet-v2 mobile: Efficient on-device photo processing with neural networks | |
| CN110400249A (en) | A kind of image processing method based on CPU | |
| DE102020106728A1 (en) | Background estimation for object segmentation using coarse level tracking | |
| CN103099635A (en) | Graphics processing unit (GPU) digital image system for microscopic captive test (CT) machine and control method of GPU digital image system for microscopic CT machine | |
| WO2022212196A1 (en) | Waveguide correction map compression |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |