CN113886042A - TDA2X vehicle gauge control platform-based parking space recognition algorithm deployment and scheduling method - Google Patents
TDA2X vehicle gauge control platform-based parking space recognition algorithm deployment and scheduling method Download PDFInfo
- Publication number
- CN113886042A CN113886042A CN202111140154.7A CN202111140154A CN113886042A CN 113886042 A CN113886042 A CN 113886042A CN 202111140154 A CN202111140154 A CN 202111140154A CN 113886042 A CN113886042 A CN 113886042A
- Authority
- CN
- China
- Prior art keywords
- parking space
- cpu
- tda2x
- control platform
- scheduling method
- 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
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000001514 detection method Methods 0.000 claims description 26
- 238000006243 chemical reaction Methods 0.000 claims description 22
- 230000006870 function Effects 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 18
- 238000010586 diagram Methods 0.000 claims description 5
- 238000002054 transplantation Methods 0.000 abstract description 3
- 238000013461 design Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 2
- 230000014509 gene expression Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
- G06T2207/30264—Parking
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method for deploying and scheduling a parking space recognition algorithm based on a TDA2X vehicle level control platform. The invention has clear structural logic, better universality and multiplexing support, fully exerts the strength of a multiprocessor, obviously shortens the deployment time of the whole algorithm from a PC to a TDA2X vehicle scale control platform, and greatly reduces the transplantation risk.
Description
Technical Field
The invention relates to the technical field of automatic parking, in particular to a method for deploying and scheduling a parking space recognition algorithm based on a TDA2X vehicle scale control platform.
Background
When a linear parking space recognition algorithm is deployed on the basis of a TDA2X vehicle scale control platform, multi-core hardware characteristics are not fully considered, a CPU is not specially customized according to operational characteristics on a used hardware platform, and a special use frame is not designed for deploying the linear parking space recognition algorithm on the use platform, so that a series of problems of long deployment time, low operation efficiency, poor real-time performance and the like are caused.
Disclosure of Invention
In view of the above, the invention aims to provide a method for deploying and scheduling a parking space recognition algorithm based on a TDA2X vehicle scale control platform, which solves the problems of long deployment time, low operation efficiency and poor real-time performance.
The technical scheme adopted by the invention is as follows:
a parking space identification algorithm deployment and scheduling method based on a TDA2X vehicle scale control platform comprises the following steps:
inputting original pictures acquired by a plurality of vehicle-mounted cameras into a look-around splicing module running on a first CPU;
the all-round splicing module splices the original pictures into a top-view splicing picture and inputs the top-view splicing picture to a parking space detection module running on a second CPU through a conversion module;
the parking space detection module obtains initial parking space information based on the overlook splicing diagram and inputs the initial parking space information to a coordinate conversion module running on a first CPU through a conversion module;
the coordinate conversion module converts the coordinate of the initial parking space information into a world coordinate, and the parking space information based on the world coordinate system is input to a third CPU through the conversion module, and the third CPU is used for outputting and displaying a parking space image.
In at least one possible implementation manner, the method further includes, for a deployment manner of the second CPU:
automatically generating a C language code of a parking space based on a pre-constructed parking space model;
and replacing the original image processing function in the C language code with the image processing function supported by the second CPU.
In at least one possible implementation manner, the automatically generating the C language code of the parking space includes: and automatically generating a parking space C code through a Matlab Coder tool based on a pre-constructed parking space model.
In at least one possible implementation, the second CPU has a DSP core; the replacing of the original image processing function in the C language code with the image processing function supported by the second CPU includes: and replacing the original image processing function with an image processing function supported by the DSP core.
In at least one possible implementation, the method further includes invoking a policy for code of the second CPU:
an input queue and an output queue are created in advance;
copying a top view splicing diagram output by a first CPU into an input queue as an input;
acquiring a pixel format of the overlook splicing map from the input queue, and judging whether the pixel format type meets a preset format or not;
if yes, continuing to judge whether picture data or video data already exist in the cache of the input queue;
if the parking space detection intermediate result exists, applying for caching the parking space detection intermediate result;
acquiring a top view mosaic image, storing the top view mosaic image in the cache, and calling the picture in the cache as the input of a parking space detection module;
after being processed by the parking space detection module, the initial parking space information is output to an output queue.
In at least one possible implementation manner, the preset format includes YUV422 or YUV 420.
In at least one possible implementation, the vehicle-mounted camera is a fisheye camera.
In at least one possible implementation, the number of the fisheye cameras is at least 4.
The invention has the main design concept that based on the requirements of a vehicle scale control platform and according to the respective working characteristics of multiple processors, a parking space recognition algorithm is deployed on different processors, and a chain frame is constructed to realize multi-core cooperative scheduling, thereby effectively improving the operation efficiency and the real-time property. The invention has clear structural logic, better universality and multiplexing support, fully exerts the strength of a multiprocessor, obviously shortens the deployment time of the whole algorithm from a PC to a TDA2X vehicle scale control platform, and greatly reduces the transplantation risk.
Drawings
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described with reference to the accompanying drawings, in which:
fig. 1 is a flowchart of a parking space identification algorithm deployment scheduling method based on a TDA2X vehicle scale level control platform according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative only and should not be construed as limiting the invention.
The invention provides an embodiment of a parking space identification algorithm deployment scheduling method based on a TDA2X vehicle scale level control platform, which is specifically shown in FIG. 1 and comprises the following steps:
step S1, inputting the original pictures collected by the plurality of fisheye cameras into a panoramic stitching module running on a first CPU;
s2, the all-round splicing module splices the original pictures into a top-view splicing picture and inputs the top-view splicing picture into a parking space detection module running on a second CPU through a conversion module;
step S3, the parking space detection module obtains initial parking space information based on the overlook splicing chart and inputs the information to a coordinate conversion module running on a first CPU through a conversion module;
and step S4, the coordinate conversion module converts the coordinates of the initial parking space information into world coordinates, and the parking space information based on the world coordinate system is input to a third CPU through the conversion module, and the third CPU is used for outputting and displaying the parking space image.
Further, the method further comprises, for a deployment of the second CPU:
automatically generating a C language code of a parking space based on a pre-constructed parking space model;
and replacing the original image processing function in the C language code with the image processing function supported by the DSP core.
Further, the method also includes invoking a policy for code of the second CPU:
an input queue and an output queue are created in advance;
copying a top view splicing diagram output by a first CPU into an input queue as an input;
acquiring a pixel format of the overlook splicing map from the input queue, and judging whether the pixel format type meets YUV422 or YUV 420;
if yes, continuing to judge whether picture data or video data already exist in the cache of the input queue;
if the parking space detection intermediate result exists, applying for caching the parking space detection intermediate result;
acquiring a top view mosaic image, storing the top view mosaic image in the cache, and calling the picture in the cache as the input of a parking space detection module;
after being processed by the parking space detection module, the initial parking space information is output to an output queue.
Based on the foregoing embodiments, specific description is made with reference to the following:
inputting pictures acquired by the four fisheye cameras into a ring-view splicing module, splicing the four original pictures into a top-view splicing picture by the ring-view splicing module, wherein the ring-view splicing module runs on a first CPU (such as an A15 core), and the first CPU can be used as a main processing unit;
the all-round-looking splicing module inputs the spliced overlook splicing map into the parking space detection module through the conversion module, the parking space detection module operates in a second CPU (such as a DSP core), and the strong item of the second CPU can operate a complex image processing algorithm;
the parking space detection module inputs the detected initial parking space information to the coordinate conversion module through the conversion module, the coordinate conversion module is used for converting the coordinate of the initial parking space information into a world coordinate, and the coordinate conversion module can also run on a first CPU (such as an A15 core);
the coordinate conversion module inputs the parking space information based on the world coordinate system to a third CPU (such as an M4 core) through the conversion module, and the third CPU is mainly used for outputting and displaying images.
The aforementioned translation module functions to transfer data between different Cores (CPUs).
In particular to the deployment of the parking space detection task, reference may be made to the following:
based on a pre-constructed parking space model, a parking space C code can be automatically generated through tools such as Matlab Coder and the like, and then a function related to image processing is replaced by an image processing function supported by a DSP core, so that the performance and the operation efficiency are improved. And further designs the following parking space detection code using frame:
firstly, an input queue and an output queue structure are created, a top view spliced picture running on a first CPU (A15) is taken as input and copied into the input queue, an image format is obtained from the input queue, whether the pixel format type meets YUV422 or YUV420 is judged, if any image format is met, whether picture or video data already exists in a buffer memory of the input queue is judged, and if yes, a buffer memory for storing an intermediate result of a parking space detection algorithm is applied; obtaining the overlook spliced picture to the cache, and calling the picture in the cache as the input of the parking space detection module; and finally, outputting initial parking space information to an output queue after the parking space information is processed by a parking space detection algorithm.
In conclusion, the main design concept of the invention is that based on the requirements of the vehicle scale control platform and according to the respective working characteristics of the multiple processors, the parking space recognition algorithm is deployed on the different processors, and the chain framework is constructed to realize multi-core cooperative scheduling, so that the operation efficiency and the real-time performance are effectively improved. The invention has clear structural logic, better universality and multiplexing support, fully exerts the strength of a multiprocessor, obviously shortens the deployment time of the whole algorithm from a PC to a TDA2X vehicle scale control platform, and greatly reduces the transplantation risk.
In the embodiments of the present invention, "at least one" means one or more, "a plurality" means two or more. "and/or" describes the association relationship of the associated objects, and means that there may be three relationships, for example, a and/or B, and may mean that a exists alone, a and B exist simultaneously, and B exists alone. Wherein A and B can be singular or plural. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c may represent: a, b, c, a and b, a and c, b and c or a and b and c, wherein a, b and c can be single or multiple.
The structure, features and effects of the present invention have been described in detail with reference to the embodiments shown in the drawings, but the above embodiments are merely preferred embodiments of the present invention, and it should be understood that technical features related to the above embodiments and preferred modes thereof can be reasonably combined and configured into various equivalent schemes by those skilled in the art without departing from and changing the design idea and technical effects of the present invention; therefore, the invention is not limited to the embodiments shown in the drawings, and all the modifications and equivalent embodiments that can be made according to the idea of the invention are within the scope of the invention as long as they are not beyond the spirit of the description and the drawings.
Claims (8)
1. A parking space recognition algorithm deployment and scheduling method based on a TDA2X vehicle scale control platform is characterized by comprising the following steps:
inputting original pictures acquired by a plurality of vehicle-mounted cameras into a look-around splicing module running on a first CPU;
the all-round splicing module splices the original pictures into a top-view splicing picture and inputs the top-view splicing picture to a parking space detection module running on a second CPU through a conversion module;
the parking space detection module obtains initial parking space information based on the overlook splicing diagram and inputs the initial parking space information to a coordinate conversion module running on a first CPU through a conversion module;
the coordinate conversion module converts the coordinate of the initial parking space information into a world coordinate, and the parking space information based on the world coordinate system is input to a third CPU through the conversion module, and the third CPU is used for outputting and displaying a parking space image.
2. The TDA2X vehicle scale level control platform-based parking space identification algorithm deployment scheduling method of claim 1, further comprising the following steps for the deployment mode of the second CPU:
automatically generating a C language code of a parking space based on a pre-constructed parking space model;
and replacing the original image processing function in the C language code with the image processing function supported by the second CPU.
3. The TDA2X vehicle scale level control platform-based parking space identification algorithm deployment scheduling method of claim 2, wherein the automatically generating C language codes of parking spaces comprises: and automatically generating a parking space C code through a Matlab Coder tool based on a pre-constructed parking space model.
4. The TDA2X vehicle scale level control platform-based parking space identification algorithm deployment scheduling method of claim 2, wherein the second CPU has a DSP core; the replacing of the original image processing function in the C language code with the image processing function supported by the second CPU includes: and replacing the original image processing function with an image processing function supported by the DSP core.
5. The TDA2X vehicle scale level control platform-based parking space identification algorithm deployment scheduling method of claim 1, further comprising calling a policy for a code of a second CPU:
an input queue and an output queue are created in advance;
copying a top view splicing diagram output by a first CPU into an input queue as an input;
acquiring a pixel format of the overlook splicing map from the input queue, and judging whether the pixel format type meets a preset format or not;
if yes, continuing to judge whether picture data or video data already exist in the cache of the input queue;
if the parking space detection intermediate result exists, applying for caching the parking space detection intermediate result;
acquiring a top view mosaic image, storing the top view mosaic image in the cache, and calling the picture in the cache as the input of a parking space detection module;
after being processed by the parking space detection module, the initial parking space information is output to an output queue.
6. The TDA2X vehicle scale control platform-based parking space recognition algorithm deployment scheduling method of claim 5, wherein the preset format comprises YUV422 or YUV 420.
7. The TDA2X vehicle scale control platform-based parking space recognition algorithm deployment scheduling method according to any one of claims 1 to 6, wherein the vehicle-mounted camera is a fisheye camera.
8. The TDA2X vehicle scale level control platform-based parking space identification algorithm deployment scheduling method of claim 7, wherein the number of the fisheye cameras is at least 4.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111140154.7A CN113886042A (en) | 2021-09-28 | 2021-09-28 | TDA2X vehicle gauge control platform-based parking space recognition algorithm deployment and scheduling method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111140154.7A CN113886042A (en) | 2021-09-28 | 2021-09-28 | TDA2X vehicle gauge control platform-based parking space recognition algorithm deployment and scheduling method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113886042A true CN113886042A (en) | 2022-01-04 |
Family
ID=79007219
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111140154.7A Pending CN113886042A (en) | 2021-09-28 | 2021-09-28 | TDA2X vehicle gauge control platform-based parking space recognition algorithm deployment and scheduling method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113886042A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114038235A (en) * | 2021-11-29 | 2022-02-11 | 安徽江淮汽车集团股份有限公司 | Intelligent parking space detection method based on vehicle gauge level controller |
-
2021
- 2021-09-28 CN CN202111140154.7A patent/CN113886042A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114038235A (en) * | 2021-11-29 | 2022-02-11 | 安徽江淮汽车集团股份有限公司 | Intelligent parking space detection method based on vehicle gauge level controller |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11481862B2 (en) | System and method for real-time, simultaneous object detection and semantic segmentation | |
CN102693528B (en) | Noise suppressed in low light images | |
WO2021248423A1 (en) | Artificial intelligence resource scheduling method and apparatus, storage medium, and chip | |
CN113096201B (en) | Embedded video image deep learning method, equipment and storage medium | |
CN112184587B (en) | Edge data enhancement model, and efficient edge data enhancement method and system based on model | |
CN113939845A (en) | Method, system and computer readable medium for improving image color quality | |
WO2019128735A1 (en) | Imaging processing method and device | |
CN113886042A (en) | TDA2X vehicle gauge control platform-based parking space recognition algorithm deployment and scheduling method | |
CN111709879B (en) | Image processing method, image processing device and terminal equipment | |
CN111797715A (en) | Parking space detection method and device, electronic equipment and storage medium | |
JP2022002376A (en) | Image processing apparatus, image processing method, and program | |
WO2011019330A1 (en) | System and method for region-of-interest-based artifact reduction in image sequences | |
CN111062894A (en) | Artificial intelligence processor and artificial intelligence analysis device | |
CN1955933A (en) | Data processing apparatus and method | |
EP1152331A3 (en) | Parallel task processing system and method | |
WO2020039897A1 (en) | Station monitoring system and station monitoring method | |
CN111428732A (en) | YUV image recognition method, system and computer equipment | |
CN210694184U (en) | Artificial intelligence analysis device and artificial intelligence processor | |
CN115115538A (en) | Video processing method, device, equipment and storage medium | |
US8254711B2 (en) | System and method for processing digital noise in a digital image | |
CN112313944A (en) | Image processing method, device, equipment and storage medium | |
EP3394834B1 (en) | Key lights direction detection | |
CN118114776A (en) | Forward reasoning processing method, chip and medium | |
CN113095231B (en) | Video identification method, system, device and storage medium based on classified object | |
CN113034520B (en) | Incremental ortho-image data slicing service method and device |
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 |