CN114038235A - Intelligent parking space detection method based on vehicle gauge level controller - Google Patents
Intelligent parking space detection method based on vehicle gauge level controller Download PDFInfo
- Publication number
- CN114038235A CN114038235A CN202111436299.1A CN202111436299A CN114038235A CN 114038235 A CN114038235 A CN 114038235A CN 202111436299 A CN202111436299 A CN 202111436299A CN 114038235 A CN114038235 A CN 114038235A
- Authority
- CN
- China
- Prior art keywords
- parking space
- vehicle
- processing unit
- controller
- level controller
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 75
- 238000004148 unit process Methods 0.000 claims abstract description 5
- 238000006243 chemical reaction Methods 0.000 claims description 13
- 238000000926 separation method Methods 0.000 claims description 7
- 238000010586 diagram Methods 0.000 claims description 6
- 238000013461 design Methods 0.000 claims description 5
- 238000000034 method Methods 0.000 claims description 5
- 238000011161 development Methods 0.000 abstract description 5
- 238000002054 transplantation Methods 0.000 abstract description 4
- 230000000694 effects Effects 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/14—Traffic control systems for road vehicles indicating individual free spaces in parking areas
- G08G1/141—Traffic control systems for road vehicles indicating individual free spaces in parking areas with means giving the indication of available parking spaces
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/168—Driving aids for parking, e.g. acoustic or visual feedback on parking space
Abstract
The invention discloses an intelligent parking space detection method based on a vehicle standard level controller, which fully ensures the safety and reliability of a parking system based on the vehicle standard level controller, and particularly, a scheduling relation among processing units is configured on the vehicle standard level controller in advance corresponding to a space identification task, a first processing unit of the vehicle standard level controller acquires a overlooking splicing picture of a space to be parked, the overlooking splicing picture is forwarded to a second processing unit of the vehicle standard level controller, the second processing unit processes the overlooking splicing picture based on a space identification algorithm, a space identification result is output to a third processing unit of the vehicle standard level controller, and finally, the third processing unit outputs a display target space to a user according to the space identification result. The invention gives full play to the strength of each processing unit of the vehicle gauge-level controller, and can accelerate the deployment time of the whole algorithm from PC to TDA2X while accelerating the processing speed, thereby greatly improving the development speed and reducing the risk of transplantation.
Description
Technical Field
The invention relates to the technical field of intelligent parking, in particular to an intelligent parking stall detection method based on a vehicle gauge controller.
Background
With the wide application of intelligent parking, the requirement of a user on intelligent parking is not limited to a space parking space, but the user needs to safely park a vehicle into the parking space with higher accuracy, so that the control system needs to accurately judge whether an available and suitable empty parking space exists. The existing intelligent parking mostly adopts a scheme of matching a laser radar with a visual algorithm, and through analysis and research, the existing parking space identification algorithm is too complex, so that the transportability is poor, the development cycle is influenced, and the safety and the reliability can not be reliably ensured.
Disclosure of Invention
In view of the above, the present invention aims to provide an intelligent parking space detection method based on a vehicle scale controller, so as to solve the problems that the existing automatic parking space recognition algorithm for parking space has poor safety and reliability, high complexity, slow development speed and difficulty in transplantation.
The technical scheme adopted by the invention is as follows:
an intelligent parking space detection method based on a vehicle gauge level controller comprises the following steps:
corresponding to the parking space identification task, configuring a scheduling relation among all processing units on a vehicle gauge level controller in advance;
acquiring a top view splicing diagram of the parking space to be parked through a first processing unit preset by a vehicle gauge level controller;
forwarding the overlooking splicing diagram acquired by the first processing unit to a second processing unit preset by a vehicle gauge controller by using a preset first conversion module;
the second processing unit processes the overlook spliced graph based on a preset parking space recognition algorithm and outputs a parking space recognition result to a third processing unit preset by the vehicle gauge controller;
and the third processing unit outputs and displays the target parking space to the user according to the parking space recognition result.
In at least one possible implementation manner, the processing the overhead mosaic based on a preset parking space recognition algorithm includes:
predicting to obtain parking space information, wherein the parking space information comprises: the method comprises the following steps of (1) enabling an entrance line vertex close to a vehicle head, an entrance line vertex far away from the vehicle head, an entrance line inclination angle, a separation line inclination angle and parking place attribute information to be obtained;
judging whether the parking space is a real parking space or not according to the parking space information;
if not, outputting a parking space identification result representing the wrong parking space identification;
if yes, continuously judging whether the parking space information meets the set error requirement or not according to the parking space information;
if not, outputting a parking space identification result representing a parking space identification error;
and if so, outputting a parking space identification result representing correct parking space identification.
In at least one possible implementation manner, the processing the overhead mosaic image based on a preset parking space recognition algorithm specifically includes:
judging whether the current parking space to be parked is a real parking space or not according to the parking space attribute information;
if so, judging whether the vertex of the inlet line close to the vehicle head, the vertex of the inlet line far away from the vehicle head, the inclination angle of the inlet line and the inclination angle of the separation line are all within a preset error range;
if so, outputting a parking space identification result representing correct parking space identification.
In at least one possible implementation manner, the outputting, to a user, a display target parking space according to the parking space recognition result includes:
and after receiving a parking space identification result representing that the parking space identification is correct, the third processing unit draws and outputs an image of the target parking space.
In at least one possible implementation manner, the overhead mosaic is synthesized by using a preset all-round mosaic algorithm from parking space pictures acquired by at least four fisheye cameras on the whole vehicle.
In at least one possible implementation manner, the obtaining, by the first processing unit, a top mosaic of parking spaces includes: the intelligent parking system captures a spliced top view displayed on a vehicle-mounted screen in the parking process through data write-back and stores the spliced top view in a cache, and the first processing unit reads the spliced top view from the cache.
In at least one possible implementation manner, the configuring, in advance, a scheduling relationship between the processing units on the vehicle-specification-level controller includes: and a TDA controller is adopted as the vehicle gauge level controller, and the scheduling relation among all processors in the TDA controller is set in advance according to the link & chain architecture design of the TDA controller.
In at least one possible implementation manner, the forwarding the top view mosaic acquired by the first processing unit to a second processing unit preset by a vehicle gauge level controller includes: and forwarding the top mosaic acquired by the M4 core to a plurality of parallel EVE cores by using an M4 core and EVE core conversion module controlled by TDA.
In at least one possible implementation manner, the outputting the parking space recognition result to the third processing unit preset by the vehicle gauge level controller includes: and transmitting the parking space identification result output by the EVE core to a third processing unit after data conversion through a preset second conversion module.
In at least one possible implementation thereof, the third processing unit includes an a15 processor.
The invention has the main design concept that based on a vehicle gauge controller, the safety and the reliability of a parking system are fully ensured, specifically, corresponding to a parking space identification task, scheduling relations among processing units are configured on the vehicle gauge controller in advance, a first processing unit of the vehicle gauge controller acquires a top view splicing map of a parking space to be parked, the top view splicing map is forwarded to a second processing unit of the vehicle gauge controller, the second processing unit processes the top view splicing map based on a preset parking space identification algorithm, a parking space identification result is output to a third processing unit of the vehicle gauge controller, and finally, the third processing unit outputs a display target parking space to a user according to the parking space identification result. The invention gives full play to the strength of each processing unit of the vehicle gauge-level controller, and can accelerate the deployment time of the whole algorithm from PC to TDA2X while accelerating the processing speed, thereby greatly improving the development speed and reducing the risk of transplantation.
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 an intelligent parking space detection method based on a vehicle-specification-level controller 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 an intelligent parking space detection method based on a vehicle gauge level controller, and specifically, as shown in fig. 1, the method comprises the following steps:
step S1, corresponding to the parking space identification task, configuring the scheduling relation among the processing units on the vehicle gauge level controller in advance;
in practical operation, a TDA controller may be adopted, and the scheduling relationship between the processors of the TDA may be designed according to the link & chain architecture.
S2, acquiring a top view splicing diagram of the parking space to be parked through a first processing unit preset by the vehicle gauge controller;
the overlook splicing diagram is obtained by synthesizing parking space pictures acquired by at least four fisheye cameras on the whole vehicle through a preset look-around splicing algorithm.
Specifically, the MP5 parking screen information can be captured by the M4 core of the TDA controller through data write-back and stored in the buffer, and then the overhead view information can be read out from the buffer according to the starting position and the length and width of the overhead view image of the MP5 screen during parking.
Step S3, forwarding the top view splicing map acquired by the first processing unit to a preset second processing unit by using a preset first conversion module;
for example, the TDA-controlled M4 core and EVE core conversion module may be used to transfer the overhead parking space image information acquired by the M4 core to the EVE core, and the TDA controller usually has 4 EVE cores capable of processing in parallel, so in some embodiments of the present invention, the parking space identification processing related to deep learning is performed by the EVE core.
Step S4, after receiving the top-view mosaic, the second processing unit performs processing based on a preset parking space recognition algorithm (preferably, a lightweight algorithm), and outputs a parking space recognition result to a preset third processing unit;
as described above, in actual operation, the output result of the second processing unit may be transmitted to the third processing unit after being subjected to data conversion by the preset second conversion module, and for example, the processing result of the EVE core may be input into the a15 core by using the conversion modules of the EVE core and the a15 core.
In practical operation, the second processing unit may predict the parking space information through a feature extraction network (such as RepVGG), where the related parking space information includes, but is not limited to: the parking space attribute information comprises the length of an entrance line, an entrance line vertex P1 (the entrance line vertex closer to the vehicle head), P2 (the entrance line vertex farther from the vehicle head), an entrance line center point P3 (the center point of the connection line of P1 and P2), an entrance line inclination angle (the included angle between the clockwise and anticlockwise rotation of the x axis of a Cartesian coordinate system and a preset direction, wherein the preset direction is the preset direction from P1 towards the longitudinal direction of the parking space), a separation line inclination angle (the included angle between the clockwise rotation of the x axis of the Cartesian coordinate system and the vector from P3 to P2), and parking space attribute information (such as horizontal parking spaces, vertical parking spaces, skew parking spaces and the like).
Further, whether the vehicle is a real vehicle can be judged according to the predicted vehicle attribute information, if not, the identification result can represent vehicle identification errors, if so, whether the P1, the P2, the entrance line inclination angle and the separation line inclination angle are within a preset error range or not is continuously judged (for example, the errors of the P1 and the P2 are within 10 pixels, and the errors of the entrance line inclination angle and the separation line inclination angle are within 10 degrees), and if the error requirements are met, the vehicle identification result representing the correct vehicle identification is output.
And step S5, the third processing unit outputs and displays the target parking space to the user according to the parking space identification result.
Specifically, the a15 processor may draw and output an image of the target parking space according to the processing result sent by the EVE core, for example, when receiving a parking space recognition result that the EVE core outputs a correct parking space recognition representation; if the parking space recognition result which represents the parking space recognition error is output by the EVE core, the recognition error prompt information can be triggered, and the like, and the invention is not limited.
In summary, the main design concept of the present invention is that, based on a vehicle controller, the safety and reliability of a parking system are fully ensured, and specifically, corresponding to a parking space identification task, a scheduling relationship between processing units is configured on the vehicle controller in advance, a first processing unit of the vehicle controller obtains a top view mosaic image of a parking space to be parked, and forwards the top view mosaic image to a second processing unit of the vehicle controller, the second processing unit processes the top view mosaic image based on a preset parking space identification algorithm, and outputs a parking space identification result to a third processing unit of the vehicle controller, and finally the third processing unit outputs a display target parking space to a user according to the parking space identification result. The invention gives full play to the strength of each processing unit of the vehicle gauge-level controller, and can accelerate the deployment time of the whole algorithm from PC to TDA2X while accelerating the processing speed, thereby greatly improving the development speed and reducing the risk of transplantation.
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 (10)
1. An intelligent parking space detection method based on a vehicle gauge-level controller is characterized by comprising the following steps:
corresponding to the parking space identification task, configuring a scheduling relation among all processing units on a vehicle gauge level controller in advance;
acquiring a top view splicing diagram of the parking space to be parked through a first processing unit preset by a vehicle gauge level controller;
forwarding the overlooking splicing diagram acquired by the first processing unit to a second processing unit preset by a vehicle gauge controller by using a preset first conversion module;
the second processing unit processes the overlook spliced graph based on a preset parking space recognition algorithm and outputs a parking space recognition result to a third processing unit preset by the vehicle gauge controller;
and the third processing unit outputs and displays the target parking space to the user according to the parking space recognition result.
2. The vehicle scale controller-based intelligent parking space detection method according to claim 1, wherein the processing of the overhead mosaic based on a preset space recognition algorithm comprises:
predicting to obtain parking space information, wherein the parking space information comprises: the method comprises the following steps of (1) enabling an entrance line vertex close to a vehicle head, an entrance line vertex far away from the vehicle head, an entrance line inclination angle, a separation line inclination angle and parking place attribute information to be obtained;
judging whether the parking space is a real parking space or not according to the parking space information;
if not, outputting a parking space identification result representing the wrong parking space identification;
if yes, continuously judging whether the parking space information meets the set error requirement or not according to the parking space information;
if not, outputting a parking space identification result representing a parking space identification error;
and if so, outputting a parking space identification result representing correct parking space identification.
3. The vehicle-scale-controller-based intelligent parking space detection method according to claim 2, wherein the processing of the overhead mosaic based on the preset space recognition algorithm specifically comprises:
judging whether the current parking space to be parked is a real parking space or not according to the parking space attribute information;
if so, judging whether the vertex of the inlet line close to the vehicle head, the vertex of the inlet line far away from the vehicle head, the inclination angle of the inlet line and the inclination angle of the separation line are all within a preset error range;
if so, outputting a parking space identification result representing correct parking space identification.
4. The vehicle specification controller-based intelligent parking space detection method according to claim 1, wherein the outputting of the display target space to the user according to the space recognition result comprises:
and after receiving a parking space identification result representing that the parking space identification is correct, the third processing unit draws and outputs an image of the target parking space.
5. The vehicle scale controller-based intelligent parking space detection method according to claim 1, wherein the overhead view mosaic is synthesized by preset look-around mosaic algorithms from space pictures acquired by at least four fisheye cameras on the whole vehicle.
6. The vehicle-scale-controller-based intelligent parking space detection method according to claim 1, wherein the step of acquiring the top-view mosaic of the parking spaces by the first processing unit comprises the following steps: the intelligent parking system captures a spliced top view displayed on a vehicle-mounted screen in the parking process through data write-back and stores the spliced top view in a cache, and the first processing unit reads the spliced top view from the cache.
7. The intelligent parking space detection method based on the vehicle gauge level controller as claimed in any one of claims 1 to 6, wherein the configuring of the scheduling relationship among the processing units on the vehicle gauge level controller in advance comprises: and a TDA controller is adopted as the vehicle gauge level controller, and the scheduling relation among all processors in the TDA controller is set in advance according to the link & chain architecture design of the TDA controller.
8. The vehicle-scale-controller-based intelligent parking space detection method according to claim 7, wherein the step of forwarding the overhead view mosaic acquired by the first processing unit to a second processing unit preset by a vehicle-scale-controller comprises the steps of: and forwarding the top mosaic acquired by the M4 core to a plurality of parallel EVE cores by using an M4 core and EVE core conversion module controlled by TDA.
9. The intelligent parking space detection method based on the vehicle-specification-level controller according to claim 8, wherein the outputting of the space recognition result to the third processing unit preset by the vehicle-specification-level controller comprises: and transmitting the parking space identification result output by the EVE core to a third processing unit after data conversion through a preset second conversion module.
10. The intelligent parking space detection method based on the vehicle scale controller as claimed in claim 8, wherein the third processing unit comprises an A15 processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111436299.1A CN114038235A (en) | 2021-11-29 | 2021-11-29 | Intelligent parking space detection method based on vehicle gauge level controller |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111436299.1A CN114038235A (en) | 2021-11-29 | 2021-11-29 | Intelligent parking space detection method based on vehicle gauge level controller |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114038235A true CN114038235A (en) | 2022-02-11 |
Family
ID=80145862
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111436299.1A Pending CN114038235A (en) | 2021-11-29 | 2021-11-29 | Intelligent parking space detection method based on vehicle gauge level controller |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114038235A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109398226A (en) * | 2018-10-12 | 2019-03-01 | 百度在线网络技术（北京）有限公司 | The outer reminding method of the vehicle of vehicle parking state and device |
| CN109649384A (en) * | 2019-02-15 | 2019-04-19 | 华域汽车系统股份有限公司 | A kind of parking assistance method |
| CN111415523A (en) * | 2020-03-06 | 2020-07-14 | 北京智行者科技有限公司 | Autonomous passenger-riding-substituting parking method and system |
| CN112052782A (en) * | 2020-08-31 | 2020-12-08 | 安徽江淮汽车集团股份有限公司 | Around-looking-based parking space identification method, device, equipment and storage medium |
| CN113537105A (en) * | 2021-07-23 | 2021-10-22 | 北京经纬恒润科技股份有限公司 | Parking space detection method and device |
| CN113886042A (en) * | 2021-09-28 | 2022-01-04 | 安徽江淮汽车集团股份有限公司 | TDA2X vehicle gauge control platform-based parking space recognition algorithm deployment and scheduling method |
-
2021
- 2021-11-29 CN CN202111436299.1A patent/CN114038235A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109398226A (en) * | 2018-10-12 | 2019-03-01 | 百度在线网络技术（北京）有限公司 | The outer reminding method of the vehicle of vehicle parking state and device |
| CN109649384A (en) * | 2019-02-15 | 2019-04-19 | 华域汽车系统股份有限公司 | A kind of parking assistance method |
| CN111415523A (en) * | 2020-03-06 | 2020-07-14 | 北京智行者科技有限公司 | Autonomous passenger-riding-substituting parking method and system |
| CN112052782A (en) * | 2020-08-31 | 2020-12-08 | 安徽江淮汽车集团股份有限公司 | Around-looking-based parking space identification method, device, equipment and storage medium |
| CN113537105A (en) * | 2021-07-23 | 2021-10-22 | 北京经纬恒润科技股份有限公司 | Parking space detection method and device |
| CN113886042A (en) * | 2021-09-28 | 2022-01-04 | 安徽江淮汽车集团股份有限公司 | TDA2X vehicle gauge control platform-based parking space recognition algorithm deployment and scheduling method |
Non-Patent Citations (3)
| Title |
|---|
| 于继成;王强;赵目龙;焦育成;: "智能驾驶-AI芯片的算力研究", 汽车文摘 * |
| 单祥茹;: "TI最新ADAS SoC重新定义驾驶体验", 中国电子商情(基础电子) * |
| 韩霜;: "TI最新TDA2x SoC系列重新定义高级驾驶员辅助系统", 世界电子元器件 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11275962B2 (en) | Method of controlling image acquisition and other related tools | |
| CN110070572B (en) | Method and system for generating range images using sparse depth data | |
| JP5783885B2 (en) | Information presentation apparatus, method and program thereof | |
| KR20190128724A (en) | Target recognition methods, devices, storage media and electronic devices | |
| CN110060230B (en) | Three-dimensional scene analysis method, device, medium and equipment | |
| WO2019006389A1 (en) | Dense optical flow processing in a computer vision system | |
| CN113378834B (en) | Object detection method, device, apparatus, storage medium, and program product | |
| CN109635639B (en) | Method, device, equipment and storage medium for detecting position of traffic sign | |
| US10462448B2 (en) | Image generation system and image generation method | |
| CN110651475B (en) | Hierarchical data organization for compact optical streaming | |
| WO2024060708A1 (en) | Target detection method and apparatus | |
| CN114038235A (en) | Intelligent parking space detection method based on vehicle gauge level controller | |
| WO2020039897A1 (en) | Station monitoring system and station monitoring method | |
| JP2023171742A (en) | Flying object controller, flying object control method and program | |
| CN112101113A (en) | Lightweight unmanned aerial vehicle image small target detection method | |
| JP2023517365A (en) | Parking space detection method, device, device and storage medium | |
| CN108734066A (en) | A kind of images of gestures acquisition device and method | |
| CN115065604B (en) | Equipment topology map generation method, generation device, storage medium and terminal equipment | |
| US20070242876A1 (en) | Image Processing Apparatus, Image Processing Method, and Program | |
| JP7107597B2 (en) | STATION MONITORING DEVICE, STATION MONITORING METHOD AND PROGRAM | |
| CN112351181A (en) | Intelligent camera based on CMOS chip and ZYNQ system | |
| JP2021131377A (en) | Generation device, method for generation, and program | |
| CN114419451B (en) | Method and device for identifying inside and outside of elevator, electronic equipment and storage medium | |
| CN116645468B (en) | Human body three-dimensional modeling method, method and device for training human body structure to generate model | |
| US20230109494A1 (en) | Methods and devices for building a training dataset |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220211 |
|
| RJ01 | Rejection of invention patent application after publication |