CN107274674B - Parking area management method, device and system based on video identification - Google Patents
Parking area management method, device and system based on video identification Download PDFInfo
- Publication number
- CN107274674B CN107274674B CN201710507665.5A CN201710507665A CN107274674B CN 107274674 B CN107274674 B CN 107274674B CN 201710507665 A CN201710507665 A CN 201710507665A CN 107274674 B CN107274674 B CN 107274674B
- Authority
- CN
- China
- Prior art keywords
- parking area
- edge feature
- determining
- feature map
- frame
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
- G06V10/507—Summing image-intensity values; Histogram projection analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
Abstract
The invention relates to a parking area management method based on video identification, which comprises the following steps: acquiring a plurality of frames of street view images including a parking area in a video; determining whether a parking area image in the street view image is stable or not according to the multiple frames of street view images; and if so, identifying the vehicle information in the parking area. In addition, the invention also relates to a parking area management device and system based on video identification. By adopting the method, the interference of illumination, pedestrians or vehicles coming and going to the target parking space recognition can be effectively eliminated, and the recognition accuracy is improved.
Description
Technical Field
The invention relates to the field of Intelligent Transportation (ITS) and particularly relates to a method, a device and a System for managing a parking area based on video identification.
Background
With the continuous development of cities, the quantity of private cars is continuously increased, the number of parking lots is limited, the phenomenon of random parking and random parking in roads has increasingly prominent negative effects on the cities, and the phenomenon is also a big factor for causing urban traffic congestion. In order to alleviate traffic jam, effectively utilize in-road resources and standardize in-road parking, relevant government departments have introduced relevant policies for in-road parking, and a certain number of parking spaces are drawn in a road as legal parking areas, and a vehicle owner needs to pay a certain parking fee while parking in the areas.
In order to avoid the problems of missed charging and wrong charging, the vehicles parked in the parking spaces need to be accurately detected, identified, timed (whether the same vehicle is parked within a certain time period or not) and forensics. To solve these problems, the use of video-based vehicle detection technology is an effective solution. However, pedestrians or vehicles often pass through the parking space within the parking space visual angle range (the region of interest) to block the vehicles in the parking space, so that the judgment of the vehicles in the parking space is interfered, including the judgment of whether the vehicles stop and whether the same vehicle is detected between the front frame and the rear frame of the video. In order to solve the problem of interference caused by the passing of pedestrians or vehicles in the parking space visual angle range, whether pedestrians or vehicles pass through the parking space area needs to be detected.
Disclosure of Invention
In order to solve the problem of interference caused by illumination, pedestrians or vehicles in the identification process, the invention provides a parking area management method based on video identification, which comprises the following steps:
acquiring a plurality of frames of street view images including a parking area in a video;
determining whether a parking area image in the street view image is stable or not according to the multiple frames of street view images;
and if so, identifying the vehicle information in the parking area.
Further, the step of determining whether the parking area image in the street view image is stable according to the plurality of frames of street view images includes:
judging whether the edge characteristics of the parking area image are stable or not;
and/or the presence of a gas in the gas,
and judging whether the illumination of the parking area image is stable.
Further, the step of determining whether the edge feature of the parking area image is stable specifically includes:
selecting an initial frame from a plurality of frames of street view images;
determining an initial edge feature map of a parking area in the initial frame, and calculating a projection histogram in the vertical direction of the initial edge feature map;
determining a projection histogram in the vertical direction of a background edge feature map according to the projection histogram in the vertical direction of the initial edge feature map;
acquiring a current frame in the street view image, determining a current edge feature map of a parking area in the current frame, and calculating a projection histogram in the vertical direction of the current edge feature map;
and if the change of the projection histogram in the vertical direction of the current edge feature map and the change of the projection histogram in the vertical direction of the background edge feature map are larger than a first preset value, determining that the edge feature of the parking area image corresponding to the current frame is unstable, otherwise, determining that the parking area image is stable.
Further, the step of determining a projection histogram in the vertical direction of the background edge feature map according to the projection histogram in the vertical direction of the initial edge feature map specifically includes:
and correcting the projection histogram in the vertical direction of the initial edge feature map by combining the updating rate according to the projection histogram in the vertical direction of the edge feature map corresponding to each frame of street view image after the initial frame to obtain the projection histogram in the vertical direction of the background edge feature map.
Further, the specific step of judging whether the illumination of the parking area image is stable is as follows:
selecting an initial frame from a plurality of frames of street view images;
determining an illumination background frame according to the initial frame, and calculating the pixel average value of the illumination background frame;
acquiring a current frame in the street view image, and determining a pixel average value of the current frame;
and if the variation of the pixel average value of the current frame and the pixel average value of the illumination background frame is larger than a second preset value, determining that the illumination of the parking area image corresponding to the current frame is unstable, otherwise, determining that the illumination is stable.
In another aspect, the present invention further provides a parking area management apparatus based on video recognition, including:
the acquisition module is used for acquiring a plurality of frames of street view images including a parking area in a video;
the stability judgment module is used for determining whether the parking area image in the street view image is stable or not according to the multiple frames of street view images;
and the identification module is used for identifying the vehicle information in the target parking area if the judgment result of the stability judgment module is stable.
Further, the stability determining module includes:
the first judging unit is used for judging whether the edge characteristics of the parking area image are stable or not;
and/or the presence of a gas in the gas,
and the second judging unit is used for judging whether the illumination of the parking area image is stable or not.
In one aspect, the present invention also provides a video recognition-based parking area management apparatus comprising a processor and a memory for storing program instructions, the program instructions being loaded by the processor and executing steps implementing the method as described above.
In one aspect, the invention also provides a memory having stored thereon program instructions which are loaded by a processor and which perform the steps of implementing the method as described above.
In one aspect, the present invention further provides a video recognition-based parking area management system, including at least one parking area management device as described above, and at least one server connected to the parking area management device for managing recognized vehicles.
By adopting the method, whether the image is stable is judged before vehicle information identification is carried out, and image identification is not carried out when the image is unstable. The vehicle information is identified only when the image is stable, so that the interference of illumination, pedestrians or vehicles coming and going to the target parking space identification can be effectively eliminated, the identification accuracy is improved, and the management dispute caused by the identification error is effectively eliminated.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts. In the drawings:
FIG. 1 is a flow chart of a method for managing parking areas based on video recognition according to the present invention;
FIG. 2 is a flow chart of the present invention for determining whether the edge feature of the parking area image is stable;
fig. 3 is a flowchart of determining whether the illumination of the parking area image is stable according to the present invention.
Detailed Description
In an on-road parking system for city management, in order to avoid the problems of charge omission and charge error, vehicles parked in parking spaces need to be accurately detected and identified, for example, a video identification system is adopted. In the video, when there is pedestrian, the car of passing, or illumination changes, profile characteristic and illumination in the image have certain change, through detecting these changes, can be effectual whether stable in judgement parking stall region.
In embodiment 1, the parking area management method based on video recognition provided by the present invention includes:
acquiring a plurality of frames of street view images including a parking area in a video;
determining whether a parking area image in the street view image is stable or not according to the multiple frames of street view images; specifically, judging whether the parking area image in the street view image is stable or not, including judging whether the edge feature of the parking area image is stable or not; and/or judging whether the illumination of the parking area image is stable.
Specifically, the step of determining whether the edge feature of the parking area image is stable specifically includes:
selecting an initial frame from a plurality of frames of street view images;
determining an initial edge feature map of a parking area in the initial frame, and calculating a projection histogram in the vertical direction of the initial edge feature map; the initial edge feature graph is obtained by performing edge processing on the picture of the initial frame.
Determining a projection histogram in the vertical direction of a background edge feature map according to the projection histogram in the vertical direction of the initial edge feature map; the mode of obtaining the projection histogram in the vertical direction of the initial edge feature map is to accumulate pixel values in the edge feature map by columns, and for the pictures with R rows by C columns, the result after projection by columns is 1 row x C columns. The histogram is then calculated in the same way.
Specifically, according to a projection histogram in the vertical direction of an edge feature map corresponding to each frame of street view image after an initial frame, the projection histogram in the vertical direction of the initial edge feature map is corrected by combining an update rate, so as to obtain a projection histogram in the vertical direction of a background edge feature map.
It will be appreciated that each update after the initial frame generates a new projected histogram N in the vertical direction of the background edge feature map. And when the next frame pair is used for correction, directly correcting the projection histogram N in the vertical direction of the background edge feature map to generate a projection histogram N +1 in the vertical direction of the background edge feature map, and so on. Because the update rate is combined, actually not every frame is processed, and it is possible to perform the processing after several frames according to a preset rule, which is not limited herein.
Acquiring a current frame in the street view image, determining a current edge feature map of a parking area in the current frame, and calculating a projection histogram in the vertical direction of the current edge feature map;
and if the change of the projection histogram in the vertical direction of the current edge feature map and the change of the projection histogram in the vertical direction of the background edge feature map are larger than a first preset value, determining that the edge feature of the parking area image corresponding to the current frame is unstable, otherwise, determining that the parking area image is stable.
It is easy to understand that the specific steps of judging whether the illumination of the parking area image is stable are as follows:
selecting an initial frame from a plurality of frames of street view images;
determining an illumination background frame according to the initial frame, and calculating the pixel average value of the illumination background frame;
acquiring a current frame in the street view image, and determining a pixel average value of the current frame;
and if the variation of the pixel average value of the current frame and the pixel average value of the illumination background frame is larger than a second preset value, determining that the illumination of the parking area image corresponding to the current frame is unstable, otherwise, determining that the illumination is stable. It can be understood that, if the edge features of the image in the parking area are similar to each other, the illumination background frame also needs to be updated, specifically, the average value of the pixel values of the initial frame is calculated, the initial frame is updated to a new background frame according to the weight difference, the updated value is compared with the average value of the pixel values calculated by the current video frame, the initial frame is also the first illumination background frame, and then the illumination background frame is continuously updated according to the method, and it can be understood that the update of the illumination background frame also has an update rate. The first preset value and the second preset value have a large difference according to different use environments, so that adjustment needs to be performed according to an actual installation environment, and no limitation is made herein.
By the method, if the image is judged to be stable, the vehicle information in the target parking area is identified, and further the vehicle is charged or the parking space is recorded according to the identified vehicle information.
By adopting the method, whether the image is stable is judged before vehicle information identification is carried out, and image identification is not carried out when the image is unstable. The vehicle information is identified only when the image is stable, so that the interference of illumination, pedestrians or vehicles coming and going to the target parking space identification can be effectively eliminated, the identification accuracy is improved, and the management dispute caused by the identification error is effectively eliminated.
In yet another embodiment, the method is illustrated.
Firstly, initializing, obtaining an edge feature map for an initial frame street view image I (0), calculating a projection histogram Hc (0) in the vertical direction of the edge feature map, taking the obtained edge feature projection histogram H (0) as an initial value of the projection histogram in the vertical direction of the background edge feature map, and initializing the mean value HM (0) and the variance HV (0) of the difference Hd (0) between the projection histogram Hc (0) in the vertical direction of the current frame edge feature map and the projection histogram H (0) in the vertical direction of the background edge feature map to zero. The image difference is specifically a difference result obtained by subtracting corresponding pixels of one image from corresponding pixels of the other image. The mean and variance after the difference are used as a group of data of all pixels in the difference result, and the statistical values (mean and variance) of the group of data are obtained.
Meanwhile, the pixel average value IMc (0) of the street view image I (0) of the initial frame is calculated, and the pixel average value IM (0) of the illumination background frame is initialized to IMc (0).
Updating background data, and updating a projection histogram in the vertical direction of the background edge feature map. For a newly acquired frame of image, an edge feature projection histogram Hc (n +1) is obtained, which is used to update the projection histogram H (n +1) ═ 1-w1 × H (n) + w1 × Hc (n +1) in the vertical direction of the background edge feature map, where w1 is the update rate.
And (6) difference. Hd (n +1) — Hc (n +1) -H (n +1), and the mean HMc (n +1) and the variance HVc (n +1) after the difference are obtained.
Updating of the difference mean and variance. HM (n +1) ═ HM (1-w1) × (n) + w1 × HMc (n +1), HV (n +1) ═ HV (1-w1) × HV (n) + w1 × HVc (n +1), the value of HV (n +1) is controlled in the range of [ HTLow, HTUp ], if HV (n +1) < HTLow, HV (n +1) ═ HTLow; if HV (n +1) > HTUp, HV (n +1) ═ HTUp;
updating of the pixel average of the light background frame. IM (n +1) ═ (1-w1) × IM (n) + w1 × IMc (n + 1);
and stability judgment, when ITlow < IMc (n))/IM (n)) < ITup, or when the ratio of the number of absolute values abs (Hd (n))) of the projection histogram difference, which is smaller than a threshold HT (n)) (HM (n)) + k (HV (n)), in the histogram sequence of the difference histogram sequence length, is smaller than the threshold HT, the histogram sequence is judged to be stable, otherwise, the histogram sequence length is judged to be unstable. Wherein, ITlow and ITup are threshold values of illumination intensity change, k is a multiple coefficient, and HT is an instability threshold value (or sensitivity).
According to the actual use condition, can use marginal feature stability or illumination stability alone, to daytime, singly use marginal feature stability to judge can judge the stability of parking stall, and to the condition at night, then need add illumination stability and judge.
In still another embodiment, a parking area management apparatus based on video recognition is further disclosed, including:
the acquisition module is used for acquiring a plurality of frames of street view images including a parking area in a video;
the stability judgment module is used for determining whether the parking area image in the street view image is stable or not according to the multiple frames of street view images;
and the identification module is used for identifying the vehicle information in the target parking area if the judgment result of the stability judgment module is stable.
The stability judgment module includes:
the first judging unit is used for judging whether the edge characteristics of the parking area image are stable or not;
and/or the presence of a gas in the gas,
and the second judging unit is used for judging whether the illumination of the parking area image is stable or not.
Also disclosed is a video recognition-based parking area management device comprising a processor and a memory for storing program instructions which are loaded by the processor and execute steps implementing the method of embodiment 1.
Also disclosed is a memory having stored thereon program instructions which are loaded by a processor and which perform the steps of implementing the method according to embodiment 1.
The parking area management system based on video identification comprises at least one parking area management device and at least one server connected with the parking area management device and used for managing the identified vehicles.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (6)
1. A parking area management method based on video recognition is characterized by comprising the following steps:
acquiring a plurality of frames of street view images including a parking area in a video;
determining whether a parking area image in the street view image is stable or not according to the multiple frames of street view images;
if the vehicle information is stable, identifying the vehicle information in the parking area;
the step of determining whether the parking area image in the street view image is stable according to the plurality of frames of street view images comprises the following steps:
judging whether the edge characteristics of the parking area image are stable or not;
the step of judging whether the edge features of the parking area image are stable specifically comprises:
selecting an initial frame from a plurality of frames of street view images;
determining an initial edge feature map of a parking area in the initial frame, and calculating a projection histogram in the vertical direction of the initial edge feature map;
determining a projection histogram in the vertical direction of a background edge feature map according to the projection histogram in the vertical direction of the initial edge feature map;
acquiring a current frame in the street view image, determining a current edge feature map of a parking area in the current frame, and calculating a projection histogram in the vertical direction of the current edge feature map;
if the change of the projection histogram in the vertical direction of the current edge feature map and the change of the projection histogram in the vertical direction of the background edge feature map are larger than a first preset value, determining that the edge feature of the parking area image corresponding to the current frame is unstable, otherwise, determining that the parking area image is stable;
and/or the presence of a gas in the gas,
judging whether the illumination of the parking area image is stable or not;
the specific steps of judging whether the illumination of the parking area image is stable are as follows:
selecting an initial frame from a plurality of frames of street view images;
determining an illumination background frame according to the initial frame, and calculating the pixel average value of the illumination background frame;
acquiring a current frame in the street view image, and determining a pixel average value of the current frame;
and if the variation of the pixel average value of the current frame and the pixel average value of the illumination background frame is larger than a second preset value, determining that the illumination of the parking area image corresponding to the current frame is unstable, otherwise, determining that the illumination is stable.
2. The method according to claim 1, wherein the step of determining the projection histogram in the vertical direction of the background edge feature map according to the projection histogram in the vertical direction of the initial edge feature map specifically comprises:
and correcting the projection histogram in the vertical direction of the initial edge feature map by combining the updating rate according to the projection histogram in the vertical direction of the edge feature map corresponding to each frame of street view image after the initial frame to obtain the projection histogram in the vertical direction of the background edge feature map.
3. A video recognition-based parking area management apparatus, comprising:
the acquisition module is used for acquiring a plurality of frames of street view images including a parking area in a video;
the stability judgment module is used for determining whether the parking area image in the street view image is stable or not according to the multiple frames of street view images;
the identification module is used for identifying the vehicle information in the parking area if the judgment result of the stability judgment module is stable;
the stability judgment module includes:
the first judging unit is used for judging whether the edge characteristics of the parking area image are stable or not;
the step of the first judging unit being configured to judge whether the edge feature of the parking area image is stable specifically is:
selecting an initial frame from a plurality of frames of street view images;
determining an initial edge feature map of a parking area in the initial frame, and calculating a projection histogram in the vertical direction of the initial edge feature map;
determining a projection histogram in the vertical direction of a background edge feature map according to the projection histogram in the vertical direction of the initial edge feature map;
acquiring a current frame in the street view image, determining a current edge feature map of a parking area in the current frame, and calculating a projection histogram in the vertical direction of the current edge feature map;
if the change of the projection histogram in the vertical direction of the current edge feature map and the change of the projection histogram in the vertical direction of the background edge feature map are larger than a first preset value, determining that the edge feature of the parking area image corresponding to the current frame is unstable, otherwise, determining that the parking area image is stable;
and/or the presence of a gas in the gas,
a second judging unit for judging whether the illumination of the parking area image is stable;
the specific steps of the second judging unit for judging whether the illumination of the parking area image is stable are as follows:
selecting an initial frame from a plurality of frames of street view images;
determining an illumination background frame according to the initial frame, and calculating the pixel average value of the illumination background frame;
acquiring a current frame in the street view image, and determining a pixel average value of the current frame;
and if the variation of the pixel average value of the current frame and the pixel average value of the illumination background frame is larger than a second preset value, determining that the illumination of the parking area image corresponding to the current frame is unstable, otherwise, determining that the illumination is stable.
4. A video recognition based parking area management apparatus comprising a processor and a memory, said memory storing program instructions which are loaded by said processor and executed to implement the steps of the method according to any one of claims 1-2.
5. A memory having stored thereon program instructions that are loaded by a processor and execute steps implementing the method according to any of claims 1-2.
6. A video recognition based parking area management system comprising at least one parking area management device according to any one of claims 3 to 4, and at least one server connected to said parking area management device for managing recognized vehicles.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710507665.5A CN107274674B (en) | 2017-06-28 | 2017-06-28 | Parking area management method, device and system based on video identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710507665.5A CN107274674B (en) | 2017-06-28 | 2017-06-28 | Parking area management method, device and system based on video identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107274674A CN107274674A (en) | 2017-10-20 |
| CN107274674B true CN107274674B (en) | 2020-04-24 |
Family
ID=60070374
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710507665.5A Active CN107274674B (en) | 2017-06-28 | 2017-06-28 | Parking area management method, device and system based on video identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107274674B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113014744B (en) * | 2019-12-19 | 2022-07-22 | 合肥君正科技有限公司 | Method for detecting shielding of monitoring picture in vehicle |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9197861B2 (en) * | 2012-11-15 | 2015-11-24 | Avo Usa Holding 2 Corporation | Multi-dimensional virtual beam detection for video analytics |
| CN105869182B (en) * | 2016-06-17 | 2018-10-09 | 北京精英智通科技股份有限公司 | A kind of parking stall condition detection method and system |
| CN106251695B (en) * | 2016-09-29 | 2019-07-09 | 成都通甲优博科技有限责任公司 | Destination parking stall intelligent recommendation system and method based on parking space state monitoring |
| CN106529471B (en) * | 2016-11-08 | 2019-08-23 | 广东安居宝数码科技股份有限公司 | Vehicle face-positioning method and system |
-
2017
- 2017-06-28 CN CN201710507665.5A patent/CN107274674B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN107274674A (en) | 2017-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6180482B2 (en) | Methods, systems, products, and computer programs for multi-queue object detection and analysis (multi-queue object detection and analysis) | |
| US11380104B2 (en) | Method and device for detecting illegal parking, and electronic device | |
| JP6819205B2 (en) | Parking space detectors, electronics and methods | |
| CN112417943A (en) | Advanced Driver Assistance System (ADAS) operation with algorithmic skyline detection | |
| CN109961057B (en) | Vehicle position obtaining method and device | |
| Felisa et al. | Robust monocular lane detection in urban environments | |
| US20130266187A1 (en) | Video-based method for parking angle violation detection | |
| CN107591005B (en) | Parking area management method, server and system combining dynamic and static detection | |
| JP2018010634A (en) | Parking space state detection method, detection apparatus, and electronic device | |
| CN107748882B (en) | Lane line detection method and device | |
| US20150117705A1 (en) | Hybrid Parking Detection | |
| CN112101272A (en) | Traffic light detection method and device, computer storage medium and road side equipment | |
| Blumer et al. | Cost-effective single-camera multi-car parking monitoring and vacancy detection towards real-world parking statistics and real-time reporting | |
| CN112149471B (en) | Loop detection method and device based on semantic point cloud | |
| CN107274674B (en) | Parking area management method, device and system based on video identification | |
| CN109410598B (en) | Traffic intersection congestion detection method based on computer vision | |
| US11137256B2 (en) | Parking area map refinement using occupancy behavior anomaly detector | |
| CN104463170A (en) | Unlicensed vehicle detecting method based on multiple detection under gate system | |
| CN110008957B (en) | District illegal stop warning method based on license plate recognition and image variance algorithm | |
| Sanberg et al. | Color-based free-space segmentation using online disparity-supervised learning | |
| CN108847035B (en) | Traffic flow evaluation method and device | |
| Bong et al. | Car-park occupancy information system | |
| EP3627378A1 (en) | Enhancing the detection of non-structured objects on a driveway | |
| CN115761668A (en) | Camera stain recognition method and device, vehicle and storage medium | |
| JP2020095631A (en) | Image processing device and image processing method |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |