CN112750312A - Information detection method, device, equipment and storage medium - Google Patents
Information detection method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN112750312A CN112750312A CN202011595163.0A CN202011595163A CN112750312A CN 112750312 A CN112750312 A CN 112750312A CN 202011595163 A CN202011595163 A CN 202011595163A CN 112750312 A CN112750312 A CN 112750312A
- Authority
- CN
- China
- Prior art keywords
- target
- image
- license plate
- vehicle
- sub
- 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
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
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/08—Detecting or categorising vehicles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/09—Recognition of logos
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
Abstract
The embodiment of the application provides an information detection method, an information detection device, information detection equipment and a storage medium, wherein the method comprises the following steps: acquiring a plurality of target images, detecting each target image in sequence, and judging whether the target image contains a warning sign or not, wherein the target image is an image of a position area where an accident vehicle is located; if the warning sign is not contained in each target image, acquiring a first sub-image of each target image containing a vehicle; and detecting the vehicles in each first sub-image, and if the number of the target images of the first sub-images containing the same vehicle is greater than or equal to a preset number, determining that the same vehicle is a target vehicle which is an accident vehicle without a warning sign. The embodiment of the application can overcome the problems that the detection method in the prior art is low in efficiency and wastes human resources.
Description
Technical Field
The embodiment of the application relates to the technical field of images, in particular to an information detection method, an information detection device, information detection equipment and a storage medium.
Background
Along with the increase of vehicles, traffic accidents are increased, especially, the vehicles run at a high speed, and if the accident vehicles are not provided with warning signs, other vehicles easily cause accidents again at the accident site. Therefore, the setting of the warning flag is very necessary, and the detection of the warning flag is also important.
At present, the conventional method for judging whether a warning sign is set after a traffic accident occurs at a high speed is to manually check whether a shot picture (such as shot by an unmanned aerial vehicle) contains the warning sign, and then determine whether an accident vehicle violates a rule according to a detection result.
However, the detection method in the prior art is not only inefficient, but also wastes human resources.
Disclosure of Invention
The embodiment of the application provides an information detection method, an information detection device, information detection equipment and a storage medium, and aims to solve the problems that the detection method in the prior art is low in efficiency and wastes human resources.
In a first aspect, an embodiment of the present application provides an information detection method, including:
acquiring a plurality of target images, detecting each target image in sequence, and judging whether the target image contains a warning sign or not, wherein the target image is an image of a position area where an accident vehicle is located;
if the warning sign is not contained in each target image, acquiring a first sub-image of each target image containing a vehicle;
and detecting the vehicles in each first sub-image, and if the number of the target images of the first sub-images containing the same vehicle is greater than or equal to a preset number, determining that the same vehicle is a target vehicle which is an accident vehicle without a warning sign.
In one possible design, the acquiring a first sub-image of each target image containing a vehicle includes:
for each target image, identifying the position of a vehicle in the target image;
and generating a first sub-image from a preset area image of the position of the vehicle in the target image.
In one possible design, the detecting the vehicle in each of the first sub-images includes:
for each first sub-image, identifying whether the first sub-image contains a license plate;
if the license plate is contained, detecting license plate information corresponding to the license plate;
and if the license plate information is detected, determining the number of target images of the first sub-images containing the same vehicle according to the license plate information.
In a possible design, if the license plate is included, detecting license plate information corresponding to the license plate includes:
if the license plate is contained, generating a second sub-image from a preset area image of the position of the license plate in the first sub-image;
and detecting license plate information corresponding to the license plate in the second sub-image.
In one possible design, the determining, according to the license plate information, the number of target images to which a first sub-image containing the same vehicle belongs includes:
determining a target image to which the license plate information belongs according to the license plate information;
determining the number of target images containing the same license plate information according to the target images to which the license plate information belongs;
and if the number of the target images containing the same license plate information is smaller than a preset number, detecting other target images except the target image containing the same license plate information according to the license plate information, and determining the number of the target images containing the same vehicle.
In a possible design, the detecting, according to the license plate information, other target images except a target image containing the same license plate information, and determining the number of target images containing the same vehicle includes:
taking the first sub-image which does not contain the license plate in the other target images and the first sub-image corresponding to the second sub-image which does not detect the license plate information as target sub-images;
respectively comparing the vehicles in the first sub-images corresponding to the target images containing the license plate information with the vehicles in each target sub-image, and determining whether the vehicles in the target sub-images are the vehicles in the first sub-images corresponding to the target images containing the license plate information;
if yes, determining a target image to which the target sub-image belongs, and counting the number of the target images to which the target sub-image belongs according to the target images containing the same license plate information, wherein the number of the target images to which the target sub-image belongs is the number of the target images containing the same vehicle.
In one possible design, the method further includes:
and reporting the license plate information corresponding to the target vehicle to a target terminal, so that the target terminal executes violation processing operation on the target vehicle.
In a second aspect, an embodiment of the present application provides an information detecting apparatus, including:
the system comprises a first acquisition module, a second acquisition module and a warning module, wherein the first acquisition module is used for acquiring a plurality of target images, detecting each target image in sequence and judging whether the target image contains a warning sign or not, and the target image is an image of a position area where an accident vehicle is located;
the second acquisition module is used for acquiring a first sub-image containing a vehicle in each target image when the warning sign is not contained in each target image;
and the target vehicle determining module is used for detecting the vehicles in each first sub-image, and if the number of the target images of the first sub-images containing the same vehicle is greater than or equal to the preset number, determining that the same vehicle is the target vehicle, and the target vehicle is an accident vehicle without a warning sign.
In a third aspect, an embodiment of the present application provides an information detection apparatus, including: at least one processor and memory;
the memory stores computer-executable instructions;
the at least one processor executes computer-executable instructions stored by the memory to cause the at least one processor to perform the information detection method as described above in the first aspect and various possible designs of the first aspect.
In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where computer-executable instructions are stored, and when a processor executes the computer-executable instructions, the information detection method according to the first aspect and various possible designs of the first aspect are implemented.
In the information detection method, apparatus, device, and storage medium provided in this embodiment, a plurality of target images are first acquired, each target image is sequentially detected, whether each target image contains a warning sign is determined, if none of the plurality of target images contains the warning sign, a vehicle with violation is indicated, and detection of a target vehicle, which is a vehicle with violation (an accident vehicle without a warning sign), from the plurality of images is started; in order to narrow the detection range and save resources, first sub-images containing vehicles can be acquired from each image, then each first sub-image is detected, the number of target images containing the same vehicles in all the first sub-images, namely the number of the same vehicles distributed in which target images or the number of the same vehicles appearing in which target images are determined, if the number of the obtained target images is larger than the preset number, the same vehicles are indicated to be accident vehicles without warning signs, image detection and information comparison are performed through a machine, whether warning signs are set or not and illegal vehicle detection is achieved, the detection efficiency is improved, and meanwhile human resources are saved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to these drawings without inventive exercise.
Fig. 1 is an application scenario diagram of an information detection method provided in an embodiment of the present application;
fig. 2 is a schematic flowchart of an information detection method according to an embodiment of the present application;
fig. 3 is a schematic flowchart of an information detection method according to yet another embodiment of the present application;
fig. 4 is a schematic flowchart of an information detection method according to another embodiment of the present application;
fig. 5 is a schematic structural diagram of an information detection apparatus according to an embodiment of the present application;
fig. 6 is a schematic diagram of a hardware structure of an information detection apparatus according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
At present, the conventional method for judging whether a warning sign is set after a traffic accident occurs at a high speed is to manually check whether a shot picture (such as shot by an unmanned aerial vehicle) contains the warning sign, and then determine whether an accident vehicle violates a rule according to a detection result. However, the detection method in the prior art is not only inefficient, but also wastes human resources.
In order to solve the problems, the technical idea of the application is to provide a method for judging whether an accident vehicle is provided with a warning sign or not based on a deep learning algorithm, firstly, whether each image in each group of images contains the warning sign or not is detected, if the warning sign is not detected, the position of the vehicle in each image is detected by using a detection model, then, the image of the position of the vehicle is intercepted, the intercepted vehicle image is detected, the frequency of the accident vehicle appearing in the group of images is determined, and if the frequency exceeds a certain frequency, the accident vehicle is not provided with the warning sign, namely, the accident vehicle is an illegal vehicle, so that the efficient detection is realized through the on-line detection of machine learning, and the human resources are saved.
In practical application, referring to fig. 1, fig. 1 is an application scenario diagram of the information detection method provided in the present application, and an execution subject of the present application may be a terminal device, such as a server 10. The server 10 may acquire a plurality of images (which may be target images) from the acquisition device 20 (which may be a shooting device loaded by an unmanned aerial vehicle) through an interface, sequentially detect each target image, determine whether the plurality of target images contain a set warning mark, and determine whether an illegal vehicle, that is, an accident vehicle without a set warning mark exists; if each target image does not contain the warning sign, the accident vehicle at the accident site is proved to have the illegal behavior, and the accident vehicle with the illegal behavior needs to be further detected; in order to reduce redundant detection operation and determine an accident vehicle without a warning sign, a target image can be intercepted to obtain first sub-images containing vehicles, the first sub-images are further detected to determine whether each first sub-image contains the same vehicle, if so, the number of the target images containing the same vehicle is determined to be equal to the preset number, if so, the same vehicle is the accident vehicle without the warning sign, and the violation treatment is carried out on the accident vehicle without the warning sign. Therefore, the information detection method provided by the application realizes the detection of whether the warning sign is set and the detection of the illegal vehicle through the image detection and the information comparison of machine learning, improves the detection efficiency and saves human resources at the same time.
It should be noted that the server may not need to directly obtain the image to be detected from the acquisition device, and may also obtain the image to be detected (i.e. the target image) by other means, such as obtaining the image from a database, etc., which is not specifically limited herein, and fig. 1 is merely an example.
The following describes the technical solutions of the present application and how to solve the above technical problems with specific embodiments. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.
Fig. 2 is a schematic flow chart of an information detection method according to an embodiment of the present application. As shown in fig. 2, the method may include:
s201, acquiring a plurality of target images, detecting each target image in sequence, and judging whether the target images contain warning signs or not, wherein the target images are images of a position area where an accident vehicle is located.
In this embodiment, a plurality of target images may be collected by an image collecting device, and the plurality of target images are used as a group of images to be detected, where the target image is an image of a location area where an accident vehicle is located, and each target image is an image containing a vehicle. And then, the server utilizes the detection model to detect warning signs based on the collected multiple target images, judges whether warning signs are arranged in the position areas where the accident vehicles are located in the group of images to be detected, and further determines whether the accident vehicles have illegal behaviors.
Specifically, each image in a group of images is detected by using a detection model, the vehicle and the warning mark in the image are detected, and whether the image contains the warning mark or not is judged. And respectively intercepting the small images of the vehicles detected in each image to obtain the small images of the vehicles, and providing the small images to be detected for further detecting the illegal vehicles under the scene that all target images do not contain the warning signs.
S202, if each target image does not contain the warning sign, acquiring a first sub-image of each target image containing a vehicle.
In this embodiment, if no warning sign is set in each target image in the group of images to be detected, it indicates that an illegal vehicle, that is, an accident vehicle without a warning sign, exists in the group of images to be detected, and therefore, the server detects the vehicle in each target image by using the detection model, and intercepts the area image in which the contained vehicle is located, so as to obtain a plurality of first sub-images. The first sub-image is a sub-image of the target image including the vehicle.
Specifically, since no warning sign is set in each image, in order to find out an illegal vehicle, that is, an accident vehicle without a warning sign, the vehicle in each target image needs to be detected, and in order to increase the processing speed and improve the detection efficiency, only the image in the preset area where the vehicle is located in each target image can be detected, so that redundant detection operations are reduced, and resources can be saved.
S203, detecting the vehicles in each first sub-image, and if the number of the target images of the first sub-images containing the same vehicle is greater than or equal to the preset number, determining that the same vehicle is a target vehicle, wherein the target vehicle is an accident vehicle without a warning sign.
In this embodiment, the server detects each first sub-image by using the first detection model for locking the position of the vehicle, determines whether all the first sub-images contain the same vehicle, determines the number of target images containing the same vehicle from the plurality of target images if the first sub-images contain the same vehicle, compares the number of target images containing the same vehicle with a preset number, and determines that the same vehicle is the target vehicle, that is, the accident vehicle without the warning sign, if the number of target images containing the same vehicle is greater than the preset number. The detection of the online illegal vehicle in machine learning is realized, the detection is different from manual detection, the detection efficiency is improved, and human resources are saved.
In the information detection method provided by the embodiment, a plurality of target images are acquired, each target image is sequentially detected, whether each target image contains a warning sign or not is judged, if the warning signs are not contained in the plurality of target images, vehicles in violation are indicated, and detection of vehicles in violation (accident vehicles without warning signs) namely target vehicles is started from the plurality of images; in order to narrow the detection range and save resources, first sub-images containing vehicles can be acquired from each image, then each first sub-image is detected, the number of target images containing the same vehicles in all the first sub-images, namely the number of the same vehicles distributed in which target images or the number of the same vehicles appearing in which target images are determined, if the number of the obtained target images is larger than the preset number, the same vehicles are indicated to be accident vehicles without warning signs, through image detection of machine learning and information comparison, detection of whether warning signs are set or not and detection of illegal vehicles are achieved, the detection efficiency is improved, and meanwhile, human resources are saved.
In a possible design, referring to fig. 3, fig. 3 is a schematic flow chart of an information detection method according to still another embodiment of the present application. The present embodiment describes how to acquire the first sub-image containing the vehicle in each target image based on the above-mentioned embodiments. The acquiring of the first sub-image containing the vehicle in each target image may include:
s301, aiming at each target image, identifying the position of the vehicle in the target image.
S302, generating a first sub-image from a preset area image of the position of the vehicle in the target image.
In this embodiment, the position of the vehicle in each target image is identified through the first detection model, then a preset area image of the position of the vehicle is captured from each target image, and the captured small vehicle image is used as the first sub-image.
Specifically, before intercepting a small vehicle image, a detection model is used to detect each image in a group of images (i.e. a group of images to be detected, generally 3 images), and a warning sign and a vehicle in the images are detected; then, whether the warning signs are detected in the three images is judged, and if one warning sign is detected, the next group of pictures are detected without performing violation punishment on any vehicle in the group of images (for example, S301 to S302).
If no one of the three images detects the warning sign, respectively intercepting each detected vehicle small image in the three images to obtain the vehicle small image. Therefore, the position of the vehicle in each target image can be locked through the first detection model, and further, in order to detect the vehicle in the image emphatically, only the preset area image of the position of the vehicle in the target image can be detected, the preset area image of the position of the vehicle needs to be intercepted, a first sub-image is generated, image data is provided for further detecting the information (such as license plate and license plate number) of the vehicle, resources are saved, and redundant operation is reduced.
In a possible design, referring to fig. 4, fig. 4 is a schematic flow chart of an information detection method according to another embodiment of the present application. The present embodiment describes how to determine the number of target images containing the same vehicle in detail on the basis of the above-described embodiments. The detecting the vehicle in each of the first sub-images may include:
s401, aiming at each first sub-image, identifying whether the first sub-image contains a license plate.
In practical application, a license plate is not shown in an image or license plate information (such as a license plate number) is not clearly seen or is not shown in the displayed license plate due to shooting speed or illumination in the image shot by the unmanned aerial vehicle, so that the license plate and/or the license plate information and the like need to be further detected for the intercepted first sub-image through the detection model.
S402, if the license plate is contained, detecting license plate information corresponding to the license plate.
S403, if the license plate information is detected, determining the number of target images of the first sub-images containing the same vehicle according to the license plate information.
In the embodiment, the server detects the position of the license plate in each first subimage by using a second detection model for locking the position of the license plate, and judges whether the license plate is contained in the first subimage; if one or some first sub-images contain the license plate, the license plate number on the license plate is identified by using an image identification technology, and whether the license plate number of the license plate can be identified is judged. If the license plate or the license plate information is not detected, reserving the corresponding first sub-image for further counting the number of target images containing the same vehicle; if the license plate number is recognized, the number of target images containing the same vehicle is counted according to the recognized license plate number.
Specifically, the license plate detection model (i.e., the second detection model) is used to detect the license plate of the vehicle thumbnail captured in S302, and the detected license plate is captured to obtain a license plate thumbnail, i.e., the second sub-image. And (4) still keeping the small images of the vehicles without the detected license plates, and carrying out the following operation of containing the number of the target images of the same vehicles.
In a possible design, if the license plate is included, how to detect the license plate information corresponding to the license plate can be realized through the following steps:
step a1, if the license plate is contained, generating a second sub-image from a preset area image of the position of the license plate in the first sub-image.
Step a2, detecting license plate information corresponding to the license plate in the second sub-image.
In this embodiment, by using the license plate detection model, if the position of the license plate of the vehicle in the captured small vehicle image is detected, in order to further detect the license plate number of the license plate and reduce redundant detection operations, a preset region image of the position of the license plate can be captured from the first sub-image, and then the small vehicle plate image, that is, the second sub-image, is obtained. The server can perform image recognition on each second subimage by using the license plate recognition model, and detect license plate information corresponding to the license plate in the second subimage, such as information of the license plate number and the like.
Specifically, the license plate recognition model is used for recognizing the license plate small image intercepted in the step a1, so as to obtain a corresponding license plate number, and find a suspected target vehicle (namely a suspected accident vehicle without a warning sign). And meanwhile, reserving the vehicles which do not recognize the corresponding license plate number, and performing the following operation of containing the number of target images of the same vehicle.
In a possible design, the present embodiment provides a detailed description of how to determine the number of target images containing the same vehicle based on the above embodiments, and can be implemented by the following steps:
step b1, determining a target image to which the license plate information belongs according to the license plate information.
In this embodiment, based on the detected license plate information, each second sub-image to which each license plate information belongs is first determined, then the first sub-image to which each second sub-image belongs is found out, and then the target image to which each first sub-image belongs is determined.
Example 1, at least one vehicle thumbnail (i.e., a first target sub-image) may be truncated in each target image. For example, there are 3 target images, a target image 1, a target image 2 and a target image 3, where two first sub-images, namely, a vehicle thumbnail 11 and a vehicle thumbnail 12, are captured from the target image 1; a first sub-image, namely a vehicle small map 21, is intercepted in the target image 2, and a first sub-image, namely a vehicle small map 31, is intercepted in the target image 3; a license plate small image 111 is captured on the vehicle small image 11, a license plate small image 112 is captured on the vehicle small image 12, a license plate small image is not captured in the vehicle small image 21, and a license plate small image 311 is captured in the vehicle small image 31; in addition, the license plate information 1110 is recognized in the license plate thumbnail 111, the license plate information 1111 is recognized in the ticket thumbnail 112, and the license plate information is not recognized in the license plate thumbnail 311.
Specifically, the license plate information 1110 and the license plate information 1111 belong to license plates on different license plates, and therefore the license plate information 1110 and the license plate information 1111 are different vehicles, but since the license plate information 1110 and the license plate information 1111 belong to the same first sub-image, that is, the vehicle thumbnail 11, the license plate information 1110 and the license plate information 1111 belong to the same target image 1, and therefore, when other target vehicles need to be detected, other vehicles except the target image 1 are detected. That is, if the number of target images to which the first sub-image containing the same license plate information belongs is smaller than the preset number, the number of target images containing the same vehicle needs to be determined by further detecting the residual images corresponding to other target images not containing license plate information or license plates according to the license plate information.
Exemplary 2, at least one vehicle thumbnail (i.e., the first target sub-image) may be truncated in each target image. For example, there are 3 target images, a target image 1, a target image 2 and a target image 3, where two first sub-images, namely, a vehicle thumbnail 11 and a vehicle thumbnail 12, are captured from the target image 1; two first sub-images, namely a vehicle small image 21 and a vehicle small image 22, are intercepted in the target image 2, and a first sub-image, namely a vehicle small image 31, is intercepted in the target image 3; a license plate small image 111 is captured on the vehicle small image 11, a license plate small image 112 is captured on the vehicle small image 12, a license plate small image is not captured in the vehicle small image 21, a license plate small image 221 is captured in the vehicle small image 22, and a license plate small image 311 is captured in the vehicle small image 31; in addition, license information 1110 can be recognized in the license plate thumbnail 111, license information 1111 can be recognized in the ticket thumbnail 112, license information 2210 can be recognized in the license plate thumbnail 221, and license information 3110 can be recognized in the license plate thumbnail 311.
Specifically, license plate information 1110, license plate information 1111, license plate information 2210 and license plate information 3110 are respectively compared to determine whether the license plate information is the same, wherein license plate information 1110 and license plate information 2210 are the same, license plate information 1110, license plate information 1111 and license plate information 3110 are not the same, and then license plate information 1110 and license plate information 2210 are the same, and license plate information 1110 belongs to a vehicle small map 11, vehicle small map 11 belongs to a target vehicle 1, license plate information 2210 belongs to a vehicle small map 22, and vehicle small map 22 belongs to a target vehicle, then license plate information 1110 and license plate information 2210 belong to two different target images, so that the number of target images of a first sub-image to which the same license plate information belongs is equal to a preset number, and it is determined that the vehicle corresponding to the license plate information is the target vehicle; in addition, since the license plate information 1110, the license plate information 1111, and the license plate information 3110 are different and occur once, further detection needs to be performed on other target images except for the target image to which the license plate information belongs, that is, the remaining images corresponding to other target images not including the license plate information or the license plate, according to the license plate information, so as to determine the number of target images including the same vehicle.
B2, determining the number of target images containing the same license plate information according to the target images to which the license plate information belongs;
b3, if the number of the target images containing the same license plate information is smaller than the preset number, detecting other target images except the target image containing the same license plate information according to the license plate information, and determining the number of the target images containing the same vehicle.
In this embodiment, first, a target image to which an image containing the same license plate information belongs may be determined, and if the effective number is greater than or equal to a preset number, it is not necessary to further detect other images (i.e., other first sub-images and other second sub-images except for other first sub-images containing the same license plate information) by counting the effective number of the target image containing the same license plate information (i.e., the number of the target image containing the same license plate information), and it may be directly determined that a vehicle corresponding to the same license plate information contains the target vehicle; if the effective number is less than the preset number, it is necessary to further determine whether the other images without license plate information contain the same vehicle.
For example, the number of target images containing the same license plate information may be directly detected, and if at least two (for example, the preset number is 2) target images containing the same license plate information are detected in the multiple target images, it is determined that the number of the target images containing the same license plate information is greater than or equal to a preset threshold, and the vehicle corresponding to the same license plate information is determined to be a target vehicle, that is, an accident vehicle without a warning identifier, and the accident vehicle needs to be subjected to violation processing, and the license plate number of the vehicle may be reported and processed by a relevant department.
In a possible design, the present embodiment, based on the above embodiment, detects other target images except the target image containing the same license plate information, and determines the number of target images containing the same vehicle. The method can be realized by the following steps:
and c1, taking the first sub-images which do not contain the license plate in the other target images and the first sub-images corresponding to the second sub-images which do not detect the license plate information as target sub-images.
Step c2, comparing the vehicles in the first sub-images corresponding to the target images containing the license plate information with the vehicles in each target sub-image respectively, and determining whether the vehicles in the target sub-images are the vehicles in the first sub-images corresponding to the target images containing the license plate information.
And c3, if yes, determining the target images of the target sub-images, and counting the number of the target images of the target sub-images according to the target images containing the same license plate information, wherein the number of the target images of the target sub-images is the number of the target images containing the same vehicle.
The purpose of keeping the small vehicle image without the detected license plate and keeping the vehicle without the license plate corresponding to the recognized license plate is to execute steps c1 to c 3.
In this embodiment, for each detected license plate information, all target images except for the license plate information are sequentially detected by the following steps that no license plate is detected or no vehicle corresponding to the license plate number is recognized, that is, target sub-images: and aiming at the vehicles which do not detect the license plate or do not recognize the corresponding license plate, sequentially matching the vehicles which recognize the license plate by using the reid model, thereby finding out whether the vehicles which recognize the license plate are contained in other images or not, and if so, finding out the vehicle. So far, the license plates corresponding to all vehicles in a plurality of images (such as 3 images) are determined, and the times of the same vehicle appearing in the three images are also determined.
The license plate information corresponding to the target vehicle can be reported to the target terminal, so that the target terminal can execute violation processing operation on the target vehicle. For example, the vehicles for which no license plate is determined in any of the three figures are directly ignored because it is not possible to determine which vehicle violates the license plate. Vehicles that appear only once in the three figures are directly ignored and no penalty is imposed because vehicles that may be passing by appear only once. And (4) carrying out violation penalty on the target vehicles which appear for 2 times or more in the three graphs.
Therefore, the method for judging whether the accident vehicle is provided with the warning sign or not based on the deep learning in the embodiment utilizes the computer vision technology of the deep learning algorithm, is stable and efficient, has good reusability, and saves a large amount of manpower.
Fig. 5 is a schematic structural diagram of an information detecting apparatus according to an embodiment of the present application, corresponding to the information detecting method according to the foregoing embodiment. For convenience of explanation, only portions related to the embodiments of the present application are shown. As shown in fig. 5, the information detection apparatus 50 includes: a first acquisition module 501, a second acquisition module 502, and a target vehicle determination module 503. The first obtaining module 501 is configured to obtain a plurality of target images, sequentially detect each target image, and determine whether the target image contains a warning sign, where the target image is an image of a location area where an accident vehicle is located; a second obtaining module 502, configured to obtain a first sub-image of each target image including a vehicle when the warning sign is not included in each target image; the target vehicle determining module 503 is configured to detect vehicles in each first sub-image, and if the number of target images to which the first sub-images containing the same vehicle belong is greater than or equal to a preset number, determine that the same vehicle is a target vehicle, where the target vehicle is an accident vehicle without a warning flag.
The information detection apparatus provided in this embodiment uses the first obtaining module 501, the second obtaining module 502, and the target vehicle determining module 503 to obtain a plurality of target images, sequentially detect each target image, determine whether each target image contains a warning sign, if none of the plurality of target images contains the warning sign, indicate that there is a vehicle in violation, and start to detect a target vehicle, which is an illegal vehicle (an accident vehicle without a warning sign), from the plurality of images; in order to narrow the detection range and save resources, first sub-images containing vehicles can be acquired from each image, then each first sub-image is detected, the number of target images containing the same vehicles in all the first sub-images, namely the number of the same vehicles distributed in which target images or the number of the same vehicles appearing in which target images are determined, if the number of the obtained target images is larger than the preset number, the same vehicles are indicated to be accident vehicles without warning signs, image detection and information comparison are performed through a machine, whether warning signs are set or not and illegal vehicle detection is achieved, the detection efficiency is improved, and meanwhile human resources are saved.
In one possible design, the first obtaining module is specifically configured to:
for each target image, identifying the position of a vehicle in the target image; and generating a first sub-image from a preset area image of the position of the vehicle in the target image.
In one possible design, the target vehicle determination module 503 is specifically configured to:
for each first sub-image, identifying whether the first sub-image contains a license plate; if the license plate is contained, detecting license plate information corresponding to the license plate; and if the license plate information is detected, determining the number of target images of the first sub-images containing the same vehicle according to the license plate information.
In one possible design, the target vehicle determination module 503 is further specifically configured to:
when the license plate is contained, generating a second sub-image from a preset area image of the position of the license plate in the first sub-image; and detecting license plate information corresponding to the license plate in the second sub-image.
In one possible design, the target vehicle determination module 503 is further specifically configured to:
determining a target image to which the license plate information belongs according to the license plate information; determining the number of target images containing the same license plate information according to the target images to which the license plate information belongs; and if the number of the target images containing the same license plate information is smaller than a preset number, detecting other target images except the target image containing the same license plate information according to the license plate information, and determining the number of the target images containing the same vehicle.
In one possible design, the target vehicle determination module 503 is further specifically configured to:
taking the first sub-image which does not contain the license plate in the other target images and the first sub-image corresponding to the second sub-image which does not detect the license plate information as target sub-images; respectively comparing the vehicles in the first sub-images corresponding to the target images containing the license plate information with the vehicles in each target sub-image, and determining whether the vehicles in the target sub-images are the vehicles in the first sub-images corresponding to the target images containing the license plate information; if yes, determining a target image to which the target sub-image belongs, and counting the number of the target images to which the target sub-image belongs according to the target images containing the same license plate information, wherein the number of the target images to which the target sub-image belongs is the number of the target images containing the same vehicle.
In one possible design, the apparatus further includes: an information reporting module; and the information reporting module is used for reporting the license plate information corresponding to the target vehicle to a target terminal so that the target terminal can execute violation processing operation on the target vehicle.
The apparatus provided in the embodiment of the present application may be configured to implement the technical solution of the method embodiment, and the implementation principle and the technical effect are similar, which are not described herein again in the embodiment of the present application.
Fig. 6 is a schematic diagram of a hardware structure of an information detection apparatus according to an embodiment of the present application. As shown in fig. 6, the present embodiment provides an information detecting apparatus 60 including: at least one processor 601 and memory 602. The processor 601 and the memory 602 are connected by a bus 603.
In a specific implementation, at least one processor 601 executes computer-executable instructions stored by the memory 602 to cause the at least one processor 601 to perform the methods of the above-described method embodiments.
For a specific implementation process of the processor 601, reference may be made to the above method embodiments, which implement the principle and the technical effect similarly, and details of this embodiment are not described herein again.
In the embodiment shown in fig. 6, it should be understood that the Processor may be a Central Processing Unit (CPU), other general purpose processors, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the present invention may be embodied directly in a hardware processor, or in a combination of the hardware and software modules within the processor.
The memory may comprise high speed RAM memory and may also include non-volatile storage NVM, such as at least one disk memory.
The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended ISA (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, the buses in the figures of the present application are not limited to only one bus or one type of bus.
The embodiment of the invention also provides a computer-readable storage medium, wherein a computer execution instruction is stored in the computer-readable storage medium, and when a processor executes the computer execution instruction, the information detection method of the embodiment of the method is realized.
The computer-readable storage medium may be implemented by any type of volatile or non-volatile memory device or combination thereof, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disk. Readable storage media can be any available media that can be accessed by a general purpose or special purpose computer.
An exemplary readable storage medium is coupled to the processor such the processor can read information from, and write information to, the readable storage medium. Of course, the readable storage medium may also be an integral part of the processor. The processor and the readable storage medium may reside in an Application Specific Integrated Circuits (ASIC). Of course, the processor and the readable storage medium may also reside as discrete components in the apparatus.
Those of ordinary skill in the art will understand that: all or a portion of the steps of implementing the above-described method embodiments may be performed by hardware associated with program instructions. The program may be stored in a computer-readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. An information detection method, comprising:
acquiring a plurality of target images, detecting each target image in sequence, and judging whether the target image contains a warning sign or not, wherein the target image is an image of a position area where an accident vehicle is located;
if the warning sign is not contained in each target image, acquiring a first sub-image of each target image containing a vehicle;
and detecting the vehicles in each first sub-image, and if the number of the target images of the first sub-images containing the same vehicle is greater than or equal to a preset number, determining that the same vehicle is a target vehicle which is an accident vehicle without a warning sign.
2. The method of claim 1, wherein said obtaining a first sub-image of each of said target images containing a vehicle comprises:
for each target image, identifying the position of a vehicle in the target image;
and generating a first sub-image from a preset area image of the position of the vehicle in the target image.
3. The method according to claim 1 or 2, wherein the detecting the vehicle in each of the first sub-images comprises:
for each first sub-image, identifying whether the first sub-image contains a license plate;
if the license plate is contained, detecting license plate information corresponding to the license plate;
and if the license plate information is detected, determining the number of target images of the first sub-images containing the same vehicle according to the license plate information.
4. The method of claim 3, wherein detecting license plate information corresponding to the license plate if the license plate is included comprises:
if the license plate is contained, generating a second sub-image from a preset area image of the position of the license plate in the first sub-image;
and detecting license plate information corresponding to the license plate in the second sub-image.
5. The method of claim 4, wherein the determining the number of target images to which the first sub-image containing the same vehicle belongs according to the license plate information comprises:
determining a target image to which the license plate information belongs according to the license plate information;
determining the number of target images containing the same license plate information according to the target images to which the license plate information belongs;
and if the number of the target images containing the same license plate information is smaller than a preset number, detecting other target images except the target image containing the same license plate information according to the license plate information, and determining the number of the target images containing the same vehicle.
6. The method of claim 5, wherein the detecting, according to the license plate information, target images other than the target image containing the same license plate information to determine the number of target images containing the same vehicle comprises:
taking the first sub-image which does not contain the license plate in the other target images and the first sub-image corresponding to the second sub-image which does not detect the license plate information as target sub-images;
respectively comparing the vehicles in the first sub-images corresponding to the target images containing the license plate information with the vehicles in each target sub-image, and determining whether the vehicles in the target sub-images are the vehicles in the first sub-images corresponding to the target images containing the license plate information;
if yes, determining a target image to which the target sub-image belongs, and counting the number of the target images to which the target sub-image belongs according to the target images containing the same license plate information, wherein the number of the target images to which the target sub-image belongs is the number of the target images containing the same vehicle.
7. The method of claim 1, further comprising:
and reporting the license plate information corresponding to the target vehicle to a target terminal, so that the target terminal executes violation processing operation on the target vehicle.
8. An information detecting apparatus, characterized by comprising:
the system comprises a first acquisition module, a second acquisition module and a warning module, wherein the first acquisition module is used for acquiring a plurality of target images, detecting each target image in sequence and judging whether the target image contains a warning sign or not, and the target image is an image of a position area where an accident vehicle is located;
the second acquisition module is used for acquiring a first sub-image containing a vehicle in each target image when the warning sign is not contained in each target image;
and the target vehicle determining module is used for detecting the vehicles in each first sub-image, and if the number of the target images of the first sub-images containing the same vehicle is greater than or equal to the preset number, determining that the same vehicle is the target vehicle, and the target vehicle is an accident vehicle without a warning sign.
9. An information detecting apparatus characterized by comprising: at least one processor and memory;
the memory stores computer-executable instructions;
the at least one processor executing the computer-executable instructions stored by the memory causes the at least one processor to perform the information detection method of any one of claims 1 to 7.
10. A computer-readable storage medium having stored thereon computer-executable instructions that, when executed by a processor, implement the information detection method according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011595163.0A CN112750312A (en) | 2020-12-28 | 2020-12-28 | Information detection method, device, equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011595163.0A CN112750312A (en) | 2020-12-28 | 2020-12-28 | Information detection method, device, equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112750312A true CN112750312A (en) | 2021-05-04 |
Family
ID=75646787
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011595163.0A Pending CN112750312A (en) | 2020-12-28 | 2020-12-28 | Information detection method, device, equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112750312A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109522796A (en) * | 2018-10-12 | 2019-03-26 | 李雨宸 | Fault car reminding method and Related product in high speed |
CN110444019A (en) * | 2019-08-01 | 2019-11-12 | 安徽科力信息产业有限责任公司 | Vehicle abnormality parking origin cause of formation detection method and system based on vehicle-mounted parking warning sign |
JP2019212188A (en) * | 2018-06-08 | 2019-12-12 | スズキ株式会社 | Road sign recognition device |
CN110991320A (en) * | 2019-11-29 | 2020-04-10 | 北京百度网讯科技有限公司 | Road condition detection method and device, electronic equipment and storage medium |
CN111640301A (en) * | 2020-05-25 | 2020-09-08 | 北京百度网讯科技有限公司 | Method, system and device for detecting fault vehicle, electronic equipment and storage medium |
CN112101223A (en) * | 2020-09-16 | 2020-12-18 | 北京百度网讯科技有限公司 | Detection method, device, equipment and computer storage medium |
-
2020
- 2020-12-28 CN CN202011595163.0A patent/CN112750312A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2019212188A (en) * | 2018-06-08 | 2019-12-12 | スズキ株式会社 | Road sign recognition device |
CN109522796A (en) * | 2018-10-12 | 2019-03-26 | 李雨宸 | Fault car reminding method and Related product in high speed |
CN110444019A (en) * | 2019-08-01 | 2019-11-12 | 安徽科力信息产业有限责任公司 | Vehicle abnormality parking origin cause of formation detection method and system based on vehicle-mounted parking warning sign |
CN110991320A (en) * | 2019-11-29 | 2020-04-10 | 北京百度网讯科技有限公司 | Road condition detection method and device, electronic equipment and storage medium |
CN111640301A (en) * | 2020-05-25 | 2020-09-08 | 北京百度网讯科技有限公司 | Method, system and device for detecting fault vehicle, electronic equipment and storage medium |
CN112101223A (en) * | 2020-09-16 | 2020-12-18 | 北京百度网讯科技有限公司 | Detection method, device, equipment and computer storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111476234B (en) | License plate character shielding recognition method and device, storage medium and intelligent equipment | |
CN111369801B (en) | Vehicle identification method, device, equipment and storage medium | |
CN111612104B (en) | Vehicle loss assessment image acquisition method, device, medium and electronic equipment | |
CN112464030B (en) | Suspicious person determination method and suspicious person determination device | |
CN112749622B (en) | Emergency lane occupation recognition method and device | |
CN113313400A (en) | Accident warning method, accident warning device and storage medium | |
CN110889388A (en) | Violation identification method, device, equipment and storage medium | |
CN114155488A (en) | Method and device for acquiring passenger flow data, electronic equipment and storage medium | |
CN111368617B (en) | Vehicle access data processing method and device | |
CN115810134A (en) | Image acquisition quality inspection method, system and device for preventing car insurance from cheating | |
CN111768630A (en) | Violation waste image detection method and device and electronic equipment | |
CN113408364B (en) | Temporary license plate recognition method, system, device and storage medium | |
CN112115946B (en) | License plate fake-license plate identification method and device, storage medium and electronic equipment | |
CN111860512B (en) | Vehicle identification method, device, electronic equipment and computer readable storage medium | |
CN111724607B (en) | Steering lamp use detection method and device, computer equipment and storage medium | |
CN111985304A (en) | Patrol alarm method, system, terminal equipment and storage medium | |
CN111985438A (en) | Static face processing method, device and equipment | |
CN112750312A (en) | Information detection method, device, equipment and storage medium | |
CN113160578B (en) | License plate information confirmation method, system and computer storage medium | |
CN115984786A (en) | Vehicle damage detection method and device, terminal and storage medium | |
CN113642521B (en) | Traffic light identification quality evaluation method and device and electronic equipment | |
CN112597924B (en) | Electric bicycle track tracking method, camera device and server | |
CN111191603B (en) | Method and device for identifying people in vehicle, terminal equipment and medium | |
CN113469179A (en) | Image identification method and device, electronic equipment and storage medium | |
CN111369803B (en) | Marginal bayonet detection method and device and computer readable storage medium |
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 | ||
AD01 | Patent right deemed abandoned | ||
AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20220830 |