CN111845647A - Automobile camera cleaning system and method - Google Patents

Automobile camera cleaning system and method Download PDF

Info

Publication number
CN111845647A
CN111845647A CN201910351091.6A CN201910351091A CN111845647A CN 111845647 A CN111845647 A CN 111845647A CN 201910351091 A CN201910351091 A CN 201910351091A CN 111845647 A CN111845647 A CN 111845647A
Authority
CN
China
Prior art keywords
image
target
type
camera
shielding object
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.)
Granted
Application number
CN201910351091.6A
Other languages
Chinese (zh)
Other versions
CN111845647B (en
Inventor
胡连强
汪茂盛
谭方培
张健
周庆伟
周锋
王燕文
钱锋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
SAIC Motor Corp Ltd
Original Assignee
SAIC Motor Corp Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by SAIC Motor Corp Ltd filed Critical SAIC Motor Corp Ltd
Priority to CN201910351091.6A priority Critical patent/CN111845647B/en
Publication of CN111845647A publication Critical patent/CN111845647A/en
Application granted granted Critical
Publication of CN111845647B publication Critical patent/CN111845647B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60S—SERVICING, CLEANING, REPAIRING, SUPPORTING, LIFTING, OR MANOEUVRING OF VEHICLES, NOT OTHERWISE PROVIDED FOR
    • B60S1/00—Cleaning of vehicles
    • B60S1/02—Cleaning windscreens, windows or optical devices
    • B60S1/56—Cleaning windscreens, windows or optical devices specially adapted for cleaning other parts or devices than front windows or windscreens
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
    • B60R11/00—Arrangements for holding or mounting articles, not otherwise provided for
    • B60R11/04—Mounting of cameras operative during drive; Arrangement of controls thereof relative to the vehicle
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60S—SERVICING, CLEANING, REPAIRING, SUPPORTING, LIFTING, OR MANOEUVRING OF VEHICLES, NOT OTHERWISE PROVIDED FOR
    • B60S1/00—Cleaning of vehicles
    • B60S1/02—Cleaning windscreens, windows or optical devices
    • B60S1/46—Cleaning windscreens, windows or optical devices using liquid; Windscreen washers
    • B60S1/48—Liquid supply therefor
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60S—SERVICING, CLEANING, REPAIRING, SUPPORTING, LIFTING, OR MANOEUVRING OF VEHICLES, NOT OTHERWISE PROVIDED FOR
    • B60S1/00—Cleaning of vehicles
    • B60S1/02—Cleaning windscreens, windows or optical devices
    • B60S1/54—Cleaning windscreens, windows or optical devices using gas, e.g. hot air
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60S—SERVICING, CLEANING, REPAIRING, SUPPORTING, LIFTING, OR MANOEUVRING OF VEHICLES, NOT OTHERWISE PROVIDED FOR
    • B60S1/00—Cleaning of vehicles
    • B60S1/02—Cleaning windscreens, windows or optical devices
    • B60S1/56—Cleaning windscreens, windows or optical devices specially adapted for cleaning other parts or devices than front windows or windscreens
    • B60S1/566—Cleaning windscreens, windows or optical devices specially adapted for cleaning other parts or devices than front windows or windscreens including wiping devices

Abstract

The invention provides a cleaning system and a cleaning method for an automobile camera. The electronic control unit can determine the type of the shielding object by analyzing a target image shot by the target camera, so that the working states of the cleaning device and the blowing device are controlled according to the type of the shielding object. Based on the invention, the automatic cleaning of the camera can be realized, the driving safety is improved, and the maintenance work of drivers and passengers on the camera is reduced.

Description

Automobile camera cleaning system and method
Technical Field
The invention relates to the technical field of automobiles, in particular to an automobile camera cleaning system and method.
Background
In the current automobile market, for a camera installed outside a vehicle, a driver and a passenger need to check the cleanliness of the surface by themselves and artificially clean to ensure the normal function of the camera. Therefore, driving safety is influenced, and maintenance work of drivers and passengers on the camera is increased.
Disclosure of Invention
In view of the above, the present invention provides a system and a method for cleaning a camera of an automobile. The technical scheme is as follows:
An automotive camera cleaning system, the system comprising:
the cleaning device comprises an electronic control unit, a cleaning device and a purging device, wherein the electronic control unit is electrically connected with the cleaning device and the purging device respectively;
the electronic control unit is used for acquiring a target image shot by a target camera; analyzing the target image to determine the type of the shielding object corresponding to the target camera; controlling the working states of the cleaning device and the blowing device according to the type of the shielding object so as to clean the target camera;
the electronic control unit, configured to analyze the target image and determine a type of a mask corresponding to the target camera, is specifically configured to:
calculating the global fuzziness of the target image; judging whether the global ambiguity is greater than a specified global ambiguity threshold value; if the global fuzziness is larger than the designated global fuzziness threshold, segmenting the target image to obtain sub-images; extracting a feature vector of the sub-image, and performing feature classification on the feature vector to obtain a first shelter type corresponding to the sub-image; determining a shielding object type corresponding to the target camera according to a first shielding object type corresponding to the sub-image; if the global ambiguity is not greater than the designated global ambiguity, acquiring other images shot by other cameras except the target camera; splicing the target image and the other images to obtain a spliced image; processing the spliced image by using a MobileNet-SSD deep convolution neural network model to obtain a detection frame positioned on the spliced image and a second shelter type corresponding to the detection frame; selecting an optimal detection frame from the detection frames by utilizing an NMS algorithm; and under the condition that the position of the optimal detection frame on the spliced image corresponds to the target image, determining the type of the shielding object corresponding to the target camera according to the type of the second shielding object corresponding to the optimal detection frame.
Preferably, the system further comprises:
the vehicle-mounted sensor is electrically connected with the electronic control unit;
the electronic control unit is further configured to acquire environment data through the vehicle-mounted sensor, and execute the analysis of the target image to determine a type of a shielding object corresponding to the target camera when the environment data meets a specified condition.
Preferably, the system further comprises:
the mobile terminal is in communication connection with the electronic control unit;
the electronic control unit is also used for sending the cleaning result of the target camera to the mobile terminal;
and the mobile terminal is used for prompting based on the cleaning result of the target camera.
Preferably, the cleaning device includes:
the washing machine comprises a washing kettle containing washing liquid, a water pump and a first spray head, wherein the water pump is electrically connected with the electronic control unit, and the washing liquid is sprayed out of the first spray head under the action of the water pump.
Preferably, the first spray head is a gas-liquid dual-purpose spray head.
Preferably, the purge device includes:
the air pump is electrically connected with the electronic control unit, and air is sprayed out of the second spray head under the action of the air pump.
Preferably, the second spray head is a gas-liquid dual-purpose spray head.
Preferably, a heater is arranged on the second spray head.
A method for cleaning a vehicle camera, applied to the electronic control unit in the vehicle camera cleaning system, the method comprising:
acquiring a target image shot by a target camera;
analyzing the target image to determine the type of the shielding object corresponding to the target camera;
controlling the working states of the cleaning device and the blowing device according to the type of the shielding object so as to clean the target camera;
wherein the analyzing the target image to determine the type of the shielding object corresponding to the target camera comprises:
calculating the global fuzziness of the target image;
judging whether the global ambiguity is greater than a specified global ambiguity threshold value;
if the global fuzziness is larger than the designated global fuzziness threshold, segmenting the target image to obtain sub-images;
extracting a feature vector of the sub-image, and performing feature classification on the feature vector to obtain a first shelter type corresponding to the sub-image;
determining a shielding object type corresponding to the target camera according to a first shielding object type corresponding to the sub-image;
If the global ambiguity is not greater than the designated global ambiguity, acquiring other images shot by other cameras except the target camera;
splicing the target image and the other images to obtain a spliced image;
processing the spliced image by using a MobileNet-SSD deep convolution neural network model to obtain a detection frame positioned on the spliced image and a second shelter type corresponding to the detection frame;
selecting an optimal detection frame from the detection frames by utilizing an NMS algorithm;
and under the condition that the position of the optimal detection frame on the spliced image corresponds to the target image, determining the type of the shielding object corresponding to the target camera according to the type of the second shielding object corresponding to the optimal detection frame.
The invention provides a cleaning system and a cleaning method for an automobile camera. The electronic control unit can determine the type of the shielding object by analyzing a target image shot by the target camera, so that the working states of the cleaning device and the blowing device are controlled according to the type of the shielding object. Based on the invention, the automatic cleaning of the camera can be realized, the driving safety is improved, and the maintenance work of drivers and passengers on the camera is reduced.
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 embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a schematic structural diagram of an automobile camera cleaning system according to an embodiment of the present invention;
fig. 2 is another schematic structural diagram of an automobile camera cleaning system according to an embodiment of the present invention;
FIG. 3 is a schematic view of another embodiment of a cleaning system for a camera of an automobile;
fig. 4 is a flowchart of a method for cleaning a camera of an automobile according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the 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 invention.
An embodiment of the present invention provides an automobile camera cleaning system, a schematic structural diagram of which is shown in fig. 1 (a straight line in fig. 1 represents an electrical connection), including:
the electronic control unit 10, the cleaning device 20 and the purging device 30, wherein the electronic control unit 10 is electrically connected with the cleaning device 20 and the purging device 30 respectively.
In this embodiment, the electronic control unit 10 may be connected to the cleaning device 20 and the purge device 30 by hard wire harnesses, respectively.
An electronic control unit 10 for acquiring a target image photographed by a target camera; analyzing the target image to determine the type of the shielding object corresponding to the target camera; controlling the working states of the cleaning device 20 and the purging device 30 according to the type of the shielding object to clean the target camera;
the electronic control unit for analyzing the target image to determine the type of the shielding object corresponding to the target camera is specifically configured to:
calculating the global fuzziness of the target image; judging whether the global ambiguity is greater than a specified global ambiguity threshold value or not; if the global fuzziness is larger than a designated global fuzziness threshold, dividing the target image to obtain sub-images; extracting a feature vector of the subimage, and performing feature classification on the feature vector to obtain a first shelter type corresponding to the subimage; determining a shielding object type corresponding to the target camera according to the first shielding object type corresponding to the sub-image; if the global ambiguity is not greater than the designated global ambiguity, acquiring other images shot by other cameras except the target camera; splicing the target image and other images to obtain a spliced image; processing the spliced image by using a MobileNet-SSD deep convolution neural network model to obtain a detection frame positioned on the spliced image and a second shelter type corresponding to the detection frame; selecting an optimal detection frame from the detection frames by utilizing an NMS algorithm; and under the condition that the position of the optimal detection frame on the spliced image corresponds to the target image, determining the type of the shielding object corresponding to the target camera according to the type of the second shielding object corresponding to the optimal detection frame.
In this embodiment, when receiving a target image sent by a target camera, the electronic control unit 10 may calculate image features such as a degree of blur and color distribution in blocks, and process the image features with an evaluation function to obtain a global degree of blur of the target image, where the global degree of blur represents a degree of blur of the image.
1) And if the global ambiguity is greater than a specified global ambiguity threshold value, indicating that the cleanliness of the outer surface of the target camera is poor. In order to improve the detection efficiency, the target image may be divided into a plurality of sub-images, the first shelter type corresponding to each sub-image is identified, and then the shelter type corresponding to the target camera is determined by integrating the first shelter types corresponding to all the sub-images, for example, the first shelter type with the largest occurrence frequency may be determined as the shelter type corresponding to the target camera, specifically:
for each sub-image, firstly extracting the edge features of the sub-image, then extracting the characteristic vectors except the edge features by applying a Laplacian, and finally performing feature classification on the characteristic vectors by using an SVM (support vector machine) model to obtain a first shelter type corresponding to the sub-image. Wherein the SVM model is trained using a large number of samples in advance to provide the SVM model with the ability to trend the predicted shade type toward the actual shade type.
It should be noted that, to improve the detection effectiveness, before segmenting the target image, the target image may be preprocessed, such as performing Gamma correction, and further, for example, extracting an effective region, etc. It should be understood that the above is only an example of the pretreatment, and other pretreatment methods not listed are also within the scope of the present invention.
2) And if the global ambiguity is greater than a specified global ambiguity threshold value, the cleanliness of the outer surface of the target camera is better. In order to reduce the waste of computing resources, other images shot by other cameras can be spliced with the target image, so that the effects of detecting the types of the shielding objects of the cameras at one time and reducing the detection time are achieved.
Further, the mosileNet-SSD deep convolution neural network model is used for processing the spliced image to obtain a detection frame positioned on the spliced image and a second shelter type corresponding to the detection frame. In addition, in the embodiment, the MobileNet-SSD deep convolutional neural network model is trained by using a large number of samples in advance, so that the MobileNet-SSD deep convolutional neural network model has the capability of making the predicted detection frame tend to the actual detection frame and the predicted shelter type tend to the actual shelter type.
Further, selecting the best detection box from the detection boxes using NMS (Non-maximum suppression) algorithm that the mask type can represent the mask type of the other detection boxes. And if the position of the optimal detection frame on the stitched image corresponds to the target image (that is, the position of the optimal detection frame on the stitched image belongs to the range of the target image on the stitched image, in whole or in part), the second mask type corresponding to the optimal detection frame may be used as the mask type corresponding to the target camera.
It should be noted that, in order to improve the detection effectiveness, before the mosilenet-SSD deep convolutional neural network model is used to process the stitched image, the stitched image may be preprocessed, for example, Gamma correction is performed, and then, for example, an effective region is extracted. It should be understood that the above is only an example of the pretreatment, and other pretreatment methods not listed are also within the scope of the present invention.
It should be further noted that, in order to improve the detection accuracy, if the position portion of the optimal detection frame on the stitched image belongs to the range where the target image is located on the stitched image, the MobileNet-SSD deep convolutional neural network model may be reused to process the target image to obtain a new detection frame located on the target image and a third mask type corresponding to the new detection frame, and the third mask types corresponding to all new detection frames are integrated to determine the mask type corresponding to the target camera, for example, the third mask type with the largest occurrence number may be determined as the mask type corresponding to the target camera.
It should be noted that the type of the shielding object may be any one or more of dust, rain, oil stain, ice, snow, frost, and the like. It is understood that other types of coverings not listed are also within the scope of the present embodiment.
Further, after determining the type of the covering corresponding to the target camera, the operating states of the washing device 20 and the purging device 30 are controlled in association with the previously stored correspondence between the different types of the outer surface coverings and the cleaning strategy. The following description will be given taking the types of the shelter as dust and rain, respectively, as examples:
1) if the type of the shielding object is dust, the electronic control unit 10 starts the cleaning device 20 to clean the outer surface of the target camera, and then starts the purging device 30 to purge the outer surface of the target camera.
2) If the shelter type is rain, the electronic control unit 10 activates the purging device 30 to purge the outer surface of the target camera.
In practical applications, the cleaning device 20 in this embodiment includes a washing kettle containing a washing liquid, a water pump and a first nozzle, the water pump is electrically connected to the electronic control unit 10, and the washing liquid is sprayed from the first nozzle under the action of the water pump. Specifically, be provided with the first pipeline that is used for transmitting the washing liquid between water pump and the first shower nozzle, electronic control unit 10 control opening of water pump stops, and when the water pump started, can be with washing liquid pump to first shower nozzle through this first pipeline, by first shower nozzle blowout.
Of course, to prevent the first nozzle from being mistakenly sprayed, the electronic control unit 10 may be electrically connected to the first nozzle for controlling the start and stop of the first nozzle, and the cleaning solution may be sprayed from the first nozzle only when the first nozzle is started.
The purging device 30 in this embodiment includes an air pump and a second nozzle, the air pump is electrically connected to the electronic control unit 10, and air is ejected from the second nozzle under the action of the air pump. Specifically, a second pipeline for transmitting air is arranged between the air pump and the second nozzle, the electronic control unit 10 controls the start and stop of the air pump, and when the air pump is started, the air pump can be pumped to the second nozzle through the second pipeline and is sprayed out of the second nozzle.
Of course, to prevent the second nozzle from being mistakenly sprayed, the electronic control unit 10 may be electrically connected to the second nozzle for controlling the start and stop of the second nozzle, and only when the second nozzle is started, the air may be sprayed from the second nozzle.
In specific application, the first spray head can be a gas-liquid dual-purpose spray head, and the second spray head can also be a gas-liquid dual-purpose spray head. Moreover, if the first nozzle and the second nozzle both use gas-liquid dual-purpose nozzles, the cleaning rotating device and the purging device 30 can share the same gas-liquid dual-purpose nozzle, and the tail of the gas-liquid dual-purpose nozzle is provided with a first pipeline and a second pipeline.
In addition, in order to improve the blowing efficiency of the blowing device 30, a heater may be provided on the second nozzle, so that the second nozzle blows hot air to improve the evaporation rate of the cleaning liquid or water.
It should be noted that, in the present embodiment, the cleaning device 20 may also adopt an existing device having a cleaning function, and the purging device 30 may also adopt an existing device having a purging function.
In other embodiments, to reduce false alarms and eliminate the influence of extreme environments, based on the automobile camera cleaning system shown in fig. 1, the system further includes the following components, and the structural schematic diagram is shown in fig. 2 (the straight line in fig. 2 represents the electrical connection):
and the vehicle-mounted sensor 40, wherein the vehicle-mounted sensor 40 is electrically connected with the electronic control unit 10.
In this embodiment, the electronic control unit 10 may be connected to the in-vehicle sensor 40 through a CAN bus. The vehicle-mounted sensor 40 may be a brightness sensor, a rainfall sensor, or the like, which is not limited in this embodiment and can be selected according to actual needs. It is understood that other on-board sensors 40, not listed, are also within the scope of the present embodiment.
And the electronic control unit 10 is also used for acquiring environment data through the vehicle-mounted sensor 40 and analyzing the target image to determine the type of the shielding object corresponding to the target camera under the condition that the environment data meets specified conditions.
In this embodiment, the electronic control unit 10 may determine the environmental data around the target camera based on the sensor signal fed back by the in-vehicle sensor 40, for example, the brightness sensor may be used to determine the environmental brightness, and for example, the rainfall sensor may be used to determine the environmental rainfall. The following description will be made by taking the ambient brightness and the ambient rainfall as examples, respectively:
1) for the ambient brightness, if the ambient brightness is smaller than the specified brightness threshold, it can be determined that the current environment of the target camera is a dark environment, and at this time, analyzing the target image shot by the target camera cannot accurately determine the type of the shielding object of the target camera. Therefore, when the ambient brightness is greater than the specified brightness threshold, the basis for identifying the type of the shielding object of the target camera is provided, and the target image is analyzed to determine the type of the shielding object corresponding to the target camera and the subsequent steps are also executed.
2) For the environmental rainfall, if the environmental rainfall is greater than the specified rainfall threshold, the rainstorm environment of the current environment of the target camera can be determined, and at this time, the effect of activating the purging device 30 to purge the target camera is not great. Therefore, when the environmental rainfall is less than the designated rainfall threshold, the necessity of cleaning the target camera is met, and the steps of analyzing the target image to determine the type of the shielding object corresponding to the target camera and the subsequent steps are also executed.
Further, when the environmental rainfall is smaller than the designated rainfall threshold, in order to improve the purging efficiency, the electronic control unit 10 may further adopt any one or more of a mode of improving the purging intensity, a mode of improving the purging frequency, and a mode of improving the purging temperature.
In other embodiments, to improve human-computer interaction capability, on the basis of the automobile camera cleaning system shown in fig. 1, the system further includes the following components, and the schematic structural diagram is shown in fig. 3 (the straight line segment in fig. 3 represents an electrical connection, and the straight line segment of a bidirectional arrow represents a communication connection):
and the mobile terminal 50 is in communication connection with the electronic control unit 10, and the mobile terminal 50 is in communication connection with the electronic control unit 10.
In this embodiment, the mobile terminal 50 may establish a communication connection with the electronic control unit 10 in a bluetooth mode, a hot spot mode, or a wifi mode, which is not limited in this embodiment. It is understood that other communication methods not listed are also within the protection scope of the present embodiment.
And the electronic control unit 10 is also used for sending the cleaning result of the target camera to the mobile terminal 50.
In this embodiment, the electronic control unit 10 may send the cleaning result to the mobile terminal 50, and may also synthesize the historical cleaning result to diagnose the target camera, for example, the surface of the target camera is still not clean after the target camera is cleaned and/or purged for multiple times, and send the diagnosis result of "the target camera may be damaged" to the mobile terminal 50.
And the mobile terminal 50 is used for prompting based on the cleaning result of the target camera.
In this embodiment, the mobile terminal 50 may prompt in a specific manner, such as a prompt sound, and further, in a dialog box, for example. At this time, the driver and the crew can clean and/or purge the surface of the target camera by manual operation or by using the APP to control the electronic control unit 10.
The cleaning system for the automobile camera comprises an electronic control unit, a cleaning device and a blowing device. The electronic control unit can determine the type of the shielding object by analyzing a target image shot by the target camera, so that the working states of the cleaning device and the blowing device are controlled according to the type of the shielding object. Based on the invention, the automatic cleaning of the camera can be realized, the driving safety is improved, and the maintenance work of drivers and passengers on the camera is reduced.
Based on the automobile camera cleaning system provided by the embodiment, the embodiment of the invention provides an automobile camera cleaning method, the method is applied to an electronic control unit in the automobile camera cleaning system, a flow chart of the method is shown in fig. 4, and the method comprises the following steps:
And S10, acquiring the target image shot by the target camera.
And S20, analyzing the target image to determine the type of the shielding object corresponding to the target camera.
And S30, controlling the working states of the cleaning device and the blowing device according to the type of the shielding object so as to clean the target camera.
In step S20, "analyzing the target image to determine the type of the shielding object corresponding to the target camera", may include the following steps:
calculating the global fuzziness of the target image;
judging whether the global ambiguity is greater than a specified global ambiguity threshold value or not;
if the global fuzziness is larger than a designated global fuzziness threshold, dividing the target image to obtain sub-images;
extracting a feature vector of the subimage, and performing feature classification on the feature vector to obtain a first shelter type corresponding to the subimage;
determining a shielding object type corresponding to the target camera according to the first shielding object type corresponding to the sub-image;
if the global ambiguity is not greater than the designated global ambiguity, acquiring other images shot by other cameras except the target camera;
splicing the target image and other images to obtain a spliced image;
processing the spliced image by using a MobileNet-SSD deep convolution neural network model to obtain a detection frame positioned on the spliced image and a second shelter type corresponding to the detection frame;
Selecting an optimal detection frame from the detection frames by utilizing an NMS algorithm;
and under the condition that the position of the optimal detection frame on the spliced image corresponds to the target image, determining the type of the shielding object corresponding to the target camera according to the type of the second shielding object corresponding to the optimal detection frame.
In other embodiments, to reduce false alarms and eliminate the influence of extreme environments, on the basis of the method for cleaning the car camera shown in fig. 4, the method further includes the following steps:
the environmental data is acquired by the in-vehicle sensor, and in the case where the environmental data satisfies the specified condition, step S20 is executed.
In other embodiments, to improve the man-machine interaction capability, on the basis of the method for cleaning the automobile camera shown in fig. 4, the method further includes the following steps:
and sending the cleaning result of the target camera to the mobile terminal so that the mobile terminal can prompt based on the cleaning result of the target camera.
According to the cleaning method for the automobile camera, the type of the shielding object can be determined by analyzing the target image shot by the target camera, so that the working states of the cleaning device and the blowing device are controlled according to the type of the shielding object. Based on the invention, the automatic cleaning of the camera can be realized, the driving safety is improved, and the maintenance work of drivers and passengers on the camera is reduced.
The foregoing detailed description of the system and method for cleaning a camera of an automobile according to the present invention is provided, and the specific examples are used herein to explain the principles and embodiments of the present invention, and the descriptions of the above examples are only used to help understand the method and the core ideas of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
It should be noted that, in the present specification, the embodiments are all described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments may be referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
It is further noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include or include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (9)

1. An automotive camera cleaning system, the system comprising:
the cleaning device comprises an electronic control unit, a cleaning device and a purging device, wherein the electronic control unit is electrically connected with the cleaning device and the purging device respectively;
the electronic control unit is used for acquiring a target image shot by a target camera; analyzing the target image to determine the type of the shielding object corresponding to the target camera; controlling the working states of the cleaning device and the blowing device according to the type of the shielding object so as to clean the target camera;
the electronic control unit, configured to analyze the target image and determine a type of a mask corresponding to the target camera, is specifically configured to:
Calculating the global fuzziness of the target image; judging whether the global ambiguity is greater than a specified global ambiguity threshold value; if the global fuzziness is larger than the designated global fuzziness threshold, segmenting the target image to obtain sub-images; extracting a feature vector of the sub-image, and performing feature classification on the feature vector to obtain a first shelter type corresponding to the sub-image; determining a shielding object type corresponding to the target camera according to a first shielding object type corresponding to the sub-image; if the global ambiguity is not greater than the designated global ambiguity, acquiring other images shot by other cameras except the target camera; splicing the target image and the other images to obtain a spliced image; processing the spliced image by using a MobileNet-SSD deep convolution neural network model to obtain a detection frame positioned on the spliced image and a second shelter type corresponding to the detection frame; selecting an optimal detection frame from the detection frames by utilizing an NMS algorithm; and under the condition that the position of the optimal detection frame on the spliced image corresponds to the target image, determining the type of the shielding object corresponding to the target camera according to the type of the second shielding object corresponding to the optimal detection frame.
2. The system of claim 1, further comprising:
the vehicle-mounted sensor is electrically connected with the electronic control unit;
the electronic control unit is further configured to acquire environment data through the vehicle-mounted sensor, and execute the analysis of the target image to determine a type of a shielding object corresponding to the target camera when the environment data meets a specified condition.
3. The system of claim 1, further comprising:
the mobile terminal is in communication connection with the electronic control unit;
the electronic control unit is also used for sending the cleaning result of the target camera to the mobile terminal;
and the mobile terminal is used for prompting based on the cleaning result of the target camera.
4. The system of claim 1, wherein the cleaning device comprises:
the washing machine comprises a washing kettle containing washing liquid, a water pump and a first spray head, wherein the water pump is electrically connected with the electronic control unit, and the washing liquid is sprayed out of the first spray head under the action of the water pump.
5. The system of claim 4, wherein the first showerhead is a dual-use gas-liquid showerhead.
6. The system of claim 1, wherein the purge device comprises:
the air pump is electrically connected with the electronic control unit, and air is sprayed out of the second spray head under the action of the air pump.
7. The system of claim 6, wherein the second showerhead is a dual-use gas-liquid showerhead.
8. The system of claim 6, wherein the second showerhead is provided with a heater.
9. A cleaning method for an automobile camera, which is applied to the electronic control unit in the cleaning system for the automobile camera as claimed in any one of claims 1 to 8, and the method comprises the following steps:
acquiring a target image shot by a target camera;
analyzing the target image to determine the type of the shielding object corresponding to the target camera;
controlling the working states of the cleaning device and the blowing device according to the type of the shielding object so as to clean the target camera;
wherein the analyzing the target image to determine the type of the shielding object corresponding to the target camera comprises:
calculating the global fuzziness of the target image;
judging whether the global ambiguity is greater than a specified global ambiguity threshold value;
If the global fuzziness is larger than the designated global fuzziness threshold, segmenting the target image to obtain sub-images;
extracting a feature vector of the sub-image, and performing feature classification on the feature vector to obtain a first shelter type corresponding to the sub-image;
determining a shielding object type corresponding to the target camera according to a first shielding object type corresponding to the sub-image;
if the global ambiguity is not greater than the designated global ambiguity, acquiring other images shot by other cameras except the target camera;
splicing the target image and the other images to obtain a spliced image;
processing the spliced image by using a MobileNet-SSD deep convolution neural network model to obtain a detection frame positioned on the spliced image and a second shelter type corresponding to the detection frame;
selecting an optimal detection frame from the detection frames by utilizing an NMS algorithm;
and under the condition that the position of the optimal detection frame on the spliced image corresponds to the target image, determining the type of the shielding object corresponding to the target camera according to the type of the second shielding object corresponding to the optimal detection frame.
CN201910351091.6A 2019-04-28 2019-04-28 Automobile camera cleaning system and method Active CN111845647B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910351091.6A CN111845647B (en) 2019-04-28 2019-04-28 Automobile camera cleaning system and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910351091.6A CN111845647B (en) 2019-04-28 2019-04-28 Automobile camera cleaning system and method

Publications (2)

Publication Number Publication Date
CN111845647A true CN111845647A (en) 2020-10-30
CN111845647B CN111845647B (en) 2021-07-27

Family

ID=72966426

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910351091.6A Active CN111845647B (en) 2019-04-28 2019-04-28 Automobile camera cleaning system and method

Country Status (1)

Country Link
CN (1) CN111845647B (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113415257A (en) * 2021-07-02 2021-09-21 广东汇天航空航天科技有限公司 Sensor dedusting and demisting treatment method and system
CN113635864A (en) * 2021-07-27 2021-11-12 际络科技(上海)有限公司 Cleaning method and device for vehicle-mounted laser radar, electronic equipment and storage medium
CN113709374A (en) * 2021-09-01 2021-11-26 寒武纪行歌(南京)科技有限公司 Camera cleaning control method, device, equipment and system
CN114655166A (en) * 2022-04-11 2022-06-24 阿波罗智能技术(北京)有限公司 Cleaning method and device for sensor, electronic equipment and storage medium
CN114713531A (en) * 2022-04-25 2022-07-08 深圳智优停科技有限公司 Convolutional neural network training method, lens contamination type judgment method, lens wiper control method, storage medium and shooting system
WO2023029633A1 (en) * 2021-09-01 2023-03-09 寒武纪行歌(南京)科技有限公司 Camera cleaning control method and apparatus, device, and system

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005295152A (en) * 2004-03-31 2005-10-20 Canon Inc Accessory device
WO2014007153A1 (en) * 2012-07-03 2014-01-09 クラリオン株式会社 Vehicle surrounding monitoring device
WO2014017178A1 (en) * 2012-07-25 2014-01-30 ソニー株式会社 Cleaning apparatus, cleaning method, and image pickup apparatus
CN104412573A (en) * 2012-07-03 2015-03-11 歌乐株式会社 On-board device
CN106572349A (en) * 2016-11-18 2017-04-19 维沃移动通信有限公司 Camera cleanliness detection method and mobile terminal
CN107745697A (en) * 2017-11-16 2018-03-02 北京图森未来科技有限公司 A kind of auto cleaning system and method
CN107792019A (en) * 2017-09-27 2018-03-13 北京图森未来科技有限公司 A kind of camera clean method and system
CN108416337A (en) * 2018-04-28 2018-08-17 北京小米移动软件有限公司 User is reminded to clean the method and device of camera lens
CN108898592A (en) * 2018-06-22 2018-11-27 北京小米移动软件有限公司 Prompt method and device, the electronic equipment of camera lens degree of fouling
DE102017213019A1 (en) * 2017-07-28 2019-01-31 Robert Bosch Gmbh Method for controlling at least one washing device of at least one sensor arranged on an outer contour of a vehicle
CN109344688A (en) * 2018-08-07 2019-02-15 江苏大学 The automatic identifying method of people in a kind of monitor video based on convolutional neural networks

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005295152A (en) * 2004-03-31 2005-10-20 Canon Inc Accessory device
WO2014007153A1 (en) * 2012-07-03 2014-01-09 クラリオン株式会社 Vehicle surrounding monitoring device
CN104412573A (en) * 2012-07-03 2015-03-11 歌乐株式会社 On-board device
WO2014017178A1 (en) * 2012-07-25 2014-01-30 ソニー株式会社 Cleaning apparatus, cleaning method, and image pickup apparatus
CN106572349A (en) * 2016-11-18 2017-04-19 维沃移动通信有限公司 Camera cleanliness detection method and mobile terminal
DE102017213019A1 (en) * 2017-07-28 2019-01-31 Robert Bosch Gmbh Method for controlling at least one washing device of at least one sensor arranged on an outer contour of a vehicle
CN107792019A (en) * 2017-09-27 2018-03-13 北京图森未来科技有限公司 A kind of camera clean method and system
CN107745697A (en) * 2017-11-16 2018-03-02 北京图森未来科技有限公司 A kind of auto cleaning system and method
CN108416337A (en) * 2018-04-28 2018-08-17 北京小米移动软件有限公司 User is reminded to clean the method and device of camera lens
CN108898592A (en) * 2018-06-22 2018-11-27 北京小米移动软件有限公司 Prompt method and device, the electronic equipment of camera lens degree of fouling
CN109344688A (en) * 2018-08-07 2019-02-15 江苏大学 The automatic identifying method of people in a kind of monitor video based on convolutional neural networks

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113415257A (en) * 2021-07-02 2021-09-21 广东汇天航空航天科技有限公司 Sensor dedusting and demisting treatment method and system
CN113635864A (en) * 2021-07-27 2021-11-12 际络科技(上海)有限公司 Cleaning method and device for vehicle-mounted laser radar, electronic equipment and storage medium
CN113635864B (en) * 2021-07-27 2023-08-15 际络科技(上海)有限公司 Vehicle-mounted laser radar cleaning method and device, electronic equipment and storage medium
CN113709374A (en) * 2021-09-01 2021-11-26 寒武纪行歌(南京)科技有限公司 Camera cleaning control method, device, equipment and system
WO2023029633A1 (en) * 2021-09-01 2023-03-09 寒武纪行歌(南京)科技有限公司 Camera cleaning control method and apparatus, device, and system
CN114655166A (en) * 2022-04-11 2022-06-24 阿波罗智能技术(北京)有限公司 Cleaning method and device for sensor, electronic equipment and storage medium
CN114655166B (en) * 2022-04-11 2023-12-29 阿波罗智能技术(北京)有限公司 Sensor cleaning method and device, electronic equipment and storage medium
CN114713531A (en) * 2022-04-25 2022-07-08 深圳智优停科技有限公司 Convolutional neural network training method, lens contamination type judgment method, lens wiper control method, storage medium and shooting system

Also Published As

Publication number Publication date
CN111845647B (en) 2021-07-27

Similar Documents

Publication Publication Date Title
CN111845647B (en) Automobile camera cleaning system and method
CN112581440B (en) Method and device for maintaining image quality of vehicle-mounted camera and vehicle-mounted camera
US11565659B2 (en) Raindrop recognition device, vehicular control apparatus, method of training model, and trained model
CN110999273B (en) Attachment detection device and vehicle system having same
CN104412573B (en) Car-mounted device
US7860275B2 (en) Method for detecting rain on a windscreen
JP6374238B2 (en) In-vehicle device
JP4131700B2 (en) Transmission detector for window part and cleaning device for visual field area of window part
WO2018150661A1 (en) Onboard image-capture device
CN106657716A (en) Field of view clearing method and device for electronic in-car rearview mirror
CN106945637B (en) A kind of rain blowing controlled system and its method based on automobile data recorder
CN108696746A (en) Video camera misalignment determines method and system
CN106218587A (en) A kind of method and apparatus opening defrosting function
CN112109664A (en) Control method and system
CN104732489A (en) Image processing apparatus and image processing method for removing rain streaks from image data
JPH11326541A (en) Image pickup environmental assumer
JP2008137548A (en) Vehicular washer device
JP3636955B2 (en) Raindrop detection device for vehicles
JP2821041B2 (en) Image processing method
JP6058307B2 (en) Wiper control device
CN110539694B (en) Rainfall warning method and device and vehicle
US20190253584A1 (en) Detection apparatus, imaging apparatus, moveable body, and detection method
JP7334757B2 (en) Sensor cleaning system, sensor cleaning method, and vehicle
CN113581134B (en) Vehicle, cleaning method and device thereof, storage medium and vehicle-mounted controller
JPH10148618A (en) Foreign matter detecting method in rain drop detecting device

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant