WO2020220185A1 - 清洗方法、清洗控制系统、计算机可读存储介质、清洗系统、光学传感器及可移动平台 - Google Patents

清洗方法、清洗控制系统、计算机可读存储介质、清洗系统、光学传感器及可移动平台 Download PDF

Info

Publication number
WO2020220185A1
WO2020220185A1 PCT/CN2019/084952 CN2019084952W WO2020220185A1 WO 2020220185 A1 WO2020220185 A1 WO 2020220185A1 CN 2019084952 W CN2019084952 W CN 2019084952W WO 2020220185 A1 WO2020220185 A1 WO 2020220185A1
Authority
WO
WIPO (PCT)
Prior art keywords
image data
dirty
cleaning
reaches
image
Prior art date
Application number
PCT/CN2019/084952
Other languages
English (en)
French (fr)
Inventor
王婷
黄祎伦
孙毅峰
Original Assignee
深圳市大疆创新科技有限公司
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 深圳市大疆创新科技有限公司 filed Critical 深圳市大疆创新科技有限公司
Priority to PCT/CN2019/084952 priority Critical patent/WO2020220185A1/zh
Priority to CN201980008957.1A priority patent/CN111684487B/zh
Publication of WO2020220185A1 publication Critical patent/WO2020220185A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G06T7/001—Industrial image inspection using an image reference approach
    • 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
    • B60S1/60—Cleaning windscreens, windows or optical devices specially adapted for cleaning other parts or devices than front windows or windscreens for signalling devices, e.g. reflectors
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/01—Arrangements or apparatus for facilitating the optical investigation
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/94—Investigating contamination, e.g. dust
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/50—Constructional details
    • H04N23/55—Optical parts specially adapted for electronic image sensors; Mounting thereof
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30168—Image quality inspection

Definitions

  • This application relates to the cleaning field, and in particular to a cleaning method, a cleaning control system, a computer-readable storage medium, a cleaning system, an optical sensor, and a movable platform.
  • Optical sensors include sensors that measure based on optical principles. They can be used in products such as industry, automobiles, electronics, and retail automation, and because optical sensors have the characteristics of non-contact, fast response, and reliable performance, they are especially obtained in industrial automation devices and robots. widely used. In use, the light-transmitting surface of the optical sensor will inevitably be dirty, so it is necessary to clean the light-transmitting surface of the optical sensor.
  • This application provides an improved cleaning method, cleaning control system, computer-readable storage medium, cleaning system, optical sensor, and movable platform.
  • a cleaning method is provided, which is applied to an optical sensor, the optical sensor is provided with a light-transmitting surface located on the outer side, and the optical sensor is used to obtain an external image through the light-transmitting surface, and The corresponding image data is generated according to the acquired external image;
  • the cleaning method includes: acquiring the image data generated by the optical sensor; comparing the image data with image reference data to determine whether the image data includes Dirty image data; if the image data includes the dirty image data, determine whether the dirty image data reaches the dirty degree threshold; and if it is determined that the dirty image data reaches the dirty degree threshold, Then the cleaning mechanism is controlled to clean the transparent surface.
  • a cleaning control system which is characterized in that it includes one or more processors for implementing the cleaning method.
  • a computer-readable storage medium which is characterized in that a program is stored thereon, and when the program is executed by a processor, a cleaning method is implemented.
  • a cleaning system including: a cleaning mechanism; and a cleaning control system.
  • an optical sensor including: a light-transmitting surface; an image processing module for acquiring an external image passing through the light-transmitting surface, and generating a corresponding image according to the acquired external image Image data; and cleaning system.
  • a movable platform including: a body; a power system provided on the body for providing power to the movable platform; an optical sensor provided on the body; And cleaning system.
  • This application can effectively determine whether dirt affects the imaging quality, and clean the translucent surface in time when the dirt affects the imaging quality.
  • Fig. 1 shows a flowchart of an embodiment of the cleaning method of the present application.
  • Fig. 2 shows a schematic block diagram of an embodiment of the cleaning system of the present application.
  • FIG. 3 is a perspective schematic diagram of an embodiment of the cleaning execution device of the optical sensor and the cleaning mechanism of the present application.
  • Fig. 4 is a three-dimensional exploded schematic diagram of the optical sensor and the cleaning execution device shown in Fig. 3.
  • Fig. 5 shows a block diagram of an embodiment of the cleaning control system of the cleaning system of the present application.
  • Fig. 6 is a block diagram of a module of an embodiment of the optical sensor of the present application.
  • FIG. 7 is a schematic diagram of an embodiment of the mobile platform of this application.
  • the cleaning method of the embodiment of the present application is applied to an optical sensor.
  • the optical sensor is provided with a light-transmitting surface on the outer side.
  • the optical sensor is used to obtain an external image through the light-transmitting surface and generate corresponding image data according to the obtained external image.
  • Optical sensors may include various forms of radar, such as lidar, millimeter wave radar, etc., monocular cameras, binocular cameras, infrared sensors, ultraviolet sensors, and the like.
  • the optical sensor can be applied to movable platforms such as ordinary vehicles, autonomous vehicles, and unmanned aerial vehicles.
  • the cleaning method includes: acquiring image data generated by the optical sensor; comparing the image data with image reference data to determine whether the image data includes dirty image data; if the image data includes dirty image data, determining whether the dirty image data is dirty Stain level threshold; and if it is determined that the stained image data reaches the stain level threshold, control the cleaning mechanism to clean the light-transmitting surface.
  • the transparent surface can be automatically cleaned without manual intervention, the cleaning process is simple and efficient, and does not affect other modules. Especially when the number of optical sensors is large and/or the installation position is hidden, this cleaning method is particularly convenient.
  • the cleaning method of the embodiment of the present application includes if the image data includes dirty image data, determining whether the dirty image data reaches the dirty degree threshold, and if the dirty image data reaches the dirty degree threshold, cleaning the light-transmitting surface.
  • the stained image data reaches the stain level threshold, it means that the stain affects the image quality, and the light-transmitting surface needs to be cleaned. This can effectively determine whether the stain affects the image quality.
  • the stain affects the image quality, clean it in time. Glossy surface to ensure image quality, and can avoid or reduce the waste of resources such as cleaning media by cleaning the translucent surface when the degree of contamination does not affect the image quality.
  • the cleaning control system of the embodiment of the present application includes one or more processors for implementing the cleaning method.
  • the computer-readable storage medium of the embodiment of the present application has a program stored thereon, and when the program is executed by a processor, the cleaning method is implemented.
  • the cleaning system of the embodiment of the present application includes a cleaning mechanism and a cleaning control system.
  • the optical sensor of the embodiment of the present application includes a light-transmitting surface, an image processing module, and a cleaning system.
  • the image processing module is used to obtain an external image passing through the transparent surface, and generate corresponding image data according to the obtained external image.
  • the movable platform of the embodiment of the present application includes a body, a power system, an optical sensor, and a cleaning system.
  • the power system is arranged in the body to provide power for the movable platform.
  • the optical sensor is arranged in the body.
  • FIG. 1 shows a flowchart of an embodiment of a cleaning method 100.
  • the cleaning method 100 is applied to an optical sensor.
  • the optical sensor is provided with a light-transmitting surface located on the outer side.
  • the optical sensor is used to acquire an external image through the light-transmitting surface and generate corresponding image data according to the acquired external image.
  • the optical sensor may include a lens, and the lens includes a light-transmitting surface. External light can pass through the light-transmitting surface and enter the optical sensor, which can convert the light signal into an electrical signal.
  • the optical sensor may perform image processing on the acquired external image through a vision algorithm, and convert the external image into image data.
  • optical sensors can be used for cameras, video cameras, cameras, and the like.
  • the cleaning method 100 includes steps 101-104.
  • step 101 image data generated by the optical sensor is acquired.
  • the image data may include pixel values.
  • step 102 the image data is compared with the image reference data to determine whether the image data includes dirty image data.
  • the image reference data may be pre-stored in the database.
  • different external images acquired by the optical sensor can be collected, and corresponding image data can be generated, and a large amount of image data can be analyzed and processed and recorded in the database as at least part of the image reference data.
  • the image reference data includes at least one of stained image reference data and stain-free image reference data.
  • multiple unstained external images can be collected, and corresponding unstained image data can be generated, and a large number of unstained image data can be processed and recorded in the database as the unstained image data.
  • Stain image reference data can be collected, and corresponding image data can be generated, and a large amount of image data can be processed and recorded in a database as reference data for soiled images.
  • the image reference data in the database can be updated during the execution of the method.
  • the image quality is closely related to the pixel value.
  • the pixel value can directly reflect the image quality.
  • the image reference data may include stain-free image reference data and/or other data different from the stained image reference data. Compare the pixel value of the dirty image data with the pixel value of the image reference data, if not equal, determine that the image data is dirty image data. In this way, it can be simply and accurately determined whether the image data is dirty image data.
  • the image reference data may include a plurality of discrete data, and when the pixel values of the image data and each discrete data are not equal, the image data is determined to be dirty image data.
  • the image reference data may include a data range. The pixel values of the data in the data range of the image data and the image reference data are not equal, that is, when the image data is not in the data range, it is determined that the image data is dirty image data.
  • the image reference data may include stain-free image reference data and/or other data different from the stained image reference data. In this way, a certain error margin is allowed, especially when the amount of image reference data is not large enough, it can be more accurately determined whether the image data is dirty image data.
  • the image reference data may include a plurality of discrete data. When the pixel value difference between the image data and each discrete data reaches the difference threshold, it is determined that the image data is dirty image data.
  • the image reference data may include a data range. When the difference between the end pixel values of the data range of the image data and the image reference data reaches the difference threshold, it is determined that the image data is dirty image data.
  • the dirty image reference data is different from the dirty image reference data and can be used to distinguish between dirty images and non-stained images.
  • the image data when the pixel values of the image data and the stain-free image reference data are not equal, indicating that the image data does not match the stain-free image reference data, it is determined that the image data is dirty image data. In another embodiment, when the data difference between the pixel values of the image data and the stain-free image reference data reaches the difference threshold, indicating that the image data does not match the stain-free image reference data, it is determined that the image data is dirty Image data. See above for detailed description. If at least one image data of the plurality of image data generated by the optical sensor does not match the stain-free image reference data, it is determined that the plurality of image data generated by the optical sensor includes dirty image data.
  • the image data is compared with the dirty image reference data to determine whether the image data includes dirty image data. If the image data matches the dirty image reference data, it is determined that the image data is dirty image data; otherwise, it is determined that the image data is dirty image data. By comparing with the dirty image reference data, it can be directly determined whether the image data is dirty image data. In one embodiment, when the pixel values of the image data and the dirty image reference data are not equal, indicating that the image data does not match the dirty image reference data, the image data is determined to be non-stained image data, otherwise the image data is determined to be Dirty image data. In an embodiment, the dirty image reference data may include a plurality of discrete data.
  • the image data is determined to be dirty-free image data, otherwise it is dirty.
  • the dirty image reference data may include a data range. The pixel values of the data in the data range of the image data and the dirty image reference data are not equal, that is, when the image data is not in the data range, it is determined that the image data is non-stained image data. When the image data is within the data range of the dirty image reference data, it is determined that the image data is dirty image data.
  • the image reference data may include a plurality of discrete data. When the difference between the pixel value of the image data and each discrete data reaches the dirty data difference threshold, it is determined that the image data is a dirty image Data, otherwise it is dirty image data. In another embodiment, the image reference data may include a data range.
  • the image data When the difference between the end pixel values of the data range of the image data and the image reference data reaches the dirty data difference threshold, it is considered that the image data does not match the dirty image reference data, and the image data is determined to be dirty image data, otherwise it is dirty Dirty image data.
  • At least one image data matches the dirty image reference data, it is determined that the plurality of image data generated by the optical sensor includes the dirty image data. If all the image data does not match the dirty image reference data, it can be determined that the multiple image data generated by the optical sensor does not include the dirty image data.
  • the image data is compared with the dirty image reference data and the dirty image reference data respectively. If the image data matches the dirty image reference data and does not match the dirty image reference data, it is determined that the image data is dirty image data. If the image data matches the dirty image reference data and does not match the dirty image reference data, it is determined that the image data is dirty image data.
  • the method described above can be referred to to determine whether the image data matches the stained image reference data, and whether the image data matches the stain-free image reference data.
  • the degree of similarity between the image data and the stain-free image reference data can be determined according to the degree of similarity between the image data and the stain-free image reference data.
  • the similarity of the image reference data determines that the image data matches the image reference data with higher similarity. If the image data has a high degree of similarity with the stain-free image reference data, the image data is determined to be stain-free image data; otherwise, the image data is determined to be stained image data.
  • the image data includes the data range, and the "approximation" may include the smallest data difference between the end pixel values of the image data and the image reference data, the smallest data difference between the image data and the dirty image reference data, and the image data The smaller of the smallest data difference with the non-stained image reference data indicates a higher degree of similarity.
  • the image data includes discrete data, and the "approximation degree" may include an average value of data differences between pixel values of the discrete data.
  • the image data may be determined to be stain-free image data or stained image data according to other factors. After it is determined that the image data is dirty image data or dirty image data, the corresponding image reference data can be updated.
  • the image data is compared with the image reference data in real time, and it is determined in a timely manner whether the image data includes dirty image data. This helps to timely judge whether the optical sensor is dirty, and improves the efficiency of judging and processing the dirt.
  • step 103 if the image data includes stained image data, it is determined whether the stained image data reaches the stain level threshold.
  • the dirty image data is filtered out from the image data, and based on the filtered dirty image data, it is determined whether the dirty image data reaches the dirt degree threshold.
  • the image data is selected, and all the dirty image data in the generated multiple image data are selected, so that one or more dirty image data can be obtained.
  • the filtered stained image data reaches the stain level threshold.
  • all dirty image data can be filtered out from all image data.
  • the filtered dirty image data can be used in subsequent steps, reducing the amount of data processing in subsequent steps and increasing the processing speed.
  • a large number of reliability tests can be used to classify the degree of dirtiness. Different degree of dirtiness corresponds to different image data, corresponding to different image data ranges, and the dirtiness degree threshold can be determined according to the range of image data. In one embodiment, the minimum value of the image data range corresponding to the degree of dirt that needs cleaning may be set as the dirt degree threshold. It can be determined whether the stained image data reaches the stain level threshold by comparing the stained image data with the stain level threshold.
  • the stained image data is compared with the image reference data to determine whether the stained image data reaches the stain level threshold.
  • the image reference data may include multiple pieces of data, and the dirty image data can be compared with multiple pieces of data to more accurately determine whether the dirty image data reaches the stain level threshold.
  • the image reference data may include stained image reference data and/or stain-free image reference data, and stained image data may be compared with stained image reference data and/or stain-free image reference data to determine stained image data Whether the dirt level threshold is reached.
  • the image reference data includes dirty image reference data.
  • the stained image data is compared with the stained image reference data to determine whether the stained image data reaches the stain level threshold. In this way, it can be directly and simply determined whether the dirty image data reaches the dirty degree threshold.
  • the image reference data may include stained image reference data representing different degrees of stains.
  • the stained image data is compared with multiple stained image reference data corresponding to different degrees of stains to determine the stain reached by the current translucent surface. Degree of pollution.
  • the first difference threshold may be stored in the database in advance. In one embodiment, the first difference threshold may be updated during method execution.
  • the first difference threshold if the number of the difference between the dirty image data and the dirty image reference data reaches the first difference threshold reaches the first number threshold, it is determined that the dirty image data reaches the dirty degree threshold.
  • the number of the first difference threshold reaches a certain number, it means that the dirt affects the image quality to the extent that it needs to be cleaned. If the number of the first difference threshold is small, it means that the dirt has a low impact and does not need to be cleaned. Can accurately determine whether the dirt affects the image quality.
  • the first number threshold can be stored in the database in advance. In one embodiment, the number threshold can be updated during method execution.
  • the first number threshold is the total number of dirty image data. In another embodiment, the first number threshold may be less than the total number of dirty image data. For example, the first number threshold can exceed half of the total number of dirty image data, so that in more than half of the dirty image data, the difference between each dirty image data and each dirty image reference data reaches the first A difference threshold is used to determine that the stained image data reaches the stain level threshold.
  • the first number threshold can be set according to actual applications.
  • the difference between the sum of the plurality of dirty image data and the sum of the plurality of dirty image reference data reaches the second difference threshold, it is determined that the dirty image data reaches the dirt degree threshold.
  • the difference between the sum of all the filtered dirty image data and the sum of the multiple dirty image reference data reaches the second difference threshold, and it is determined that the dirty image data reaches the dirt degree threshold. Therefore, when multiple dirty image data each does not reach the corresponding dirty degree threshold, but the total dirt reaches the dirty degree threshold, it means that the overall dirt affects the image quality and needs to be cleaned, so that the dirt can be cleaned in time .
  • the image reference data includes stain-free image reference data.
  • the stained image data is compared with non-stained image reference data to determine whether the stained image data reaches the stain level threshold.
  • Contamination-free image reference data is easier to collect and more accurate, which simplifies the data collection process and more accurately determines whether the dirty image data reaches the contamination level threshold.
  • the third difference threshold if the difference between the plurality of dirty image data and the non-stained image data respectively reaches the third difference threshold, it is determined that the dirty image data reaches the dirty degree threshold. If the difference between multiple stained image data and non-stained image data respectively reaches the third difference threshold, it means that the stained image data and the non-stained image data are quite different, and the degree of staining reaches the level that requires cleaning. Cleaning, so that cleaning can be performed in time, and the waste caused by cleaning without cleaning can be avoided or reduced.
  • the third difference threshold may be stored in the database in advance. In one embodiment, the third difference threshold may be updated during method execution.
  • the second number threshold can be stored in the database in advance. In one embodiment, the two number threshold can be updated during method execution. In one embodiment, if the difference between each dirty image data and each non-stained image reference data reaches the third difference threshold, it is determined that the dirty image data reaches the dirty degree threshold.
  • the second number threshold is the total number of dirty image data. In another embodiment, the second number threshold may be less than the total number of dirty image data.
  • the difference between the sum of the plurality of dirty image data and the sum of the plurality of non-stained image reference data reaches the fourth difference threshold, it is determined that the dirty image data has reached the dirt level threshold.
  • the difference between the sum of all filtered dirty image data and the sum of multiple dirty image reference data reaches the third difference threshold, and it is determined that the dirty image data reaches the dirt degree threshold. Therefore, when multiple dirty image data each does not reach the corresponding dirty degree threshold, but the total dirt reaches the dirty degree threshold, it means that the overall dirt affects the image quality and needs to be cleaned, so that the dirt can be cleaned in time .
  • the cleaning method 100 includes determining the stain type according to the stain image data.
  • Different stain types can correspond to different stain image data. Dirt types indicate different types of dirt, such as leaves, soil, liquid, ash layer, etc.
  • the pixel value of the dirty image data is compared with the pixel value of the image reference data to determine the type of stain.
  • the image reference data may include image reference data corresponding to different stain types. External images can be acquired under different working conditions, the generated image data can be collected, processed and analyzed, and the type of contamination and corresponding image reference data can be determined, which can be correspondingly stored in the database.
  • the stain type is determined by comparing the stain image data with the reference stain image data corresponding to multiple stain types. When the difference between the dirty image data and the dirty image reference data corresponding to one of the dirty types is less than the type difference threshold, it is determined that the dirty image data belongs to the dirty type.
  • the filtered stained image data can be classified into the corresponding stain type. It is further determined whether the stained image data corresponding to the stain type reaches the stain level threshold corresponding to the stain type.
  • Different dirt types can correspond to different dirt degree thresholds. For stains with the same number of pixels, different types of stains have different effects on image quality. For example, rain has less impact on image quality than soil.
  • the dirty image reference data corresponding to the type is used to determine whether the dirty image data reaches the dirty degree threshold, which can more effectively determine whether the dirty affects the image quality.
  • the stained image data corresponding to the stain type is compared with the stain level threshold corresponding to the stain type to determine whether the stain image data corresponding to the stain type reaches the corresponding stain level threshold.
  • the dirty image data corresponding to the dirty type is compared with the dirty image reference data corresponding to the dirty type to determine whether the dirty image data corresponding to the dirty type reaches the corresponding dirty degree threshold. . In one embodiment, if the difference between the dirty image data corresponding to the dirty type and the dirty image reference data corresponding to the dirty type reaches the fifth difference threshold respectively, the dirty image data corresponding to the dirty type is determined The corresponding dirt level threshold is reached. In an embodiment, if the difference between the dirty image data corresponding to the dirty type and the dirty image reference data reaches the fifth difference threshold and the number reaches the third number threshold, it is determined that the dirty image data reaches the dirty level Threshold.
  • the difference between the sum of the plurality of dirty image data corresponding to the dirty type and the sum of the plurality of dirty image reference data corresponding to the dirty type reaches the sixth difference threshold, it is determined that the dirty The stained image data reaches the stain level threshold.
  • the stain level threshold corresponding to the stain types if there are multiple types of stains, it is determined whether the stained image data corresponding to the multiple types of stains reaches the stain level threshold corresponding to the stain types.
  • the above-mentioned method can be used to separately determine whether the dirty image data corresponding to each type of dirty reaches the corresponding dirty degree threshold. Determining whether the stain level threshold is reached for different types of stains can more accurately determine whether the stains affect the image quality and need to be cleaned.
  • the stain degree threshold it is determined whether the sum of stain image data corresponding to the multiple stain types reaches the stain degree threshold.
  • different stain types have different effects on image quality, and the influence coefficient can be set according to the degree of the impact of different stain types on image quality.
  • the sum of stained image data may be the sum of stained image data corresponding to multiple stain types multiplied by corresponding coefficients. Therefore, the stained image data corresponding to various stain types have not reached the corresponding staining degree threshold, but the sum of the stained image data has reached the staining degree threshold, indicating that the overall staining affects the image quality and needs to be cleaned. In this way, the dirt can be cleaned in time.
  • step 104 if it is determined that the stained image data reaches the stain level threshold, the cleaning mechanism is controlled to clean the light-transmitting surface.
  • the stained image data reaches the stain level threshold, indicating that the stain affects the image quality, and the translucent surface needs to be cleaned.
  • the cleaning mechanism is controlled to clean the light-transmitting surface to realize the cleaning of the light-transmitting surface. If the stained image data does not reach the stain level threshold, the transparent surface is not cleaned.
  • the cleaning mechanism is controlled to clean the light-transmitting surface.
  • Different stain types have different effects on the image.
  • the transparent surface is cleaned, so that the transparent surface can be cleaned more timely. And it can more effectively avoid or reduce the waste of resources such as cleaning media by cleaning the light-transmitting surface when the degree of contamination does not affect the image quality.
  • the cleaning mechanism is controlled to clean the transparent surface. In one embodiment, if the stained image data corresponding to at least one stain type reaches the stain level threshold corresponding to the stain type, the cleaning mechanism is controlled to clean the light-transmitting surface. In another embodiment, if the stained image data corresponding to multiple stain types all reach the stain level threshold corresponding to the stain type, the cleaning mechanism is controlled to clean the light-transmitting surface.
  • the stain level threshold is used to control the cleaning mechanism to clean the transparent surface.
  • the light-transmitting surface can be automatically cleaned without manual intervention, the cleaning process is simple and efficient, and does not affect other modules. Especially when the number of optical sensors is large and/or the installation location is hidden, the cleaning method 100 is particularly convenient. Moreover, in the cleaning method of the embodiment of the present application, it is determined whether the dirty image data reaches the dirty degree threshold, and if the dirty image data reaches the dirty degree threshold, the light-transmitting surface is cleaned.
  • cleaning the light-transmitting surface can clean the light-transmitting surface in time to ensure the image quality, and at the same time, it can avoid or reduce when the degree of dirt does not affect the image quality.
  • Cleaning the translucent surface causes waste of resources such as cleaning media.
  • FIG. 2 shows a schematic block diagram of an embodiment of a cleaning system 500.
  • the cleaning system 500 includes a cleaning mechanism 200 and a cleaning control system 300. 1 and 2, in one embodiment, the step 104 of controlling the cleaning mechanism to clean the light-transmitting surface includes: controlling the fluid conveying device 201 of the cleaning mechanism 200 to push the cleaning medium into the pipeline 202 so that the cleaning medium passes through the pipeline 202 Convey to the nozzle 2031-203N, and spray through the nozzle 2031-203N to clean the transparent surface.
  • the cleaning medium includes liquid and/or gas. In one embodiment, the cleaning medium includes at least one of the following: water, glass water, and air.
  • the cleaning medium includes liquid
  • the fluid conveying device 201 includes a pump for pumping the liquid into the pipeline 202.
  • the cleaning medium includes gas
  • the fluid delivery device 201 may include a compressor or a blower.
  • the fluid delivery device 201 can be connected to the cleaning control system 300, and the cleaning control system 300 can execute the cleaning method 100 and can control the fluid delivery device 201.
  • the fluid delivery device 201 is controlled to work, and after the cleaning is completed, the fluid delivery device 201 is controlled to stop working.
  • the step 104 of controlling the cleaning mechanism to clean the light-transmitting surface includes: controlling the switching device 2041-204N provided on the pipeline 202 to open to allow the cleaning medium to flow through.
  • the switching device 2041-204N can be connected to the cleaning control system 300, and the cleaning control system 300 can control the opening and closing of the switching device 2041-204N.
  • the control switch device 2041-204N is opened, and after the cleaning is completed, the control switch device 2041-204N is closed.
  • the switching devices 2041-204N may include solenoid valves.
  • the cleaning mechanism 200 includes at least one pipeline 2021-202N connecting the fluid delivery device 201, at least one nozzle 2031-203N, and connecting the fluid delivery device 201 and the nozzle 2031-203N.
  • At least one pipeline 2021-202N is equipped with a switch device 2041-204N, which is used to control the opening and closing of the pipeline, so as to control the cleaning medium to flow through the corresponding pipeline 2021-202N and spray from the corresponding nozzle 2031-203N , To clean the corresponding translucent surface.
  • At least one nozzle 2031-203N corresponds to the light-transmitting surface of the plurality of optical sensors 401-40N.
  • the switch device 2041-204N on the corresponding pipeline 2021-202N is controlled to open.
  • Multiple pipelines 2021-202N are respectively provided with switch devices 2041-204N that control the on and off of the pipelines.
  • the corresponding switch devices 2041-204N are controlled Open, so that the cleaning medium can flow through the corresponding pipeline 2021-202N and sprayed from the corresponding nozzle 2031-203N to clean the corresponding light-transmitting surface.
  • the switch device 2041-204N corresponding to the transparent surface that does not need to be cleaned can be kept closed. In this way, the light-transmitting surface of one or more of the multiple optical sensors is automatically cleaned.
  • FIG. 3 is a perspective view of an embodiment of the cleaning execution device 205 of the optical sensor 400 and the cleaning mechanism 200.
  • FIG. 4 shows a three-dimensional exploded schematic diagram of an embodiment of the optical sensor 400 and the cleaning execution device 205.
  • the optical sensor 400 includes a light-transmitting surface 401 provided on the outside. In the illustrated embodiment, the optical sensor 400 includes a lens. The optical sensor 400 can be assembled in the cleaning execution device 205.
  • the cleaning execution device 205 includes a fixed housing 206, a nozzle 203 assembled on the fixed housing 206, and a pipe joint 207 communicating with the nozzle 203.
  • the optical sensor 400 can be fixedly assembled to the fixed housing 206.
  • the fixed housing 206 is provided with a mounting hole 208, the optical sensor 400 is inserted into the mounting hole 208, and the light-transmitting surface 401 is exposed from the mounting hole 208.
  • the nozzle 203 is assembled on one side of the mounting hole 208, and the nozzle 203 is arranged corresponding to the transparent surface 401, and the transparent surface 401 is within the spray range of the nozzle 203.
  • the cleaning medium sprayed from the nozzle 203 can be sprayed onto the transparent surface 401 to clean the transparent surface 401.
  • the fixed housing 206 is formed with a nozzle fixing hole 209 outside the mounting hole 208, and the nozzle 203 is fixedly installed in the nozzle fixing hole 209.
  • the nozzle 203 includes an ejection port 210 from which the cleaning medium is ejected.
  • the ejection outlet 210 faces the mounting hole 208, and can be inclined from one side of the light-transmitting surface 401 to the light-transmitting surface 401, so that the sprayed cleaning medium can be sprayed onto the light-transmitting surface 401.
  • the ejection outlet 210 is a fan-shaped opening, so that the cleaning medium is ejected in a fan shape, and the ejection force is stronger and the coverage area is wider, so that the transparent surface 401 can be effectively cleaned.
  • the pipe joint 207 can connect the nozzle 203 and the pipe (not shown).
  • the pipe joint 207 is fixedly assembled to the fixed housing 206.
  • the fixed housing 206 is provided with a channel 211 communicating with the nozzle 203, and the pipe joint 207 can communicate with the channel 211 and further communicate with the nozzle 203.
  • the channel 211 communicates with the nozzle fixing hole 209.
  • the pipe joint 207 may be directly connected to the nozzle 203.
  • FIG. 5 shows a block diagram of an embodiment of the cleaning control system 300.
  • the cleaning control system 300 includes one or more processors 301 for implementing the cleaning method 100.
  • the processor 301 of the cleaning control system 300 can implement the cleaning method described above. In some embodiments,
  • the cleaning control system 300 may include a computer-readable storage medium 304, which may store a program that can be called by the processor 301, and may include a non-volatile storage medium.
  • the cleaning control system 300 may include a memory 303 and an interface 302. In some embodiments, the cleaning control system 300 may also include other hardware according to actual applications.
  • the computer-readable storage medium 304 of the embodiment of the present application has a program stored thereon, and when the program is executed by the processor 301, the cleaning method 100 is implemented.
  • This application can take the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program codes.
  • Computer-readable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be achieved by any method or technology.
  • the information can be computer-readable instructions, data structures, program modules, or other data.
  • Examples of computer-readable storage media include, but are not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only Memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical storage , Magnetic cassette tape, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by computing devices.
  • PRAM phase change memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • RAM random access memory
  • ROM read-only Memory
  • EEPROM electrically erasable programmable read-only memory
  • flash memory or other memory technologies
  • CD-ROM compact disc
  • DVD digital versatile disc
  • Magnetic cassette tape magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by computing devices.
  • FIG. 6 shows a block diagram of an embodiment of the optical sensor 400.
  • the optical sensor 400 includes a light-transmitting surface 401 (as shown in FIG. 3 ), an image processing module 402 and a cleaning system 500.
  • the image processing module 402 is configured to obtain an external image passing through the light-transmitting surface 401, and generate corresponding image data according to the obtained external image.
  • the image processing module 402 can generate image data through a vision algorithm.
  • the cleaning system 500 may be the cleaning system of the above-mentioned embodiment.
  • the cleaning control system 300 of the cleaning system 500 and the image processing module 402 can be installed on the same control circuit board.
  • FIG. 7 shows a schematic diagram of an embodiment of the movable platform 700.
  • the movable platform 700 may include a mobile car, an unmanned aerial vehicle, a car, a robot, or other movable devices.
  • the movable platform 700 includes a body 701, a power system 702, an optical sensor 703, and a cleaning system 500.
  • the power system 702 is installed in the body and used to provide power to the movable platform 700.
  • the power system 702 may include an electric motor.
  • the optical sensor 703 is provided in the body 701 and can be used to capture images.
  • the movable platform 700 can use the image taken by the optical sensor 703 for distance measurement, tracking and the like.
  • the cleaning system 500 may be the cleaning system of the above-mentioned embodiment.
  • the cleaning system 500 can clean the transparent surface of the optical sensor 703.
  • the cleaning control system 300 of the cleaning system 500 may be installed outside the movable platform, such as in a control center.
  • Control centers such as supercomputing centers, computing platforms, etc.
  • the optical sensor 703 generates image data and sends it to the cleaning control system 300.
  • the movable platform 700 further includes a GPS device for obtaining location information of the movable platform 700. If it is determined that the dirty image data reaches the dirty degree threshold, the cleaning mechanism is controlled to clean the light-transmitting surface, and the dirty image data and current position information are stored. For example, the current location information of the movable platform 700 is acquired through the GPS device, and when it is determined that the stained image data reaches the stain level threshold, the stained image data and the current location information of the movable platform 700 are stored at the same time. It is helpful for the user to reasonably plan the travel route according to the relationship between the position of the movable platform 700 and the optical sensor 703.
  • each part of this application can be implemented by hardware, software or a combination thereof.
  • multiple steps or methods can be implemented by software or hardware stored in a memory and executed by a suitable instruction execution system.
  • a suitable instruction execution system For example, if it is implemented by hardware, it can be implemented by any one of the following technologies or a combination of them: discrete logic circuits with logic gates for realizing logic functions on data signals, and dedicated logic gates with suitable combinational logic gates Integrated circuit, programmable gate array (PGA), field programmable gate array (FPGA), etc.
  • a person of ordinary skill in the art can understand that all or part of the steps carried in the implementation method described above can be completed by a program instructing relevant hardware.
  • the program can be stored in a computer-readable storage medium. When it includes one of the steps of the method embodiment or a combination thereof.

Abstract

一种清洗方法、清洗控制系统、计算机可读存储介质、清洗系统、光学传感器及可移动平台。清洗方法应用于光学传感器，光学传感器设有位于外侧的透光面，光学传感器用于通过透光面获取外部图像，并根据获取的外部图像生成相应的图像数据。清洗方法包括：获取光学传感器生成的图像数据（101）；将图像数据与图像参考数据进行比较，确定图像数据是否包括脏污图像数据（102）；若图像数据包括脏污图像数据，则确定脏污图像数据是否达到脏污程度阈值（103）；及若确定脏污图像数据达到脏污程度阈值，则控制清洗机构清洗透光面（104）。

Description

清洗方法、清洗控制系统、计算机可读存储介质、清洗系统、光学传感器及可移动平台 技术领域
本申请涉及清洗领域，尤其涉及一种清洗方法、清洗控制系统、计算机可读存储介质、清洗系统、光学传感器及可移动平台。
背景技术
光学传感器包括依据光学原理进行测量的传感器，可用于工业、汽车、电子和零售自动化等产品中，而且由于光学传感器具有非接触、响应快、性能可靠等特点，尤其在工业自动化装置和机器人中获得广泛应用。在使用中，光学传感器的透光面不可避免的会出现脏污的状况，因此需要对光学传感器的透光面进行清洗。
发明内容
本申请提供改进的清洗方法、清洗控制系统、计算机可读存储介质、清洗系统、光学传感器及可移动平台。
根据本申请实施例的一个方面，提供一种清洗方法，应用于光学传感器，所述光学传感器设有位于外侧的透光面，所述光学传感器用于通过所述透光面获取外部图像，并根据获取的所述外部图像生成相应的图像数据；所述清洗方法包括：获取所述光学传感器生成的所述图像数据；将所述图像数据与图像参考数据进行比较，确定所述图像数据是否包括脏污图像数据；若所述图像数据包括所述脏污图像数据，则确定所述脏污图像数 据是否达到脏污程度阈值；及若确定所述脏污图像数据达到所述脏污程度阈值，则控制清洗机构清洗所述透光面。
根据本申请实施例的另一个方面，提供一种清洗控制系统，其特征在于，包括一个或多个处理器，用于实现清洗方法。
根据本申请实施例的另一个方面，提供一种计算机可读存储介质，其特征在于，其上存储有程序，该程序被处理器执行时，实现清洗方法。
根据本申请实施例的另一个方面，提供一种清洗系统，包括：清洗机构；及清洗控制系统。
根据本申请实施例的另一个方面，提供一种光学传感器，包括：透光面；图像处理模块，用于获取透过所述透光面的外部图像，根据获取的所述外部图像生成相应的图像数据；及清洗系统。
根据本申请实施例的另一个方面，提供一种可移动平台，包括：机体；动力系统，设于所述机体，用于为所述可移动平台提供动力；光学传感器，设于所述机体；及清洗系统。
本申请可以有效确定脏污是否影响成像质量，在脏污影响成像质量时，及时地清洗透光面。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。
图1所示为本申请清洗方法的一个实施例的流程图。
图2所示为本申请清洗系统的一个实施例的原理框图。
图3所示为本申请光学传感器和清洁机构的清洁执行装置的一个实施例的立体示意图。
图4所示为图3所示的光学传感器和清洁执行装置的立体分解示意图。
图5所示为本申请清洗系统的清洗控制系统的一个实施例的模块框图。
图6所示为本申请光学传感器的一个实施例的模块框图。
图7所示为本申请可移动平台的一个实施例的示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
这里将详细地对示例性实施例进行说明，其示例表示在附图中。下面的描述涉及附图时，除非另有表示，不同附图中的相同数字表示相同或相似的要素。以下示例性实施例中所描述的实施方式并不代表与本申请相一致的所有实施方式。相反，它们仅是与如所附权利要求书中所详述的、本申请的一些方面相一致的装置和方法的例子。
在本申请使用的术语是仅仅出于描述特定实施例的目的，而非旨在限制本申请。在本申请和所附权利要求书中所使用的单数形式的“一种”、“所述”和“该”也旨在包括多数形式，除非上下文清楚地表示其他含义。还应当理解，本文中使用的术语“和/或”是指并包含一个或多个相关联的列出 项目的任何或所有可能组合。除非另行指出，“前部”、“后部”、“下部”和/或“上部”等类似词语只是为了便于说明，而并非限于一个位置或者一种空间定向。“连接”或者“相连”等类似的词语并非限定于物理的或者机械的连接，而且可以包括电性的连接，不管是直接的还是间接的。“多个”或者“若干”等类似词语表示两个及两个以上。
本申请实施例的清洗方法，应用于光学传感器，光学传感器设有位于外侧的透光面，光学传感器用于通过透光面获取外部图像，并根据获取的外部图像生成相应的图像数据。光学传感器可以包括各种形式的雷达，例如激光雷达、毫米波雷达等，单目摄像头、双目摄像头、红外传感器、紫外传感器等。该光学传感器可应用于普通车辆、自动驾驶车辆、无人飞行器等可移动平台。
清洗方法包括：获取光学传感器生成的图像数据；将图像数据与图像参考数据进行比较，确定图像数据是否包括脏污图像数据；若图像数据包括脏污图像数据，则确定脏污图像数据是否达到脏污程度阈值；及若确定脏污图像数据达到脏污程度阈值，则控制清洗机构清洗透光面。
通过本申请实施例的清洗方法，可以对透光面进行自动清洗，无需人工进行干预，清洗过程简单高效，且不会对其它模块造成影响。尤其当光学传感器数量多和/或安装位置较隐蔽时，该清洗方法显得尤为便利。
相关技术中，通过获取图像的像素，得知与脏污对应的像素的数量，进而根据脏污对应的像素的数量确定是否对透光面进行清洗。然而，这种方法无法有效确定脏污是否会影响成像质量，因此不能对透光面进行有效的清洗。本申请实施例的清洗方法包括若图像数据包括脏污图像数据，确定脏污图像数据是否达到脏污程度阈值，若脏污图像数据达到脏污程度阈值，则对透光面进行清洗。脏污图像数据达到脏污程度阈值时，表示脏污影响到图像质量，需要对透光面进行清洗，如此可以有效确定脏污是否影响成像质量，在脏污影响成像质量时，及时地清洗透光面，以保证图像质 量，且可以避免或降低在脏污程度没有影响图像质量时，对透光面进行清洗而造成清洁介质等资源的浪费。
本申请实施例的清洗控制系统包括一个或多个处理器，用于实现清洗方法。本申请实施例的计算机可读存储介质，其上存储有程序，该程序被处理器执行时，实现清洗方法。本申请实施例的清洗系统包括清洗机构及清洗控制系统。本申请实施例的光学传感器包括透光面、图像处理模块及清洗系统。图像处理模块用于获取透过透光面的外部图像，根据获取的外部图像生成相应的图像数据。本申请实施例的可移动平台包括机体、动力系统、光学传感器和清洗系统。动力系统设于机体，用于为可移动平台提供动力。光学传感器设于机体。
图1所示为清洗方法100的一个实施例的流程图。清洗方法100应用于光学传感器，光学传感器设有位于外侧的透光面，光学传感器用于通过透光面获取外部图像，并根据获取的外部图像生成相应的图像数据。光学传感器可以包括镜头，镜头包括透光面。外部光线可以穿过透光面，进入光学传感器内，光学传感器可以将光信号转换为电信号。在一些实施例中，光学传感器可以通过视觉算法对获取的外部图像进行图像处理，将外部图像转换为图像数据。在一些实施例中，光学传感器可以用于相机、摄像机、摄像头等。
清洗方法100包括步骤101-104。在步骤101中，获取光学传感器生成的图像数据。图像数据可以包括像素值。
在步骤102中，将图像数据与图像参考数据进行比较，确定图像数据是否包括脏污图像数据。
在一些实施例中，图像参考数据可以预先存储于数据库中。在产品设计时，可以对光学传感器获取的不同的外部图像进行搜集，并生成相应的图像数据，对大量的图像数据进行分析处理并记录在数据库中，作为至 少一部分的图像参考数据。
在一些实施例中，图像参考数据包括脏污图像参考数据和无脏污图像参考数据中的至少一者。在一些实施例中，可以对多张无脏污的外部图像进行搜集，并生成相应的无脏污的图像数据，对大量的无脏污的图像数据进行处理并记录在数据库中，作为无脏污图像参考数据。在一些实施例中，可以对多张不同脏污程度的外部图像进行搜集，并生成相应的图像数据，对大量的图像数据进行处理并记录在数据库中，作为脏污图像参考数据。在一些实施例中，可以在方法执行过程中更新数据库中的图像参考数据。
在一些实施例中，通过将图像数据的像素值和图像参考数据的像素值进行对比，确定图像数据是否包括脏污图像数据。图像质量与像素值关系紧密，像素值可以直接体现图像质量，通过比对像素值，可以准确、简单地确定图像数据是否包括脏污图像数据。
在一些实施例中，当图像数据与图像参考数据的像素值不相等时，则确定图像数据为脏污图像数据。图像参考数据可以包括无脏污图像参考数据和/或其他与脏污图像参考数据不同的数据。比较脏污图像数据的像素值和图像参考数据的像素值，若不相等，确定图像数据为脏污图像数据。如此可以简单准确地确定图像数据是否为脏污图像数据。在一个实施例中，图像参考数据可以包括多个离散的数据，图像数据与每个离散的数据的像素值均不相等时，则确定图像数据为脏污图像数据。在另一个实施例中，图像参考数据可以包括数据范围。图像数据与图像参考数据的数据范围内的数据的像素值均不相等，即图像数据不在数据范围内时，确定图像数据为脏污图像数据。
在另一些实施例中，当图像数据与图像参考数据的像素值的数据差值达到差值阈值时，则确定图像数据为脏污图像数据。图像参考数据可以包括无脏污图像参考数据和/或其他与脏污图像参考数据不同的数据。如 此，允许一定的误差裕度，尤其是在图像参考数据的数据量不够多时，可以较准确地确定图像数据是否为脏污图像数据。在一个实施例中，图像参考数据可以包括多个离散的数据，图像数据与每个离散的数据的像素值的差值均达到差值阈值时，则确定图像数据为脏污图像数据。在另一个实施例中，图像参考数据可以包括数据范围。图像数据与图像参考数据的数据范围的端点像素值的差值达到差值阈值时，确定图像数据为脏污图像数据。
在一些实施例中，通过将图像数据和图像参考数据的脏污图像参考数据和无脏污图像参考数据中的至少一者进行对比，确定图像数据是否包括脏污图像数据。脏污图像参考数据和无脏污图像参考数据不同，可以用于区分脏污图像和无脏污图像。在一个实施例中，可以通过将图像数据的像素值和图像参考数据的脏污图像参考数据的像素值和无脏污图像参考数据的像素值中的至少一者进行对比，确定图像数据是否包括脏污图像数据。
在一个实施例中，通过将图像数据和无脏污图像参考数据进行对比，确定图像数据是否包括脏污图像数据。若图像数据与无脏污图像参考数据匹配，确定图像数据为无脏污图像数据，否则确定图像数据为脏污图像数据。不同的脏污状况较多较复杂，相对于脏污图像数据的采集，无脏污图像数据的采集更准确、方便，因此通过将图像数据和无脏污图像参考数据进行对比，可以更准确地确定图像数据是否包括脏污图像数据。在一个实施例中，当图像数据与无脏污图像参考数据的像素值不相等时，说明图像数据与无脏污图像参考数据不匹配，则确定图像数据为脏污图像数据。在另一个实施例中，当图像数据与无脏污图像参考数据的像素值的数据差值达到差值阈值时，说明图像数据与无脏污图像参考数据不匹配，则确定图像数据为脏污图像数据。详细描述参见上文。若光学传感器生成的多个图像数据中至少一个图像数据与无脏污图像参考数据不匹配，确定光学传感器生成的多个图像数据包括脏污图像数据。
在另一个实施例中，通过将图像数据和脏污图像参考数据进行对比， 确定图像数据是否包括脏污图像数据。若图像数据与脏污图像参考数据匹配，确定图像数据为脏污图像数据，否则确定图像数据为无脏污图像数据。通过与脏污图像参考数据进行对比，可以直接确定图像数据是否为脏污图像数据。在一个实施例中，当图像数据与脏污图像参考数据的像素值不相等时，说明图像数据与脏污图像参考数据不匹配，则确定图像数据为无脏污图像数据，否则确定图像数据为脏污图像数据。在一个实施例中，脏污图像参考数据可以包括多个离散的数据，图像数据与每个离散的数据的像素值均不相等时，则确定图像数据为无脏污图像数据，否则为脏污图像数据。在另一个实施例中，脏污图像参考数据可以包括数据范围。图像数据与脏污图像参考数据的数据范围内的数据的像素值均不相等，即图像数据不在数据范围内时，确定图像数据为无脏污图像数据。图像数据在脏污图像参考数据的数据范围内时，确定图像数据为脏污图像数据。
在另一些实施例中，当图像数据与脏污图像参考数据的像素值的数据差值达到脏污数据差值阈值时，认为图像数据与脏污图像参考数据不匹配，则确定图像数据为无脏污图像数据，否则确定图像数据为脏污图像数据。在一个实施例中，图像参考数据可以包括多个离散的数据，图像数据与每个离散的数据的像素值的差值均达到脏污数据差值阈值时，则确定图像数据为无脏污图像数据，否则为脏污图像数据。在另一个实施例中，图像参考数据可以包括数据范围。图像数据与图像参考数据的数据范围的端点像素值的差值达到脏污数据差值阈值时，认为图像数据与脏污图像参考数据不匹配，确定图像数据为无脏污图像数据，否则为脏污图像数据。
若至少一个图像数据与脏污图像参考数据匹配，确定光学传感器生成的多个图像数据包括脏污图像数据。若所有的图像数据均与脏污图像参考数据不匹配，可以确定光学传感器生成的多个图像数据不包括脏污图像数据。
在另一个实施例中，通过将图像数据和图像参考数据的脏污图像参 考数据和无脏污图像参考数据进行对比，确定图像数据是否包括脏污图像数据。图像数据分别与脏污图像参考数据和无脏污图像参考数据进行对比。若图像数据与脏污图像参考数据匹配，且与无脏污图像参考数据不匹配，确定图像数据为脏污图像数据。若图像数据与无脏污图像参考数据匹配，且与脏污图像参考数据不匹配，确定图像数据为无脏污图像数据。可以参考上文所述的方法来确定图像数据是否与脏污图像参考数据匹配，且确定图像数据是否与无脏污图像参考数据匹配。
在一个实施例中，若图像数据与无脏污图像参考数据不匹配，且与脏污图像参考数据不匹配，可以根据图像数据与无脏污图像参考数据的近似度，和图像数据与脏污图像参考数据的近似度，确定图像数据与其中近似度较高的图像参考数据匹配。若图像数据与无脏污图像参考数据的近似度高，确定图像数据为无脏污图像数据，否则确定图像数据为脏污图像数据。在一个实施例中，图像数据包括数据范围，“近似度”可以包括图像数据与图像参考数据的端点像素值的最小数据差值，图像数据与脏污图像参考数据的最小数据差值和图像数据与无脏污图像参考数据的最小数据差值中较小的表示近似度较高。在另一个实施例中，图像数据包括离散的数据，“近似度”可以包括与离散的数据的像素值的数据差值的平均值。
在另一个实施例中，若图像数据与无脏污图像参考数据不匹配，且与脏污图像参考数据不匹配，可以根据其他因素确定图像数据为无脏污图像数据或脏污图像数据。在确定了图像数据为无脏污图像数据或脏污图像数据后，可以更新对应的图像参考数据。
将图像数据和图像参考数据的脏污图像参考数据和无脏污图像参考数据进行对比，可以确定图像数据是否为脏污图像数据，进而可以确定光学传感器生成的多个图像数据是否包括脏污图像数据。如此，可以更准确地确定图像数据是否包括脏污图像数据。
在一些实施例中，根据获取到光学传感器生成的图像数据，实时将 图像数据与图像参考数据进行比较，以及时判断确定所述图像数据是否包括脏污图像数据。从而有助于及时对光学传感器是否有脏污进行判断，提高对脏污进行判断和处理的效率。
在步骤103中，若图像数据包括脏污图像数据，则确定脏污图像数据是否达到脏污程度阈值。
在一个实施例中，从图像数据中筛选出脏污图像数据，并根据筛选出的脏污图像数据，确定脏污图像数据是否达到脏污程度阈值。在确定图像数据为脏污图像数据时，将该图像数据选出，将生成的多个图像数据中的脏污图像数据均选出，从而可以获得一个或多个脏污图像数据。进而确定筛选出的脏污图像数据是否达到脏污程度阈值。在一个实施例中，可以从所有图像数据中筛选出所有脏污图像数据。在另一个实施例中，可以从透光面部分区域对应的图像数据中，例如透光面中心区域对应的图像数据中，筛选出脏污图像数据，即筛选出部分区域内的脏污对应的脏污图像数据。如此，筛选出的脏污图像数据可以用于后续步骤中，降低后续步骤的数据处理量，提高处理速度。
在一个实施例中，可以通过大量可靠性测试，对脏污的程度进行划分，不同的脏污程度对应的图像数据不同，对应不同的图像数据范围，可以根据图像数据范围确定脏污程度阈值。在一个实施例中，可以设定脏污达到需要清洗的脏污程度对应的图像数据范围的最小值为脏污程度阈值。可以通过比对脏污图像数据和脏污程度阈值，确定脏污图像数据是否达到脏污程度阈值。
在一个实施例中，将脏污图像数据与图像参考数据进行比较，以确定脏污图像数据是否达到脏污程度阈值。图像参考数据可以包括多个数据，脏污图像数据与多个数据进行比较，可以更准确地判断脏污图像数据是否达到脏污程度阈值。图像参考数据可以包括脏污图像参考数据和/或无脏污图像参考数据，可以将脏污图像数据与脏污图像参考数据和/或无脏污图像 参考数据进行比较，以确定脏污图像数据是否达到脏污程度阈值。
在一个实施例中，图像参考数据包括脏污图像参考数据。将脏污图像数据与脏污图像参考数据进行比较，以确定脏污图像数据是否达到脏污程度阈值。如此可以直接简单地确定脏污图像数据是否达到脏污程度阈值。图像参考数据可以包括表示不同脏污程度的脏污图像参考数据，脏污图像数据与多个不同脏污程度对应的脏污图像参考数据进行比较，可以确定当前透光面的脏污达到的脏污程度。在一个实施例中，若多个脏污图像数据与脏污图像参考数据的差值分别达到第一差值阈值，确定脏污图像数据达到脏污程度阈值。如此留出一定的误差裕度。第一差值阈值可以预先存储于数据库中。在一个实施例中，第一差值阈值可以在方法执行中更新。
在一个实施例中，若脏污图像数据与脏污图像参考数据的差值达到第一差值阈值的个数达到第一个数阈值，确定脏污图像数据达到脏污程度阈值。在第一差值阈值的个数达到一定数量时，说明脏污影响图像质量达到需要清洗的程度，若第一差值阈值的个数较少时，说明脏污影响较低，无需清洗，如此可以准确地确定脏污是否影响图像质量。第一个数阈值可以预先存储于数据库中。在一个实施例中，个数阈值可以在方法执行中更新。在一个实施例中，若每个脏污图像数据与每个脏污图像参考数据的差值均达到第一差值阈值，确定脏污图像数据达到脏污程度阈值。第一个数阈值为脏污图像数据的总个数。在另一个实施例中，第一个数阈值可以小于脏污图像数据的总个数。例如，第一个数阈值可以超过脏污图像数据的总个数的一半，从而超过半数的脏污图像数据中，每个脏污图像数据与每个脏污图像参考数据的差值均达到第一差值阈值，确定脏污图像数据达到脏污程度阈值。上述仅是一个例子，并不限于上面的例子，第一个数阈值可以根据实际应用设定。
在另一个实施例中，若多个脏污图像数据的总和与多个脏污图像参考数据的总和的差值达到第二差值阈值，则确定脏污图像数据达到脏污程 度阈值。在一个实施例中，筛选出的所有脏污图像数据的总和与多个脏污图像参考数据的总和的差值达到第二差值阈值，确定脏污图像数据达到脏污程度阈值。从而在多个脏污图像数据各自未达到对应的脏污程度阈值，但脏污总体达到脏污程度阈值，则说明脏污总体影响到图像质量，需要进行清洗，如此可以及时对脏污进行清洗。
在另一个实施例中，图像参考数据包括无脏污图像参考数据。将脏污图像数据与无脏污图像参考数据进行比较，以确定脏污图像数据是否达到脏污程度阈值。无脏污图像参考数据较容易采集且较精确，如此可以简化数据采集过程，且较准确地确定脏污图像数据是否达到脏污程度阈值。在一个实施例中，若多个脏污图像数据与无脏污图像数据的差值分别达到第三差值阈值，确定脏污图像数据达到脏污程度阈值。若多个脏污图像数据与无脏污图像数据的差值分别达到第三差值阈值，说明脏污图像数据与无脏污图像数据相差较大，脏污程度达到需要清洗的程度，需进行清洗，如此可以及时进行清洗，且可以避免或降低无需清洗的情况下进行清洗造成的浪费。第三差值阈值可以预先存储于数据库中。在一个实施例中，第三差值阈值可以在方法执行中更新。
在一个实施例中，若多个脏污图像数据与无脏污图像参考数据的差值达到第三差值阈值的个数达到第二个数阈值，确定脏污图像数据达到脏污程度阈值。若达到第三差值阈值的脏污图像数据数量较多，说明脏污程度达到需要清洗的程度，需进行清洗，如此可以及时进行清洗，且可以避免或降低无需清洗的情况下进行清洗造成的浪费。第二个数阈值可以预先存储于数据库中。在一个实施例中二个数阈值可以在方法执行中更新。在一个实施例中，若每个脏污图像数据与每个无脏污图像参考数据的差值均达到第三差值阈值，确定脏污图像数据达到脏污程度阈值。第二个数阈值为脏污图像数据的总个数。在另一个实施例中，第二个数阈值可以小于脏污图像数据的总个数。
在另一个实施例中，若多个脏污图像数据的总和与多个无脏污图像参考数据的总和的差值达到第四差值阈值，则确定脏污图像数据达到脏污程度阈值。在一个实施例中，筛选出的所有脏污图像数据的总和与多个脏污图像参考数据的总和的差值达到第三差值阈值，确定脏污图像数据达到脏污程度阈值。从而在多个脏污图像数据各自未达到对应的脏污程度阈值，但脏污总体达到脏污程度阈值，则说明脏污总体影响到图像质量，需要进行清洗，如此可以及时对脏污进行清洗。
在一个实施例中，清洗方法100包括根据脏污图像数据，确定脏污类型。不同的脏污类型可对应不同的脏污图像数据。脏污类型表示不同类型的脏污，例如树叶、泥土、液体、灰层等。在一个实施例中，将脏污图像数据的像素值与图像参考数据的像素值进行比较，确定脏污类型。图像参考数据可以包括不同脏污类型对应的图像参考数据。可以在不同工况下获取外部图像，对生成的图像数据进行采集，处理分析，确定脏污类型和对应的图像参考数据，可以对应存储于数据库中。
通过对比脏污图像数据与多种脏污类型分别对应的脏污图像参考数据，确定脏污类型。脏污图像数据与其中一种脏污类型对应的脏污图像参考数据的差值小于类型差值阈值时，确定脏污图像数据属于该脏污类型。可以将筛选出的脏污图像数据归类至对应的脏污类型中。进一步确定脏污类型对应的脏污图像数据是否达到脏污类型对应的脏污程度阈值。不同的脏污类型可以对应不同的脏污程度阈值。对于相同像素数目的脏污，不同类型的脏污对图像质量的影响不同，例如雨水相对于泥土对图像质量的影响较小，所以通过确定脏污图像数据所属的脏污类型，并根据脏污类型对应的脏污图像参考数据来确定脏污图像数据是否达到脏污程度阈值，可以更加有效地确定脏污是否影响图像质量。
在一个实施例中，将脏污类型对应的脏污图像数据与该脏污类型对应的脏污程度阈值进行对比，确定脏污类型对应的脏污图像数据是否达到 对应的脏污程度阈值。
在另一个实施例中，将脏污类型对应的脏污图像数据与该脏污类型对应的脏污图像参考数据进行比较，确定脏污类型对应的脏污图像数据是否达到对应的脏污程度阈值。在一个实施例中，若脏污类型对应的脏污图像数据与该脏污类型对应的脏污图像参考数据的差值分别达到第五差值阈值，确定该脏污类型对应的脏污图像数据达到对应的脏污程度阈值。在一个实施例中，若脏污类型对应的脏污图像数据与脏污图像参考数据的差值达到第五差值阈值的个数达到第三个数阈值，确定脏污图像数据达到脏污程度阈值。在另一个实施例中，若脏污类型对应的多个脏污图像数据的总和与该脏污类型对应的多个脏污图像参考数据的总和的差值达到第六差值阈值，则确定脏污图像数据达到脏污程度阈值。
在一个实施例中，若脏污类型为多种，则确定多种脏污类型对应的脏污图像数据是否达到脏污类型对应的脏污程度阈值。可以通过上文所述的方法，分别确定每种脏污类型对应的脏污图像数据是否达到对应的脏污程度阈值。针对不同的脏污类型确定是否达到脏污程度阈值，可以更准确地确定脏污是否影响图像质量，需要进行清洗。
在另一个实施例中，若脏污类型为多种，确定多种脏污类型对应的脏污图像数据的总和是否达到脏污程度阈值。在一个实施例中，不同脏污类型对图像质量的影响不同，可以根据不同脏污类型对图像质量的影响程度设置影响系数。脏污图像数据的总和可以为多种脏污类型对应的脏污图像数据乘以对应的系数之后的总和。从而在多种脏污类型对应的脏污图像数据各自未达到对应的脏污程度阈值，但脏污图像数据的总和达到脏污程度阈值，则说明脏污总体影响到图像质量，需要进行清洗，如此可以及时对脏污进行清洗。
在步骤104中，若确定脏污图像数据达到脏污程度阈值，则控制清洗机构清洗透光面。
脏污图像数据达到脏污程度阈值，说明脏污影响到图像质量，需要对透光面进行清洗。以此，控制清洗机构清洗透光面，实现对透光面的清洗。若脏污图像数据未达到脏污程度阈值，则不对透光面进行清洗。
在一个实施例中，若脏污类型对应的脏污图像数据达到脏污类型对应的脏污程度阈值，则控制清洗机构清洗透光面。不同的脏污类型对图像的影响不同，在脏污类型对应的脏污图像数据达到脏污类型对应的脏污程度阈值时，对透光面进行清洗，如此可以更及时地清洗透光面，且可以更有效地避免或降低在脏污程度没有影响图像质量时，对透光面进行清洗而造成清洁介质等资源的浪费。
在一个实施例中，脏污类型为多种，若脏污类型对应的脏污图像数据达到脏污类型对应的脏污程度阈值，则控制清洗机构进行清洗透光面。在一个实施例中，若至少一种脏污类型对应的脏污图像数据达到脏污类型对应的脏污程度阈值，则控制清洗机构进行清洗透光面。在另一个实施例中，若多种脏污类型对应的脏污图像数据均达到脏污类型对应的脏污程度阈值，则控制清洗机构进行清洗透光面。
在一个实施例中，脏污类型为多种，确定多种脏污类型对应的脏污图像数据的总和是否达到脏污程度阈值；若多种脏污类型对应的脏污图像数据总和达到脏污程度阈值，则控制清洗机构清洗透光面。
通过本申请实施例的清洗方法100，可以对透光面进行自动清洗，无需人工进行干预，清洗过程简单高效，且不会对其它模块造成影响。尤其当光学传感器数量多和/或安装位置较隐蔽时，该清洗方法100显得尤为便利。而且本申请实施例的清洗方法中，确定脏污图像数据是否达到脏污程度阈值，若脏污图像数据达到脏污程度阈值，则对透光面进行清洗。如此可以有效确定脏污是否影响成像质量，且在脏污影响成像质量时，清洗透光面，可以及时清洗透光面，保证图像质量，同时可以避免或降低在脏污程度没有影响图像质量时，对透光面进行清洗而造成清洁介质等资源的 浪费。
图2所示为清洗系统500的一个实施例的原理框图。清洗系统500包括清洗机构200及清洗控制系统300。参考图1和2，在一个实施例中，控制清洗机构进行清洗透光面的步骤104包括：控制清洗机构200的流体输送装置201将清洁介质推送入管路202，以使清洁介质通过管路202输送至喷嘴2031-203N，并通过喷嘴2031-203N喷出，来清洗透光面。在一些实施例中，清洁介质包括液体和/或气体。在一个实施例中，清洁介质包括如下至少一种：水、玻璃水、空气。
在一个实施例中，清洁介质包括液体，流体输送装置201包括泵，用于将液体泵入管道202中。在另一个实施例中，清洁介质包括气体，流体输送装置201可以包括压缩机或鼓风机等。流体输送装置201可以与清洗控制系统300连接，清洗控制系统300可以执行清洗方法100，可以控制流体输送装置201。在需要清洗时，控制流体输送装置201工作，清洗完成后，控制流体输送装置201停止工作。
在一个实施例中，控制清洗机构进行清洗透光面的步骤104包括：控制管路202上设置的开关装置2041-204N打开，使清洁介质流过。开关装置2041-204N可以与清洗控制系统300连接，清洗控制系统300可以控制开关装置2041-204N的开闭。在需要清洗时，控制开关装置2041-204N打开，清洗完成后，控制开关装置2041-204N关闭。开关装置2041-204N可以包括电磁阀。
在一个实施例中，清洗机构200包括连接流体输送装置201、至少一个喷嘴2031-203N、以及连通流体输送装置201与喷嘴2031-203N的至少一条管路2021-202N。至少一条管路2021-202N上设有开关装置2041-204N，用于控制管路的打开和闭合，以控制清洁介质可以流过对应的管路2021-202N，从对应的喷嘴2031-203N喷出，来清洁对应的透光面。至少一个喷嘴2031-203N对应多个光学传感器401-40N的透光面。在透光 面对应的脏污图像数据达到脏污程度阈值时，控制对应的管路2021-202N上的开关装置2041-204N打开。多条管路2021-202N上分别设有控制管路通断的开关装置2041-204N，在其中一个或多个光学传感器401-40N的透光面需要清洗时，控制对应的开关装置2041-204N打开，使清洁介质可以流过对应的管路2021-202N，从对应的喷嘴2031-203N喷出，来清洁对应的透光面。无需清洁的透光面对应的开关装置2041-204N可以保持关闭。如此自动地实现对多个光学传感器中的一个或多个光学传感器的透光面进行清洗。
图3所示为光学传感器400和清洁机构200的清洁执行装置205的一个实施例的立体示意图。图4所示为光学传感器400和清洁执行装置205的一个实施例的立体分解示意图。光学传感器400包括设于外侧的透光面401。在图示实施例中，光学传感器400包括镜头。光学传感器400可组装于清洁执行装置205。
清洁执行装置205包括固定壳体206、组装于固定壳体206的喷嘴203和与喷嘴203连通的管路接头207。光学传感器400可固定组装于固定壳体206。在一个实施例中，固定壳体206设有安装孔208，光学传感器400穿设于安装孔208内，透光面401从安装孔208露出。喷嘴203组装于安装孔208的一侧，喷嘴203与透光面401相对应设置，透光面401在喷嘴203的喷射范围内。从喷嘴203喷出的清洁介质可以喷至透光面401上，对透光面401进行清洗。在一个实施例中，固定壳体206于安装孔208的外侧形成有喷嘴固定孔209，喷嘴203固定安装于喷嘴固定孔209内。
喷嘴203包括喷出口210，清洁介质从喷出口210喷出。喷出口210面向安装孔208，可以从透光面401的一侧倾斜面向透光面401，使喷出的清洁介质可以喷到透光面401上。在一个实施例中，喷出口210呈扇形开口，使清洁介质呈扇形喷出，喷出的力度更强、覆盖面积更广，如此可以有效地清洁透光面401。
管路接头207可以连通喷嘴203和管路(未图示)。在一个实施例中，管路接头207固定组装于固定壳体206。固定壳体206开设有与喷嘴203连通的通道211，管路接头207可以与通道211连通，进而与喷嘴203连通。在一个实施例中，通道211与喷嘴固定孔209连通。在另一个实施例中，管路接头207可以直接与喷嘴203连接。
图5所示为清洗控制系统300的一个实施例的模块框图。清洗控制系统300包括一个或多个处理器301，用于实现清洗方法100。清洗控制系统300的处理器301可以实现上文所述的清洗方法。在一些实施例中，
清洗控制系统300可以包括计算机可读存储介质304，计算机可读存储介质可以存储有可被处理器301调用的程序，可以包括非易失性存储介质。在一些实施例中，清洗控制系统300可以包括内存303和接口302。在一些实施例中，清洗控制系统300还可以根据实际应用包括其他硬件。
本申请实施例的计算机可读存储介质304，其上存储有程序，该程序被处理器301执行时，实现清洗方法100。
本申请可采用在一个或多个其中包含有程序代码的存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。计算机可读存储介质包括永久性和非永久性、可移动和非可移动媒体，可以由任何方法或技术来实现信息存储。信息可以是计算机可读指令、数据结构、程序的模块或其他数据。计算机可读存储介质的例子包括但不限于：相变内存(PRAM)、静态随机存取存储器(SRAM)、动态随机存取存储器(DRAM)、其他类型的随机存取存储器(RAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、快闪记忆体或其他内存技术、只读光盘只读存储器(CD-ROM)、数字多功能光盘(DVD)或其他光学存储、磁盒式磁带，磁带磁磁盘存储或其他磁性存储设备或任何其他非传输介质，可用于存储可以被计算设备访问的信息。
图6所示为光学传感器400的一个实施例的模块框图。光学传感器400包括透光面401(如图3所示)、图像处理模块402及清洗系统500。图像处理模块402用于获取透过透光面401的外部图像，根据获取的外部图像生成相应的图像数据。图像处理模块402可以通过视觉算法生成图像数据。清洗系统500可以为上文所述实施例的清洗系统。清洗系统500的清洗控制系统300可以与图像处理模块402设置于同一控制电路板上。
图7所示为可移动平台700的一个实施例的示意图。可移动平台700可以包括移动小车、无人飞行器、汽车、机器人或其他可移动装置。可移动平台700包括机体701、动力系统702、光学传感器703和清洗系统500。动力系统702设于机体，用于为可移动平台700提供动力。在一些实施例中，动力系统702可以包括电机。光学传感器703设于机体701，可以用来拍摄图像。在一些实施例中，可移动平台700可以利用光学传感器703拍摄的图像进行测距、追踪跟随等。清洗系统500可以为上文所述实施例的清洗系统。清洗系统500可以清洗光学传感器703的透光面。
在其他一些实施例中，清洗系统500的清洗控制系统300可以设置于可移动平台外，例如设于控制中心。控制中心例如超级计算中心、计算平台等。光学传感器703生成图像数据，发送给清洗控制系统300。
在一个实施例中，可移动平台700还包括GPS装置，用于获取可移动平台700的位置信息。若确定脏污图像数据达到脏污程度阈值，控制清洗机构清洗透光面，并将脏污图像数据以及当前的位置信息进行存储。例如，通过GPS装置获取可移动平台700的当前位置信息，确定脏污图像数据达到脏污程度阈值时，同时存储脏污图像数据与可移动平台700的当前位置信息。有利于用户根据可移动平台700的位置与光学传感器703之间的关系，对出行路线进行合理规划。
应当理解，本申请的各部分可以用硬件、软件或它们的组合来实现。在上述实施例中，多个步骤或方法可以用存储在存储器中且由合适的指令 执行系统执行的软件或硬件来实现。例如，如果用硬件来实现，可用下列技术中的任一项或他们的组合来实现：具有用于对数据信号实现逻辑功能的逻辑门电路的离散逻辑电路，具有合适的组合逻辑门电路的专用集成电路，可编程门阵列(PGA)，现场可编程门阵列(FPGA)等。
本技术领域的普通技术人员可以理解实现上述实施方法携带的全部或部分步骤是可以通过程序来指令相关的硬件完成，所述的程序可以存储于一种计算机可读存储介质中，该程序在执行时，包括方法实施例的步骤之一或其组合。
需要说明的是，在本文中，诸如“第一”和“第二”等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
以上对本发明实施例所提供的方法和装置进行了详细介绍，本文中应用了具体个例对本发明的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本发明的方法及其核心思想；同时，对于本领域的一般技术人员，依据本发明的思想，在具体实施方式及应用范围上均会有改变之处，综上所述，本说明书内容不应理解为对本发明的限制。
本专利文件披露的内容包含受版权保护的材料。该版权为版权所有人所有。版权所有人不反对任何人复制专利与商标局的官方记录和档案中所存在的该专利文件或者该专利披露。

Claims (31)

  1. 一种清洗方法，应用于光学传感器，所述光学传感器设有位于外侧的透光面，所述光学传感器用于通过所述透光面获取外部图像，并根据获取的所述外部图像生成相应的图像数据；其特征在于，所述清洗方法包括：
    获取所述光学传感器生成的所述图像数据；
    将所述图像数据与图像参考数据进行比较，确定所述图像数据是否包括脏污图像数据；
    若所述图像数据包括所述脏污图像数据，则确定所述脏污图像数据是否达到脏污程度阈值；及
    若确定所述脏污图像数据达到所述脏污程度阈值，则控制清洗机构清洗所述透光面。
  2. 根据权利要求1所述的清洗方法，其特征在于，所述将所述图像数据与图像参考数据进行比较，确定所述图像数据是否包括脏污图像数据，包括：
    通过将所述图像数据的像素值和所述图像参考数据的像素值进行对比，确定所述图像数据是否包括脏污图像数据。
  3. 根据权利要求1所述的清洗方法，其特征在于，所述将所述图像数据与图像参考数据进行比较，确定所述图像数据是否包括脏污图像数据，包括：
    通过将所述图像数据和所述图像参考数据的脏污图像参考数据和无脏污图像参考数据中的至少一者进行对比，确定所述图像数据是否包括脏污图像数据。
  4. 根据权利要求2所述的清洗方法，其特征在于，所述将所述图像数据与图像参考数据进行比较，确定所述图像数据是否包括脏污图像数据，包括：
    当所述图像数据与所述图像参考数据的像素值的数据差值达到差值阈值 时，则确定所述图像数据为所述脏污图像数据。
  5. 根据权利要求2所述的清洗方法，其特征在于，
    当所述图像数据与所述图像参考数据的像素值不相等时，则确定所述图像数据为所述脏污图像数据。
  6. 根据权利要求1所述的清洗方法，其特征在于，包括：
    从所述图像数据中筛选出所述脏污图像数据；
    根据筛选出的所述脏污图像数据，确定所述脏污图像数据是否达到所述脏污程度阈值。
  7. 根据权利要求1所述的清洗方法，其特征在于，所述若所述图像数据包括所述脏污图像数据，则确定所述脏污图像数据是否达到脏污程度阈值，包括：
    将所述脏污图像数据与所述图像参考数据进行比较，以确定所述脏污图像数据是否达到脏污程度阈值。
  8. 根据权利要求7所述的清洗方法，其特征在于，包括：
    将所述脏污图像数据与所述脏污图像参考数据进行比较，以确定所述脏污图像数据是否达到脏污程度阈值。
  9. 根据权利要求8所述的清洗方法，其特征在于，所述将所述脏污图像数据与所述脏污图像参考数据进行比较，以确定所述脏污图像数据是否达到脏污程度阈值，包括：
    若多个所述脏污图像数据与所述脏污图像参考数据的差值分别达到第一差值阈值，确定所述脏污图像数据达到脏污程度阈值。
  10. 根据权利要求9所述的清洗方法，其特征在于，所述将所述脏污图像数据与所述脏污图像参考数据进行比较，以确定所述脏污图像数据是否达到脏污程度阈值，包括：
    若所述脏污图像数据与所述脏污图像参考数据的差值达到第一差值阈值的个数达到第一个数阈值，确定所述脏污图像数据达到脏污程度阈值；或者，
    若多个所述脏污图像数据的总和与多个所述脏污图像参考数据的总和的差值达到第二差值阈值，则确定所述脏污图像数据达到脏污程度阈值。
  11. 根据权利要求7所述的清洗方法，其特征在于，所述图像参考数据包括无脏污图像参考数据，
    所述将所述脏污图像数据与所述图像参考数据进行比较，以确定所述脏污图像数据是否达到脏污程度阈值，包括：
    将所述脏污图像数据与所述无脏污图像参考数据进行比较，以确定所述脏污图像数据是否达到脏污程度阈值。
  12. 根据权利要求11所述的清洗方法，其特征在于，所述将所述脏污图像数据与所述无脏污图像参考数据进行比较，以确定所述脏污图像数据是否达到脏污程度阈值，包括：
    若多个所述脏污图像数据与所述无脏污图像数据的差值分别达到第三差值阈值，确定所述脏污图像数据达到脏污程度阈值。
  13. 根据权利要求12所述的清洗方法，其特征在于，所述将所述脏污图像数据与所述无脏污图像参考数据进行比较，以确定所述脏污图像数据是否达到脏污程度阈值，包括：
    若多个所述脏污图像数据与所述无脏污图像参考数据的差值达到第三差值阈值的个数达到第二个数阈值，确定所述脏污图像数据达到脏污程度阈值；或者，
    若多个所述脏污图像数据的总和与多个所述无脏污图像参考数据的总和的差值达到第四差值阈值，则确定所述脏污图像数据达到脏污程度阈值。
  14. 根据权利要求1所述的清洗方法，其特征在于，所述清洗方法包括：根据所述脏污图像数据，确定脏污类型。
  15. 根据权利要求14所述的清洗方法，其特征在于，所述根据所述脏污图像数据，确定脏污类型，包括：
    将所述脏污图像数据的像素值与所述图像参考数据的像素值进行比较，确定所述脏污类型。
  16. 根据权利要求14所述的清洗方法，其特征在于，所述若所述图像数据包括所述脏污图像数据，则确定所述脏污图像数据是否达到脏污程度阈值，包括：
    确定所述脏污类型对应的所述脏污图像数据是否达到所述脏污类型对应的所述脏污程度阈值。
  17. 根据权利要求16所述的清洗方法，其特征在于，所述若确定所述脏污图像数据达到所述脏污程度阈值，则控制清洗机构清洗所述透光面，包括：
    若所述脏污类型对应的所述脏污图像数据达到所述脏污类型对应的所述脏污程度阈值，则控制所述清洗机构清洗所述透光面。
  18. 根据权利要求16所述的清洗方法，其特征在于，包括：
    若所述脏污类型为多种，则确定多种所述脏污类型对应的所述脏污图像数据是否达到所述脏污类型对应的脏污程度阈值；
    若脏污类型对应的所述脏污图像数据达到所述脏污类型对应的脏污程度阈值，则控制所述清洗机构进行清洗所述透光面。
  19. 根据权利要求16所述的清洗方法，其特征在于，包括：
    若所述脏污类型为多种，确定多种所述脏污类型对应的所述脏污图像数据的总和是否达到所述脏污程度阈值；
    若多种脏污类型对应的所述脏污图像数据总和达到所述脏污程度阈值，则控制所述清洗机构清洗所述透光面。
  20. 根据权利要求14所述的清洗方法，其特征在于，所述根据所述脏污图像数据，确定脏污类型，包括：
    通过对比所述脏污图像数据与多种所述脏污类型分别对应的脏污图像参考数据，确定所述脏污类型。
  21. 根据权利要求1所述的清洗方法，其特征在于，所述若确定所述脏污图像数据达到所述脏污程度阈值，则控制清洗机构清洗所述透光面，包括：
    控制所述清洗机构的流体输送装置将清洁介质推送入管路，以使所述清洁介质通过所述管路输送至喷嘴，并通过所述喷嘴喷出，来清洗所述透光面。
  22. 根据权利要求21所述的清洗方法，其特征在于，所述清洁介质包括液体和/或气体。
  23. 根据权利要求22所述的清洗方法，其特征在于，所述清洁介质包括如下至少一种：水、玻璃水、空气。
  24. 根据权利要求21所述的清洗方法，其特征在于，所述若确定所述脏污图像数据达到所述脏污程度阈值，则控制清洗机构清洗所述透光面，包括：
    控制所述管路上设置的开关装置打开，使所述清洁介质流过。
  25. 根据权利要求24所述的清洗方法，其特征在于，所述清洗机构包括连接所述流体输送装置和多个所述喷嘴的多条所述管路，多个所述喷嘴对应多个所述光学传感器的所述透光面；
    所述控制所述管路上设置的开关装置打开，包括：
    在所述透光面对应的所述脏污图像数据达到所述脏污程度阈值时，控制对应的所述管路上的开关装置打开。
  26. 一种清洗控制系统，其特征在于，包括一个或多个处理器，单独的或共同的工作，用于实现如权利要求1-25中任一项所述的清洗方法。
  27. 一种计算机可读存储介质，其特征在于，其上存储有程序，该程序被处理器执行时，实现如权利要求1-25中任意一项所述的清洗方法。
  28. 一种清洗系统，其特征在于，包括：
    清洗机构；及
    权利要求26所述的清洗控制系统，所述清洗控制系统用于控制所述清洗机构对所述透光面进行清洗。
  29. 根据权利要求28所述的清洗系统，其特征在于，所述清洗机构包括：
    流体输送装置，与所述清洗控制系统相连接；
    喷嘴，与所述光学传感器的所述透光面相对应，使所述透光面在所述喷嘴的喷射范围内；
    管路，用于连通所述流体输送装置与所述喷嘴；及，
    开关装置，安装在所述管路，用于控制管路的导通；
    当所述清洗控制系统需要控制所述清洗机构对所述透光面进行清洗时，所述清洗控制系统控制所述流体输送装置输送所述清洗介质给所述管体，并控制所述开关装置导通所述管路，使得清洁介质流过对应的所述管路，从对应的所述喷嘴喷出，以清洗对应的所述透光面。
  30. 一种光学传感器，其特征在于，包括：
    透光面；
    图像处理模块，用于获取透过所述透光面的外部图像，根据获取的所述外部图像生成相应的图像数据；及
    权利要求28或29所述的清洗系统，其中，所述喷嘴与所述透光面相对应设置，以实现当所述清洗介质从所述喷嘴喷出时，对应的所述透光面在所述喷嘴的喷射范围内。
  31. 一种可移动平台，其特征在于，包括：
    机体；
    动力系统，设于所述机体，用于为所述可移动平台提供动力；以及
    权力要求30所述的光学传感器。
PCT/CN2019/084952 2019-04-29 2019-04-29 清洗方法、清洗控制系统、计算机可读存储介质、清洗系统、光学传感器及可移动平台 WO2020220185A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2019/084952 WO2020220185A1 (zh) 2019-04-29 2019-04-29 清洗方法、清洗控制系统、计算机可读存储介质、清洗系统、光学传感器及可移动平台
CN201980008957.1A CN111684487B (zh) 2019-04-29 2019-04-29 清洗方法、清洗控制系统、计算机可读存储介质、清洗系统、光学传感器及可移动平台

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/084952 WO2020220185A1 (zh) 2019-04-29 2019-04-29 清洗方法、清洗控制系统、计算机可读存储介质、清洗系统、光学传感器及可移动平台

Publications (1)

Publication Number Publication Date
WO2020220185A1 true WO2020220185A1 (zh) 2020-11-05

Family

ID=72433274

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/084952 WO2020220185A1 (zh) 2019-04-29 2019-04-29 清洗方法、清洗控制系统、计算机可读存储介质、清洗系统、光学传感器及可移动平台

Country Status (2)

Country Link
CN (1) CN111684487B (zh)
WO (1) WO2020220185A1 (zh)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113014194A (zh) * 2021-03-10 2021-06-22 优兔创新有限公司 太阳能面板污迹清理方法、装置、计算机设备和存储介质
CN113409279A (zh) * 2021-06-24 2021-09-17 北京车和家信息技术有限公司 激光雷达清洗系统的效果评估方法、装置、设备和介质
US11347243B2 (en) * 2017-08-08 2022-05-31 Ford Global Technologies, Llc Vehicle inspection systems and methods
CN114589160A (zh) * 2022-01-25 2022-06-07 深圳大方智能科技有限公司 一种室内施工用摄像头防护方法
CN115022547A (zh) * 2022-06-09 2022-09-06 小米汽车科技有限公司 车载摄像头清洗方法、装置、车辆、存储介质及芯片
CN116100703A (zh) * 2022-12-30 2023-05-12 玫瑰塑胶(昆山)有限公司 一种应用再生材料的塑胶制品制造方法和系统

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11209282B2 (en) * 2019-07-02 2021-12-28 Ford Global Technologies, Llc Vehicle cleanliness detection and carwash recommendation

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1677682A (zh) * 2004-03-30 2005-10-05 原相科技股份有限公司 能选取最佳距离封装光学感测模块的封装设备及其方法
CN102029976A (zh) * 2009-09-29 2011-04-27 株式会社电装 车载光学传感器盖板以及车载光学传感器装置
US20160307881A1 (en) * 2015-04-20 2016-10-20 Advanced Semiconductor Engineering, Inc. Optical sensor module and method for manufacturing the same
CN106575366A (zh) * 2014-07-04 2017-04-19 光实验室股份有限公司 关于检测和/或指示脏镜头状况的方法和装置
CN208213718U (zh) * 2018-03-24 2018-12-11 北醒（北京）光子科技有限公司 一种光学测距传感器清洁系统

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2949520B1 (en) * 2014-05-27 2018-03-21 Fico Transpar, S.A. System for cleaning a vehicle-mounted optic lens
CN106231297B (zh) * 2016-08-29 2019-03-19 深圳天珑无线科技有限公司 摄像头的检测方法及装置
US10112585B1 (en) * 2017-06-20 2018-10-30 Ford Global Technologies, Llc Vehicle cleanliness detection systems and methods
CN109204239A (zh) * 2017-07-05 2019-01-15 宁波恒帅微电机有限公司 车用光学传感器主动清洁装置
CN108668080B (zh) * 2018-06-22 2020-09-08 北京小米移动软件有限公司 提示镜头脏污程度的方法及装置、电子设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1677682A (zh) * 2004-03-30 2005-10-05 原相科技股份有限公司 能选取最佳距离封装光学感测模块的封装设备及其方法
CN102029976A (zh) * 2009-09-29 2011-04-27 株式会社电装 车载光学传感器盖板以及车载光学传感器装置
CN106575366A (zh) * 2014-07-04 2017-04-19 光实验室股份有限公司 关于检测和/或指示脏镜头状况的方法和装置
US20160307881A1 (en) * 2015-04-20 2016-10-20 Advanced Semiconductor Engineering, Inc. Optical sensor module and method for manufacturing the same
CN208213718U (zh) * 2018-03-24 2018-12-11 北醒（北京）光子科技有限公司 一种光学测距传感器清洁系统

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11347243B2 (en) * 2017-08-08 2022-05-31 Ford Global Technologies, Llc Vehicle inspection systems and methods
CN113014194A (zh) * 2021-03-10 2021-06-22 优兔创新有限公司 太阳能面板污迹清理方法、装置、计算机设备和存储介质
CN113409279A (zh) * 2021-06-24 2021-09-17 北京车和家信息技术有限公司 激光雷达清洗系统的效果评估方法、装置、设备和介质
CN114589160A (zh) * 2022-01-25 2022-06-07 深圳大方智能科技有限公司 一种室内施工用摄像头防护方法
CN114589160B (zh) * 2022-01-25 2023-05-16 深圳大方智能科技有限公司 一种室内施工用摄像头防护方法
CN115022547A (zh) * 2022-06-09 2022-09-06 小米汽车科技有限公司 车载摄像头清洗方法、装置、车辆、存储介质及芯片
CN116100703A (zh) * 2022-12-30 2023-05-12 玫瑰塑胶(昆山)有限公司 一种应用再生材料的塑胶制品制造方法和系统
CN116100703B (zh) * 2022-12-30 2023-11-14 玫瑰塑胶(昆山)有限公司 一种应用再生材料的塑胶制品制造方法和系统

Also Published As

Publication number Publication date
CN111684487B (zh) 2024-03-19
CN111684487A (zh) 2020-09-18

Similar Documents

Publication Publication Date Title
WO2020220185A1 (zh) 清洗方法、清洗控制系统、计算机可读存储介质、清洗系统、光学传感器及可移动平台
US11047146B2 (en) Pool cleaner with laser range finder system and method
US10705535B2 (en) Systems and methods for performing simultaneous localization and mapping using machine vision systems
EP3224649B1 (en) Systems and methods for performing simultaneous localization and mapping using machine vision systems
US20190210570A1 (en) Sensor cleaning and cooling
US20170305010A1 (en) Systems and Methods for Performing Occlusion Detection
CN109466512A (zh) 车辆传感器系统
CN110499727B (zh) 一种基于多传感器的贴边清扫方法和清扫车
Schlegl et al. Combined capacitive and ultrasonic distance measurement for automotive applications
Talele et al. Detection of real time objects using TensorFlow and OpenCV
WO2020127151A1 (en) Method for improved object detection
Nagarajan et al. Obstacle detection and avoidance for mobile robots using monocular vision
An et al. Ceiling vision-based active SLAM framework for dynamic and wide-open environments
Paul et al. A novel approach to steel rivet detection in poorly illuminated steel structural environments
KR102467990B1 (ko) 로봇 청소기
KR20230090218A (ko) 청소 결정을 위한 오염율 계산 방법 및 장치
Zheng Feature recognition and obstacle detection for drive assistance in indoor environments
WO2020059292A1 (ja) 自律走行掃除機
GB2554162A (en) Method and control device for processing an image representing at least one halo and an imaging system
KR102203438B1 (ko) 이동 로봇 및 이동 로봇의 제어방법
Hesch et al. Simultaneous localization and mapping using an omni-directional camera
JP7403423B2 (ja) ロボット掃除機
CN116784732A (zh) 垃圾收集方法及设备
Gómez et al. Deep Learning based Litter Identification and Adaptive Cleaning using Self-reconfigurable Pavement Sweeping Robot
Her et al. Steering angle determination using cart and voting

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19927499

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19927499

Country of ref document: EP

Kind code of ref document: A1