WO2021036824A1 - 信息采集装置、方法、巡检机器人和存储介质 - Google Patents

信息采集装置、方法、巡检机器人和存储介质 Download PDF

Info

Publication number
WO2021036824A1
WO2021036824A1 PCT/CN2020/109214 CN2020109214W WO2021036824A1 WO 2021036824 A1 WO2021036824 A1 WO 2021036824A1 CN 2020109214 W CN2020109214 W CN 2020109214W WO 2021036824 A1 WO2021036824 A1 WO 2021036824A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
information collection
instruction
light
brightness parameter
Prior art date
Application number
PCT/CN2020/109214
Other languages
English (en)
French (fr)
Inventor
许哲涛
Original Assignee
北京海益同展信息科技有限公司
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 北京海益同展信息科技有限公司 filed Critical 北京海益同展信息科技有限公司
Priority to US17/634,470 priority Critical patent/US20220303447A1/en
Priority to EP20857764.3A priority patent/EP4024848A4/en
Publication of WO2021036824A1 publication Critical patent/WO2021036824A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/10—Image acquisition
    • G06V10/12—Details of acquisition arrangements; Constructional details thereof
    • G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
    • G06V10/141—Control of illumination
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/70—Circuitry for compensating brightness variation in the scene
    • H04N23/74—Circuitry for compensating brightness variation in the scene by influencing the scene brightness using illuminating means
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
    • G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/98—Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns
    • G06V10/993—Evaluation of the quality of the acquired pattern
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60—Control of cameras or camera modules
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60—Control of cameras or camera modules
    • H04N23/64—Computer-aided capture of images, e.g. transfer from script file into camera, check of taken image quality, advice or proposal for image composition or decision on when to take image
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/70—Circuitry for compensating brightness variation in the scene
    • H04N23/71—Circuitry for evaluating the brightness variation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/07—Target detection

Definitions

  • the present disclosure relates to the field of robotics technology, and in particular to an information collection device, method, inspection robot, and storage medium.
  • IDC Internet Data Center
  • IDC inspection robot is a method that uses machine inspection instead of manual inspection. It has the characteristics of high efficiency, simplicity and uninterrupted work.
  • the IDC robot collects the working status of the host by automatically moving in the machine room and taking images during the movement.
  • a technical problem to be solved by the embodiments of the present disclosure is: how to improve the accuracy and efficiency of information collection.
  • an information collection device including: a photographing module configured to photograph an image in response to acquiring a photographing instruction; and a light source drive configured to respond to acquiring a supplementary light instruction, Drive the light source connected to the light source driver to emit light according to the brightness parameter in the light supplement instruction; and the image processor, which is connected to the camera module and the light source driver, and is configured to perform when the shooting quality of the image captured by the camera module does not meet the preset conditions , Adjust the brightness parameter according to the image, and send a shooting instruction to the shooting module, and send a supplementary light instruction including the adjusted brightness parameter to the light source driver.
  • the image processor is further configured to adjust the brightness parameter according to the contrast and grayscale information of the image.
  • the image processor is further configured to: if the contrast of the image is lower than the preset value of contrast, if the proportion of pixels in the low grayscale area of the image is lower than the preset value of low grayscale, Decrease the brightness parameter; if the proportion of pixels in the high grayscale area of the image is higher than the preset value of high grayscale, increase the brightness parameter; or if the contrast of the image is lower than the preset value of contrast, if the image is gray If the average value of the gray scale is greater than the preset upper limit of gray scale, reduce the brightness parameter; if the average gray scale of the image is less than the preset lower limit of gray scale, increase the brightness parameter.
  • the image processor is further configured to, in response to receiving an information acquisition instruction from the main controller of the robot, send a shooting instruction to the shooting module and send a supplementary light instruction to the light source driver.
  • the image processor is further configured to recognize target information in the image when the shooting quality meets a preset condition.
  • the image processor is further configured to identify the category to which the device in the image belongs; determine the target area in the image; when the shooting quality meets a preset condition, adopt the category corresponding to the category to which the device in the image belongs.
  • the model recognizes the target information in the target area.
  • the image processor is further configured to use an adjustment step corresponding to the category of the device in the image to adjust the brightness parameter when the shooting quality of the target area of the image does not meet the preset condition.
  • the information collection device further includes: a fill light controller configured to obtain a fill light instruction from the image processor; if the fill light instruction does not include a brightness parameter, determine the brightness parameter and determine The brightness parameter of is added to the fill light command; and the fill light command is sent to the light source drive circuit.
  • a fill light controller configured to obtain a fill light instruction from the image processor; if the fill light instruction does not include a brightness parameter, determine the brightness parameter and determine The brightness parameter of is added to the fill light command; and the fill light command is sent to the light source drive circuit.
  • the information collection device further includes: a light intensity sensor configured to sense the intensity of the ambient light and send the intensity of the ambient light to the supplement light controller, so that the supplement light controller determines according to the intensity of the ambient light Brightness parameter.
  • the information collection device further includes a light source, which is connected to the light source drivingly, and is configured to emit light under the driving of the light source driving.
  • a patrol robot including: any one of the aforementioned information collection devices; and a main controller configured to receive and store target information in images from the information collection device .
  • the main controller is further configured to send an information acquisition instruction to the image processor of the information acquisition device, where the information acquisition instruction is used to instruct the image processor to send a shooting instruction to the shooting module and to send compensation to the light source driver.
  • Light instructions are used to send an information acquisition instruction to the image processor of the information acquisition device, where the information acquisition instruction is used to instruct the image processor to send a shooting instruction to the shooting module and to send compensation to the light source driver.
  • the main controller is further configured to instruct the inspection robot to drive along a preset path, where the path includes one or more information collection points; in response to the inspection robot reaching the information collection point, the image processing The device sends information acquisition instructions.
  • the main controller is further configured to control the inspection robot to stop or decelerate when the inspection robot reaches the information collection point; in response to receiving the target information in the image from the image processor, control the inspection The robot travels to the next information collection point at a preset speed.
  • an information collection method including: determining the shooting quality of the acquired image; in the case that the shooting quality does not meet a preset condition, adjusting the brightness used by the light source driving module according to the image Parameters, and send a shooting instruction to the shooting module, and send a supplementary light instruction including the adjusted brightness parameter to the light source driver, wherein the shooting instruction is used to instruct the photography module to shoot an image, and the supplementary light instruction is used to instruct the light source driver to drive the light source to emit light.
  • the brightness parameter is adjusted according to the contrast and gray information of the image.
  • adjusting the brightness parameter according to the contrast and grayscale information of the image includes: when the contrast of the image is lower than the preset value of the contrast, if the proportion of pixels in the low grayscale area of the image is lower than the low grayscale If the pixel ratio of the high-gray-scale area of the image is higher than the preset value of high-gray-scale, increase the brightness parameter; or when the contrast of the image is lower than the preset value of the contrast If the average gray value of the image is greater than the preset gray upper limit, reduce the brightness parameter; if the average gray value of the image is less than the preset gray lower limit, increase the brightness parameter.
  • the information collection method further includes: in response to receiving an information acquisition instruction from the main controller of the inspection robot, sending a shooting instruction to the shooting module and sending a supplementary light instruction to the light source driver.
  • the information collection method further includes: recognizing target information in the image when the shooting quality meets a preset condition.
  • the information collection method further includes: identifying the category to which the device in the image belongs, and determining the target area in the image; and when the shooting quality meets a preset condition, adopting the category corresponding to the device in the image The model recognizes the target information in the target area.
  • the adjustment step size corresponding to the category of the device in the image is used to adjust the brightness parameter.
  • the information collection method further includes: the main controller instructs the inspection robot to drive along a preset path, wherein the path includes one or more information collection points; in response to the inspection robot reaching the information collection point, the main controller The controller sends an information acquisition instruction.
  • the information collection method further includes: when the inspection robot arrives at the information collection point, the main controller controls the inspection robot to stop or decelerate; in response to acquiring the target information, the main controller controls the inspection robot according to a preset Speed to the next information collection point.
  • an information collection device including: a memory; and a processor coupled to the memory, and the processor is configured to execute any one of the foregoing based on instructions stored in the memory Information collection method.
  • a computer-readable storage medium having a computer program stored thereon, wherein, when the program is executed by a processor, any one of the foregoing information collection methods is implemented.
  • Fig. 1 is a schematic structural diagram of an information collection device according to some embodiments of the present disclosure.
  • Fig. 2 is a schematic structural diagram of an information collection device according to other embodiments of the present disclosure.
  • Fig. 3 is a schematic structural diagram of an information collection device according to still other embodiments of the present disclosure.
  • Fig. 4 is a schematic structural diagram of an information collection device according to still other embodiments of the present disclosure.
  • Fig. 5 is a schematic structural diagram of an information collection device according to still other embodiments of the present disclosure.
  • Fig. 6 is a schematic structural diagram of a patrol robot according to some embodiments of the present disclosure.
  • FIG. 7 is a schematic flowchart of an information collection method according to some embodiments of the present disclosure.
  • 8A and 8B are schematic flowcharts of methods for adjusting brightness parameters according to some embodiments of the present disclosure.
  • FIG. 9 is a schematic flowchart of a method for identifying target information according to some embodiments of the present disclosure.
  • FIG. 10 is a schematic flowchart of a polling method according to some embodiments of the present disclosure.
  • Fig. 11 is a schematic structural diagram of an information collection device according to still other embodiments of the present disclosure.
  • Fig. 12 is a schematic structural diagram of an information collection device according to still other embodiments of the present disclosure.
  • the inventor found that when the environment of the computer room is dark, the quality of the acquired image is low. This will reduce the recognition rate of the image, thereby making the collected information inaccurate and the collection efficiency relatively low.
  • the embodiments of the present disclosure can provide a more accurate and efficient information collection solution.
  • Fig. 1 is a schematic structural diagram of an information collection device according to some embodiments of the present disclosure.
  • the information collection device 10 of this embodiment includes a photographing module 110, a light source driver 120, and an image processor 130, wherein the image processor 130 is connected to the photographing module 110 and the light source driver 120, respectively.
  • the photographing module 110 is configured to photograph an image in response to acquiring a photographing instruction.
  • the light source driver 120 is configured to drive the light source connected to the light source drive to emit light according to the brightness parameter in the light supplement instruction in response to acquiring the light supplement instruction.
  • the photographing module 110 can take a picture in an environment with light.
  • the image processor 130 is configured to send a shooting command to the shooting module 110 and a supplementary light command to the light source driver 120 in response to receiving an information acquisition command from the main controller of the robot. Thereby, the image processor can instruct the shooting module and the light source driver to work together, so as to realize the acquisition of images under light conditions. According to needs, other devices or components may also send a shooting instruction to the shooting module 110.
  • the brightness parameter is set by the image processor 130, or set by other components in the information collection device 10 or by default.
  • the environment of the computer room is complex and changeable. If a fixed light intensity is used to fill light, there may still be situations where the quality of the captured image is low.
  • the image processor 130 in the information collection device of the embodiment of the present disclosure can further determine whether the captured image is available.
  • the image processor 130 is further configured to adjust the brightness parameters according to the image when the shooting quality of the image captured by the shooting module 110 does not meet a preset condition, that is, the shooting quality is poor, and to send shooting instructions to the shooting module and to the light source.
  • the driver sends the fill light command including the brightness parameter.
  • the image processor 130 is further configured to recognize the target information in the image when the shooting quality meets a preset condition, that is, the shooting quality is good.
  • the image processor 130 is further configured to determine the quality parameter of the image according to the contrast and gray information of the acquired image to determine whether the image quality meets the preset condition.
  • the contrast is relatively low, it is difficult to distinguish features such as edges and corners in the image during the recognition process, which makes it difficult to accurately recognize the information in the image. Therefore, the method of determining the quality parameter based on the contrast is adopted. For example, the contrast is directly used as the quality parameter, or the quality parameter is determined through a preset correspondence between the contrast and the quality parameter.
  • the contrast of the image may be low. If the ambient light intensity is low, it may cause more noise in the image, which is not conducive to the recognition of the target information in the image; if the ambient light intensity is too high, it may cause excessive saturation in some areas of the image, and it is not conducive to image segmentation and other processing . Therefore, in some embodiments, when the contrast of the image is lower than the preset value of the contrast, the gray level information of the image is further extracted to re-determine the brightness parameter.
  • the grayscale information Take the grayscale information as an example of the grayscale value of the image on each grayscale.
  • the pixel ratio of the low-gray-scale area of the image is lower than the preset value of the low-gray-scale, that is, when the low-gray-scale area of the grayscale histogram of the image is partially missing, it indicates that the current ambient light intensity is too high and the image is too high.
  • the gray information Take the gray information as the average gray value of the image as an example.
  • the average gray value of the image is greater than the preset gray upper limit, it indicates that the current ambient light intensity is too large and the image is too bright; when the image gray average is less than the preset gray lower limit, it indicates the current ambient light intensity The intensity is too small and the image is too dark.
  • the gray information of the image is obtained; the brightness parameter is re-determined according to the contrast and gray information of the image.
  • the corresponding relationship between the contrast, grayscale information and the adjustment value of the brightness parameter is preset, and the corresponding relationship is embodied by a correspondence table, formula, etc.; then, the adjustment value of the brightness parameter is used to compare the adjustment value of the brightness parameter used in the previous shooting. Brightness parameters are adjusted. When the image is too bright, the adjustment value of the brightness parameter is a negative number; when the image is too dark, the adjustment value of the brightness parameter is a positive number.
  • the image processor 130 is further configured to recognize the category of the device in the image through methods such as target recognition; determine the target area in the image; and use the image The model corresponding to the category to which the device belongs in identifies the target information in the target area.
  • the target information to be collected includes, for example, the information indicated by the indicator light, the display screen, and the label.
  • the target information may be expressed in different forms. Therefore, by determining the category of the device in the image to determine the model of the chassis, the target area can be determined according to the location or salient features of the target information of the model. For example, for indicator lights, the target information is recognized by a model with color recognition function; and for display screens or labels, the target information is recognized by a model with text recognition function.
  • the shooting quality of the image when determining the shooting quality of the image, it is based on the shooting quality of the target area of the image. For example, the contrast and gray information of the target area are calculated, so that the adjusted brightness parameters can be more accurate.
  • the adjustment step size corresponding to the category of the device in the image is used to adjust the brightness parameter.
  • the indicator light has fewer details, so you can use a larger step to adjust the brightness; another example, the text information has more details, so you can use a smaller step to adjust the brightness to avoid over-adjustment and miss the best brightness.
  • the photographing module and the light source driver can work together, so as to realize the image collection under the light condition.
  • the image processor detects that the quality of the captured photos is relatively low, by instructing the shooting module and the light source driving module to work again after adjusting the brightness parameters, higher quality images can be obtained, and the accuracy of image recognition can be improved. Sex. Therefore, the embodiments of the present disclosure improve the accuracy and efficiency of information collection.
  • a supplemental light controller is used to further control the driving of the light source.
  • Fig. 2 is a schematic structural diagram of an information collection device according to other embodiments of the present disclosure.
  • the information collection device 20 of this embodiment includes a photographing module 210, a light source driver 220, an image processor 230, and a light supplement controller 240.
  • the image processor 230 is connected to the photographing module 210 and the light supplement controller 240 respectively, and the light source driver 220 is connected to the light supplement controller 240.
  • the fill light controller 240 is configured to obtain the fill light command from the image processor; if the brightness parameter is not included in the fill light command, determine the brightness parameter and add the determined brightness parameter to the fill light command; and The fill light command is sent to the light source driving circuit.
  • the image processor 230 after the image processor 230 receives the information acquisition instruction from the main controller of the robot, if it is the first time to send the supplementary light instruction, since the current ambient light intensity is uncertain, it chooses not to set the supplementary light instruction.
  • the brightness parameters When the fill light controller 240 detects that the brightness parameter is not included in the fill light command, it determines the preset default brightness value as the value of the brightness parameter; or, obtains the current intensity of the ambient light from other modules, and determines the intensity of the current ambient light according to the current ambient light intensity. The intensity determines the brightness parameter, for example, by looking up a preset comparison table between the intensity of the ambient light and the brightness parameter value.
  • the information collection device 20 further includes a light intensity sensor 250 configured to sense the intensity of the ambient light and send the intensity of the ambient light to the supplement light controller 240, so that the supplement light controller 240 responds to the ambient light The intensity of determines the brightness parameter. Therefore, when the image processor does not instruct the brightness parameter, the supplementary light at the time of shooting is performed according to the intensity of the ambient light.
  • the information collection device 20 may further include a light source 260.
  • the light source 260 is connected to the light source driver 220 and is configured to emit light under the driving of the light source driver 220.
  • Fig. 3 is a schematic structural diagram of an information collection device according to still other embodiments of the present disclosure.
  • the information collection device 30 of this embodiment includes a photographing module 310, a light source driver 320, an image processor 330, and a memory 370.
  • the memory 370 is connected to the image processor 330 and is configured to store calculation data of the image processor 330. As a result, the computational efficiency of the image processor has been improved.
  • Fig. 4 is a schematic structural diagram of an information collection device according to still other embodiments of the present disclosure.
  • the information collection device 40 of this embodiment includes a photographing module 410, a light source driver 420, an image processor 430, and a physical layer controller 480. It is connected to the Ethernet interface of the image processor 430 and is configured to pass Ethernet The network realizes the communication between the image processor 430 and the main controller of the robot.
  • the physical layer controller 480 is a physical interface transceiver, which forms an Ethernet controller together with the Ethernet media intervention controller of the image processor 430, and realizes the function of the network port. Therefore, the information collection device 40 can flexibly communicate with other devices via Ethernet.
  • Fig. 5 is a schematic structural diagram of an information collection device according to still other embodiments of the present disclosure.
  • the information collection device 50 of this embodiment includes: a camera 510, a light source drive circuit 520, an NVIDIA Tegra image processor 530, a single-chip microcomputer 540, a light intensity sensor 550, a light source 560, and a fourth-generation low-power dual-data Rate (Low Power Double Data Rate 4, abbreviated as: LPDDR4) memory 570 and a port physical layer (Port Physical Layer, abbreviated as: PHY) controller 580.
  • LPDDR4 Low Power Double Data Rate 4
  • PHY port physical layer
  • the USB interface 5301 of the Tegra image processor 530 is connected to one end of the USB data line 591, and the other end of the USB data line 591 is connected to the camera 510.
  • the shooting instruction of the Tegra image processor 530 is transmitted to the camera 510 through the USB data line 591, and the image taken by the camera 510 is transmitted to the Tegra image processor 530 through the USB data line 591.
  • the CAN peripheral interface 5302 of the Tegra image processor 530 is connected to the first CAN transceiver 592, the single-chip microcomputer 540 as a light supplement controller is connected to the second CAN transceiver 593, between the first CAN transceiver 592 and the second CAN transceiver 593 It is connected via CAN bus 594.
  • the first CAN transceiver 592 and the second CAN transceiver 593 are configured to convert the logic level on the device side and the differential level on the CAN bus side.
  • the single-chip microcomputer 540 is respectively connected with the light intensity sensor 550 and the light source driving circuit 520, and the light source driving circuit 520 is connected with the light source 560.
  • the Ethernet interface 5303 of the Tegra image processor 530 is connected to the port physical layer controller 580.
  • the Tegra image processor 530 is also connected to the LPDDR4 memory 570.
  • the main controller of the inspection robot sends an information acquisition instruction to the Tegra image processor 530 via the Ethernet.
  • the Tegra image processor 530 After receiving the information acquisition instruction, the Tegra image processor 530 sends a shooting instruction to the camera 510 through the USB interface 5301, and sends a supplementary light instruction to the single-chip microcomputer 540 through the CAN peripheral interface 5302.
  • the single-chip microcomputer 540 finds that there is no brightness parameter in the light supplement command, so it obtains the intensity of the ambient light from the light intensity sensor 550, and determines the brightness parameter according to the intensity of the ambient light. After the single-chip microcomputer 540 adds the brightness parameter to the light supplement instruction, it sends the light supplement instruction to the light source driving circuit 520.
  • the light source driving circuit 520 drives the light source 560 to emit light according to the brightness parameter in the supplemental light command. Therefore, when the camera 510 captures an image according to the shooting instruction, the light source 560 performs supplementary light.
  • the camera 510 transmits the captured image to the Tegra image processor 530, and then the Tegra image processor 530 determines the shooting quality of the image and determines whether the shooting quality is greater than a preset value.
  • the shooting quality is greater than the preset value
  • the target information in the image is recognized, and the recognized information is sent to the main controller of the inspection robot through the Ethernet interface 5303.
  • the brightness parameter is re-determined according to the image, and the shooting instruction is sent to the camera 510, and the supplementary light instruction including the brightness parameter is sent to the light source driving circuit 520.
  • the single-chip microcomputer 540 finds that there is a brightness parameter in the light supplement command, and then sends the light supplement command to the light source driving circuit 520.
  • the light source driving circuit 520 drives the light source 560 to emit light according to the brightness parameter in the supplemental light command. Therefore, when the camera 510 captures an image according to the shooting instruction, the light source 560 performs supplementary light, and the supplementary light effect is adjusted according to the previous photographing quality and the photographed image.
  • the camera 510 transmits the image taken again to the Tegra image processor 530, and then the Tegra image processor 530 continues to process according to the aforementioned processing logic. If the shooting quality of the captured image obtained this time is still not greater than the preset value, the brightness parameter is re-determined again and the shooting and fill light process is triggered until an image with the shooting quality greater than the preset value is obtained.
  • the main controller After the main controller has obtained the target information of the current position, it drives the inspection robot to move to the next position and repeats the above process.
  • the light intensity sensor is used to obtain preliminary brightness parameters for supplementary light, and in the case where the quality of the initial shooting is low, the brightness parameters are adjusted and the shooting is performed again. Therefore, the target information in the high-quality image can be identified, and the accuracy and efficiency of information collection in the automatic inspection process of the computer room are improved.
  • Fig. 6 is a schematic structural diagram of a patrol robot according to some embodiments of the present disclosure.
  • the inspection robot 60 of this embodiment includes an information collection device 61 and a main controller 62.
  • the main controller 62 is configured to receive and store the target information in the image from the information collection device 61.
  • the main controller 62 is further configured to send an information acquisition instruction to the image processor of the information acquisition device, so that the image processor, in response to the information acquisition instruction, sends a photography instruction to the photographing module and sends compensation to the light source driver. Light instructions.
  • the main controller 62 is further configured to store a preset path and information collection points on the path; instruct the inspection robot to drive along the preset path; in response to reaching the information collection point, send a message to the information collection device
  • the image processor sends an information acquisition instruction.
  • the main controller 62 is further configured to control the patrol robot to stop or decelerate when the patrol robot reaches the information collection point; in response to receiving the target information in the image from the image processor, control the patrol The inspection robot travels to the next information collection point at a preset speed.
  • FIG. 7 is a schematic flowchart of an information collection method according to some embodiments of the present disclosure. As shown in FIG. 7, the information collection method of this embodiment includes steps S702 to S710.
  • step S702 in response to receiving an information acquisition instruction from the main controller of the robot, a shooting instruction is sent to the shooting module, and a supplementary light instruction is sent to the light source driver, so that the shooting module can shoot an image in response to the acquisition of the shooting instruction, and the light source drive The light source is driven to emit light in response to the acquisition of the supplementary light instruction.
  • step S702 is selectively executed as needed.
  • step S704 the shooting quality of the acquired image is determined.
  • the photographing quality of the image photographed by the photographing module is determined according to the contrast information of the image.
  • step S706 it is determined whether the shooting quality satisfies a preset condition.
  • step S708 when the shooting quality meets a preset condition, the target information in the image is recognized.
  • step S710 when the shooting quality does not meet the preset condition, adjust the brightness parameter used by the light source driving module according to the image, and send a shooting instruction to the shooting module and a supplementary light instruction including the adjusted brightness parameter to the light source driver.
  • step S710 after performing step S710, return to step S704 until the target information is identified when the shooting quality meets the preset condition.
  • the method in the foregoing embodiment instructs the photographing module and the light source driver to work together, so as to realize image collection under light conditions. Moreover, when the image processor detects that the quality of the photographed photo is relatively low, it instructs the photographing module and the light source driving module to work again after adjusting the brightness parameters to obtain a higher-quality image, thereby improving the accuracy of image recognition. Thus, the accuracy and efficiency of information collection are improved.
  • the shooting quality is not greater than the preset value, when the intensity of the ambient light is too high or too low, the contrast of the image may be low.
  • the gray level information of the image is acquired, and the brightness parameter is re-determined according to the contrast and gray level information of the image. The embodiments of adjusting the brightness parameter are described below with reference to FIGS. 8A and 8B respectively.
  • FIG. 8A shows a schematic flowchart of a method for adjusting a brightness parameter according to some embodiments of the present disclosure. As shown in FIG. 8A, the method of this embodiment includes steps S802 to S806.
  • step S802 the contrast of the image is compared with the preset value of the contrast. If the contrast of the image is lower than the preset value of the contrast, steps S804 and S806 are executed.
  • the contrast of the image is lower than the preset value of the contrast, it is determined that there is no need to take the image again.
  • step S804 if the proportion of pixels in the low-gray-scale area of the image is lower than the preset value of the low-gray-scale, the brightness parameter is reduced.
  • step S806 if the proportion of pixels in the high-gray-scale area of the image is higher than the preset value of the high-gray-scale, the brightness parameter is increased.
  • the brightness parameter can be reasonably re-determined, so that a higher quality image can be obtained in the next shooting.
  • FIG. 8B shows a schematic flowchart of a method for adjusting a brightness parameter according to some embodiments of the present disclosure. As shown in FIG. 8B, the method of this embodiment includes steps S812 to S818.
  • step S812 the contrast of the image is compared with the preset value of the contrast. If the contrast of the image is lower than the preset value of the contrast, step S814 is executed.
  • step S814 the average gray value of the image is compared with the preset gray upper limit and gray lower limit, respectively. If the average gray value of the image is greater than the preset gray upper limit, step S816 is executed; if the average gray value of the image is less than the preset gray lower limit, step S818 is executed.
  • step S816 the brightness parameter is reduced.
  • step S818 the brightness parameter is increased.
  • the brightness parameter can be reasonably re-determined, so that a higher quality image can be obtained in the next shooting.
  • FIG. 9 is a schematic flowchart of a method for identifying target information according to some embodiments of the present disclosure. As shown in FIG. 9, the target information identification method of this embodiment includes steps S902 to S906.
  • step S902 the category to which the device in the image belongs is identified.
  • sample images of devices of each category are obtained in advance, and categories are marked for the device regions in the images; then, the sample images are used to train a machine learning model for classification, such as a target detection model YOLO3 model. Therefore, the trained model can be used to identify the category to which the device in the image belongs.
  • a machine learning model for classification such as a target detection model YOLO3 model. Therefore, the trained model can be used to identify the category to which the device in the image belongs.
  • step S904 the target area in the image is determined.
  • a target detection algorithm is used to determine the target area in the image.
  • step S906 when the shooting quality meets the preset condition, the model corresponding to the category to which the device in the image belongs is used to identify the target information in the target area.
  • the adjustment step size corresponding to the category of the device in the image is used to adjust the brightness parameter.
  • FIG. 10 is a schematic flowchart of a polling method according to some embodiments of the present disclosure. As shown in FIG. 10, the inspection method of this embodiment includes steps S1002 to S1012.
  • step S1002 the main controller instructs the inspection robot to travel according to a preset path, where the path includes one or more information collection points.
  • step S1004 in response to the inspection robot reaching the information collection point, the main controller sends an information acquisition instruction to the image processor in the information collection device.
  • step S1006 the image processor, in response to receiving an information acquisition instruction from the main controller of the robot, sends a shooting instruction to the shooting module, and sends a supplementary light instruction to the light source driver, so that the shooting module captures the image in response to the acquisition of the shooting instruction ,
  • the light source drive drives the light source to emit light in response to the acquisition of the supplementary light command.
  • step S1008 the image processor determines whether to re-determine the brightness parameter according to the photographing quality of the photographed image, and re-fill light and photograph. In the case where the shooting quality of the image is greater than the preset value, the target information in the image is recognized.
  • step S1010 the image processor sends the identified target information to the main controller.
  • step S1012 in response to acquiring the target information, the main controller controls the inspection robot to travel to the next information collection point. In some embodiments, after performing step S1012, return to step S1004.
  • the main controller controls the patrol robot to stop or decelerate to obtain a higher quality image.
  • the main controller controls the inspection robot to drive to the next information collection point at a preset speed, so as to improve the efficiency of information collection.
  • the inspection robot automatically completes the high-quality image collection process, making the collected information more accurate and the collection process more efficient.
  • Fig. 11 is a schematic structural diagram of an information collection device according to still other embodiments of the present disclosure.
  • the information collection device 110 of this embodiment includes a memory 1110 and a processor 1120 coupled to the memory 1110.
  • the processor 1120 is configured to execute any one of the foregoing implementations based on instructions stored in the memory 1110.
  • the information collection method in the example is a schematic structural diagram of an information collection device according to still other embodiments of the present disclosure.
  • the information collection device 110 of this embodiment includes a memory 1110 and a processor 1120 coupled to the memory 1110.
  • the processor 1120 is configured to execute any one of the foregoing implementations based on instructions stored in the memory 1110.
  • the information collection method in the example is a schematic structural diagram of an information collection device according to still other embodiments of the present disclosure.
  • the information collection device 110 of this embodiment includes a memory 1110 and a processor 1120 coupled to the memory 1110.
  • the processor 1120 is configured to execute any one of the foregoing implementations based on instructions stored
  • the memory 1110 may include, for example, a system memory, a fixed non-volatile storage medium, and the like.
  • the system memory stores, for example, an operating system, an application program, a boot loader (Boot Loader), and other programs.
  • Fig. 12 is a schematic structural diagram of an information collection device according to still other embodiments of the present disclosure.
  • the information collection device 120 of this embodiment includes a memory 1210 and a processor 1220, and may also include an input/output interface 1230, a network interface 1240, a storage interface 1250, and the like. These interfaces 1230, 1240, 1250 and the memory 1210 and the processor 1220 may be connected via a bus 1260, for example.
  • the input and output interface 1230 provides a connection interface for input and output devices such as a display, a mouse, a keyboard, and a touch screen.
  • the network interface 1240 provides a connection interface for various networked devices.
  • the storage interface 1250 provides connection interfaces for external storage devices such as SD cards and U disks.
  • the embodiments of the present disclosure also provide a computer-readable storage medium on which a computer program is stored, characterized in that, when the program is executed by a processor, any one of the foregoing information collection methods is implemented.
  • the embodiments of the present disclosure can be provided as a method, a system, or a computer program product. Therefore, the present disclosure may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable non-transitory storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. .
  • These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing equipment to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction device.
  • the device implements the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.
  • These computer program instructions can also be loaded on a computer or other programmable data processing equipment, so that a series of operation steps are executed on the computer or other programmable equipment to produce computer-implemented processing, so as to execute on the computer or other programmable equipment.
  • the instructions provide steps for implementing the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram.

Abstract

本公开涉及一种信息采集装置、方法、巡检机器人和存储介质，涉及机器人技术领域。信息采集装置包括：拍摄模块，被配置为响应于获取拍摄指令而拍摄图像；光源驱动，被配置为响应于获取到补光指令，按照补光指令中的亮度参数驱动光源驱动连接的光源发光；图像处理器，与拍摄模块和光源驱动连接，被配置为确定拍摄模块拍摄的图像的拍摄质量；判断拍摄质量是否大于预设值；在拍摄质量大于预设值的情况下，识别图像中的目标信息；在拍摄质量不大于预设值的情况下，根据图像重新确定亮度参数，并向拍摄模块发送拍摄指令、向光源驱动发送包括亮度参数的补光指令。本公开能够提高信息采集的准确率和采集效率。

Description

信息采集装置、方法、巡检机器人和存储介质
相关申请的交叉引用
本申请是以CN申请号为CN201910799851.X，申请日为2019年8月28日的申请为基础，并主张其优先权，该CN申请的公开内容在此作为整体引入本申请中。
技术领域
本公开涉及机器人技术领域，特别涉及一种信息采集装置、方法、巡检机器人和存储介质。
背景技术
随着信息技术和互联网技术的发展，用于存储互联网数据业务的IDC(Internet Data Center，互联网数据中心)数据机房的数量也在快速增长。IDC数据机房的安全、高效运行，为互联网的稳定性提供了基本保障。
IDC巡检机器人是一种使用机器巡检替代人工巡检的方法，它具有高效、简单、不间断工作的特点。IDC机器人通过自动在机房内移动、并在移动过程中拍摄图像来采集主机的工作状态。
发明内容
本公开实施例所要解决的一个技术问题是：如何提高信息采集的准确率和采集效率。
根据本公开一些实施例的第一个方面，提供一种信息采集装置，包括：拍摄模块，被配置为响应于获取拍摄指令而拍摄图像；光源驱动，被配置为响应于获取到补光指令，按照补光指令中的亮度参数，驱动光源驱动连接的光源发光；以及图像处理器，与拍摄模块和光源驱动连接，被配置为在拍摄模块拍摄的图像的拍摄质量不满足预设条件的情况下，根据图像调整亮度参数，并向拍摄模块发送拍摄指令、以及向光源驱动发送包括调整的亮度参数的补光指令。
在一些实施例中，图像处理器进一步被配置为：根据图像的对比度和灰度信息调整亮度参数。
在一些实施例中，图像处理器进一步被配置为：在图像的对比度低于对比度的预 设值的情况下，如果图像的低灰阶区域的像素占比低于低灰阶的预设值，降低亮度参数；如果图像的高灰阶区域的像素占比高于高灰阶的预设值，升高亮度参数；或者在图像的对比度低于对比度的预设值的情况下，如果图像的灰度平均值大于预设的灰度上限，降低亮度参数；如果图像的灰度平均值小于预设的灰度下限，升高亮度参数。
在一些实施例中，图像处理器进一步被配置为响应于接收到来自机器人的主控制器的信息获取指令，向拍摄模块发送拍摄指令、并向光源驱动发送补光指令。
在一些实施例中，图像处理器进一步被配置为在拍摄质量满足预设条件的情况下，识别图像中的目标信息。
在一些实施例中，图像处理器进一步被配置为识别图像中的设备所属的类别；确定图像中的目标区域；在拍摄质量满足预设条件的情况下，采用图像中的设备所属的类别对应的模型识别目标区域中的目标信息。
在一些实施例中，图像处理器进一步被配置为在图像的目标区域的拍摄质量不满足预设条件的情况下，使用图像中的设备所属的类别对应的调整步长来调整亮度参数。
在一些实施例中，信息采集装置还包括：补光控制器，被配置为获取来自图像处理器的补光指令；在补光指令中不包括亮度参数的情况下，确定亮度参数、并将确定的亮度参数添加到补光指令中；以及将补光指令发送给光源驱动电路。
在一些实施例中，信息采集装置还包括：光强传感器，被配置为感测环境光的强度，并将环境光的强度发送给补光控制器，以便补光控制器根据环境光的强度确定亮度参数。
在一些实施例中，信息采集装置还包括：光源，与光源驱动连接，被配置为在光源驱动的驱动下发光。
根据本公开一些实施例的第二个方面，提供一种巡检机器人，包括：前述任意一种信息采集装置；以及主控制器，被配置为接收并存储来自信息采集装置的图像中的目标信息。
在一些实施例中，主控制器进一步被配置为向信息采集装置的图像处理器发送信息获取指令，其中，信息获取指令用于指示图像处理器向拍摄模块发送拍摄指令、并向光源驱动发送补光指令。
在一些实施例中，主控制器进一步被配置为指示巡检机器人按照预设的路径行驶，其中，路径中包括一个或多个信息采集点；响应于巡检机器人到达信息采集点，向图像处理器发送信息获取指令。
在一些实施例中，主控制器进一步被配置为在巡检机器人到达信息采集点时，控制巡检机器人停止或减速；响应于接收到来自图像处理器的、图像中的目标信息，控制巡检机器人按照预设速度行驶到下一个信息采集点。
根据本公开一些实施例的第三个方面，提供一种信息采集方法，包括：确定获取的图像的拍摄质量；在拍摄质量不满足预设条件的情况下，根据图像调整光源驱动模块使用的亮度参数，并向拍摄模块发送拍摄指令、向光源驱动发送包括调整的亮度参数的补光指令，其中，拍摄指令用于指示拍摄模块拍摄图像，补光指令用于指示光源驱动来驱动光源发光。
在一些实施例中，根据图像的对比度和灰度信息调整亮度参数。
在一些实施例中，根据图像的对比度和灰度信息调整亮度参数包括：在图像的对比度低于对比度的预设值的情况下，如果图像的低灰阶区域的像素占比低于低灰阶的预设值，降低亮度参数；如果图像的高灰阶区域的像素占比高于高灰阶的预设值，升高亮度参数；或者在图像的对比度低于对比度的预设值的情况下，如果图像的灰度平均值大于预设的灰度上限，降低亮度参数；如果图像的灰度平均值小于预设的灰度下限，升高亮度参数。
在一些实施例中，信息采集方法还包括：响应于接收到来自巡检机器人的主控制器的信息获取指令，向拍摄模块发送拍摄指令、并向光源驱动发送补光指令。
在一些实施例中，信息采集方法还包括：在拍摄质量满足预设条件的情况下，识别图像中的目标信息。
在一些实施例中，信息采集方法还包括：识别图像中的设备所属的类别，以及确定图像中的目标区域；以及在拍摄质量满足预设条件的情况下，采用图像中的设备所属的类别对应的模型识别目标区域中的目标信息。
在一些实施例中，在拍摄质量不满足预设条件的情况下，使用图像中的设备所属的类别对应的调整步长来调整亮度参数。
在一些实施例中，信息采集方法还包括：主控制器指示巡检机器人按照预设的路径行驶，其中，路径中包括一个或多个信息采集点；响应于巡检机器人到达信息采集点，主控制器发送信息获取指令。
在一些实施例中，信息采集方法还包括：在巡检机器人到达信息采集点时，主控制器控制巡检机器人停止或减速；响应于获取到目标信息，主控制器控制巡检机器人按照预设速度行驶到下一个信息采集点。
根据本公开一些实施例的第四个方面，提供一种信息采集装置，包括：存储器；以及耦接至存储器的处理器，处理器被配置为基于存储在存储器中的指令，执行前述任意一种信息采集方法。
根据本公开一些实施例的第五个方面，提供一种计算机可读存储介质，其上存储有计算机程序，其中，该程序被处理器执行时实现前述任意一种信息采集方法。
通过以下参照附图对本公开的示例性实施例的详细描述，本公开的其它特征及其优点将会变得清楚。
附图说明
为了更清楚地说明本公开实施例或相关技术中的技术方案，下面将对实施例或相关技术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公开的一些实施例，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为根据本公开一些实施例的信息采集装置的结构示意图。
图2为根据本公开另一些实施例的信息采集装置的结构示意图。
图3为根据本公开又一些实施例的信息采集装置的结构示意图。
图4为根据本公开再一些实施例的信息采集装置的结构示意图。
图5为根据本公开再一些实施例的信息采集装置的结构示意图。
图6为根据本公开一些实施例的巡检机器人的结构示意图。
图7为根据本公开一些实施例的信息采集方法的流程示意图。
图8A和8B为根据本公开一些实施例的调整亮度参数的方法的流程示意图。
图9为根据本公开一些实施例的目标信息识别方法的流程示意图。
图10为根据本公开一些实施例的巡检方法的流程示意图。
图11为根据本公开再一些实施例的信息采集装置的结构示意图。
图12为根据本公开再一些实施例的信息采集装置的结构示意图。
具体实施方式
下面将结合本公开实施例中的附图，对本公开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公开一部分实施例，而不是全部的实施例。以下对至少一个示例性实施例的描述实际上仅仅是说明性的，决不作为对本公开及其 应用或使用的任何限制。基于本公开中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本公开保护的范围。
除非另外具体说明，否则在这些实施例中阐述的部件和步骤的相对布置、数字表达式和数值不限制本公开的范围。
同时，应当明白，为了便于描述，附图中所示出的各个部分的尺寸并不是按照实际的比例关系绘制的。
对于相关领域普通技术人员已知的技术、方法和设备可能不作详细讨论，但在适当情况下，所述技术、方法和设备应当被视为授权说明书的一部分。
在这里示出和讨论的所有示例中，任何具体值应被解释为仅仅是示例性的，而不是作为限制。因此，示例性实施例的其它示例可以具有不同的值。
应注意到：相似的标号和字母在下面的附图中表示类似项，因此，一旦某一项在一个附图中被定义，则在随后的附图中不需要对其进行进一步讨论。
发明人经过进一步分析后发现，当机房的环境较暗时，获取的图像质量低。这会降低图像的识别率，从而使得采集的信息不准确、采集效率比较低。本公开的实施例能够提供一种更准确、高效的信息采集方案。
图1为根据本公开一些实施例的信息采集装置的结构示意图。如图1所示，该实施例的信息采集装置10包括拍摄模块110、光源驱动120和图像处理器130，其中，图像处理器130分别与拍摄模块110、光源驱动120连接。
拍摄模块110被配置为响应于获取拍摄指令而拍摄图像。光源驱动120被配置为响应于获取到补光指令，按照补光指令中的亮度参数驱动光源驱动连接的光源发光。从而，拍摄模块110可以在具有光照的环境中拍摄照片。
在一些实施例中，图像处理器130被配置为响应于接收到来自机器人的主控制器的信息获取指令，向拍摄模块110发送拍摄指令、并向光源驱动120发送补光指令。从而，图像处理器可以指示拍摄模块和光源驱动共同工作，以实现在有光照的条件下采集图像。根据需要，也可以由其他设备或部件向拍摄模块110发送拍摄指令。
在一些实施例中，亮度参数是图像处理器130设置的，或者是信息采集装置10中的其他部件设置的或默认的。
机房的环境是复杂且多变的。如果采用固定的光照强度进行补光，可能仍然存在拍摄的图像质量较低的情况。本公开实施例的信息采集装置中的图像处理器130能够进一步判定拍摄的图像是否可用。
图像处理器130进一步被配置为在拍摄模块110拍摄的图像的拍摄质量不满足预设条件、即拍摄质量较差的情况下，根据图像调整亮度参数，并向拍摄模块发送拍摄指令、以及向光源驱动发送包括亮度参数的补光指令。
在一些实施例中，图像处理器130进一步被配置为在拍摄质量满足预设条件、即拍摄质量较好的情况下，识别图像中的目标信息。
在一些实施例中，图像处理器130进一步被配置为根据获取的图像的对比度和灰度信息确定图像的质量参数，以判断图像质量是否满足预设条件。当对比度比较低时，在识别过程中难以分辨图像中的边缘、角点等特征，导致图像中的信息难以被准确地识别出来。因此，采用根据对比度确定质量参数的方式。例如，直接将对比度作为质量参数，或者通过对比度与质量参数之间的预设的对应关系来确定质量参数。
在拍摄质量不大于预设值的情况下，当环境光的强度过高或者过低时都可能会造成图像的对比度较低。如果环境光强度低，可能导致图像的噪点较多，不利于图像中目标信息的识别；如果环境光强度过大，可能造成图像中部分区域饱和度过高，也不利于对图像进行分割等处理。因此，在一些实施例中，当图像的对比度低于对比度的预设值时，通过进一步提取图像的灰度信息、以重新确定亮度参数。
以灰度信息为图像在每个灰阶上的灰度值为例。当图像的低灰阶区域的像素占比低于低灰阶的预设值时，即图像的灰度直方图的低灰阶区域存在部分缺失时，说明当前环境光的强度过大、图像过亮；当图像的高灰阶区域的像素占比低于高灰阶的预设值时，即图像的灰度直方图的高灰阶区域存在部分缺失时，说明当前环境光的强度过小、图像过暗。
以灰度信息为图像的灰度平均值为例。当图像的灰度平均值大于预设的灰度上限时，说明当前环境光的强度过大、图像过亮；当图像的灰度平均值小于预设的灰度下限时，说明当前环境光的强度过小、图像过暗。
在一些实施例中，获取图像的灰度信息；根据图像的对比度和灰度信息重新确定亮度参数。例如，预先设置对比度、灰度信息与亮度参数的调整值之间的对应关系，该对应关系例如通过对应关系表、公式等方式体现；然后，采用亮度参数调整值对上一次拍摄中所采用的亮度参数进行调整。当图像过亮时，亮度参数的调整值为负数；当图像过暗时，亮度参数的调整值为正数。
在拍摄质量大于预设值的情况下，在一些实施例中，图像处理器130进一步被配置为通过目标识别等方法，识别图像中的设备所属的类别；确定图像中的目标区域； 并采用图像中的设备所属的类别对应的模型识别目标区域中的目标信息。
在机房中，所要采集的目标信息例如包括指示灯、显示屏、标签所表示的信息，而不同型号的机箱上，目标信息可能采用不同的形式表示。因此，通过确定图像中的设备所属的类别确定机箱的型号，能够根据该型号的目标信息所在位置或者显著特征来确定目标区域。例如对于指示灯，通过具有颜色识别功能的模型识别目标信息；又例如对于显示屏或标签，通过具有文字识别功能的模型识别目标信息。
在一些实施例中，在判断图像的拍摄质量时，依据图像的目标区域的拍摄质量。例如，计算目标区域的对比度、灰度信息等，从而调整后的亮度参数能够更准确。
在一些实施例中，在图像的目标区域的拍摄质量不满足预设条件的情况下，使用图像中的设备所属的类别对应的调整步长来调整亮度参数。例如，指示灯的细节较少，因此可以采用较大的步长调整亮度；又例如，文字信息的细节较多，因此可以采用较小的步长调整亮度，避免因过度调整而错过最佳的亮度。
通过上述实施例的方法，拍摄模块和光源驱动可以共同工作，以实现在有光照的条件下采集图像。并且，当图像处理器检测到拍摄的照片的质量比较低时，通过在调整亮度参数后指示拍摄模块和光源驱动模再次工作，能够获取更高质量的图像、继而提高了对图像进行识别的准确性。因此，本公开的实施例提高了信息采集的准确率和采集效率。
在一些实施例中，采用补光控制器来进一步控制光源驱动。图2为根据本公开另一些实施例的信息采集装置的结构示意图。如图2所示，该实施例的信息采集装置20包括拍摄模块210、光源驱动220、图像处理器230和补光控制器240。图像处理器230分别与拍摄模块210和补光控制器240连接，光源驱动220与补光控制器240连接。补光控制器240被配置为获取来自图像处理器的补光指令；在补光指令中不包括亮度参数的情况下，确定亮度参数、并将确定的亮度参数添加到补光指令中；以及将补光指令发送给光源驱动电路。
在一些实施例中，图像处理器230在接收到来自机器人的主控制器的信息获取指令后，如果是首次发送补光指令，由于不确定当前环境光的强度，因此选择不设置补光指令中的亮度参数。补光控制器240检测到补光指令中不包括亮度参数时，将预设的默认亮度值确定为亮度参数的值；或者，从其他模块中获取当前环境光的强度、并根据当前环境光的强度确定亮度参数，例如，通过查找预设的环境光的强度与亮度参数值的对照表来确定。
在一些实施例中，信息采集装置20还包括光强传感器250，被配置为感测环境光的强度，并将环境光的强度发送给补光控制器240，以便补光控制器240根据环境光的强度确定亮度参数。从而，在图像处理器未指示亮度参数的情况下，根据环境光的强度进行拍摄时的补光。
在一些实施例中，信息采集装置20还可以包括光源260。光源260与光源驱动220连接，被配置为在光源驱动220的驱动下发光。
图3为根据本公开又一些实施例的信息采集装置的结构示意图。如图3所示，该实施例的信息采集装置30包括拍摄模块310、光源驱动320、图像处理器330和内存370。内存370与图像处理器330连接，被配置为存储图像处理器330的计算数据。从而，图像处理器的计算效率得到了提升。
图4为根据本公开再一些实施例的信息采集装置的结构示意图。如图4所示，该实施例的信息采集装置40包括拍摄模块410、光源驱动420、图像处理器430和物理层控制器480，与图像处理器430的以太网接口连接，被配置为通过以太网实现图像处理器430与机器人的主控制器之间的通信。物理层控制器480是物理接口收发器，与图像处理器430的以太网媒体介入控制器共同构成了以太网控制器，实现了网口的功能。从而，信息采集装置40能够通过以太网灵活地与其他设备进行通信。
本公开的实施例提供的信息采集装置中的各个模块和器件可以由更具体的设备实现。下面参考图5描述本公开信息采集装置的一种实现方式。
图5为根据本公开再一些实施例的信息采集装置的结构示意图。如图5所示，该实施例的信息采集装置50包括：照相机510、光源驱动电路520、英伟达Tegra图像处理器530、单片机540、光强传感器550、光源560、第四代低功耗双数据速率(Low Power Double Data Rate 4，简称：LPDDR4)内存570和端口物理层(Port Physical Layer，简称：PHY)控制器580。
Tegra图像处理器530的USB接口5301连接USB数据线591的一端，USB数据线591的另一端连接照相机510。从而，Tegra图像处理器530的拍摄指令通过USB数据线591传输给照相机510，照相机510拍摄的图像通过USB数据线591传输给Tegra图像处理器530。
Tegra图像处理器530的CAN外设接口5302连接第一CAN收发器592，作为补光控制器的单片机540连接第二CAN收发器593，第一CAN收发器592和第二CAN收发器593之间通过CAN总线594连接。第一CAN收发器592和第二CAN收发器 593被配置为对设备侧的逻辑电平、CAN总线侧的差分电平进行转换。
单片机540分别与光强传感器550和光源驱动电路520连接，光源驱动电路520和光源560连接。
Tegra图像处理器530的以太网接口5303连接端口物理层控制器580。此外，Tegra图像处理器530还与LPDDR4内存570连接。
下面结合图5，描述一个示例性的信息采集过程。
当巡检机器人按照巡检路径到达预设的位置时，巡检机器人的主控制器通过以太网向Tegra图像处理器530发送信息获取指令。
Tegra图像处理器530在接收到信息获取指令之后，通过USB接口5301向照相机510发送拍摄指令、并通过CAN外设接口5302向单片机540发送补光指令。
单片机540发现补光指令中没有亮度参数，于是从光强传感器550处获得环境光的强度，并根据环境光的强度确定亮度参数。单片机540将亮度参数添加到补光指令后，将补光指令发送给光源驱动电路520。光源驱动电路520按照补光指令中的亮度参数驱动光源560发光。从而，照相机510根据拍摄指令拍摄图像时，由光源560进行补光。
照相机510将拍摄的图像传送给Tegra图像处理器530，然后Tegra图像处理器530确定图像的拍摄质量、并判断拍摄质量是否大于预设值。在拍摄质量大于预设值的情况下，识别图像中的目标信息，并将识别出的信息通过以太网接口5303发送给巡检机器人的主控制器。在拍摄质量不大于预设值的情况下，根据图像重新确定亮度参数，并向照相机510发送拍摄指令、向光源驱动电路520发送包括亮度参数的补光指令。
单片机540发现补光指令中具有亮度参数，于是将补光指令发送给光源驱动电路520。光源驱动电路520按照补光指令中的亮度参数驱动光源560发光。从而，照相机510根据拍摄指令拍摄图像时，有光源560进行补光，并且该补光效果是根据上一次的拍摄质量和拍摄的图像进行过调整的。
照相机510将再次拍摄的图像传送给Tegra图像处理器530，然后Tegra图像处理器530继续按照前述处理逻辑进行处理。如果这一次获得的拍摄图像的拍摄质量仍然不大于预设值，则再次重新确定亮度参数并触发拍摄和补光过程，直到获得拍摄质量大于预设值的图像。
当主控制器已经获取到当前位置的目标信息后，驱动巡检机器人移动到下一个位 置，并重复上述过程。
上述实施例利用光强传感器获得初步的用于补光的亮度参数，并且在初次拍摄质量较低的情况下，对亮度参数进行调整并再次拍摄。从而，可以对高质量图像中的目标信息进行识别，提高了在机房的自动巡检过程中信息采集的准确率和采集效率。
下面参考图6描述本公开巡检机器人的实施例。
图6为根据本公开一些实施例的巡检机器人的结构示意图。如图6所示，该实施例的巡检机器人60包括信息采集装置61和主控制器62。主控制器62被配置为接收并存储来自信息采集装置61的图像中的目标信息。
在一些实施例中，主控制器62还被配置为向信息采集装置的图像处理器发送信息获取指令，以便图像处理器响应于信息获取指令，向拍摄模块发送拍摄指令、并向光源驱动发送补光指令。
在一些实施例中，主控制器62还被配置为存储预设的路径和路径上的信息采集点；指示巡检机器人按照预设的路径行驶；响应于到达信息采集点，向信息采集装置的图像处理器发送信息获取指令。
在一些实施例中，主控制器62还被配置为在巡检机器人到达信息采集点时，控制巡检机器人停止或减速；响应于接收到来自图像处理器的、图像中的目标信息，控制巡检机器人按照预设速度行驶到下一个信息采集点。
下面参考图7描述本公开信息采集方法的实施例。
图7为根据本公开一些实施例的信息采集方法的流程示意图。如图7所示，该实施例的信息采集方法包括步骤S702～S710。
在步骤S702中，响应于接收到来自机器人的主控制器的信息获取指令，向拍摄模块发送拍摄指令、并向光源驱动发送补光指令，以便拍摄模块响应于获取拍摄指令而拍摄图像、光源驱动响应于获取到补光指令驱动光源发光。
在一些实施例中，步骤S702根据需要选择性地执行。
在步骤S704中，确定获取的图像的拍摄质量。
在一些实施例中，根据图像的对比度信息确定拍摄模块拍摄的图像的拍摄质量。
在步骤S706中，判断拍摄质量是否满足预设条件。
在步骤S708中，在拍摄质量满足预设条件的情况下，识别图像中的目标信息。
在步骤S710中，在拍摄质量不满足预设条件的情况下，根据图像调整光源驱动模块使用的亮度参数，并向拍摄模块发送拍摄指令、向光源驱动发送包括调整的亮度 参数的补光指令。在一些实施例中，在执行步骤S710后，回到步骤S704，直到在拍摄质量满足预设条件的情况下识别出目标信息。
上述实施例的方法指示拍摄模块和光源驱动共同工作，以实现在有光照的条件下采集图像。并且，当图像处理器检测到拍摄的照片的质量比较低时，在调整亮度参数后指示拍摄模块和光源驱动模再次工作，以获取更高质量的图像、继而提高对图像进行识别的准确性。从而，提高了信息采集的准确率和采集效率。
在拍摄质量不大于预设值的情况下，当环境光的强度过高或者过低时都可能会造成图像的对比度较低。在一些实施例中，获取图像的灰度信息，并根据图像的对比度和灰度信息重新确定亮度参数。下面分别参考图8A和8B描述调整亮度参数的实施例。
图8A示出了根据本公开一些实施例的亮度参数调整方法的流程示意图。如图8A所示，该实施例的方法包括步骤S802～S806。
在步骤S802中，将图像的对比度和对比度的预设值进行比较。如果图像的对比度低于对比度的预设值，执行步骤S804和S806。
在一些实施例中，如果图像的对比度低于对比度的预设值，判定不需要重新拍摄图像。
在步骤S804中，如果图像的低灰阶区域的像素占比低于低灰阶的预设值，降低亮度参数。
在步骤S806中，如果图像的高灰阶区域的像素占比高于高灰阶的预设值，升高亮度参数。
在图像的灰度直方图的低灰阶区域存在部分缺失时，说明当前环境光的强度过大、图像过亮；在图像的灰度直方图的高灰阶区域存在部分缺失时，说明当前环境光的强度过小、图像过暗。通过上述实施例，能够合理地重新确定亮度参数，以使得下一次拍摄能够获得质量更高的图像。
图8B示出了根据本公开一些实施例的亮度参数调整方法的流程示意图。如图8B所示，该实施例的方法包括步骤S812～S818。
在步骤S812中，将图像的对比度和对比度的预设值进行比较。如果图像的对比度低于对比度的预设值，执行步骤S814。
在步骤S814中，将图像的灰度平均值分别与预设的灰度上限和灰度下限进行比较。如果图像的灰度平均值大于预设的灰度上限，执行步骤S816；如果图像的灰度平均值小于预设的灰度下限，执行步骤S818。
在步骤S816中，降低亮度参数。
在步骤S818中，升高亮度参数。
当图像的灰度平均值大于预设的灰度上限时，说明当前环境光的强度过大、图像过亮；当图像的灰度平均值小于预设的灰度下限时，说明当前环境光的强度过小、图像过暗。通过上述实施例，能够合理地重新确定亮度参数，以使得下一次拍摄能够获得质量更高的图像。
下面参考图9描述本公开目标信息识别方法的实施例。
图9为根据本公开一些实施例的目标信息识别方法的流程示意图。如图9所示，该实施例的目标信息识别方法包括步骤S902～S906。
在步骤S902中，识别图像中的设备所属的类别。
在一些实施例中，预先获取每个类别的设备的样本图像，并且为图像中的设备区域标记类别；然后，采用样本图像训练用于分类的机器学习模型，例如目标检测模型YOLO3模型。从而，可以采用完成训练的模型识别图像中的设备所属的类别。
在步骤S904中，确定图像中的目标区域。例如，通过目标检测算法来确定图像中的目标区域。
在步骤S906中，在拍摄质量满足预设条件的情况下，采用图像中的设备所属的类别对应的模型识别目标区域中的目标信息。
在一些实施例中，在拍摄质量不满足预设条件的情况下，使用图像中的设备所属的类别对应的调整步长来调整亮度参数。
通过对不同类型的设备采用相应的模型进行识别，提高了信息采集的准确率。
下面参考图10描述本公开巡检机器人的巡检过程。
图10为根据本公开一些实施例的巡检方法的流程示意图。如图10所示，该实施例的巡检方法包括步骤S1002～S1012。
在步骤S1002中，主控制器指示巡检机器人按照预设的路径行驶，其中，路径中包括一个或多个信息采集点。
在步骤S1004中，响应于巡检机器人到达信息采集点，主控制器向信息采集装置中的图像处理器发送信息获取指令。
在步骤S1006中，图像处理器响应于接收到来自机器人的主控制器的信息获取指令，向拍摄模块发送拍摄指令、并向光源驱动发送补光指令，以便拍摄模块响应于获取拍摄指令而拍摄图像、光源驱动响应于获取到补光指令驱动光源发光。
在步骤S1008中，图像处理器根据拍摄的图像的拍摄质量确定是否重新确定亮度参数并重新补光和拍摄。在图像的拍摄质量大于预设值的情况下，识别图像中的目标信息。
在步骤S1010中，图像处理器将识别出的目标信息发送给主控制器。
在步骤S1012中，响应于获取到目标信息，主控制器控制巡检机器人行驶到下一个信息采集点。在一些实施例中，在执行步骤S1012后，回到步骤S1004。
在一些实施例中，在巡检机器人到达信息采集点时，主控制器控制巡检机器人停止或减速，以获得拍摄质量更高的图像。响应于获取到目标信息，主控制器控制巡检机器人按照预设速度行驶到下一个信息采集点，以便提高信息采集的效率。
从而，巡检机器人自动地完成高质量图像采集过程，使得采集的信息更准确、采集过程更高效。
图11为根据本公开再一些实施例的信息采集装置的结构示意图。如图11所示，该实施例的信息采集装置110包括：存储器1110以及耦接至该存储器1110的处理器1120，处理器1120被配置为基于存储在存储器1110中的指令，执行前述任意一个实施例中的信息采集方法。
其中，存储器1110例如可以包括系统存储器、固定非易失性存储介质等。系统存储器例如存储有操作系统、应用程序、引导装载程序(Boot Loader)以及其他程序等。
图12为根据本公开再一些实施例的信息采集装置的结构示意图。如图12所示，该实施例的信息采集装置120包括：存储器1210以及处理器1220，还可以包括输入输出接口1230、网络接口1240、存储接口1250等。这些接口1230，1240，1250以及存储器1210和处理器1220之间例如可以通过总线1260连接。其中，输入输出接口1230为显示器、鼠标、键盘、触摸屏等输入输出设备提供连接接口。网络接口1240为各种联网设备提供连接接口。存储接口1250为SD卡、U盘等外置存储设备提供连接接口。
本公开的实施例还提供一种计算机可读存储介质，其上存储有计算机程序，其特征在于，该程序被处理器执行时实现前述任意一种信息采集方法。
本领域内的技术人员应当明白，本公开的实施例可提供为方法、系统、或计算机程序产品。因此，本公开可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本公开可采用在一个或多个其中包含有计算机可用程 序代码的计算机可用非瞬时性存储介质(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。
本公开是参照根据本公开实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解为可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
以上所述仅为本公开的较佳实施例，并不用以限制本公开，凡在本公开的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本公开的保护范围之内。

Claims (25)

  1. 一种信息采集装置，包括：
    拍摄模块，被配置为响应于获取拍摄指令而拍摄图像；
    光源驱动，被配置为响应于获取到补光指令，按照所述补光指令中的亮度参数，驱动所述光源驱动连接的光源发光；以及
    图像处理器，与所述拍摄模块和所述光源驱动连接，被配置为在所述拍摄模块拍摄的图像的拍摄质量不满足预设条件的情况下，根据所述图像调整所述亮度参数，并向拍摄模块发送拍摄指令、以及向光源驱动发送包括调整的亮度参数的补光指令。
  2. 根据权利要求1所述的信息采集装置，其中，所述图像处理器进一步被配置为：根据所述图像的对比度和灰度信息调整所述亮度参数。
  3. 根据权利要求2所述的信息采集装置，其中，所述图像处理器进一步被配置为：
    在图像的对比度低于对比度的预设值的情况下，如果图像的低灰阶区域的像素占比低于低灰阶的预设值，降低所述亮度参数；如果图像的高灰阶区域的像素占比高于高灰阶的预设值，升高所述亮度参数；或者
    在图像的对比度低于对比度的预设值的情况下，如果图像的灰度平均值大于预设的灰度上限，降低所述亮度参数；如果图像的灰度平均值小于预设的灰度下限，升高所述亮度参数。
  4. 根据权利要求1所述的信息采集装置，其中，所述图像处理器进一步被配置为响应于接收到来自机器人的主控制器的信息获取指令，向所述拍摄模块发送拍摄指令、并向所述光源驱动发送补光指令。
  5. 根据权利要求1所述的信息采集装置，其中，所述图像处理器进一步被配置为在所述拍摄质量满足预设条件的情况下，识别所述图像中的目标信息。
  6. 根据权利要求1所述的信息采集装置，其中，所述图像处理器进一步被配置为识别所述图像中的设备所属的类别；确定所述图像中的目标区域；在所述拍摄质量满足预设条件的情况下，采用所述图像中的设备所属的类别对应的模型识别所述目标区域中的目标信息。
  7. 根据权利要求6所述的信息采集装置，其中，所述图像处理器进一步被配置为在所述图像的目标区域的拍摄质量不满足预设条件的情况下，使用所述图像中的设备所属的类别对应的调整步长来调整所述亮度参数。
  8. 根据权利要求1所述的信息采集装置，还包括：
    补光控制器，被配置为获取来自图像处理器的补光指令；在所述补光指令中不包括亮度参数的情况下，确定亮度参数、并将确定的亮度参数添加到补光指令中；以及将补光指令发送给光源驱动电路。
  9. 根据权利要求8所述的信息采集装置，还包括：
    光强传感器，被配置为感测环境光的强度，并将环境光的强度发送给所述补光控制器，以便补光控制器根据所述环境光的强度确定亮度参数。
  10. 根据权利要求1所述的信息采集装置，还包括：
    光源，与所述光源驱动连接，被配置为在所述光源驱动的驱动下发光。
  11. 一种巡检机器人，包括：
    权利要求1～10中任一项所述的信息采集装置；以及
    主控制器，被配置为接收并存储来自所述信息采集装置的图像中的目标信息。
  12. 根据权利要求11所述的巡检机器人，其中，所述主控制器进一步被配置为向所述信息采集装置的图像处理器发送信息获取指令，其中，所述信息获取指令用于指示所述图像处理器向所述拍摄模块发送拍摄指令、并向所述光源驱动发送补光指令。
  13. 根据权利要求12所述的巡检机器人，其中，所述主控制器进一步被配置为指示所述巡检机器人按照预设的路径行驶，其中，所述路径中包括一个或多个信息采集点；响应于所述巡检机器人到达信息采集点，向所述图像处理器发送所述信息获取指令。
  14. 根据权利要求13所述的巡检机器人，其中，所述主控制器进一步被配置为在所述巡检机器人到达信息采集点时，控制所述巡检机器人停止或减速；响应于接收到来自所述图像处理器的、所述图像中的目标信息，控制所述巡检机器人按照预设速度行驶到下一个信息采集点。
  15. 一种信息采集方法，包括：
    确定获取的图像的拍摄质量；
    在所述拍摄质量不满足预设条件的情况下，根据所述图像调整光源驱动模块使用的亮度参数，并向拍摄模块发送拍摄指令、向所述光源驱动发送包括调整的亮度参数的补光指令，其中，所述拍摄指令用于指示所述拍摄模块拍摄图像，所述补光指令用于指示所述光源驱动来驱动光源发光。
  16. 根据权利要求15所述的信息采集方法，其中，根据所述图像的对比度和灰度 信息调整所述亮度参数。
  17. 根据权利要求16所述的信息采集方法，其中，所述根据所述图像的对比度和灰度信息调整所述亮度参数包括：
    在图像的对比度低于对比度的预设值的情况下，如果图像的低灰阶区域的像素占比低于低灰阶的预设值，降低所述亮度参数；如果图像的高灰阶区域的像素占比高于高灰阶的预设值，升高所述亮度参数；或者
    在图像的对比度低于对比度的预设值的情况下，如果图像的灰度平均值大于预设的灰度上限，降低所述亮度参数；如果图像的灰度平均值小于预设的灰度下限，升高所述亮度参数。
  18. 根据权利要求15所述的信息采集方法，还包括：
    响应于接收到来自巡检机器人的主控制器的信息获取指令，向所述拍摄模块发送拍摄指令、并向所述光源驱动发送补光指令。
  19. 根据权利要求15所述的信息采集方法，还包括：
    在所述拍摄质量满足预设条件的情况下，识别所述图像中的目标信息。
  20. 根据权利要求19所述的信息采集方法，其中：
    所述信息采集方法还包括：识别所述图像中的设备所属的类别，以及确定所述图像中的目标区域；以及
    在所述拍摄质量满足预设条件的情况下，采用所述图像中的设备所属的类别对应的模型识别所述目标区域中的目标信息。
  21. 根据权利要求20所述的信息采集方法，其中，在所述拍摄质量不满足预设条件的情况下，使用所述图像中的设备所属的类别对应的调整步长来调整所述亮度参数。
  22. 根据权利要求16所述的信息采集方法，还包括：
    主控制器指示所述巡检机器人按照预设的路径行驶，其中，所述路径中包括一个或多个信息采集点；
    响应于所述巡检机器人到达信息采集点，主控制器发送信息获取指令。
  23. 根据权利要求22所述的信息采集方法，还包括：
    在所述巡检机器人到达信息采集点时，主控制器控制所述巡检机器人停止或减速；
    响应于获取到目标信息，主控制器控制所述巡检机器人按照预设速度行驶到下一个信息采集点。
  24. 一种信息采集装置，包括：
    存储器；以及
    耦接至所述存储器的处理器，所述处理器被配置为基于存储在所述存储器中的指令，执行如权利要求16～21中任一项所述的信息采集方法。
  25. 一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现权利要求16～21中任一项所述的信息采集方法。
PCT/CN2020/109214 2019-08-28 2020-08-14 信息采集装置、方法、巡检机器人和存储介质 WO2021036824A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US17/634,470 US20220303447A1 (en) 2019-08-28 2020-08-14 Information acquisition device, method, patrol robot and storage medium
EP20857764.3A EP4024848A4 (en) 2019-08-28 2020-08-14 INFORMATION COLLECTION APPARATUS AND METHOD, INSPECTION ROBOT AND STORAGE MEDIUM

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910799851.X 2019-08-28
CN201910799851.XA CN110460782B (zh) 2019-08-28 2019-08-28 信息采集装置、方法、巡检机器人和存储介质

Publications (1)

Publication Number Publication Date
WO2021036824A1 true WO2021036824A1 (zh) 2021-03-04

Family

ID=68489532

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/109214 WO2021036824A1 (zh) 2019-08-28 2020-08-14 信息采集装置、方法、巡检机器人和存储介质

Country Status (4)

Country Link
US (1) US20220303447A1 (zh)
EP (1) EP4024848A4 (zh)
CN (1) CN110460782B (zh)
WO (1) WO2021036824A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113311861A (zh) * 2021-05-14 2021-08-27 国家电投集团青海光伏产业创新中心有限公司 光伏组件隐裂特性的自动化检测方法及其系统
CN114040093A (zh) * 2021-10-20 2022-02-11 上海闻泰电子科技有限公司 摄像头测试装置、方法、电子设备和计算机可读存储介质

Families Citing this family (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110460782B (zh) * 2019-08-28 2021-07-20 北京海益同展信息科技有限公司 信息采集装置、方法、巡检机器人和存储介质
CN112825491B (zh) * 2019-11-21 2022-05-24 北京外号信息技术有限公司 用于实现发光装置检测的方法和系统
CN113132613A (zh) * 2019-12-31 2021-07-16 中移物联网有限公司 一种摄像头补光装置、电子设备和补光方法
CN111126417A (zh) * 2020-01-17 2020-05-08 苏州浪潮智能科技有限公司 一种数据中心机房管理方法及装置
CN111442742B (zh) * 2020-04-16 2021-11-02 创新奇智(重庆)科技有限公司 齿轮检测设备及方法
CN111866396B (zh) * 2020-08-27 2022-04-12 路邦数码有限公司 一种协助拍摄机器人
CN112562304B (zh) * 2020-11-26 2022-02-18 英博超算(南京)科技有限公司 一种应用层与传感器数据的交互系统
CN114694296A (zh) * 2020-12-31 2022-07-01 深圳怡化电脑股份有限公司 图像采集方法、装置、电子设备和存储介质
CN113709359B (zh) * 2021-07-14 2023-09-01 安克创新科技股份有限公司 图像采集装置
CN113724368B (zh) * 2021-07-23 2023-02-07 北京百度网讯科技有限公司 图像采集系统、三维重建方法、装置、设备以及存储介质
CN114005026A (zh) * 2021-09-29 2022-02-01 达闼科技（北京）有限公司 机器人的图像识别方法、装置、电子设备及存储介质
CN114167889B (zh) * 2021-11-29 2023-03-07 内蒙古易飞航空科技有限公司 基于图像ai与大数据应用的智能巡检飞行平台
CN114397301A (zh) * 2021-12-20 2022-04-26 苏州镁伽科技有限公司 用于配置光源的方法、装置、存储介质及检测设备
CN114827485A (zh) * 2022-04-25 2022-07-29 智道网联科技(北京)有限公司 车辆摄像头的补光方法及系统
CN115278063A (zh) * 2022-07-08 2022-11-01 深圳市施罗德工业集团有限公司 一种巡检方法、巡检装置及巡检机器人
CN116708752B (zh) * 2022-10-28 2024-02-27 荣耀终端有限公司 针对成像装置的成像效果测试方法、装置及系统
CN116684251B (zh) * 2023-08-01 2023-11-24 广州市羿资互联网科技有限公司 一种idc的故障监测系统及其方法
CN117615484B (zh) * 2024-01-24 2024-05-03 金品计算机科技(天津)有限公司 一种基于ai的视觉检测光源控制方法、系统、设备及介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090028387A1 (en) * 2007-07-24 2009-01-29 Samsung Electronics Co., Ltd. Apparatus and method for recognizing position of mobile robot
CN109531533A (zh) * 2018-11-30 2019-03-29 北京海益同展信息科技有限公司 一种机房巡检系统及其工作方法
CN109685709A (zh) * 2018-12-28 2019-04-26 深圳市商汤科技有限公司 一种智能机器人的照明控制方法及装置
CN109920220A (zh) * 2017-12-11 2019-06-21 深圳市海洋王照明工程有限公司 一种智能巡检装置及系统
CN110460782A (zh) * 2019-08-28 2019-11-15 北京海益同展信息科技有限公司 信息采集装置、方法、巡检机器人和存储介质

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101474064B (zh) * 2008-09-18 2011-12-07 中国计量学院机电工程学院 电子体温计检定方法及装置
CN101750848B (zh) * 2008-12-11 2011-03-30 鸿富锦精密工业（深圳）有限公司 摄像装置及其补光方法
CN104144290B (zh) * 2013-05-10 2017-10-31 上海弘视通信技术有限公司 智能交通领域双模式多目标的成像控制方法及其装置
CN103347152A (zh) * 2013-07-08 2013-10-09 华为终端有限公司 一种图像处理方法、装置及终端
CN104092953B (zh) * 2014-07-10 2018-09-04 北京智谷睿拓技术服务有限公司 调光控制方法和装置及具有拍照功能的设备
EP3195594B1 (en) * 2014-09-17 2019-08-14 SZ DJI Technology Co., Ltd. Automatic white balancing system and method
CN104199453B (zh) * 2014-09-27 2016-08-17 江苏华宏实业集团有限公司 用于巡检电力仪表的智能机器人
CN105137387B (zh) * 2015-04-01 2017-08-04 深圳龙电电气股份有限公司 室外电能表故障自动检验方法
CN106713780A (zh) * 2017-01-16 2017-05-24 维沃移动通信有限公司 一种闪光灯的控制方法及移动终端
CN109429045B (zh) * 2017-08-30 2021-11-09 深圳光峰科技股份有限公司 图像处理及显示装置、图像处理及显示方法
US10384346B2 (en) * 2017-09-08 2019-08-20 Niantic, Inc. Collision detection, estimation, and avoidance
CN109326125B (zh) * 2018-11-14 2021-07-16 太原市高远时代科技有限公司 基于嵌入式系统的图片质量诊断系统及诊断方法
CN110855900B (zh) * 2019-11-20 2021-04-20 何建军 一种基于外置闪光光源的拍照控制方法、系统和存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090028387A1 (en) * 2007-07-24 2009-01-29 Samsung Electronics Co., Ltd. Apparatus and method for recognizing position of mobile robot
CN109920220A (zh) * 2017-12-11 2019-06-21 深圳市海洋王照明工程有限公司 一种智能巡检装置及系统
CN109531533A (zh) * 2018-11-30 2019-03-29 北京海益同展信息科技有限公司 一种机房巡检系统及其工作方法
CN109685709A (zh) * 2018-12-28 2019-04-26 深圳市商汤科技有限公司 一种智能机器人的照明控制方法及装置
CN110460782A (zh) * 2019-08-28 2019-11-15 北京海益同展信息科技有限公司 信息采集装置、方法、巡检机器人和存储介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4024848A4 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113311861A (zh) * 2021-05-14 2021-08-27 国家电投集团青海光伏产业创新中心有限公司 光伏组件隐裂特性的自动化检测方法及其系统
CN114040093A (zh) * 2021-10-20 2022-02-11 上海闻泰电子科技有限公司 摄像头测试装置、方法、电子设备和计算机可读存储介质
CN114040093B (zh) * 2021-10-20 2023-10-20 黄石闻泰通讯有限公司 摄像头测试装置、方法、电子设备和计算机可读存储介质

Also Published As

Publication number Publication date
CN110460782A (zh) 2019-11-15
EP4024848A1 (en) 2022-07-06
US20220303447A1 (en) 2022-09-22
EP4024848A4 (en) 2023-10-18
CN110460782B (zh) 2021-07-20

Similar Documents

Publication Publication Date Title
WO2021036824A1 (zh) 信息采集装置、方法、巡检机器人和存储介质
KR102319177B1 (ko) 이미지 내의 객체 자세를 결정하는 방법 및 장치, 장비, 및 저장 매체
US10217195B1 (en) Generation of semantic depth of field effect
EP3736766A1 (en) Method and device for blurring image background, storage medium, and electronic apparatus
WO2022170844A1 (zh) 一种视频标注方法、装置、设备及计算机可读存储介质
WO2022116423A1 (zh) 物体位姿估计方法、装置、电子设备及计算机存储介质
WO2018161289A1 (zh) 基于深度的控制方法、基于深度的控制装置和电子装置
TW201941103A (zh) 拍攝方法、裝置和智慧型裝置
CN112287896A (zh) 一种基于深度学习的无人机航拍图像目标检测方法及系统
CN108764278A (zh) 一种基于视觉的自学习工业智能检测系统及方法
CN112040198A (zh) 一种基于图像处理的智能水表读数识别系统与方法
KR102403169B1 (ko) 이미지 분석을 통한 가이드 제공 방법 및 이를 실행시키기 위하여 기록매체에 기록된 컴퓨터 프로그램
KR102389998B1 (ko) 비식별 처리 방법 및 이를 실행시키기 위하여 기록매체에 기록된 컴퓨터 프로그램
CN114830177A (zh) 电子设备和用于控制该电子设备的方法
JP6922399B2 (ja) 画像処理装置、画像処理方法及び画像処理プログラム
CN111435429B (zh) 一种基于双目立体数据动态认知的手势识别方法及系统
EP3715834B1 (en) Grain identification method and device, and computer storage medium
CN112752031B (zh) 图像采集检测方法、装置、电子设备及存储介质
US20220292811A1 (en) Image processing device, image processing method, and program
CN113139946A (zh) 一种基于视觉的衬衫污渍定位设备
CN114170432A (zh) 一种图像处理方法、图像识别方法及相关装置
CN113284184A (zh) 面向机器人rgbd视觉感知的6d位姿估计方法及系统
CN104965595A (zh) 一种2d打印方法和装置
KR102403174B1 (ko) 중요도에 따른 데이터 정제 방법 및 이를 실행시키기 위하여 기록매체에 기록된 컴퓨터 프로그램
KR102395393B1 (ko) 센싱 환경 분석을 통한 가이드 제공 방법 및 이를 실행시키기 위하여 기록매체에 기록된 컴퓨터 프로그램

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: 20857764

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2020857764

Country of ref document: EP

Effective date: 20220328