WO2017113559A1 - 摄像头的控制方法及装置、金融设备终端 - Google Patents

摄像头的控制方法及装置、金融设备终端 Download PDF

Info

Publication number
WO2017113559A1
WO2017113559A1 PCT/CN2016/081532 CN2016081532W WO2017113559A1 WO 2017113559 A1 WO2017113559 A1 WO 2017113559A1 CN 2016081532 W CN2016081532 W CN 2016081532W WO 2017113559 A1 WO2017113559 A1 WO 2017113559A1
Authority
WO
WIPO (PCT)
Prior art keywords
camera
image
target object
preset
position point
Prior art date
Application number
PCT/CN2016/081532
Other languages
English (en)
French (fr)
Inventor
梁添才
聂芸芸
张永
刘道余
龚文川
Original Assignee
广州广电运通金融电子股份有限公司
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 广州广电运通金融电子股份有限公司 filed Critical 广州广电运通金融电子股份有限公司
Priority to US15/774,577 priority Critical patent/US10674059B2/en
Priority to EP16880367.4A priority patent/EP3399739A4/en
Priority to RU2018119912A priority patent/RU2691109C1/ru
Publication of WO2017113559A1 publication Critical patent/WO2017113559A1/zh
Priority to ZA2018/03110A priority patent/ZA201803110B/en

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60Control of cameras or camera modules
    • H04N23/61Control of cameras or camera modules based on recognised objects
    • H04N23/611Control of cameras or camera modules based on recognised objects where the recognised objects include parts of the human body
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60Control of cameras or camera modules
    • H04N23/66Remote control of cameras or camera parts, e.g. by remote control devices
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/70Circuitry for compensating brightness variation in the scene
    • H04N23/74Circuitry for compensating brightness variation in the scene by influencing the scene brightness using illuminating means

Definitions

  • the present application relates to the field of image processing technologies, and more particularly, to a camera control method and apparatus, and a financial device terminal.
  • the data processing system can identify the target object in the image and perform subsequent operations according to the recognition result.
  • a camera can be set on a device such as a self-service cash machine, and the camera sends the collected face image to the background server, and the background server performs identity recognition according to the face image, and performs operations related to the withdrawal according to the identification result.
  • the quality of the image captured by the camera can directly affect the recognition efficiency and recognition accuracy of the background server. For example, if the camera captures the image, whether the target object is complete and the target object's posture meets the requirements, it will affect the image recognition effect of the background server. Therefore, a technical solution is needed to ensure that the image captured by the camera meets the image processing requirements.
  • the present application provides a method for controlling a camera to ensure that the camera captures an image that meets image processing requirements.
  • the present application also provides a camera control device and a financial device terminal for ensuring the application and implementation of the method in practice.
  • a first aspect of the present application provides a method for controlling a camera, including:
  • the attitude parameter does not meet the preset attitude parameter standard, sending, to the camera, a fine adjustment command for controlling the second rotation of the camera until the posture of the target object in the image sent by the camera in the second rotation
  • the parameter conforms to a preset attitude parameter standard; wherein the second rotation comprises a movement in a vertical or horizontal direction and a movement in a vertical or horizontal direction.
  • a second aspect of the present application provides a camera control device, including:
  • An object detecting module configured to detect whether the target object is included in the image sent by the camera; if not, triggering a search instruction sending module;
  • a search instruction sending module configured to send, to the camera, a search instruction for controlling the camera to perform a first rotation until the target object is detected in an image sent by the camera in the first rotation;
  • a posture determining module configured to determine a posture parameter of the target object in the image
  • a fine adjustment command sending module configured to send, to the camera, a fine adjustment command for controlling the second rotation of the camera until the posture parameter does not meet the preset posture parameter standard, until the image sent by the camera in the second rotation
  • the posture parameter of the target object meets a preset attitude parameter standard; wherein the second rotation includes movement in a vertical or horizontal direction, and movement in a vertical or horizontal direction.
  • a third aspect of the present application provides a financial device terminal comprising: a camera and a control device for the camera of any of the above.
  • the present application provides an embodiment of a method for controlling a camera.
  • an image sent by a camera can be received and an image is detected. If a target object is not detected in the image, a control camera is sent to the camera. Rotating the search command until the target object is detected in the image sent by the camera, and then acquiring the attitude parameter of the target object in the image. If the posture parameter does not meet the preset attitude parameter standard, the camera is sent a fine adjustment to control the rotation and movement of the camera. Command until the pose parameter of the target object in the image conforms to the preset pose parameter standard.
  • the image collected by the camera can be detected, that is, whether the target object is included in the image and whether the posture of the target object meets the image processing requirement, and when the detection is unsuccessful, the shooting angle of the camera is advanced.
  • the line is adjusted so that the image captured by the camera meets the above requirements.
  • Embodiment 1 is a flowchart of Embodiment 1 of a method for controlling a camera provided by the present application;
  • FIG. 2 is a view showing an example of setting of a fill light provided by the present application
  • Embodiment 3 is a flowchart of Embodiment 2 of a method for controlling a camera provided by the present application;
  • FIG. 5 is a schematic diagram of a geometric adjustment process of a camera provided by the present application.
  • Embodiment 1 of a control device for a camera provided by the present application
  • FIG. 7 is a schematic structural diagram of Embodiment 2 of a control device for a camera provided by the present application.
  • Embodiment 1 of a method for controlling a camera provided by the present application is shown.
  • This embodiment can be, but is not limited to, applied to a self-service device equipped with a camera, a self-service device such as a self-service cash machine or the like. As shown in FIG. 1 , this embodiment may specifically include steps S101 to S104.
  • Step S101 Detect whether the target object is included in the image sent by the camera; if not, execute step S102.
  • the camera may acquire an image according to a certain acquisition frequency, and send the acquired image to the execution module of the embodiment.
  • the execution module of this embodiment is the control device of the camera, or it may be referred to as a controller.
  • a camera is installed above the self-service cash dispenser, and a face image of a person standing in front of the self-service cash machine can be collected and sent to the control device of the camera.
  • the camera control device After receiving the image, the camera control device detects whether the target object is included in the image.
  • the target object may be but not limited to a human face.
  • the camera control device uses a face detection algorithm to detect an image to determine whether the image includes a face image.
  • the face detection algorithm may be, but is not limited to, a HARR face detection algorithm or an LBF face detection algorithm.
  • detecting whether the target object is included in the image is detecting whether the complete target object is included, for example, detecting whether the image includes a complete face contour. If the complete target object is not detected, it means that the camera's shooting angle is deviated, and the next step is needed to adjust the shooting angle of the camera.
  • step S102 may be skipped and step S103 is performed.
  • Step S102 Send a search command for controlling the camera to perform the first rotation to the camera until the target object is detected in the image transmitted by the camera in the first rotation.
  • the search command can control the camera to rotate in any direction, but in the implementation, the camera does not capture an image containing the complete target object, generally because the height of the target object is not within the shooting angle range of the camera. Therefore, optionally, an instruction to control the camera to rotate in the vertical direction is transmitted to the camera to control the camera to rotate in the vertical direction.
  • the instruction may be referred to as a search command, and the rotation of the camera is referred to as a first rotation.
  • the height of the self-service camera is fixed, and the left and right borders of the front face of the camera are usually within the horizontal shooting angle range of the camera, but the camera does not capture the complete face image, usually due to The upper and lower boundaries of the face are not within the vertical shooting angle of the camera, and the lower bar is lower than the shooting angle or the forehead is higher than the shooting angle.
  • the camera's control device sends a search command to the camera to control the camera to rotate in the vertical direction.
  • the camera's vertical shooting angle changes to search for the target object.
  • the search command transmitted by the camera control device to the camera is not limited to one, and may be multiple.
  • the control device of the camera transmits a first search command to the camera, and the camera rotates in the vertical direction in response to the first search command.
  • the image is continuously acquired, and the collected series of images are sent to the control device of the camera.
  • the camera control device After the camera control device detects the target object in a certain image, it can send an instruction to stop the rotation to the camera.
  • control device of the camera may send a second search command to the camera to control the camera to rotate in the vertical direction again, and repeat the detection target. The process of the object until the target object is detected in an image.
  • the camera control device may send multiple search commands to the camera.
  • Vertical rotation angle can be included in each search command
  • the respective vertical rotation angles are sequentially increased, and the increase is a fixed value. Specifically, for example:
  • Search command 1 contains Indicates that the camera is in the vertical direction Degree ⁇ Degree of rotation;
  • Search command 2 contains Indicates that the camera is in the vertical direction Degree ⁇ Degree of rotation;
  • Search command 3 contains Indicates that the camera is in the vertical direction Degree ⁇ Degree of rotation
  • Search command M contains Indicates that the camera is in the vertical direction Degree ⁇ Degree of rotation.
  • the vertical rotation angle may have an upper limit and a lower limit range, for example
  • step S103 can be performed.
  • Step S103 Determine a posture parameter of the target object in the image.
  • the image of this step may be the image that the camera's control device monitors the target object. Or, as can be seen from the above search process, the camera may not follow the acquisition frequency during the rotation process.
  • the search device of the camera detects the target object in a certain image, and then sends a command to stop the rotation to the camera, the image and the image captured after the camera stops rotating can be used as the image in this step.
  • the attitude parameter of the target object is determined in the image containing the target object.
  • the posture parameter may specifically be a left-right side rotation angle, an up-and-down pitch angle, and the like of the target object. Taking a human face as an example, the posture parameter may be a left and right deflection angle and a pitch angle of the face. Because the face is deflected left and right or up and down in the image, it will affect the subsequent image processing operations, which affects the efficiency and accuracy of image processing. Therefore, it is necessary to judge the posture of the target object by the attitude parameter.
  • the attitude parameter standard is preset, and the posture parameter standard may be set according to the posture requirement of the target object in the image by the subsequent image processing module. Thereby, the determined attitude parameter of the target object is compared with the preset attitude parameter standard.
  • the preset attitude standard is that the face is deflected to a certain extent, and the pitch angle is within a certain range.
  • step S104 needs to be performed to adjust the shooting angle of the camera again to collect the posture. Image.
  • Step S104 If the attitude parameter does not meet the preset attitude parameter standard, send a fine adjustment command to the camera to control the second rotation of the camera until the attitude parameter of the target object in the image transmitted by the camera in the second rotation meets the preset attitude parameter standard.
  • the second rotation includes movement in a vertical or horizontal direction, and movement in a vertical or horizontal direction.
  • the camera may need to move and rotate, so that the attitude parameter of the target object in the captured image conforms to the preset attitude parameter standard. Therefore, it is necessary to send a fine adjustment instruction to the camera.
  • the fine-tuning command may be to control the camera to move in the vertical direction or in the horizontal direction, or to control the camera to rotate in the vertical direction or in the horizontal direction. See the description below for the direction and distance of movement, the direction and angle of rotation.
  • the present application provides an embodiment of a method for controlling a camera.
  • an image transmitted by a camera can be received and an image is detected. If a target object is not detected in the image, the control is sent to the camera.
  • the search command rotated by the camera until the target object is detected in the image sent by the camera, and then the attitude parameter of the target object in the image is acquired. If the posture parameter does not meet the preset attitude parameter standard, the camera is controlled to rotate and move the camera. Fine-tuning instructions, Until the attitude parameter of the target object in the image meets the preset attitude parameter standard.
  • the image collected by the camera can be detected, that is, whether the target object is included in the image and whether the posture of the target object meets the image processing requirement, and when the detection is unsuccessful, the shooting angle of the camera is adjusted. In order to make the image captured by the camera meet the above requirements.
  • the image processing module when processing the acquired image, not only requires the posture of the target object in the image to conform to the attitude standard, but also requires the brightness of the light in the image to conform to the brightness standard, thereby ensuring the accuracy of the image processing.
  • Embodiment 2 of the control method of the camera provides Embodiment 2 of the control method of the camera to ensure that the brightness of the light in the image collected by the camera meets the brightness requirement.
  • one or more fill light lamps need to be disposed around the camera, and the brightness of the collected image can be improved after the fill light is turned on. See Figure 2 for the setting of the fill light.
  • the fill light can be distributed around the camera in a ring shape.
  • the embodiment may further include: step S105 and step S106, in addition to the first embodiment of the control method of the camera.
  • Step S105 It is determined that the posture parameter of the target object meets the brightness parameter of the image of the preset posture parameter standard.
  • the image in this step is an image in which the posture parameter of the target object meets the preset posture parameter standard in the previous step.
  • the image in the step may be referred to as a target image.
  • the image is continuously acquired and transmitted to the control device of the camera.
  • the control device of the camera detects that the attitude parameter of the target object in a certain image conforms to the preset posture parameter standard, and may also use the subsequent image in the image as the target image in this step.
  • the image area occupied by the target object in the image may be divided into a preset number of sub-areas, and the brightness of each sub-area is calculated separately.
  • the brightness of each sub-area it may be an average of the brightness of all the pixels in the sub-area, and the average value is taken as the brightness of the sub-area.
  • the brightness of each sub-area is respectively compared with a preset brightness parameter standard (such as a brightness threshold), and if there is a sub-area that does not meet the brightness parameter standard, the table The brightness parameter of the image does not meet the preset brightness parameter standard, and the next step needs to be performed.
  • a preset brightness parameter standard such as a brightness threshold
  • Step S106 If the brightness parameter does not meet the preset brightness parameter standard, the fill light command is sent to the target fill light in the plurality of fill lights until the brightness parameter of the image sent by the camera meets the preset brightness parameter standard.
  • the sub-area that does not meet the preset brightness parameter standard is determined as the target sub-area. Since the brightness of the target sub-area does not meet the brightness parameter standard, the target sub-areas can be supplementally lighted in a targeted manner.
  • the fill light needs to be turned on to fill the light, so it is necessary to select among several fill lights to determine which fill light(s) should be lit.
  • the fill light(s) may be referred to as a target fill light.
  • the specific manner of determining the target fill light may be as follows.
  • the image area occupied by the target object can be divided into a preset number of sub-areas.
  • the face area in the image can be divided into preset numbers.
  • the sub-area as shown in FIG. 4, divides the face area in the image into eight sub-areas, which are sub-area 1, sub-area 2, sub-area 3, ... sub-area 8.
  • the division of the face area such as the number of sub-areas, whether the sub-area is divided, or the like, which is not specifically limited in this application.
  • the fill light distributed around the camera may be correspondingly divided according to the division method of the target object area.
  • the fill light around the camera is divided into the same control area as the sub-area of the target object, and each control area corresponds to a sub-area of the target object in the image.
  • the fill light around the camera is divided into eight control areas, which are control area 1, control area 2, control area 3, ... control area 8, and each control area.
  • the target sub-area that does not meet the brightness parameter standard can be determined, so that the fill light in the control area corresponding to the target sub-area can be determined as the target fill light.
  • a fill light command can be sent to the target fill light to illuminate the target fill light. In this way, the brightness of the image captured by the camera will increase.
  • the fill light command sent to the target fill light may be multiple, that is, after the fill light command is sent to the target fill light, the image transmitted by the camera is received, and the brightness parameter of the image is determined to be consistent.
  • the preset brightness parameter standard if it is met, can stop sending the fill light command to the target fill light. If it does not match, it still needs to send the fill light command to the target fill light until the brightness parameter of the image collected by the camera meets the pre-condition. Set the brightness parameter standard.
  • the fill light command may include a fill light level, and different fill light levels indicate that the brightness of the fill light is different, and the higher the fill light level, the higher the brightness of the fill light. Therefore, the amount of fill light in the plurality of fill light lamps that are transmitted can be sequentially increased, and the increase can be a fixed value.
  • Fill light command 1 the fill level is the first level
  • Fill light command 2 the fill level is the second level
  • Fill light command 3 the fill level is the third level
  • fill level is the Nth level
  • control method of the camera provided by the embodiment can illuminate the fill light around the camera to improve the brightness of the image if the brightness of the image captured by the camera does not meet the preset brightness standard.
  • the posture of the target object in the image collected by the camera conforms to the posture requirement, and the brightness of the image also meets the brightness requirement. Therefore, the image can be uploaded to the server for subsequent processing by the server. For example, the server compares the image with the pre-stored image to identify the target object in the image.
  • the server compares the image with the pre-stored image to identify the target object in the image.
  • the manner of generating the fine-tuning command in step S104 may specifically include the following steps A1 to A6.
  • ⁇ 1 and ⁇ 2 are obtained as follows:
  • (x, y), (x 1 , y 1 ), (x 2 , y 2 ), and (x 3 , y 3 ) are respectively a first preset position point, a second preset position point of the target object, The coordinates of the third preset position point and the fourth preset position point respectively in the image.
  • the first preset position point may be a left eye pupil
  • the second preset position point may be a right eye pupil
  • the third preset position point may be a nose tip
  • the fourth preset position point may be For the right eyebrow.
  • the method for obtaining the coordinates of each of the preset position points may be an ASM (Activy Shape Model) algorithm.
  • ASM Active Shape Model
  • Step A2 Calculate the formula according to the up and down pitch angle Calculate the up and down pitch angle ⁇ .
  • step A3 or step A4 is performed to move the camera horizontally and horizontally.
  • step A5 or step A6 is performed to vertically and vertically rotate the camera.
  • Step A3 If the left and right side rotation angles ⁇ ⁇ N, a fine adjustment command is generated in which the horizontal shift is leftward
  • Step A4 If the left and right side rotation angles ⁇ ⁇ -N, a fine adjustment command that shifts the horizontal to the right by
  • Step A5 If the up and down pitch angle ⁇ ⁇ N, a fine adjustment command that vertically moves
  • Step A6 If the up and down pitch angle ⁇ ⁇ -N, a fine adjustment command that vertically shifts
  • the first preset position is the left eye pupil
  • the second preset position is the right eye pupil
  • the fifth preset position is the lips
  • d is the distance of the eyes on the image
  • D is The actual distance between the eyes
  • m is the distance from the center point of the line connecting the eye to the center of the lip
  • M is the actual distance from the center of the line to the center of the lips.
  • the comparison result in steps A3 to A6 can indicate that the attitude parameter of the target object does not conform to the attitude parameter standard.
  • the present application provides a geometrical diagram of a fine tuning process of the camera, as shown in FIG.
  • the fine adjustment command is generated as described above, the fine adjustment command is sent to the camera, so that after the camera is finely adjusted for movement and rotation, an image whose posture parameter conforms to the preset posture parameter standard can be acquired.
  • control device of the camera and the financial device terminal provided by the present application. It should be noted that the following descriptions of the camera control device and the financial device terminal can be referred to the control method of the camera provided above, and are not described below.
  • the present application provides a first embodiment of a control device for a camera.
  • the embodiment may specifically include: an object detection module 601, a search instruction sending module 602, a posture determining module 603, and a fine adjustment command sending module 604. among them:
  • the object detecting module 601 is configured to detect whether the target object is included in the image sent by the camera; if not, triggering the search instruction sending module;
  • the search instruction sending module 602 is configured to send, to the camera, a search command for controlling the camera to perform the first rotation until the target object is detected in the image sent by the camera in the first rotation;
  • a posture determining module 603, configured to determine a posture parameter of the target object in the image
  • the fine adjustment command sending module 604 is configured to: if the attitude parameter does not meet the preset attitude parameter standard, send a fine adjustment command to the camera to control the second rotation of the camera until the posture parameter of the target object in the image sent by the camera in the second rotation meets A preset attitude parameter standard; wherein the second rotation includes movement in a vertical or horizontal direction, and movement in a vertical or horizontal direction.
  • the present application provides an embodiment of a camera control device.
  • the object detection module 601 can receive an image sent by a camera and detect an image. If no target object is detected in the image.
  • the search command sending module 602 sends a search command for controlling the camera rotation to the camera until the target object is detected in the image sent by the camera, and then the posture determining module 603 acquires the pose parameter of the target object in the image, if the pose parameter does not meet the pre-
  • the fine adjustment command sending module 604 sends a fine adjustment command for controlling the rotation and movement of the camera to the camera until the attitude parameter of the target object in the image conforms to the preset attitude parameter standard.
  • the image collected by the camera can be detected, that is, whether the target object is included in the image and whether the posture of the target object meets the image processing requirement, and when the detection is unsuccessful, the shooting angle of the camera is adjusted. In order to make the image captured by the camera meet the above requirements.
  • the present application provides a second embodiment of the control device for the camera.
  • a plurality of fill lamps are disposed around the camera for control.
  • the embodiment may further include an image brightness determining module 605 and a fill light command sending module 606, in addition to the first embodiment of the control device of the camera.
  • the image brightness determining module 605 is configured to determine a brightness parameter of the image of the target object that meets the preset posture parameter standard
  • the fill light command sending module 606 is configured to: if the brightness parameter does not meet the preset brightness parameter standard, send a fill light command to the target fill light in the plurality of fill lights until the brightness parameter of the image sent by the camera meets the preset brightness parameter standard.
  • control device of the camera may further include: an image uploading module. among them:
  • the image uploading module is configured to upload an image whose brightness parameter meets the preset brightness parameter standard to the server.
  • each search command sent by the search command sending module 602 to the camera includes a vertical rotation angle, and each vertical rotation angle is sequentially incremented, and the increase is a fixed value.
  • the manner in which the fine-tuning instruction sending module 604 generates the fine-tuning instruction is specifically:
  • ⁇ 1 and ⁇ 2 are obtained as follows:
  • n 1 , m 2 and m 3 are obtained as follows:
  • (x, y), (x 1 , y 1 ), (x 2 , y 2 ), and (x 3 , y 3 ) are respectively a first preset position point, a second preset position point of the target object, The coordinates of the third preset position point and the fourth preset position point in the image respectively;
  • d is the distance between the first preset position point and the second preset position point on the image
  • D is the actual distance between the first preset position point and the second preset position point
  • m is the first preset position point a distance from a center of the second preset position point and a fifth preset position point of the target object on the image
  • M is a center of the first preset position point and the second preset position point and a fifth preset of the target object The actual distance of the location point.
  • the target object occupies a plurality of image areas in the image, and each of the image areas respectively corresponds to a plurality of fill lights; correspondingly, the step of the fill light command sending module 606 determining the target fill light in the plurality of fill lights includes:
  • the fill light corresponding to the image area whose area brightness parameter is lower than the preset brightness parameter is determined as the target fill light.
  • each of the fill light commands sent by the fill light command transmitting module 606 to the target fill light includes a light fill level, and each fill light level is sequentially incremented, and the amplification is a fixed value.
  • the present application also provides a financial device terminal, including: a camera and a control device for any of the above cameras.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Studio Devices (AREA)

Abstract

本申请提供了一种摄像头的控制方法实施例,本实施例可以接收摄像头发送的图像并对图像进行检测,若未检测到目标对象,则向摄像头发送控制摄像头在垂直方向上转动的搜索指令,直至在图像中检测到目标对象,然后获取图像中目标对象的姿态参数,若该姿态参数不符合预设姿态参数标准,则向摄像头发送控制摄像头转动及移动的微调指令,直至目标对象的姿态参数符合预设姿态参数标准。可见,本实施例可以对摄像头采集到的图像进行检测,即检测图像中是否包含目标对象以及该目标对象的姿态是否符合图像处理要求,并在检测不成功时,对摄像头的拍摄角度进行调整,以使摄像头采集到的图像满足上述要求。本申请还提供了摄像头的控制装置及金融设备终端。

Description

摄像头的控制方法及装置、金融设备终端
本申请要求于2015年12月31日提交中国专利局、申请号为201511034588.3、发明名称为“摄像头的控制方法及装置、金融设备终端”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及图像处理技术领域,更具体地,是摄像头的控制方法及装置、金融设备终端。
背景技术
如今,图像识别技术已经广泛应用在各种数据处理系统中,数据处理系统可以对图像中的目标对象进行识别,并根据识别结果进行后续操作。
例如,自助取款机等设备上可以设置摄像头,摄像头将采集到的人脸图像发送至后台服务器,后台服务器根据人脸图像进行身份识别,并依据身份识别结果进行与取款相关的操作。
可以理解的是,摄像头采集到的图像的质量,可以直接影响后台服务器的识别效率及识别准确性。例如,摄像头采集到图像中,目标对象是否完整、目标对象的姿态是否符合要求,均会影响后台服务器的图像识别效果。因此,需要一种技术方案,以保证摄像头采集到的图像符合图像处理要求。
发明内容
有鉴于此,本申请提供了一种摄像头的控制方法,用以保证摄像头采集到符合图像处理要求的图像。另外,本申请还提供了一种摄像头的控制装置及金融设备终端,用以保证所述方法在实际中的应用及实现。
为实现所述目的,本申请提供的技术方案如下:
本申请的第一方面提供了一种摄像头的控制方法,包括:
检测所述摄像头发送的图像中是否包含目标对象;
若未包含,则向所述摄像头发送控制所述摄像头进行第一转动的搜索指令,直至在第一转动中的所述摄像头发送的图像中检测到所述目标对象;
确定所述图像中所述目标对象的姿态参数;
若所述姿态参数不符合预设姿态参数标准,则向所述摄像头发送控制所述摄像头进行第二转动的微调指令,直至第二转动中的所述摄像头发送的图像中所述目标对象的姿态参数符合预设姿态参数标准;其中,所述第二转动包括垂直或水平方向上的移动、以及垂直或水平方向上的移动。
本申请的第二方面提供了一种摄像头的控制装置,包括:
对象检测模块,用于检测所述摄像头发送的图像中是否包含目标对象;若未包含,则触发搜索指令发送模块;
搜索指令发送模块,用于向所述摄像头发送控制所述摄像头进行第一转动的搜索指令,直至在第一转动中的所述摄像头发送的图像中检测到所述目标对象;
姿态确定模块,用于确定所述图像中所述目标对象的姿态参数;
微调指令发送模块,用于若所述姿态参数不符合预设姿态参数标准,则向所述摄像头发送控制所述摄像头进行第二转动的微调指令,直至第二转动中的所述摄像头发送的图像中所述目标对象的姿态参数符合预设姿态参数标准;其中,所述第二转动包括垂直或水平方向上的移动、以及垂直或水平方向上的移动。
本申请的第三方面提供了一种金融设备终端,包括:摄像头以及上述任意一种的摄像头的控制装置。
由以上技术方案可知,本申请提供了一种摄像头的控制方法实施例,本实施例可以接收摄像头发送的图像并对图像进行检测,若在图像中未检测到目标对象,则向摄像头发送控制摄像头转动的搜索指令,直至在摄像头发送的图像中检测到目标对象,然后获取图像中目标对象的姿态参数,若该姿态参数不符合预设姿态参数标准,则向摄像头发送控制摄像头转动及移动的微调指令,直至图像中目标对象的姿态参数符合预设姿态参数标准。可见,本实施例可以对摄像头采集到的图像进行检测,即检测图像中是否包含目标对象以及该目标对象的姿态是否符合图像处理要求,并在检测不成功时,对摄像头的拍摄角度进 行调整,以使摄像头采集到的图像满足上述要求。
当然,实施本申请的任一产品并不一定需要同时达到以上所述的所有优点。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案,下面将对实施例或现有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申请的实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据提供的附图获得其他的附图。
图1为本申请提供的摄像头的控制方法实施例1的流程图;
图2为本申请提供的补光灯的设置方式示例图;
图3为本申请提供的摄像头的控制方法实施例2的流程图;
图4为本申请提供的人脸区域的划分方式示例意图;
图5为本申请提供的摄像头的微调过程的一种几何示意图;
图6为本申请提供的摄像头的控制装置实施例1的结构示意图;
图7为本申请提供的摄像头的控制装置实施例2的结构示意图。
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。
参见图1,其示出了本申请提供的摄像头的控制方法实施例1的流程。本实施例可以但不限定于应用在设置有摄像头的自助设备上,自助设备如自助取款机等等。如图1所示,本实施例可以具体包括步骤S101~步骤S104。
步骤S101:检测摄像头发送的图像中是否包含目标对象;若未包含,则执行步骤S102。
在实施中,摄像头可以按照一定的采集频率采集图像,并将采集到的图像发送至本实施例的执行模块。本实施例的执行模块即摄像头的控制装置,或者可将其称为控制器。
例如,自助取款机的上方安装有摄像头,可以采集站立在自助取款机前的人的脸部图像,并将该脸部图像发送至摄像头的控制装置。
摄像头的控制装置在接收到图像后,检测该图像中是否包含目标对象。其中,目标对象可以是但不限定于人脸。
以人脸为例说明,摄像头的控制装置使用人脸检测算法对图像进行检测,以判断该图像中是否包含人脸图像。人脸检测算法可以是但不限定于HARR人脸检测算法或LBF人脸检测算法。
需要说明的是,检测图像中是否包含目标对象,是检测是否包含完整的目标对象,例如,检测图像中是否包含完整的人脸轮廓。若未检测到完整的目标对象,则表示摄像头的拍摄视角有偏差,需要执行下一步骤,以调整摄像头的拍摄角度。
当然,若在图像中检测到完整的目标对象,可以跳过步骤S102,执行步骤S103。
步骤S102:向摄像头发送控制摄像头进行第一转动的搜索指令,直至在第一转动中的摄像头发送的图像中检测到目标对象。
其中,搜索指令可以控制摄像头在任意方向上转动,但是,在实施中,摄像头未采集到包含完整的目标对象的图像,一般是由于目标对象的高度并未在摄像头的拍摄角度范围内。因此,可选地,向摄像头发送控制摄像头在垂直方向上转动的指令,以控制摄像头在垂直方向上进行转动。为了便于描述,可以将该指令称为搜索指令,将摄像头的转动称为第一转动。
以自助取款机上的摄像头为例,自助摄像机的高度是固定的,且摄像头前人脸的左右边界通常在摄像头的水平拍摄角度范围内,但摄像头并未采集到完整的人脸图像,通常是由于人脸的上下边界并未在摄像头的垂直拍摄角度范围内,如下巴低于拍摄角度或者额头高于拍摄角度。
摄像头的控制装置向摄像头发送搜索指令,以控制摄像头在垂直方向上进行转动。摄像头的垂直拍摄角度变化,以搜索目标对象。
需要说明的是,摄像头的控制装置向摄像头发送的搜索指令并非局限于一条,可以是多条。具体地,摄像头的控制装置向摄像头发送第一条搜索指令,摄像头响应于第一条搜索指令,会在垂直方向上转动。摄像头在转动过程中,不断采集图像,并将采集到的一系列图像发送至摄像头的控制装置。摄像头的控制装置在某张图像中检测到目标对象后,便可以向摄像头发送停止转动的指令。
但是,若摄像头的控制装置在上述一系列图像中,并未检测到目标对象,则摄像头的控制装置可以向摄像头发送第二条搜索指令,以控制摄像头再次在垂直方向上转动,重复上述检测目标对象的过程,直至在某张图像中检测到目标对象。
由以上过程可知,摄像头的控制装置可能会向摄像头发送多条搜索指令。各条搜索指令中可以包括垂直转动角度
Figure PCTCN2016081532-appb-000001
各个垂直转动角度依次递增,且增幅为固定值。具体地,例如:
搜索指令1包含
Figure PCTCN2016081532-appb-000002
表示摄像头在垂直方向进行
Figure PCTCN2016081532-appb-000003
度~
Figure PCTCN2016081532-appb-000004
度的转动;
搜索指令2包含
Figure PCTCN2016081532-appb-000005
表示摄像头在垂直方向进行
Figure PCTCN2016081532-appb-000006
度~
Figure PCTCN2016081532-appb-000007
度的转动;
搜索指令3包含
Figure PCTCN2016081532-appb-000008
表示摄像头在垂直方向进行
Figure PCTCN2016081532-appb-000009
度~
Figure PCTCN2016081532-appb-000010
度的转动;
……
搜索指令M包含
Figure PCTCN2016081532-appb-000011
表示摄像头在垂直方向进行
Figure PCTCN2016081532-appb-000012
度~
Figure PCTCN2016081532-appb-000013
度的转动。
其中,
Figure PCTCN2016081532-appb-000014
…,
Figure PCTCN2016081532-appb-000015
依次递增,且增幅为定值。另外,垂直转动角度可以具有上限及下限范围,例如
Figure PCTCN2016081532-appb-000016
搜索指令执行规则:
第i次人脸检测失败,便执行搜索指令i+1(i<=M-1)。若从首个搜索开始,到某个搜索指令,搜索时间超过预设时间长度,如3秒,则可以提示调整目标对象的位置。
若摄像头的控制装置在接收到的图像中检测到目标对象,便可以执行步骤S103。
步骤S103:确定图像中目标对象的姿态参数。
其中,本步骤的图像可以是摄像头的控制装置监测到目标对象的那张图像。或者,由上述搜索过程可知,摄像头在转动过程中,可以按照采集频率不 断采集图像,若摄像头的搜索装置在某张图像中检测到目标对象,则向摄像头发送停止转动的指令,则该张图像及摄像头停止转动后采集到图像均可以作为本步骤中的图像。
按照预设姿态检测算法,在包含目标对象的图像中,确定目标对象的姿态参数。其中,姿态参数可以具体是目标对象的左右侧转角度、上下俯仰角度等。以人脸为例,姿态参数可以是脸部左右偏转角度及俯仰角度。因为,人脸在图像中左右偏转或者上下俯仰都会对后续的图像处理操作造成影响,即影响图像处理的效率及准确性。因此,需要通过姿态参数对目标对象的姿态进行判断。
在实施前,预先设置姿态参数标准,该姿态参数标准可以是根据后续图像处理模块对图像中目标对象的姿态要求设置的。从而将确定出的目标对象的姿态参数与预设姿态参数标准进行比较。以人脸为例,预设姿态标准为人脸左右偏转在一定的范围内,且俯仰角度在一定范围内。
若目标对象的姿态参数不符合预设姿态参数标准,则表示目标对象虽然包含在图像内,但姿态并不符合要求,则需要执行步骤S104以再次调整摄像头的拍摄角度,以采集到姿态符合要求的图像。
步骤S104:若姿态参数不符合预设姿态参数标准,则向摄像头发送控制摄像头进行第二转动的微调指令,直至第二转动中的摄像头发送的图像中目标对象的姿态参数符合预设姿态参数标准;其中,第二转动包括垂直或水平方向上的移动、以及垂直或水平方向上的移动。
其中,摄像头可能既需要移动也需要转动,才能使得拍摄到的图像中目标对象的姿态参数符合预设姿态参数标准,因此,需要向摄像头发送微调指令。
微调指令可以是控制摄像头在垂直方向上或者在水平方向上移动,也可以是控制摄像头在垂直方向上或者在水平方向上转动。有关移动的方向及距离、转动的方向及角度,详见下文说明。
由以上的技术方案可知,本申请提供了一种摄像头的控制方法实施例,本实施例可以接收摄像头发送的图像并对图像进行检测,若在图像中未检测到目标对象,则向摄像头发送控制摄像头转动的搜索指令,直至在摄像头发送的图像中检测到目标对象,然后获取图像中目标对象的姿态参数,若该姿态参数不符合预设姿态参数标准,则向摄像头发送控制摄像头转动及移动的微调指令, 直至图像中目标对象的姿态参数符合预设姿态参数标准。可见,本实施例可以对摄像头采集到的图像进行检测,即检测图像中是否包含目标对象以及该目标对象的姿态是否符合图像处理要求,并在检测不成功时,对摄像头的拍摄角度进行调整,以使摄像头采集到的图像满足上述要求。
在实际应用中,图像处理模块在对采集到的图像进行处理时,不仅要求图像中目标对象的姿态符合姿态标准,还要求图像中光线的亮度符合亮度标准,从而保证图像处理的准确性。
因此,本申请提供了摄像头的控制方法实施例2,以保证摄像头采集到的图像中光线的亮度符合亮度要求。需要说明的是,本实施例中,摄像头的周围需要设置一个或多个补光灯,补光灯点亮后可以提高采集到的图像的亮度。补光灯的设置方式可以参见图2,如图2所示,补光灯可以环形分布在摄像头的周围。
具体地,如图3所示,本实施例在上述摄像头的控制方法实施例1的基础上,还可以包括:步骤S105及步骤S106。
步骤S105:确定目标对象的姿态参数符合预设姿态参数标准的图像的亮度参数。
其中,本步骤中的图像是上一步骤中目标对象的姿态参数符合预设姿态参数标准的图像,为了便于描述,可以将该步骤中的图像称为目标图像。或者,由以上可知,摄像头在第二转动过程中,会不断采集图像并将图像发送至摄像头的控制装置。摄像头的控制装置检测到某张图像中目标对象的姿态参数符合预设姿态参数标准,也可以将该张图像中的后续图像作为本步骤中的目标图像。
确定图像的亮度参数。具体地,可以将目标对象在图像中占据的图像区域划分为预设数量的子区域,分别计算每个子区域的亮度。在计算每个子区域的亮度时,可以是计算子区域内所有像素的亮度的平均值,将平均值作为子区域的亮度。
获得每个子区域的亮度后,将各个子区域的亮度分别与预先设置的亮度参数标准(如亮度阈值)进行比较,若存在不符合亮度参数标准的子区域,则表 示图像的亮度参数不符合预设亮度参数标准,需要执行下一步骤。
步骤S106:若亮度参数未符合预设亮度参数标准,则向若干补光灯中的目标补光灯发送补光指令,直至摄像头发送的图像的亮度参数符合预设亮度参数标准。
其中,将上述亮度比较后,不符合预设亮度参数标准的子区域确定为目标子区域。由于目标子区域的亮度不符合亮度参数标准,则可以针对性地,对该些目标子区域进行补光。
可以理解的是,需要点亮补光灯才能进行补光,因此,需要在若干补光灯中进行选择,以确定应该点亮哪个(些)补光灯。为了便于描述,可以将该个(些)补光灯称为目标补光灯。
具体地,在若干补光灯中,确定目标补光灯的具体方式可以是如下方式。
由以上可知,目标对象占据的图像区域(或者称为图像中的目标对象区域)可以划分为预设数量的子区域,以人脸为例,可以将图像中的人脸区域划分为预设数量的子区域,如图4所示,将图像中的人脸区域平均划分8个子区域,分别为子区域1、子区域2、子区域3……子区域8。当然,有关人脸区域的划分如子区域的数量、是否平均划分等都可以有其他的实现方式,本申请并不做具体限定。
进一步地,可以参照目标对象区域的划分方法,对摄像头周围分布的补光灯进行相应的划分。经过划分后,摄像头周围的补光灯被划分为与目标对象的子区域相同的控制区域,每个控制区域分别与图像中目标对象的一个子区域对应。如图2所示,参照人脸区域的划分方式,将摄像头周围的补光灯划分为8个控制区域,分别为控制区域1、控制区域2、控制区域3……控制区域8,各个控制区域与图4中的各个子区域一一对应。
在上述亮度比较中,可以确定出不符合亮度参数标准的目标子区域,从而,可以将目标子区域对应的控制区域内的补光灯确定为目标补光灯。
在确定出目标补光灯后,可以向目标补光灯发送补光指令,以点亮目标补光灯。这样,摄像头拍摄到的图像的亮度便会提高。
需要说明的是,向目标补光灯发送的补光指令可以是多个,即向目标补光灯发送补光指令后,接收摄像头发送的图像,并判断图像的亮度参数是否符合 预设亮度参数标准,若符合,则可以停止向目标补光灯发送补光指令,若不符合,则仍需要向目标补光灯发送补光指令,直至摄像头采集到的图像的亮度参数符合预设亮度参数标准。
另外,发送的补光指令中可以包括补光量级,不同的补光量级表示补光灯的亮度不同,补光量级越高补光灯的亮度也相应地越高。因此,发送的多个补光灯中的补光量级可以依次递增,且增幅可以为固定值。例如:
补光指令1:补光量级为第一级;
补光指令2:补光量级为第二级;
补光指令3:补光量级为第三级;
……
补光指令N:补光量级为第N级;
补光指令执行规则:
第i次图像的亮度不符合亮度参数标准,则执行指令i+1(i<=N-1)直至亮度不符合亮度参数标准(补光量级可根据实际情况设置)。
由以上技术方案可知,本实施例提供的摄像头的控制方法,可以在摄像头采集的图像的亮度不符合预设亮度标准的情况下,点亮摄像头周围的补光灯,以提高图像的亮度。
经过上述调整后,摄像头采集到的图像中的目标对象的姿态符合姿态要求,且图像的亮度也符合亮度要求。因此,可以将该图像上传至服务器,以使服务器进行后续处理。例如,服务器将该图像与预存的图像进行图像比对,以对图像中的目标对象进行识别。当然,本领域技术人员可以理解并期望服务器的其他各种后续处理,而不脱离本申请的保护范围。
上述各个摄像头的控制方法实施例中,步骤S104中微调指令的生成方式可以具体包括如下步骤A1~步骤A6。
步骤A1:依据左右侧转角度计算公式α=|α12|,计算左右侧转角度α。
其中,α1及α2的获得方式分别为:
Figure PCTCN2016081532-appb-000017
Figure PCTCN2016081532-appb-000018
其中,m1、m2及m3的获得方式分别为:
Figure PCTCN2016081532-appb-000019
Figure PCTCN2016081532-appb-000020
其中,(x,y)、(x1,y1)、(x2,y2)及(x3,y3)分别为目标对象的第一预设位置点、第二预设位置点、第三预设位置点及第四预设位置点分别在图像中的坐标。
以目标对象为人脸为例说明,第一预设位置点可以为左眼瞳孔,第二预设位置点可以为右眼瞳孔,第三预设位置点可以为鼻尖,第四预设位置点可以为右眉心。
可选地,获得以上各个预设位置点的坐标的方法可以是ASM(Activy Shape Model)算法。
步骤A2:依据上下俯仰角度计算公式
Figure PCTCN2016081532-appb-000021
计算上下俯仰角度β。
其中,m4的获得方式分别为:
Figure PCTCN2016081532-appb-000022
在获得左右侧转角度α后,将左右侧转角度α与预设角度常数N进行比较。根据比较结果,执行步骤A3或者步骤A4,以将摄像头水平移动及水平转动。
在获得上下俯仰角度β后,将上下俯仰角度β与预设角度常数N进行比较。根据比较结果,执行步骤A5或者步骤A6,以将摄像头垂直移动及垂直转动。
步骤A3:若左右侧转角度α≥N,生成水平向左移动|DFC tanα|且向右旋转α度的微调指令;其中,N为预设常数。
步骤A4:若左右侧转角度α≤-N,生成水平向右移动|DFC tanα|且向左旋转α度的微调指令。
步骤A5:若上下俯仰角度β≥N,生成垂直向上移动|DFC tanβ|且向下旋转 β度的微调指令。
步骤A6:若上下俯仰角度β≤-N,生成垂直向下移动|DFC tanβ|且向上旋转β度的微调指令。
其中,DFC的获得方式分别为:
Figure PCTCN2016081532-appb-000023
其中,k为摄像头的性质参数;d为第一预设位置点与第二预设位置点在图像上的距离;D为第一预设位置点与第二预设位置点的实际距离;m为第一预设位置点与第二预设位置点的中心与目标对象的第五预设位置点在图像上的距离;M为第一预设位置点与第二预设位置点的中心与目标对象的第五预设位置点的实际距离。
为人脸为例进行说明,第一预设位置点为左眼瞳孔,第二预设位置点为右眼瞳孔,第五预设位置点为嘴唇,则d为双眼在图像上的距离,D为双眼的实际距离,m为双眼连线的中心点到嘴唇中心在图像上的距离,M为双眼连线的中心到嘴唇中心的实际距离。
当然,若左右侧转角度|α|<5且上下俯仰角度|β|<5,则不需要生成微调指令,直接将图像作为后续处理的目标图像。另外,可以理解的是,步骤A3~步骤A6中的比较结果可以表示目标对象的姿态参数不符合姿态参数标准。再者,为了便于理解,本申请提供了摄像头的一种微调过程的几何示意图,具体如图5所示。
按照上述方式生成微调指令后,将微调指令发送至摄像头,使得摄像头在移动及转动的微调后,可以采集到姿态参数符合预设姿态参数标准的图像。
以下对本申请提供的摄像头的控制装置及金融设备终端进行介绍,需要说明的是,下文有关摄像头的控制装置及金融设备终端的说明可以参见上文提供的摄像头的控制方法,以下并不赘述。
与上述摄像头的控制方法实施例1相对应,本申请提供了一种摄像头的控制装置实施例1。如图6所示,本实施例可以具体包括:对象检测模块601、搜索指令发送模块602、姿态确定模块603及微调指令发送模块604。其中:
对象检测模块601,用于检测摄像头发送的图像中是否包含目标对象;若未包含,则触发搜索指令发送模块;
搜索指令发送模块602,用于向摄像头发送控制摄像头进行第一转动的搜索指令,直至在第一转动中的摄像头发送的图像中检测到目标对象;
姿态确定模块603,用于确定图像中目标对象的姿态参数;
微调指令发送模块604,用于若姿态参数不符合预设姿态参数标准,则向摄像头发送控制摄像头进行第二转动的微调指令,直至第二转动中的摄像头发送的图像中目标对象的姿态参数符合预设姿态参数标准;其中,第二转动包括垂直或水平方向上的移动、以及垂直或水平方向上的移动。
由以上的技术方案可知,本申请提供了一种摄像头的控制装置实施例,本实施例中,对象检测模块601可以接收摄像头发送的图像并对图像进行检测,若在图像中未检测到目标对象,则搜索指令发送模块602向摄像头发送控制摄像头转动的搜索指令,直至在摄像头发送的图像中检测到目标对象,然后姿态确定模块603获取图像中目标对象的姿态参数,若该姿态参数不符合预设姿态参数标准,则微调指令发送模块604向摄像头发送控制摄像头转动及移动的微调指令,直至图像中目标对象的姿态参数符合预设姿态参数标准。可见,本实施例可以对摄像头采集到的图像进行检测,即检测图像中是否包含目标对象以及该目标对象的姿态是否符合图像处理要求,并在检测不成功时,对摄像头的拍摄角度进行调整,以使摄像头采集到的图像满足上述要求。
与上述摄像头的控制方法实施例2相对应,本申请提供了一种摄像头的控制装置实施例2。本实施例用于控制的摄像头的周围设置有若干补光灯。
如图7所示,本实施例在上述摄像头的控制装置实施例1的基础上,还可以包括:图像亮度确定模块605及补光指令发送模块606。
图像亮度确定模块605,用于确定目标对象的姿态参数符合预设姿态参数标准的图像的亮度参数;
补光指令发送模块606,用于若亮度参数未符合预设亮度参数标准,则向若干补光灯中的目标补光灯发送补光指令,直至摄像头发送的图像的亮度参数符合预设亮度参数标准。
另外,上述摄像头的控制装置还可以包括:图像上传模块。其中:
图像上传模块,用于将亮度参数符合预设亮度参数标准的图像上传至服务器。
在实际应用中,上述搜索指令发送模块602向摄像头发送的各个搜索指令中包括垂直转动角度,且各个垂直转动角度依次递增,增幅为固定值。
上述微调指令发送模块604生成微调指令的方式具体为:
依据左右侧转角度计算公式α=|α12|,计算左右侧转角度α;
其中,α1及α2的获得方式分别为:
Figure PCTCN2016081532-appb-000024
Figure PCTCN2016081532-appb-000025
其中,m1、m2及m3的获得方式分别为:
Figure PCTCN2016081532-appb-000026
Figure PCTCN2016081532-appb-000027
其中,(x,y)、(x1,y1)、(x2,y2)及(x3,y3)分别为目标对象的第一预设位置点、第二预设位置点、第三预设位置点及第四预设位置点分别在图像中的坐标;
依据上下俯仰角度计算公式
Figure PCTCN2016081532-appb-000028
计算上下俯仰角度β;
其中,m4的获得方式分别为:
Figure PCTCN2016081532-appb-000029
若左右侧转角度α≥N,生成水平向左移动|DFC tanα|且向右旋转α度的微调指令;其中,N为预设常数;
若左右侧转角度α≤-N,生成水平向右移动|DFC tanα|且向左旋转α度的微调指令;
若上下俯仰角度β≥N,生成垂直向上移动|DFC tanβ|且向下旋转β度的微调指令;
若上下俯仰角度β≤-N,生成垂直向下移动|DFC tanβ|且向上旋转β度的微 调指令;
其中,DFC的获得方式分别为:
Figure PCTCN2016081532-appb-000030
其中,d为第一预设位置点与第二预设位置点在图像上的距离;D为第一预设位置点与第二预设位置点的实际距离;m为第一预设位置点与第二预设位置点的中心与目标对象的第五预设位置点在图像上的距离;M为第一预设位置点与第二预设位置点的中心与目标对象的第五预设位置点的实际距离。
另外,目标对象在图像中占据多个图像区域,各个图像区域分别与若干补光灯对应;相应地,补光指令发送模块606在若干补光灯中确定目标补光灯的步骤包括:
分别确定目标对象的每个图像区域的区域亮度参数;
将区域亮度参数低于预设亮度参数的图像区域所对应的补光灯确定为目标补光灯。
另外,补光指令发送模块606向目标补光灯发送的各个补光指令中包括补光量级,且各个补光量级依次递增,增幅为固定值。
另外,本申请还提供了一种金融设备终端,包括:摄像头以及上述任意一种摄像头的控制装置。
需要说明的是,本说明书中的各个实施例均采用递进的方式描述,每个实施例重点说明的都是与其他实施例的不同之处,各个实施例之间相同相似的部分互相参见即可。
还需要说明的是,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。而且,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,并不排除在包括上述要素的过程、方法、物品或者设备中还存在另外的相同要素。
对所公开的实施例的上述说明,使本领域专业技术人员能够实现或使用本申请。对这些实施例的多种修改对本领域的专业技术人员来说将是显而易见的,本文中所定义的一般原理可以在不脱离本申请的精神或范围的情况下,在其它实施例中实现。因此,本申请将不会被限制于本文所示的这些实施例,而是要符合与本文所公开的原理和新颖特点相一致的最宽的范围。

Claims (13)

  1. 一种摄像头的控制方法,其特征在于,包括:
    检测所述摄像头发送的图像中是否包含目标对象;
    若未包含,则向所述摄像头发送控制所述摄像头进行第一转动的搜索指令,直至在第一转动中的所述摄像头发送的图像中检测到所述目标对象;
    确定所述图像中所述目标对象的姿态参数;
    若所述姿态参数不符合预设姿态参数标准,则向所述摄像头发送控制所述摄像头进行第二转动的微调指令,直至第二转动中的所述摄像头发送的图像中所述目标对象的姿态参数符合预设姿态参数标准;其中,所述第二转动包括垂直或水平方向上的移动、以及垂直或水平方向上的移动。
  2. 根据权利要求1所述的摄像头的控制方法,其特征在于,所述摄像头的周围设置有若干补光灯,该方法还包括:
    确定所述目标对象的姿态参数符合预设姿态参数标准的图像的亮度参数;
    若所述亮度参数未符合预设亮度参数标准,则向所述若干补光灯中的目标补光灯发送补光指令,直至所述摄像头发送的图像的亮度参数符合预设亮度参数标准。
  3. 根据权利要求2所述的摄像头的控制方法,其特征在于,所述目标对象在图像中占据多个图像区域,各个图像区域分别与若干补光灯对应;
    相应地,在若干所述补光灯中确定目标补光灯的步骤包括:
    分别确定所述目标对象的每个所述图像区域的区域亮度参数;
    将区域亮度参数低于预设亮度参数的图像区域所对应的补光灯确定为目标补光灯。
  4. 根据权利要求2所述的摄像头的控制方法,其特征在于,向所述目标补光灯发送的各个所述补光指令中包括补光量级,且各个所述补光量级依次递增,增幅为固定值。
  5. 根据权利要求1所述的摄像头的控制方法,其特征在于,向所述摄像头发送的各个所述搜索指令中包括垂直转动角度,且各个所述垂直转动角度依次递增,增幅为固定值。
  6. 根据权利要求1所述的摄像头的控制方法,其特征在于,所述微调指 令的生成方式具体为:
    依据左右侧转角度计算公式α=|α12|,计算左右侧转角度α;
    其中,α1及α2的获得方式分别为:
    Figure PCTCN2016081532-appb-100001
    Figure PCTCN2016081532-appb-100002
    其中,m1、m2及m3的获得方式分别为:
    Figure PCTCN2016081532-appb-100003
    Figure PCTCN2016081532-appb-100004
    其中,所述(x,y)、(x1,y1)、(x2,y2)及(x3,y3)分别为所述目标对象的第一预设位置点、第二预设位置点、第三预设位置点及第四预设位置点分别在图像中的坐标;
    依据上下俯仰角度计算公式
    Figure PCTCN2016081532-appb-100005
    计算上下俯仰角度β;
    其中,m4的获得方式分别为:
    Figure PCTCN2016081532-appb-100006
    若左右侧转角度α≥N,生成水平向左移动|DFC tanα|且向右旋转α度的微调指令;其中,N为预设常数;
    若左右侧转角度α≤-N,生成水平向右移动|DFC tanα|且向左旋转α度的微调指令;
    若上下俯仰角度β≥N,生成垂直向上移动|DFC tanβ|且向下旋转β度的微调指令;
    若上下俯仰角度β≤-N,生成垂直向下移动|DFC tanβ|且向上旋转β度的微调指令;
    其中,DFC的获得方式分别为:
    Figure PCTCN2016081532-appb-100007
    其中,d为第一预设位置点与第二预设位置点在图像上的距离;D为第一预设位置点与第二预设位置点的实际距离;m为第一预设位置点与第二预设位 置点的中心与所述目标对象的第五预设位置点在图像上的距离;M为第一预设位置点与第二预设位置点的中心与所述目标对象的第五预设位置点的实际距离。
  7. 一种摄像头的控制装置,其特征在于,包括:
    对象检测模块,用于检测所述摄像头发送的图像中是否包含目标对象;若未包含,则触发搜索指令发送模块;
    搜索指令发送模块,用于向所述摄像头发送控制所述摄像头进行第一转动的搜索指令,直至在第一转动中的所述摄像头发送的图像中检测到所述目标对象;
    姿态确定模块,用于确定所述图像中所述目标对象的姿态参数;
    微调指令发送模块,用于若所述姿态参数不符合预设姿态参数标准,则向所述摄像头发送控制所述摄像头进行第二转动的微调指令,直至第二转动中的所述摄像头发送的图像中所述目标对象的姿态参数符合预设姿态参数标准;其中,所述第二转动包括垂直或水平方向上的移动、以及垂直或水平方向上的移动。
  8. 根据权利要求7所述的摄像头的控制装置,其特征在于,所述摄像头的周围设置有若干补光灯,该装置还包括:
    图像亮度确定模块,用于确定所述目标对象的姿态参数符合预设姿态参数标准的图像的亮度参数;
    补光指令发送模块,用于若所述亮度参数未符合预设亮度参数标准,则向所述若干补光灯中的目标补光灯发送补光指令,直至所述摄像头发送的图像的亮度参数符合预设亮度参数标准。
  9. 根据权利要求8所述的摄像头的控制装置,其特征在于,所述目标对象在图像中占据多个图像区域,各个图像区域分别与若干补光灯对应;
    相应地,补光指令发送模块在若干所述补光灯中确定目标补光灯的步骤包括:
    分别确定所述目标对象的每个所述图像区域的区域亮度参数;
    将区域亮度参数低于预设亮度参数的图像区域所对应的补光灯确定为目标补光灯。
  10. 根据权利要求8所述的摄像头的控制装置,其特征在于,补光指令发送模块向所述目标补光灯发送的各个所述补光指令中包括补光量级,且各个所述补光量级依次递增,增幅为固定值。
  11. 根据权利要求7所述的摄像头的控制装置,其特征在于,搜索指令发送模块向所述摄像头发送的各个所述搜索指令中包括垂直转动角度,且各个所述垂直转动角度依次递增,增幅为固定值。
  12. 根据权利要求7所述的摄像头的控制装置,其特征在于,所述微调指令发送模块生成微调指令的方式具体为:
    依据左右侧转角度计算公式α=|α12|,计算左右侧转角度α;
    其中,α1及α2的获得方式分别为:
    Figure PCTCN2016081532-appb-100008
    Figure PCTCN2016081532-appb-100009
    其中,m1、m2及m3的获得方式分别为:
    Figure PCTCN2016081532-appb-100010
    Figure PCTCN2016081532-appb-100011
    其中,所述(x,y)、(x1,y1)、(x2,y2)及(x3,y3)分别为所述目标对象的第一预设位置点、第二预设位置点、第三预设位置点及第四预设位置点分别在图像中的坐标;
    依据上下俯仰角度计算公式
    Figure PCTCN2016081532-appb-100012
    计算上下俯仰角度β;
    其中,m4的获得方式分别为:
    Figure PCTCN2016081532-appb-100013
    若左右侧转角度α≥N,生成水平向左移动|DFC tanα|且向右旋转α度的微调指令;其中,N为预设常数;
    若左右侧转角度α≤-N,生成水平向右移动|DFC tanα|且向左旋转α度的微调指令;
    若上下俯仰角度β≥N,生成垂直向上移动|DFC tanβ|且向下旋转β度的微调指令;
    若上下俯仰角度β≤-N,生成垂直向下移动|DFC tanβ|且向上旋转β度的微调指令;
    其中,DFC的获得方式分别为:
    Figure PCTCN2016081532-appb-100014
    其中,d为第一预设位置点与第二预设位置点在图像上的距离;D为第一预设位置点与第二预设位置点的实际距离;m为第一预设位置点与第二预设位置点的中心与所述目标对象的第五预设位置点在图像上的距离;M为第一预设位置点与第二预设位置点的中心与所述目标对象的第五预设位置点的实际距离。
  13. 一种金融设备终端,其特征在于,包括:摄像头以及上述权利要求6~12任意一项所述的摄像头的控制装置。
PCT/CN2016/081532 2015-12-31 2016-05-10 摄像头的控制方法及装置、金融设备终端 WO2017113559A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
US15/774,577 US10674059B2 (en) 2015-12-31 2016-05-10 Method and device for controlling camera and financial equipment terminal
EP16880367.4A EP3399739A4 (en) 2015-12-31 2016-05-10 Method and device for controlling camera and financial equipment terminal
RU2018119912A RU2691109C1 (ru) 2015-12-31 2016-05-10 Способ и устройство для управления камерой и финансовым терминальным оборудованием
ZA2018/03110A ZA201803110B (en) 2015-12-31 2018-05-11 Method and device for controlling camera and financial equipment terminal

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201511034588.3A CN105554391B (zh) 2015-12-31 2015-12-31 摄像头的控制方法及装置、金融设备终端
CN201511034588.3 2015-12-31

Publications (1)

Publication Number Publication Date
WO2017113559A1 true WO2017113559A1 (zh) 2017-07-06

Family

ID=55833309

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/081532 WO2017113559A1 (zh) 2015-12-31 2016-05-10 摄像头的控制方法及装置、金融设备终端

Country Status (6)

Country Link
US (1) US10674059B2 (zh)
EP (1) EP3399739A4 (zh)
CN (1) CN105554391B (zh)
RU (1) RU2691109C1 (zh)
WO (1) WO2017113559A1 (zh)
ZA (1) ZA201803110B (zh)

Families Citing this family (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105554391B (zh) * 2015-12-31 2019-05-14 广州广电运通金融电子股份有限公司 摄像头的控制方法及装置、金融设备终端
CN106210529B (zh) * 2016-07-29 2017-10-17 广东欧珀移动通信有限公司 移动终端的拍摄方法以及装置
JP6794918B2 (ja) * 2017-04-28 2020-12-02 トヨタ自動車株式会社 画像送信プログラム、及び、画像送信装置
CN107155063B (zh) * 2017-06-19 2020-10-20 深圳市泰衡诺科技有限公司上海分公司 夜景拍摄控制方法、系统以及设备
CN109543491B (zh) * 2017-09-21 2023-04-28 厦门歌乐电子企业有限公司 一种疲劳驾驶监测装置和方法
CN108174115B (zh) * 2017-12-26 2020-05-26 浙江大华技术股份有限公司 一种飞碟摄像机补光方法及装置
CN108163203B (zh) * 2017-12-31 2020-10-13 深圳市道通智能航空技术有限公司 一种拍摄控制方法、装置及飞行器
CN108764044A (zh) * 2018-04-25 2018-11-06 维沃移动通信有限公司 一种补光方法、装置和移动终端
CN109151337A (zh) * 2018-10-25 2019-01-04 Oppo(重庆)智能科技有限公司 人脸识别补光方法、人脸识别补光装置及移动终端
CN111369622B (zh) * 2018-12-25 2023-12-08 中国电子科技集团公司第十五研究所 虚实叠加应用的相机世界坐标位置获取方法、装置和系统
CN109729268B (zh) * 2018-12-26 2021-03-02 武汉市澜创信息科技有限公司 一种人脸拍摄方法、装置、设备及介质
WO2021026753A1 (zh) * 2019-08-13 2021-02-18 深圳市大疆创新科技有限公司 云台控制方法、云台及计算机可读存储介质
CN110445975A (zh) * 2019-09-09 2019-11-12 河北昂帆科技有限公司 一种高清夜视高速智能摄像机
JP7383438B2 (ja) * 2019-09-27 2023-11-20 キヤノン株式会社 撮像制御装置、撮像装置、及び撮像制御方法
CN112672062B (zh) * 2020-08-21 2022-08-09 海信视像科技股份有限公司 一种显示设备及人像定位方法
CN112822392B (zh) * 2020-12-25 2022-12-13 维沃移动通信有限公司 一种控制摄像模组旋转的方法及装置
CN112712035A (zh) * 2020-12-30 2021-04-27 精英数智科技股份有限公司 采煤机跟机移架检测方法、装置及系统
CN113450478A (zh) * 2021-06-22 2021-09-28 广州小鹏汽车科技有限公司 车辆的解锁方法、车辆系统、车辆及计算机可读存储介质
CN114630048B (zh) * 2022-03-16 2023-08-22 湖州师范学院 一种自媒体短视频拍摄装置和方法
CN116977332A (zh) * 2023-09-21 2023-10-31 合肥联宝信息技术有限公司 摄像头补光灯性能测试方法、装置、电子设备及存储介质
CN117177069B (zh) * 2023-11-02 2024-01-30 百鸟数据科技(北京)有限责任公司 一种基于onvif协议的摄像头自动追踪方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060281969A1 (en) * 2005-06-02 2006-12-14 Vimicro Corporation System and method for operation without touch by operators
CN103248824A (zh) * 2013-04-27 2013-08-14 天脉聚源(北京)传媒科技有限公司 一种摄像头拍摄角度的确定方法、装置及摄像系统
CN103841313A (zh) * 2012-11-27 2014-06-04 华为技术有限公司 云台摄像机控制方法、系统及设备
CN105554391A (zh) * 2015-12-31 2016-05-04 广州广电运通金融电子股份有限公司 摄像头的控制方法及装置、金融设备终端

Family Cites Families (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE19511713A1 (de) * 1995-03-30 1996-10-10 C Vis Computer Vision Und Auto Verfahren und Vorrichtung zur automatischen Bildaufnahme von Gesichtern
US5912721A (en) * 1996-03-13 1999-06-15 Kabushiki Kaisha Toshiba Gaze detection apparatus and its method as well as information display apparatus
US20030217005A1 (en) * 1996-11-27 2003-11-20 Diebold Self Service Systems, Division Of Diebold, Incorporated Automated banking machine system and method
US7755668B1 (en) * 1998-04-09 2010-07-13 Johnston Gregory E Mobile surveillance system
GB9907515D0 (en) * 1999-04-01 1999-05-26 Ncr Int Inc Self service terminal
RU2199150C2 (ru) * 2001-02-02 2003-02-20 Курский государственный технический университет Устройство калибровки оптикоэлектронной системы
US6606458B2 (en) * 2001-09-05 2003-08-12 Nisca Corporation Automatic framing camera
US7982763B2 (en) * 2003-08-20 2011-07-19 King Simon P Portable pan-tilt camera and lighting unit for videoimaging, videoconferencing, production and recording
CN101420523A (zh) * 2007-10-23 2009-04-29 鸿富锦精密工业(深圳)有限公司 拍照装置及方法
US20100225773A1 (en) 2009-03-09 2010-09-09 Apple Inc. Systems and methods for centering a photograph without viewing a preview of the photograph
DE102009018318A1 (de) * 2009-04-22 2010-10-28 Wincor Nixdorf International Gmbh Selbstbedienungsterminal mit mindestens einer Bilddaten erzeugenden Kamera zum Erkennen von Manipulationsversuchen
JP5322287B2 (ja) * 2009-05-28 2013-10-23 パナソニック株式会社 回転台付カメラ装置
GB201103367D0 (en) * 2011-02-28 2011-04-13 360 Vision Technology Ltd Security camera
WO2012120386A1 (en) * 2011-03-04 2012-09-13 Koninklijke Philips Electronics N.V. Apparatus and method for luminance control
TWI465825B (zh) * 2012-06-27 2014-12-21 Acer Inc 影像擷取裝置與其光源輔助拍攝方法
CA3148411C (en) * 2013-10-07 2024-05-14 Google Llc Smart-home hazard detector providing context specific features and/or pre-alarm configurations

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060281969A1 (en) * 2005-06-02 2006-12-14 Vimicro Corporation System and method for operation without touch by operators
CN103841313A (zh) * 2012-11-27 2014-06-04 华为技术有限公司 云台摄像机控制方法、系统及设备
CN103248824A (zh) * 2013-04-27 2013-08-14 天脉聚源(北京)传媒科技有限公司 一种摄像头拍摄角度的确定方法、装置及摄像系统
CN105554391A (zh) * 2015-12-31 2016-05-04 广州广电运通金融电子股份有限公司 摄像头的控制方法及装置、金融设备终端

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3399739A4 *

Also Published As

Publication number Publication date
CN105554391B (zh) 2019-05-14
US10674059B2 (en) 2020-06-02
EP3399739A1 (en) 2018-11-07
ZA201803110B (en) 2019-01-30
RU2691109C1 (ru) 2019-06-11
US20180324345A1 (en) 2018-11-08
EP3399739A4 (en) 2018-12-19
CN105554391A (zh) 2016-05-04

Similar Documents

Publication Publication Date Title
WO2017113559A1 (zh) 摄像头的控制方法及装置、金融设备终端
US11102417B2 (en) Target object capturing method and device, and video monitoring device
JP6610906B2 (ja) 活動検出方法及びデバイス、並びに本人認証方法及びデバイス
US10528810B2 (en) Detecting user viewing difficulty from facial parameters
JP6700661B2 (ja) 画像処理装置、画像処理方法、及び画像処理システム
US7853050B2 (en) System and method for operation without touch by operators
WO2017161867A1 (zh) 一种调节屏幕亮度的方法、装置及智能终端
CN101893858B (zh) 一种用于控制用户眼睛与电子设备屏幕之间距离的方法
US11107246B2 (en) Method and device for capturing target object and video monitoring device
WO2016119339A1 (zh) 图像修正方法、图像修正装置和视频系统
US20170316582A1 (en) Robust Head Pose Estimation with a Depth Camera
US20170104915A1 (en) Display control apparatus, display control method, and storage medium
WO2020078440A1 (zh) 采集高清晰度面部图像的装置和摄像头云台自动俯仰调节的方法
US9098243B2 (en) Display device and method for adjusting observation distances thereof
WO2015165209A1 (zh) 穿戴式投影装置及其调焦方法、投影方法
US10986289B2 (en) Method and device for regulating imaging accuracy of motion-sensing camera
US10630890B2 (en) Three-dimensional measurement method and three-dimensional measurement device using the same
JP4968922B2 (ja) 機器制御装置及び制御方法
CN110769148A (zh) 一种基于人脸识别的摄像头自动控制方法及装置
US20230169793A1 (en) Photographing system, photographing method, and non-transitory computer-readable medium storing photographing program
CN108108024A (zh) 动态手势获取方法及装置、显示装置
JP6210946B2 (ja) エレベータかご内カメラの画角調整装置およびエレベータかご内カメラの画角調整方法
CN106664373A (zh) 一种基于自动变焦的自适应图像跟踪方法
TW201620632A (zh) 金屬板材頭端檢測方法與金屬板材軋延方法
JP6468078B2 (ja) 視線キャリブレーションプログラム、視線キャリブレーション装置、および視線キャリブレーション方法

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: 16880367

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 15774577

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2018119912

Country of ref document: RU