WO2019114617A1 - 快速抓拍的方法、装置及系统 - Google Patents

快速抓拍的方法、装置及系统 Download PDF

Info

Publication number
WO2019114617A1
WO2019114617A1 PCT/CN2018/119698 CN2018119698W WO2019114617A1 WO 2019114617 A1 WO2019114617 A1 WO 2019114617A1 CN 2018119698 W CN2018119698 W CN 2018119698W WO 2019114617 A1 WO2019114617 A1 WO 2019114617A1
Authority
WO
WIPO (PCT)
Prior art keywords
camera
distance
target
angle
ptz
Prior art date
Application number
PCT/CN2018/119698
Other languages
English (en)
French (fr)
Inventor
朱力于
杨昆
骆立俊
张德
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2019114617A1 publication Critical patent/WO2019114617A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60—Control of cameras or camera modules

Definitions

  • the present application relates to the field of video surveillance, and in particular, to a method, device and system for implementing fast capture.
  • the camera generally adopts the contrast focusing method to perform the focusing operation, and uses the intensity of the contrast texture of the image to determine whether the current focus is at the clearest position.
  • the focusing technology directly determines the acquired image information, so the accuracy is high, but since the entire focusing process needs to be analyzed, the focusing speed is slow and cannot meet the requirements of fast focusing.
  • the existing capture technology solution is a combination of a fixed focal length pan-focus lens and a wide-angle lens. After finding a target on a wide-angle lens, the pan-focus lens has a small aperture and a large depth of field to directly capture the target of a fixed distance range.
  • the pan-focus lens adopts a fixed aperture and a focal length, it can only be designed with a fixed-focus lens, and can only be used for shooting in a limited range of objects, and cannot be zoomed, and the range of use is limited.
  • the embodiment of the invention provides a method, a device and a system for quickly capturing, which can realize fast positioning, fast focusing and fast capture of an image of a target after detecting a target.
  • the embodiment of the present invention provides the following technical solutions:
  • an embodiment of the present invention provides a method for quickly capturing a video.
  • the method includes: first, detecting a target in a current picture monitored by a first camera, calculating a distance and an angle between the target and the first camera; and then, according to the obtained The distance and angle calculate the shooting parameters required for the second camera to capture the target.
  • the shooting parameters include the lens focal length and the shooting angle.
  • the target is captured after adjusting the angle of the second camera and the focal length of the lens according to the shooting parameters. Since the target is detected by the first camera and the distance of the target is calculated, thereby obtaining the shooting parameters of the second camera, the focusing time of the second camera can be reduced, and the image of the target can be captured more timely and clearly.
  • the first camera is a stereoscopic camera and the second camera is a PTZ (Pan/Tilt/Zoom) camera.
  • the PTZ camera When receiving the distance and angle calculated by the first camera, the PTZ camera can set the angle to the shooting angle of the PTZ camera, set the distance as the object distance of the target and the PTZ camera, and obtain the lens focal length of the PTZ camera according to the object distance. And adjusting the PTZ camera capture target according to the above shooting angle and lens focal length.
  • the lens focal length of the PTZ camera can be obtained by querying a comparison table of the object distance and the focal length of the lens of the PTZ camera.
  • the distance estimation error corresponding to the distance can be calculated when calculating the distance of the target, and when the object distance is obtained, the depth of field of the PTZ camera at the object distance is calculated, and the distance is compared. Estimate the error and the depth of field. If the distance estimation error is less than or equal to the depth of field, adjust the PTZ camera according to the shooting parameters and capture the target; if the distance estimation error is greater than the depth of field, adjust the PTZ camera according to the shooting parameters, automatically adjust the focal length of the lens and capture the target. Because when the depth of field is greater than or equal to the distance estimation error, the depth of field can cover the distance estimation error range of the target.
  • the capture sequence of multiple targets captures the targets according to the determined capture order, and the parameters according to the priority determination of the capture order of the multiple targets include the orientation, angle, distance, running track or whether the monitoring area is about to leave the target.
  • an embodiment of the present invention provides a fast capture system, including: a first camera, configured to acquire a current monitoring image, detect a target in the monitoring image, calculate a distance and an angle between the target and the first camera, and The distance and the angle are sent to the second camera; the second camera is configured to receive the distance and the angle sent by the first camera, calculate the shooting parameters of the second camera according to the distance and the angle, and adjust the second camera capture target according to the shooting parameter, where Shooting parameters include lens focal length and shooting angle.
  • the embodiment of the present invention further provides a fast capture device, including: a first camera for acquiring a current monitoring picture from two lenses at the same time; and a processor configured to detect a target in a current picture of the first camera and Calculating a distance and an angle between the target and the first camera, calculating a shooting parameter of the second camera according to the distance and the angle, and controlling the second camera to capture the target, wherein the shooting parameters include a shooting angle and a lens focal length; and the second camera uses Capture the target.
  • a fast capture device including: a first camera for acquiring a current monitoring picture from two lenses at the same time; and a processor configured to detect a target in a current picture of the first camera and Calculating a distance and an angle between the target and the first camera, calculating a shooting parameter of the second camera according to the distance and the angle, and controlling the second camera to capture the target, wherein the shooting parameters include a shooting angle and a lens focal length; and the second camera uses Capture the target.
  • an embodiment of the present invention further provides a computer readable storage medium, configured to store computer software instructions for use in the fast capture device and/or system, including the method for performing the method provided by the first aspect above. Program code.
  • the embodiment of the present invention discloses a fast capture method, device and system.
  • the distance and the angle of the target are quickly measured by the stereo vision technology, and then the corresponding shooting parameters of the camera are obtained, and according to the Shooting parameters Quickly adjust the camera to capture the target.
  • the method disclosed in the embodiment of the present invention can more accurately and quickly determine the shooting parameters of the camera, reduce the focusing time of the camera, and has higher accuracy, making the capturing more timely, and the image More clear.
  • FIG. 1 is a schematic diagram of a fast capture system according to an embodiment of the present invention.
  • FIG. 2 is a schematic structural diagram of a fast capture device according to an embodiment of the present invention.
  • FIG. 3 is a schematic structural diagram of a network camera according to an embodiment of the present invention.
  • FIG. 4 is a schematic diagram of a method for fast capture according to an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of calculating a distance and an angle between a target and a stereoscopic camera according to an embodiment of the present invention.
  • FIG. 6 is a schematic diagram of target coordinate conversion according to an embodiment of the present invention.
  • FIG. 7 is a schematic diagram of a method for calculating a depth of field according to an embodiment of the present invention.
  • a fast capture system 100 includes a first camera and a second camera, wherein the first camera is a stereoscopic camera 110 and the second camera is a PTZ (Pan/Tilt). /Zoom) Camera 120.
  • the first camera is a stereoscopic camera 110 and the second camera is a PTZ (Pan/Tilt). /Zoom) Camera 120.
  • the stereo vision camera 110 in the system is equipped with two lenses, which can simulate the principle of human vision, simultaneously acquire two images of the current scene from two different angles, and utilize the difference of the angles of the two images, according to the pixels between the images.
  • the matching relationship is used to calculate the actual distance and angle between the target in the image and the stereoscopic camera, and then calculate the coordinates of the target in the coordinate system of the stereoscopic camera 110.
  • the stereoscopic camera 110 in the embodiment of the present invention refers to a camera or a camera group that can be used to simultaneously acquire a current scene image from different lenses.
  • the name itself is not limited to the device, and may also be other names, such as a binocular stereoscopic camera. , binocular camera, binocular rangefinder or stereo vision rangefinder camera.
  • two ordinary cameras can be used to form a stereoscopic camera, or two ordinary camera lenses can be integrated into one device to form a stereoscopic camera.
  • the fast capture system 100 in the embodiment of the present invention may further include multiple sets of stereo vision cameras to expand the monitoring range.
  • the technical solution of the present invention will be described below only in the case of a set of stereoscopic cameras.
  • the case of a plurality of sets of stereoscopic cameras can refer to a technical solution under a set of stereoscopic cameras.
  • the PTZ camera 120 in the fast capture system 100 in the embodiment of the present invention is equipped with a pan/tilt head, which can realize omnidirectional (left/right/up and down) movement and lens zoom and zoom control for the distance between the target and the stereoscopic camera. And the angle, the PTZ camera 120 captures the shooting parameters of the target, including capturing the focal length and the shooting angle, and adjusting the image of the target after the PTZ camera 120 is adjusted according to the obtained parameters.
  • the stereoscopic camera 110 When the fast capture system 100 is in operation, the stereoscopic camera 110 simultaneously acquires the current monitoring image from the two lenses.
  • the current monitoring screen includes a first screen and a second screen, and the first screen and the second screen are respectively A picture taken by the first lens and the second lens of the stereoscopic camera 110 at the same time.
  • the stereoscopic camera 110 detects a target in the current picture and calculates the distance and angle of the target from the stereoscopic camera 110 after detecting the target. If a plurality of targets are detected, the distances and angles of the respective targets from the stereoscopic camera 110 are calculated separately.
  • the stereoscopic camera 110 transmits the distance and angle of the target to the stereoscopic camera 110 to the PTZ camera 120.
  • the shooting parameters of the PTZ camera 120 capture target are calculated according to the distance and angle of the target from the stereoscopic camera 110.
  • the PTZ camera 120 adjusts the angle of the PTZ camera 120 and the focal length of the PTZ camera 120 lens in accordance with the shooting parameters. After the adjustment is completed, the PTZ camera 120 captures an image of the target.
  • the stereoscopic camera 110 can also calculate a corresponding distance estimation error of the target and send it to the PTZ camera 120, which can also calculate the target corresponding to the object distance of the PTZ camera 120.
  • Depth of field The PTZ camera 120 compares the distance estimation error of the target and the depth of field at the object distance before adjustment. If the depth of field is greater than or equal to the distance estimation error, the PTZ camera 120 is directly adjusted according to the shooting parameters and captures the target; if the depth of field is smaller than the distance estimation error, the shooting parameters are followed. The PTZ camera 120 continues to control the auto focus after the PTZ camera 120 is adjusted.
  • the stereoscopic camera 110 calculates the distance and angle of each target from the stereoscopic camera 110, respectively.
  • the PTZ camera 120 is based on each target and stereo.
  • the distance and the angle of the visual camera 110 respectively obtain the shooting parameters of the PTZ camera 120 corresponding to the respective targets, and determine the capturing order of the plurality of targets according to the priority, respectively adjust the PTZ camera 120 to capture the images of the plurality of targets according to the determined capturing order, and determine
  • the parameters by which the priority is based include the orientation, angle, distance, trajectory of the target, or whether it is about to leave the monitoring area.
  • the fast capture system in the above embodiment separates the stereo vision camera 110 and the PTZ camera 120, and the two independently perform corresponding calculation and control functions.
  • the stereo vision camera 110 and the PTZ camera 120 can be connected by wire or wirelessly to realize information interaction between the two devices.
  • the calculation operations in the stereo vision camera 110 such as the detection target and the distance and angle between the calculation target and the stereo vision camera 110, may also be performed by the PTZ camera 120, and those skilled in the art may
  • the rapid capture system makes the above-described changes, and such modifications are considered to fall within the scope of the claims and equivalents thereof.
  • FIG. 2 is a schematic diagram showing a possible structure of a fast capture device according to an embodiment of the present invention.
  • the device includes: a first camera 210, a processor 220, a memory 230, a second camera 240, and at least one communication bus 250, wherein
  • the first camera 210 may be a stereoscopic camera 210 and the second camera 240 may be a PTZ camera 240.
  • Processor 220 can be a general purpose central processing unit CPU, microprocessor, ASIC, or one or more integrated circuits for controlling the execution of the program of the present invention.
  • the processor 220 can also be implemented using an FPGA or a DSP.
  • the memory 230 may be a volatile memory such as a random-access memory (RAM); or a non-volatile memory, a hard disk drive (HDD) or A solid-state drive (SDD); or a combination of the above types of memory and provides instructions and data to the processor.
  • RAM random-access memory
  • HDD hard disk drive
  • SDD solid-state drive
  • the bus 250 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus or the like.
  • PCI Peripheral Component Interconnect
  • EISA Extended Industry Standard Architecture
  • the bus 850 can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one thick line is shown in Figure 2, but it does not mean that there is only one bus or one type of bus.
  • the stereoscopic camera 210 includes two lenses for acquiring a current monitoring picture from two lenses at the same time.
  • the current monitoring picture includes a first picture and a second picture, and the first picture and the second picture are respectively two of the stereoscopic camera 210. The picture taken at the same time.
  • the processor 220 detects the target in the current monitoring screen of the stereo vision camera 210, calculates the distance and angle of the target from the stereoscopic camera 210, and calculates the shooting parameters corresponding to the PTZ camera 240 according to the distance and angle of the target from the stereoscopic camera 210, and the shooting parameters. Includes shooting angle and lens focal length. After obtaining the shooting parameters required by the PTZ camera 240 to capture the target, the processor 220 generates a corresponding control signal according to the shooting parameters, and controls the PTZ camera 240 to adjust the shooting angle and the lens focal length.
  • the PTZ camera 240 can be rotated up and down, left and right, and can be rotated to a corresponding angle according to the shooting angle in the shooting parameters.
  • the focal length of the PTZ camera 240 is variable, and the processor 220 can control the PTZ camera 240 to adjust the lens focal length and capture the image of the target.
  • the PTZ camera 240 can also autofocus, and the autofocus process can be completed in accordance with the autofocus command sent by the processor 220.
  • the processor 220 is further configured to calculate a corresponding distance estimation error according to the distance between the target and the stereo vision camera 210, and calculate the PTZ camera 240 according to the object distance of the target and the PTZ camera 240 and the lens focal length.
  • the processor 220 compares the relationship between the distance estimation error and the depth of field according to the corresponding depth of field. If the distance estimation error is less than or equal to the depth of field, the processor 220 controls the PTZ camera 240 to adjust and capture the image of the target according to the shooting parameters; if the distance estimation error is greater than the depth of field, After the processor 220 controls the capture camera 240 to adjust according to the shooting parameters, the PTZ camera 240 is controlled to automatically adjust the lens focal length.
  • the processor 220 detects a plurality of targets, determining a snap order of the plurality of targets according to the priority, and controlling the PTZ camera 240 to capture images of the plurality of targets according to the determined snap order, wherein the priority is determined.
  • the parameters based on the target include the orientation, angle, distance, running track or whether it is about to leave the monitoring area.
  • the fast capture device 200 may further include a communication interface 260 for transmitting an image of the captured target to the external device.
  • the fast capture device 200 in the above embodiment is a device that integrates the stereoscopic camera 210, the processor 220, and the PTZ camera 240 as a whole, and the three components of the fast capture device 200 can be separately connected and connected by wire or wireless.
  • the communication between the components can be realized, and the three components can also be combined, and the functions of the processor 220 and the memory 230 can also be implemented using one device.
  • the stereoscopic camera 210 and the PTZ camera 240 in the fast capture device 200 are mainly used to acquire a target image, and the processor 220 is used to perform calculation and control operations. It should be noted that, according to actual needs, part of the calculation or control function of the processor 220 can be implemented by the stereo vision camera 210 or the PTZ camera 240.
  • Those skilled in the art can make the above changes to the fast capture system in this embodiment. Modifications are considered to be within the scope of the appended claims and their equivalents.
  • the network camera 300 includes the structure common to the first camera and the first camera in the above embodiment, and for ease of understanding, the standard features of the network camera 300 that are not relevant to the present invention are not described.
  • the network camera 300 includes a lens 310 as a front end component of the network camera 300.
  • the lens 310 has a fixed aperture, an automatic aperture, an automatic zoom, an automatic zoom, and the like; an image sensor 320 for recording incident light, such as a complementary metal oxide semiconductor (Complementary) Metal Oxide Semiconductor (CMOS), Charge-coupled Device (CCD) or similar device; image processor 330; processor 340 for performing computational operations and controlling the camera; memory 350 for storing programs or data; A communication bus 360 for communicating information to various components and a communication interface 370 for communicating information to other nodes connected to the network over the communication network.
  • CMOS complementary metal oxide semiconductor
  • CCD Charge-coupled Device
  • Image sensor 320 receives information relating to the recorded light and processes the information by means of an A/D converter and signal processor 331, which is well known to the skilled person.
  • image sensor 320 includes an A/D converter, so no A/D converter is required in image processor 330.
  • the result produced by the A/D converter and signal processor 331 is digital image data which, in accordance with one embodiment, is processed in scaling unit 332 and image encoder 333 before being sent to processor 340.
  • the scaling unit 332 is configured to process the digital image data into at least one image of a particular size.
  • the scaling unit 332 can be arranged to generate a plurality of images of different sizes, all representing the same image/frame provided by the A/D converter and signal processor 331.
  • the functionality of scaling unit 332 is performed by image encoder 333, which in yet another embodiment does not require any scaling or resizing of the image from image sensor 320.
  • Encoder 333 is optional for performing the present invention and is arranged to encode the digital image data into a plurality of known formats for a continuous video sequence, for a limited video sequence, for a still image, or for an image/video stream Any of them.
  • image information can be encoded into MPEG1, MPEG2, MPEG4, JEPG, MJPG, bitmap, and the like.
  • Processor 340 can use an unencoded image as input data.
  • the image data is transmitted from the signal processor 331 or from the scaling unit 332 to the processor 340 without passing the image data through the image encoder 333.
  • the unencoded image may take any unencoded image format, such as BMP, PNG, PPM, PGM, PNM, and PBM, although the processor 340 may also use the encoded data as input data.
  • image data may be sent directly from signal processor 331 to processor 340 without passing through scaling unit 332 or image encoder 333. In still another embodiment, image data may be sent from scaling unit 332 to processor 340 without passing through image encoder 333.
  • the processor 340 can be a general purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more for controlling the execution of the program of the present invention. integrated circuit.
  • the processor 340 can also be implemented by using a Field Programmable Gate Array (FPGA) or a DSP. Some of the functions in image processor 330 may also be integrated on processor 340 when using DSP-based software encoding compression.
  • the processor 340 is used to manage and control the network camera 300.
  • the memory 350 is used to store application code for executing the solution of the present invention. It may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions. Random access memory (random access memory) , RAM) or other types of dynamic storage devices that can store information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (compact disc read-only memory) , CD-ROM) or other disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), disk storage media or other magnetic storage devices, or can be used to carry or store instructions or The desired program code in the form of a data structure and any other medium that can be accessed by a computer, but is not limited thereto. Memory 350 may be present independently and coupled to processor 340 via bus 360. Memory 350 can also be integrated with processor 340.
  • ROM read-only memory
  • RAM random access memory
  • EEPROM electrically erasable programm
  • Communication bus 360 can include a path for communicating information between components.
  • Communication interface 370 uses devices such as any transceiver for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), and the like.
  • devices such as any transceiver for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), and the like.
  • RAN radio access network
  • WLAN wireless local area networks
  • the stereoscopic camera in the above embodiment is provided with a plurality of lenses on the basis of the universal camera 300 to obtain the currently detected picture from the plurality of lenses; the PTZ camera in the above embodiment sets the cloud on the basis of the universal camera 300. To realize the omnidirectional (left/right/up and down) movement, adjust the shooting angle of the PTZ camera.
  • a fast capture method which uses a stereo vision camera to assist in capturing a camera focus when capturing a target, and solves the problem of excessive focusing time in the existing capture technology.
  • the fast capture method provided by the embodiment of the present invention can be applied to the fast capture system 100 in FIG. 1 and the fast capture device 200 in FIG. 2 for capturing a fast moving target in a large scene, which is described below in conjunction with FIG. A specific implementation of the method provided by the embodiment of the present invention is described.
  • the fast capture method provided by the embodiment of the present invention includes the following steps:
  • 410 Detect a target in a current picture of the first camera, and calculate a distance and an angle between the target and the first camera.
  • the first camera may be a stereoscopic camera 110, and the current picture monitored includes a first picture and a second picture, and the first picture and the second picture are respectively the first lens of the stereoscopic camera 110 and The second shot is taken at the same time.
  • the target in the picture can be detected, and the distance and angle of the target from the stereoscopic camera 210 can be calculated based on the visual difference of the target in different pictures.
  • the second camera is a PTZ camera 110.
  • the object distance and the angle of the target and the PTZ camera 120 can be obtained according to the positional relationship between the stereo vision camera 110 and the PTZ camera 120, thereby obtaining the shooting parameters of the PTZ camera 120 capture target.
  • the shooting parameters include a shooting angle and a lens focal length.
  • the shooting angle and the lens focal length are the two most important parameters of the PTZ camera 120 capturing target. According to the above steps, the shooting angle and the lens focal length of the PTZ camera 120 can be obtained, and the PTZ camera 120 can be controlled to rotate to a corresponding angle and control the PTZ camera. Zoom the lens of 120, adjust the focal length of the lens to the corresponding value, and then capture the target.
  • the PTZ camera 120 achieves the capture of the target without autofocus, avoids the long-time autofocus process, improves the capture efficiency, makes the capture more timely, and the image is clearer.
  • a method for calculating a distance between a target and a stereoscopic camera 110 is provided in an embodiment of the present invention.
  • Ol and Or are respectively the center positions of the left and right cameras of the stereoscopic camera 110, and the target points P are respectively imaged to the P1 and Pr points on the target surface of the left and right cameras, and the optical path is from the target point to the target surface through the center of the lens plane.
  • B is the baseline distance, ie the distance between the left and right cameras; f is the focal length of the stereoscopic camera; xl is the horizontal distance of the target at the left imaging point and the left camera center point; xr is the target at the right camera imaging point and The horizontal distance of the center point of the right camera; D is the distance between the target and the stereo vision camera.
  • the angular relationship between the target and the left camera, the angular relationship between the target and the right camera, and the angle between the target and the center line of the stereoscopic camera can be calculated based on the trigonometric function calculation formula based on the calculation of the target distance.
  • the distance measured by the stereo ranging technology is not absolutely accurate, and the measured distance has a certain error, which is the distance estimation error, the error range and the lens parameters of the stereoscopic camera, the pixel size of the photoreceptor, and the two lenses.
  • the spacing is related.
  • the distance estimation error increases as the target distance increases, and is proportional to the square of the distance. Thomas Luhmann gives a calculation method for distance estimation error in "Close-Range Photogrammetry and 3D Imaging” (2014).
  • the coefficient between the distance estimation error and the square of the distance can be obtained by multiple measurements, and the relationship between the measured target distance and the distance estimation error can be obtained.
  • the above calculation method is only a calculation method adopted by an embodiment of the present invention, and does not limit the scope of protection of the present invention, no matter what specific method is adopted.
  • the distance calculation method is within the protection scope of the present invention as long as the function of measuring the distance and angle of the target by the stereo ranging technique can be realized.
  • the object distance of the target and the PTZ camera 120 in step 420 and the shooting parameters of the PTZ camera 120 can be obtained in a variety of ways.
  • the distance between the stereo vision camera 110 and the target may be directly set as the object distance of the target and the PTZ camera 120, and the angle of the stereo vision camera 110 and the target is set to the shooting angle of the PTZ camera 120, according to the target and The object distance of the PTZ camera 120 results in the focal length of the PTZ camera 120.
  • the fast capture system 100 in the embodiment of the present invention is generally used to capture a target at a relatively long distance, the distance between the target and the fast capture system 100 is much larger than the distance between the stereoscopic camera 110 lens and the PTZ camera 120 lens. In the case where the shooting accuracy is not high, the distance and angle of the target from the stereoscopic camera 110 can be considered to be equal to the object distance and angle of the PTZ camera 120.
  • the stereoscopic camera 110 generally includes two lenses.
  • the center position of the two lenses is usually taken as the origin. If the PTZ camera 120 is located at the center of the two lenses of the stereoscopic camera 110, the stereoscopic camera 110 coordinates. The system coincides with the PTZ camera 120 coordinate system, at which time the distance and angle of the target from the stereoscopic camera 110 are equal to the object distance and angle of the target and the PTZ camera 120.
  • the spatial position difference between the stereo vision camera 110 and the PTZ camera 120 needs to be considered, and the object distance between the target and the PTZ camera 120 is calculated according to the spatial position difference and the distance and angle between the target and the stereoscopic camera 110. And the angle, and the angle is set to the shooting angle of the PTZ camera 120, and the shooting focal length of the PTZ camera 120 is obtained according to the object distance of the target and the PTZ camera 120.
  • the coordinates of the target in the PTZ camera 120 can be obtained by three-dimensional coordinate conversion.
  • a method for realizing coordinate conversion between a stereo vision camera and a PTZ camera according to an embodiment of the present invention, wherein coordinate O is a coordinate origin of a PTZ camera coordinate system, and coordinate O′ is a stereo vision camera coordinate system.
  • coordinate O is a coordinate origin of a PTZ camera coordinate system
  • coordinate O′ is a stereo vision camera coordinate system.
  • ⁇ is the scale factor between the two coordinate systems
  • ⁇ X, ⁇ Y and ⁇ Z are the positional differences between the PTZ camera coordinate system origin and the origin of the stereo vision camera coordinate system
  • R is the rotation matrix for coordinate transformation.
  • the coordinate axis corresponding to the coordinate system of the PTZ camera is rotated by rotating each coordinate axis of the coordinate system of the stereoscopic camera.
  • the coordinates of the target in the coordinate system of the PTZ camera 120 can be obtained, thereby obtaining the object distance of the target and the angle with the PTZ camera 120.
  • the lens focal length of the PTZ camera 120 can be obtained in a variety of ways. Specifically, as an implementation manner, after calculating the object distance of the target by storing a comparison table of the object distance and the focal length of the lens in advance, the comparison table is queried to obtain the shooting focal length of the PTZ camera 120.
  • a comparison table between the distance and the angle between the target and the stereoscopic camera 110 and the shooting parameters of the PTZ camera 120 may be pre-stored, and the pre-stored comparison table is obtained according to the distance and angle between the target and the stereoscopic camera 110. Corresponding shooting parameters.
  • the depth of field of the PTZ camera 120 at the target may also be calculated prior to adjusting the PTZ camera 120 in step 430, and the magnitude relationship between the depth of field and the distance estimation error may be compared to determine how to adjust the PTZ camera 120.
  • FIG. 7 is a schematic diagram of a method for calculating depth of field according to an embodiment of the present invention.
  • ⁇ is the allowable circle diameter
  • f is the lens focal length
  • F is the lens aperture value
  • L is the focus lens
  • ⁇ L1 is the foreground depth
  • ⁇ L2 is the back depth of field
  • ⁇ L is the depth of field
  • the depth of field is calculated as:
  • the aperture F of the camera and the diameter ⁇ of the circle of dispersion are all certain values. It can be seen from the formula that after the object distance is small to a certain value, the depth of field will be infinite, and the depth of field decreases with the increase of the object distance. small.
  • the depth of field can cover the distance estimation error range of the target, even if there is a certain error in the distance calculation, but also within the depth of field It is possible to capture a clear image, and can directly adjust the PTZ camera 120 according to the aforementioned shooting parameters and capture the target; if the depth of field is smaller than the distance estimation error, then the depth of field cannot completely cover the range estimation error range of the target, and the target may not be captured clearly. Therefore, after adjusting the PTZ camera 120 according to the aforementioned shooting parameters, it is necessary to perform an auxiliary autofocus and then capture the target.
  • the distances and angles of the respective targets from the stereoscopic camera 110 are respectively calculated, and the stereoscopic camera 110 is used according to each target.
  • the distance and the angle respectively obtain the shooting parameters of the PTZ camera 120 corresponding to the respective targets, determine the snap order of the plurality of targets according to the priority, respectively adjust the PTZ camera 120 to capture the images of the plurality of targets according to the determined snap order, and determine the priority according to the determined capture order.
  • the parameters include the orientation, angle, distance, trajectory of the target, or whether it is about to leave the monitoring area.
  • the above list is non-exhaustive and may include other parameters. It should be clarified that the above parameters are not all necessary, and several parameters can be selected as the priority consideration factors according to the actual application scenario.
  • the embodiment of the present invention further provides a computer readable storage medium for storing computer software instructions for use in the above fast capture device and/or system, comprising program code designed to execute the above method embodiment.
  • program code designed to execute the above method embodiment.
  • the embodiment of the invention also provides a computer program product.
  • the computer program product includes computer software instructions that are loadable by a processor to implement the methods of the above method embodiments.
  • embodiments of the present application can be provided as a method, apparatus (system), or computer program product.
  • the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware aspects, which are collectively referred to herein as "module” or “system.”
  • the application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.
  • the computer program is stored/distributed in a suitable medium, provided with other hardware or as part of the hardware, or in other distributed forms, such as over the Internet or other wired or wireless telecommunication systems.
  • the computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device.
  • the apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.
  • These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device.
  • the instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.

Abstract

本发明实施例提供了一种快速抓拍的方法、设备及系统，该方法包括：检测立体视觉相机监测的当前画面中的目标，计算出目标与立体视觉相机的距离和角度，根据目标与立体视觉相机的距离和角度得到目标与PTZ相机的物距和角度，并根据目标与PTZ相机的物距和角度计算拍摄参数，再按照拍摄参数调整PTZ相机的镜头焦距和拍摄角度。本发明实施例利用立体视觉相机通过双目测距技术计算出目标的距离和角度，迅速得到PTZ相机抓拍目标的拍摄参数并作出调整，省去了自动对焦的步骤，缩短了对焦的时间，实现快速抓拍目标。

Description

快速抓拍的方法、装置及系统
本申请要求于2017年12月12日提交中国国家知识产权局、申请号为201711322098.2、发明名称为“快速抓拍的方法、装置及系统”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
本申请涉及视频监控领域，特别涉及一种实现快速抓拍的方法、装置及系统。
背景技术
随着现在视频监控技术的发展，监控由单纯的记录视频向智能化发展。在许多监控场景下对目标的检测和识别的需求日益增多，这就需要快速抓拍并提取目标。目前相机普遍采用反差对焦方式进行对焦操作，采用图像的反差纹理的强度判定当前对焦是否在最清晰的位置上。该对焦技术直接判定获取的图像信息，所以精度较高，但由于需要分析整个对焦过程，因而对焦速度较慢，无法满足快速对焦的需求。
现有的抓拍技术方案是采用固定焦距的泛焦镜头和广角镜头相结合，在广角镜头上发现目标之后，利用泛焦镜头光圈小、大景深的特点直接抓拍固定距离范围的目标。但由于泛焦镜头是采用固定光圈和焦距，只能采用定焦镜头的设计，并只能适用于有限范围内的物体拍摄，无法变焦，使用范围受到局限。
发明内容
本发明实施例提供了一种快速抓拍的方法、装置及系统，能够实现检测到目标之后快速定位、快速调焦以及快速抓拍到目标的图像。
为达到上述目的，本发明实施例提供如下技术方案：
第一方面，本发明实施例提供一种快速抓拍的方法，该方法包括：首先，检测第一相机监测的当前画面中的目标，计算目标与第一相机的距离与角度；然后，根据得到的距离与角度计算第二相机抓拍目标所需要的拍摄参数，拍摄参数包括镜头焦距和拍摄角度；最后，按照拍摄参数调整第二相机的角度和镜头的焦距之后抓拍目标。由于通过第一相机检测目标并计算目标的距离，进而得到第二相机的拍摄参数，能够减少第二相机的对焦时间，更及时、更清晰地抓拍到目标的图像。
其中，第一相机为立体视觉相机，第二相机为PTZ(Pan/Tilt/Zoom)相机。
PTZ相机在收到第一相机计算的距离与角度时，可以将该角度设置为PTZ相机的拍摄角度，将该距离设置为目标与PTZ相机的物距，根据该物距得到PTZ相机的镜头焦距，进而根据上述拍摄角度和镜头焦距调整PTZ相机抓拍目标。
可选地，可以通过查询物距与PTZ相机的镜头焦距的对照表以得到PTZ相机的镜头焦距。
在一种可能的设计中，为更精确地调整PTZ相机，可以在计算目标的距离时计算该距离对应的距离估算误差，在得到物距时，计算PTZ相机在该物距的景深，比较距离估算误差和景深的大小，若距离估算误差小于等于景深，按照拍摄参数调整PTZ相机之后抓拍目标；若距离估算误差大于景深，按照拍摄参数调整PTZ相机之后，自动调整镜头的焦距并抓拍目标。因为当景深大于等于距离估算误差，则此时景深能够覆盖目标的距离估算误差范围，即使距 离计算存在一定误差，但也在景深范围内，能够拍摄到清晰的图像，可以直接按照拍摄参数调整PTZ相机抓拍目标；但若景深小于距离估算误差，此时景深无法全部覆盖目标的距离估算误差范围，很可能无法拍摄到目标清晰的图像，因此，需要在按照拍摄参数调整PTZ相机之后再进行辅助自动对焦抓拍目标，这样才能根据实际情况调整PTZ相机，得到更清晰的目标图像。
在一些情况下，可能在立体视觉相机的当前画面中检测到多个目标，计算多个目标与立体视觉相机的距离和角度，根据距离与角度计算得到PTZ相机的镜头参数，并根据优先级确定多个目标的抓拍顺序，按照确定的抓拍顺序分别抓拍目标，根据优先级确定多个目标的抓拍顺序所依据的参数包括目标所在的方位、角度、距离、运行轨迹或是否即将离开监控区域等。
第二方面，本发明实施例提供一种快速抓拍系统，包括：第一相机，用于获取当前的监测画面，检测监测画面中的目标，计算目标与第一相机的距离和角度，并将该距离和角度发送给第二相机；第二相机，用于接收第一相机发送的距离和角度，根据该距离和角度计算第二相机的拍摄参数，按照拍摄参数调整第二相机抓拍目标，其中，拍摄参数包括镜头焦距和拍摄角度。
第三方面，本发明实施例还提供一种快速抓拍装置，包括：第一相机，用于同时从两个镜头获取当前的监测画面；处理器，用于检测第一相机当前画面中的目标并计算目标与第一相机之间的距离和角度，根据该距离和角度计算第二相机的拍摄参数，并控制第二相机抓拍目标，其中，拍摄参数包括拍摄角度和镜头焦距；第二相机，用于抓拍目标。
第四方面，本发明实施例还提供一种计算机可读存储介质，用于存储为上述快速抓拍装置和/或系统所用的计算机软件指令，其包含用于执行上述第一方面提供的方法所设计的程序代码。
本发明实施例通过上述方面，公开了一种快速抓拍方法、装置及系统，在该方法中，首先通过立体视觉技术快速测量目标的距离和角度，进而获得抓拍相机相应的拍摄参数，并根据该拍摄参数快速调整抓拍相机，完成对目标的抓拍。与现有技术相比，本发明实施例公开的方法能够更为精确、快速地确定抓拍相机的拍摄参数，减少了抓拍相机的对焦时间，具有更高的准确性，使抓拍更为及时，图像更为清晰。
本发明的这些方面或其他方面在以下实施例的描述中会更加简明易懂。
附图说明
图1为本发明实施例提供的一种快速抓拍系统的示意图。
图2为本发明实施例提供的一种快速抓拍装置的结构示意图。
图3为本发明实施例提供的网络摄像机的结构示意图。
图4为本发明实施例提供的一种快速抓拍的方法的示意图。
图5为本发明实施例提供的一种计算目标与立体视觉相机距离和角度的示意图。
图6为本发明实施例提供的一种目标坐标转换的示意图。
图7为本发明实施例提供的一种计算景深的方法的示意图。
具体实施方式
下面将结合附图，对本发明实施例中的技术方案进行描述。方法实施例中的具体操作方法也可以应用于装置实施例或系统实施例中。其中，在本发明的描述中，除非另有说明，“多 个”的含义是两个或两个以上。
如图1所示，为本发明实施例提供的一种快速抓拍的系统100，包括第一相机和第二相机，其中，第一相机为立体视觉相机110，第二相机为PTZ(Pan/Tilt/Zoom)相机120。
该系统中的立体视觉相机110装配有两个镜头，能够模拟人类视觉原理，从两个不同的角度同时获取当前场景的两幅图像，利用两幅图像拍摄角度的差异，根据图像之间像素的匹配关系，计算出图像中的目标与立体视觉相机之间的实际距离与角度，进而计算出目标在立体视觉相机110坐标系中的坐标。本发明实施例中的立体视觉相机110指能够用来同时从不同镜头获取当前场景图像的相机或相机组，这个名字本身对设备不构成限定，也可以为其他名字，如：双目立体视觉相机、双目相机、双目测距相机或立体视觉测距相机等。实际中，可以利用两个普通的摄像机来组成一组立体视觉相机，也可以将两个普通相机镜头集成在一个设备上组成立体视觉相机。
本发明实施例中的快速抓拍系统100中还可以包括多组立体视觉相机，以扩大监测范围。下面仅以一组立体视觉相机的情况来说明本发明技术方案，多组立体视觉相机的情形可以参照一组立体视觉相机下的技术方案。
本发明实施例中的快速抓拍系统100中的PTZ相机120装配有云台，能够实现全方位(左右/上下)移动及镜头变倍、变焦控制，用于根据目标与立体视觉相机之间的距离和角度，得到PTZ相机120抓拍该目标的拍摄参数，包括拍摄焦距与拍摄角度，并按照得到的参数调整PTZ相机120之后抓拍目标的图像。
上述快速抓拍系统100工作时，立体视觉相机110同时从两个镜头获取当前的监测画面，当前监测画面包含第一画面和第二画面，所述第一画面和所述第二画面分别为所述立体视觉相机110的第一镜头和第二镜头在同一时刻拍摄的画面。立体视觉相机110检测当前画面中的目标，并在检测到目标之后计算该目标与立体视觉相机110的距离和角度。若检测到多个目标，则分别计算各个目标与立体视觉相机110的距离和角度。接着，立体视觉相机110将目标与立体视觉相机110的距离和角度发送给PTZ相机120。PTZ相机120接收到立体视觉相机110发送的距离和角度之后，根据目标与立体视觉相机110的距离和角度，计算得到PTZ相机120抓拍目标的拍摄参数。得到拍摄参数之后，PTZ相机120根据拍摄参数调整PTZ相机120的角度及PTZ相机120镜头的焦距。完成调整之后，PTZ相机120抓拍目标的图像。
在需要更加精确地抓拍目标的图像的情况下，立体视觉相机110还可以计算目标的对应的距离估算误差并发送给PTZ相机120，PTZ相机120还可以计算目标与PTZ相机120的物距对应的景深。PTZ相机120在调整前比较目标的距离估算误差和物距处的景深，若景深大于等于距离估算误差，直接按照拍摄参数调整PTZ相机120并抓拍目标；若景深小于距离估算误差，则按照拍摄参数调整PTZ相机120之后继续控制PTZ相机120自动对焦。
在某些情况下，有可能在立体视觉相机110的当前监测画面中检测到多个目标，立体视觉相机110分别计算各个目标与立体视觉相机110的距离和角度，PTZ相机120根据各个目标与立体视觉相机110的距离和角度分别获得各个目标对应的PTZ相机120的拍摄参数，并根据优先级确定多个目标的抓拍顺序，按照确定的抓拍顺序分别调整PTZ相机120抓拍多个目标的图像，确定优先级所依据的参数包括目标所在的方位、角度、距离、运行轨迹或是否即将离开监控区域等。
上述实施例中的快速抓拍系统是将立体视觉相机110和PTZ相机120分离设置，两者独立完成相应的运算和控制功能。立体视觉相机110和PTZ相机120之间可以通过有线或无线形式连接，实现两设备之间的信息交互。需要说明的是，立体视觉相机110中的计算操作， 如检测目标和计算目标与立体视觉相机110之间的距离和角度等，也可以由PTZ相机120执行，本领域技术人员可以对本实施例中的快速抓拍系统做出上述改动，这样的改动应当被视为属于本发明的权利要求及其等同技术的范围之内。
图2示出了本发明实施例提供的快速抓拍装置的一种可能的结构示意图，该装置包括：第一相机210、处理器220、存储器230、第二相机240和至少一个通信总线250，其中第一相机210可以为立体视觉相机210，第二相机240可以为PTZ相机240。
处理器220可以是一个通用中央处理器CPU，微处理器，ASIC，或一个或多个用于控制本发明方案程序执行的集成电路。处理器220也可以采用FPGA或DSP来实现。
存储器230，可以是易失性存储器(volatile memory)，例如随机存取存储器(random-access memory，RAM)；或者非易失性存储器(non-volatile memory)，硬盘(hard disk drive，HDD)或固态硬盘(solid-state drive，SDD)；或者上述种类的存储器的组合，并向处理器提供指令和数据。
总线250可以是外设部件互连标准(Peripheral Component Interconnect，PCI)总线或扩展工业标准结构(Extended Industry Standard Architecture，EISA)总线等。所述总线850可以分为地址总线、数据总线、控制总线等。为便于表示，图2中仅用一条粗线表示，但并不表示仅有一根总线或一种类型的总线。
立体视觉相机210包括两个镜头，用于同时从两个镜头获取当前的监测画面，当前监测画面包含第一画面和第二画面，第一画面和第二画面分别为立体视觉相机210的两个镜头在同一时刻拍摄的画面。
处理器220检测立体视觉相机210当前监测画面中的目标，并计算目标与立体视觉相机210的距离和角度，根据目标与立体视觉相机210的距离和角度计算PTZ相机240对应的拍摄参数，拍摄参数包括拍摄角度和镜头焦距。处理器220在得到PTZ相机240抓拍目标所需要的拍摄参数之后，根据拍摄参数生成相应的控制信号，控制PTZ相机240调整拍摄角度和镜头焦距。
PTZ相机240可以上下左右转动，能够根据拍摄参数中的拍摄角度旋转到相应角度，PTZ相机240的焦距是可变的，处理器220可以控制PTZ相机240调整镜头焦距并抓拍目标的图像。PTZ相机240还能自动对焦，可以根据处理器220发送的自动对焦的命令完成自动对焦过程。
在一种可能的实现方式中，处理器220还用于根据目标与立体视觉相机210的距离计算对应的距离估算误差，根据目标与PTZ相机240的物距以及镜头焦距计算PTZ相机240在该物距对应的景深，处理器220比较距离估算误差与景深的大小关系，若距离估算误差小于等于景深，处理器220按照拍摄参数控制PTZ相机240调整并抓拍目标的图像；若距离估算误差大于景深，处理器220按照拍摄参数控制抓拍相机240调整之后，再控制PTZ相机240自动调整镜头焦距。
在某些可能的情况下，若处理器220检测到多个目标，根据优先级确定多个目标的抓拍顺序，按照确定的抓拍顺序控制PTZ相机240抓拍多个目标的图像，其中，确定优先级所依据的参数包括目标所在的方位、角度、距离、运行轨迹或是否即将离开监控区域等。
可选地，快速抓拍装置200还可以包括通信接口260，通信接口260用于将抓拍的目标的图像传输给外部装置。
上述实施例中的快速抓拍装置200是将立体视觉相机210、处理器220和PTZ相机240三者作为一个整体的装置，作为快速抓拍装置200的三部件可以分离设置而通过有线或无线 形式连接以实现部件之间的通信，也可以将三部件合并设置，还可以使用一个设备来实现处理器220与存储器230的功能。快速抓拍装置200中的立体视觉相机210和PTZ相机240主要用于获取目标图像，处理器220则用于执行计算和控制操作。需要说明的是，根据实际需要，处理器220的部分计算或控制功能可以由立体视觉相机210或PTZ相机240实现，本领域技术人员可以对本实施例中的快速抓拍系统做出上述改动，这样的改动应当被视为属于本发明的权利要求及其等同技术的范围之内。
如图3所示，为上述快速抓拍系统100及快速抓拍装置200中的第一相机和第二相机通用的网络摄像机300的结构。网络摄像机300包括了上述实施例中的第一相机和第一相机共有的结构，为了便于理解，不再描述网络摄像机300与本发明无关的标准特征。网络摄像机300包括作为网络摄像机300前端部件的镜头310，镜头310有固定光圈、自动光圈、自动变焦、自动变倍等种类；用于记录入射光的图像传感器320，例如互补金属氧化物半导体(Complementary Metal Oxide Semiconductor，CMOS)、电荷耦合器件(Charge-coupled Device，CCD)或相似器件；图像处理器330；用于执行计算操作和控制摄像机的处理器340；用于存储程序或数据的存储器350；用于各部件传递信息的通信总线360以及用于通过通信网络向连接至该网络的其他节点传递信息的通信接口370。
图像传感器320接收与所记录的光有关的信息，并借助A/D转换器和信号处理器331对该信息进行处理，其中A/D转换器和信号处理器331是技术人员公知的。在一些实施例中，例如当图像传感器320是CMOS传感器时，图像传感器320包括A/D转换器，因此图像处理器330中不需要任何A/D转换器。由A/D转换器和信号处理器331产生的结果是数字图像数据，根据一个实施例，该数字图像数据在被发送到处理器340之前，在缩放单元332和图像编码器333中被处理。缩放单元332用于将该数字图像数据处理为至少一个特定大小的图像。然而，缩放单元332可以被布置为生成多个不同大小的图像，所有图像均表示由A/D转换器和信号处理器331提供的同一图像/帧。根据另一实施例，缩放单元332的功能由图像编码器333来执行，在再一实施例中，不需要对来自图像传感器320的图像执行任何缩放或大小调整。
编码器333对于执行本发明来说是可选的，被布置为针对连续的视频序列、针对有限的视频序列、针对静态图像或针对图像/视频流将该数字图像数据编码为多种已知格式中的任意一种。举例来说，可以将图像信息编码为MPEG1、MPEG2、MPEG4、JEPG、MJPG、位映像等。处理器340可以使用未经编码的图像作为输入数据。在这种情况下，图像数据从信号处理器331或从缩放单元332传输到处理器340，而不通过图像编码器333传递该图像数据。未经编码的图像可以采用任何未经编码的图像格式，例如BMP、PNG、PPM、PGM、PNM和PBM，尽管处理器340也可以使用经过编码的数据作为输入数据。
在本发明的一个实施例中，可以将图像数据从信号处理器331直接发送至处理器340，而不通过缩放单元332或图像编码器333。在再一实施例中，可以将图像数据从缩放单元332发送至处理器340，而不通过图像编码器333。
处理器340可以是一个通用中央处理器(central processing unit，CPU)，微处理器，特定应用集成电路(application-specific integrated circuit，ASIC)，或一个或多个用于控制本发明方案程序执行的集成电路。处理器340也可以采用现场可编程门阵列(Field Programmable Gate Array，FPGA)或DSP来实现。在采用基于DSP的软件编码压缩时，也可以将图像处理器330中的部分功能集成在处理器340上。处理器340用于管理和控制网络摄像机300。
存储器350用于存储执行本发明方案的应用程序代码，可以是只读存储器(read-only  memory，ROM)或可存储静态信息和指令的其他类型的静态存储设备，随机存取存储器(random access memory，RAM)或者可存储信息和指令的其他类型的动态存储设备，也可以是电可擦可编程只读存储器(electrically erasable programmable read-only memory，EEPROM)、只读光盘(compact disc read-only memory，CD-ROM)或其他光盘存储、光碟存储(包括压缩光碟、激光碟、光碟、数字通用光碟、蓝光光碟等)、磁盘存储介质或者其他磁存储设备、或者能够用于携带或存储具有指令或数据结构形式的期望的程序代码并能够由计算机存取的任何其他介质，但不限于此。存储器350可以是独立存在，通过总线360与处理器340相连接。存储器350也可以和处理器340集成在一起。
通信总线360可包括一通路，在各组件之间传送信息。
通信接口370使用任何收发器一类的装置，用于与其他设备或通信网络通信，如以太网，无线接入网(radio access network，RAN)，无线局域网(wireless local area networks，WLAN)等。
上述实施例中的立体视觉相机在通用摄像机300的基础上设置了多个镜头，以实现从多个镜头获得当前检测的画面；上述实施例中的PTZ相机在通用摄像机300的基础上设置了云台，以实现全方位(左右/上下)移动，调整PTZ相机的拍摄角度。
以下将结合附图，对本发明实施例提供的快速抓拍系统和方法做进一步说明。
如图4所示，为本发明实施例提供的一种快速抓拍的方法，该方法在抓拍目标时，采用立体视觉相机协助抓拍相机调焦，解决了现有抓拍技术中对焦时间过长的问题。本发明实施例提供的快速抓拍方法可应用于图1中的快速抓拍系统100以及图2中的快速抓拍装置200上，用于对大场景中的快速移动的目标的抓拍，下面结合图1来阐述本发明实施例提供的方法的具体实现。本发明实施例提供的快速抓拍方法包括以下步骤：
410，检测第一相机当前画面中的目标，计算目标与第一相机的距离和角度。
其中，第一相机可以为立体视觉相机110，其监测的当前画面包含第一画面和第二画面，所述第一画面和所述第二画面分别为所述立体视觉相机110的第一镜头和第二镜头在同一时刻拍摄的画面。可以检测出画面中的目标，并根据目标在不同的画面中的视觉差，计算出目标与立体视觉相机210的距离和角度。
420，根据目标与第一相机的距离和角度计算第二相机的拍摄参数。
在快速抓拍系统中，第二相机为PTZ相机110。在得到了目标与立体视觉相机110的距离和角度之后，可以根据立体视觉相机110与PTZ相机120的位置关系得到目标与PTZ相机120的物距和角度，进而获得PTZ相机120抓拍目标的拍摄参数，其中，拍摄参数包括拍摄角度和镜头焦距。
430，按照拍摄参数调整第二相机的拍摄角度和镜头焦距再抓拍目标。
拍摄角度和镜头焦距是PTZ相机120抓拍目标最重要的两个参数，根据上述步骤得到PTZ相机120的拍摄角度和镜头焦距，可以控制PTZ相机120的云台转动到对应的角度，并控制PTZ相机120的镜头变焦，将镜头焦距调整到对应的值，再抓拍目标。
通过以上步骤，实现了PTZ相机120不需要自动对焦而完成对目标的抓拍，避免了长时间的自动对焦过程，提高了抓拍效率，使抓拍更为及时，图像更为清晰。
下面结合附图，进一步说明以上步骤的具体实现方法。
如图5所示，为本发明实施例提供了一种计算目标与立体视觉相机110的距离的方法。其中Ol、Or分别为立体视觉相机110的左右相机的靶面中心位置，目标点P分别在左右相机的靶面成像到Pl、Pr点，光路从目标点经镜头平面的中心到靶面成像点，按照相似三角形原 理，可以得到：
(B+(xl-xr))/D＝B/(Z-f)
简化后得到：
D＝(f×(B+xl-xr))/(xl-xr)
其中，B为基线距离，即左右相机之间的距离；f为立体视觉相机焦距；xl为目标在左侧成像点与左侧相机中心点的水平距离；xr为目标在右侧相机成像点与右侧相机中心点的水平距离；D为目标与立体视觉相机的距离。
目标与左侧相机的角度关系、目标与右侧相机的角度关系以及目标与立体视觉相机中心线的角度，则在目标距离计算的基础上按照三角函数计算公式即可算出。
但立体测距技术测得的距离并不是绝对准确的，测得的距离具有一定的误差，该误差为距离估算误差，误差范围与立体视觉相机的镜头参数、感光器的像素大小以及两个镜头的间距相关。距离估算误差随着目标距离的增加而增加，与距离的平方成正比关系。Thomas·Luhmann在《Close-Range Photogrammetry and 3D Imaging》(2014)一书中给出了一种距离估算误差的计算方法。此外，还可以通过多次测量，得到距离估算误差与距离的平方之间的系数，进而得到测得的目标的距离与距离估算误差的关系。
采用立体测距技术测量目标距离的计算方法有多种，上述计算方法仅是本发明一个实施例采用的一种计算方法，并不对本发明的保护范围起到限定作用，无论采用何种具体的距离计算方法，只要能实现通过立体测距技术测量目标的距离和角度的功能，则均在本发明保护范围之内。
在获得了目标与立体视觉相机110的距离和角度，可以通过多种方式得到步骤420中的目标与PTZ相机120的物距以及PTZ相机120的拍摄参数。
在一种实现方式中，可以直接将立体视觉相机110与目标的距离设置为目标与PTZ相机120的物距，将立体视觉相机110与目标的角度设置为PTZ相机120的拍摄角度，根据目标与PTZ相机120的物距得到PTZ相机120的拍摄焦距。由于本发明实施例中的快速抓拍系统100通常用来拍摄较远距离的目标，目标与快速抓拍系统100之间的距离远大于立体视觉相机110镜头与PTZ相机120镜头之间的距离，在对拍摄精度要求不高的情况下，可以认为目标与立体视觉相机110的距离和角度跟目标与PTZ相机120的物距和角度相等。此外，立体视觉相机110一般包含两个镜头，在建立坐标系时通常以两个镜头的中心位置为原点，如果PTZ相机120位于立体视觉相机110两个镜头的中心位置，则立体视觉相机110坐标系与PTZ相机120坐标系重合，此时目标与立体视觉相机110的距离和角度跟目标与PTZ相机120的物距和角度相等。
在另一种实现方式中，需要考虑立体视觉相机110与PTZ相机120之间的空间位置差，根据该空间位置差以及目标与立体视觉相机110的距离和角度计算目标与PTZ相机120的物距和角度，并将该角度设置为PTZ相机120的拍摄角度，再根据目标与PTZ相机120的物距得到PTZ相机120的拍摄焦距。
对于立体视觉相机110与PTZ相机120存在空间位置差的，可以通过三维坐标转换得到目标在PTZ相机120中的坐标。
如图6所示，为本发明实施例提供的一种实现立体视觉相机与PTZ相机之间坐标转换的方式，其中坐标O为PTZ相机坐标系的坐标原点，坐标O′为立体视觉相机坐标系的坐标原点，这两个坐标系中各点的关系如下式所示：
Figure PCTCN2018119698-appb-000001
其中，λ为两个坐标系间的尺度比例因素，ΔX、ΔY和ΔZ为PTZ相机坐标系坐标原点与立体视觉相机坐标系坐标原点之间的位置差，R为坐标转换时的旋转矩阵，用于实现将立体视觉相机的坐标系各坐标轴旋转到PTZ相机的坐标系对应的坐标轴。
R＝R(ε Y)R(ε YX)R(ε YZ)
展开为：
Figure PCTCN2018119698-appb-000002
由上式可以得到目标在PTZ相机120坐标系中的坐标，进而得到目标的物距以及与PTZ相机120的角度。
在得到目标与PTZ相机120的物距之后，可以有多种方式得到PTZ相机120的镜头焦距。具体地，作为一种实现方式，可以通过预先存储物距与镜头焦距的对照表，计算得到目标的物距之后，查询对照表以得到PTZ相机120的拍摄焦距。
更进一步，可以预先存储目标与立体视觉相机110之间的距离和角度跟PTZ相机120的拍摄参数的对照表，根据目标与立体视觉相机110之间的距离和角度查询上述预先存储的对照表获得对应的拍摄参数。
可选地，在步骤430中调整PTZ相机120之前还可以计算PTZ相机120在目标处的景深，并比较景深与距离估算误差的大小关系以决定如何调整PTZ相机120。
图7为本发明实施例提出的一种计算景深的方法示意图。
其中：δ为容许弥散圆直径，f为镜头焦距，F为镜头的拍摄光圈值，L为对焦镜头，ΔL1为前景深，ΔL2为后景深，ΔL为景深，景深的计算公式为：
Figure PCTCN2018119698-appb-000003
在一定的应用场景下，相机的光圈F及弥散圆直径δ都为确定值，从公式可以看出，在物距小到一定值后，景深将会无穷大，并且景深随着物距增大而减小。
比较PTZ相机120在目标处的景深与目标的距离估算误差，若景深大于等于距离估算误差，则此时景深能够覆盖目标的距离估算误差范围，即使距离计算存在一定误差，但也在景深范围内，能够拍摄到清晰的图像，可以直接按照前述拍摄参数调整PTZ相机120并抓拍目标；若景深小于距离估算误差，则此时景深无法全部覆盖目标的距离估算误差范围，很可能无法拍摄到目标清晰的图像，因此，按照前述拍摄参数调整PTZ相机120之后还需要进行辅助自动对焦再抓拍目标。
在具体实现中，作为一种实施例，若在立体视觉相机110的监测画面中检测到多个目标，分别计算各个目标与立体视觉相机110的距离和角度，并根据各个目标与立体视觉相机110的距离和角度分别获得各个目标对应的PTZ相机120的拍摄参数，根据优先级确定多个目标的抓拍顺序，按照确定的抓拍顺序分别调整PTZ相机120抓拍多个目标的图像，确定优先级所依据的参数包括目标所在的方位、角度、距离、运行轨迹或是否即将离开监控区域等。上述列举为非穷尽式，还可以包括其他参数。需要明确的是，上述参数并非都是必须的，可以根据实际应用场景选取若干个参数作为优先级的考量因素。
上述方法实施例是结合图1中的快速抓拍系统100来阐述，在图2中的快速抓拍装置200 中实现上述方法实施例的具体步骤与上述记载类似，本领域技术人员应当清楚地理解，在此不再赘述。
本发明实施例还提供一种计算机可读存储介质，用于存储为上述快速抓拍装置和/或系统所用的计算机软件指令，其包含用于执行上述方法实施例所设计的程序代码。通过执行存储的程序代码，可以实现通过立体视觉技术快速测量目标的距离和角度，进而获得抓拍图像的拍摄参数，快速调整PTZ相机，完成对目标的抓拍，能够缩短抓拍时间，保证抓拍的实时性。
本发明实施例还提供了计算机程序产品。该计算机程序产品包括计算机软件指令，该计算机软件指令可通过处理器进行加载来实现上述方法实施例中的方法。
尽管在此结合各实施例对本发明进行了描述，然而，在实施所要求保护的本发明过程中，本领域技术人员通过查看所述附图、公开内容、以及所附权利要求书，可理解并实现所述公开实施例的其他变化。在权利要求中，“包括”(comprising)一词不排除其他组成部分或步骤，“一”或“一个”不排除多个的情况。单个处理器或其他单元可以实现权利要求中列举的若干项功能。相互不同的从属权利要求中记载了某些措施，但这并不表示这些措施不能组合起来产生良好的效果。
本领域技术人员应明白，本申请的实施例可提供为方法、装置(系统)、或计算机程序产品。因此，本申请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，这里将它们都统称为“模块”或“系统”。而且，本申请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存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。计算机程序存储/分布在合适的介质中，与其它硬件一起提供或作为硬件的一部分，也可以采用其他分布形式，如通过Internet或其它有线或无线电信系统。
本申请是参照本发明实施例的方法、装置(设备)和计算机程序产品的流程图和/或方框图来描述的。应理解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。
尽管结合具体特征及其实施例对本发明进行了描述，显而易见的，在不脱离本发明的精神和范围的情况下，可对其进行各种修改和组合。相应地，本说明书和附图仅仅是所附权利要求所界定的本发明的示例性说明，且视为已覆盖本发明范围内的任意和所有修改、变化、组合或等同物。显然，本领域的技术人员可以对本发明进行各种改动和变型而不脱离本发明的精神和范围。这样，倘若本发明的这些修改和变型属于本发明权利要求及其等同技术的范围之内，则本发明也意图包含这些改动和变型在内。

Claims (14)

  1. 一种快速抓拍的方法，其特征在于，所述方法包括：
    检测第一相机当前画面中的目标，计算所述目标与所述第一相机之间的距离与角度；
    根据所述距离与角度计算第二相机的拍摄参数，所述拍摄参数包括镜头焦距和拍摄角度；
    按照所述拍摄参数调整所述第二相机抓拍所述目标。
  2. 根据权利要求1所述的方法，其特征在于，所述第一相机为立体视觉相机，所述第二相机为PTZ(Pan/Tilt/Zoom)相机。
  3. 根据权利要求2所述的方法，其特征在于，所述根据所述距离与角度计算所述PTZ相机的拍摄参数包括：
    将所述距离设置为所述目标与所述PTZ相机的物距；
    将所述角度设置为所述PTZ相机的拍摄角度；
    根据所述物距得到所述PTZ相机的镜头焦距。
  4. 根据权利要求3所述的方法，其特征在于，根据所述物距得到所述PTZ相机的镜头焦距，包括：
    获取物距与PTZ相机的镜头焦距的对照表；
    根据所述物距查询所述对照表得到所述物距对应的所述PTZ相机的镜头焦距。
  5. 根据权利要求2至4所述的方法，其特征在于：
    根据所述距离与角度计算所述PTZ相机的拍摄参数，还包括：根据所述距离计算所述距离对应的距离估算误差，以及根据所述物距计算所述PTZ相机在所述物距对应的景深；
    所述调整所述PTZ相机之前，还包括比较所述距离估算误差与所述景深的大小；
    若所述距离估算误差小于等于所述景深，按照所述拍摄参数调整所述PTZ相机之后抓拍所述目标；
    若所述距离估算误差大于所述景深，按照所述镜头参数调整所述PTZ相机之后，自动调整镜头焦距并抓拍所述目标。
  6. 根据权利要求1至5所述的方法，其特征在于，若在所述立体视觉相机的当前画面中检测到多个目标，根据所述距离与角度计算所述PTZ相机的镜头参数之后还包括，根据优先级确定所述多个目标的抓拍顺序，按照确定的抓拍顺序分别抓拍所述目标；
    所述根据优先级确定所述多个目标的抓拍顺序所依据的参数包括目标所在的方位、角度、距离、运行轨迹或是否即将离开监控区域。
  7. 一种快速抓拍系统，其特征在于，包括：
    第一相机，用于获取当前的监测画面，检测监测画面中的目标，计算所述目标与所述第一相机的距离和角度，并将所述距离和角度发送给第二相机；
    第二相机，用于接收所述第一相机发送的所述距离和角度，根据所述距离和角度计算所述第二相机的拍摄参数，按照所述拍摄参数调整所述第二相机抓拍所述目标，其中，所述拍摄参数包括镜头焦距和拍摄角度。
  8. 根据权利要求7所述的系统，其特征在于，所述第一相机为立体视觉相机，所述第二相机为PTZ(Pan/Tilt/Zoom)相机。
  9. 根据权利要求8所述的快速抓拍系统，其特征在于：
    所述立体视觉相机还用于根据所述距离计算所述距离对应的距离估算误差，并将所述距离估算误差发送给所述PTZ相机；
    所述PTZ相机还用于接收所述立体视觉相机发送的所述距离估算误差，并根据所述目标与所述PTZ相机的物距计算所述PTZ相机在所述物距对应的景深；
    所述PTZ相机比较所述距离估算误差与所述景深的大小：
    若所述距离估算误差小于等于所述景深，所述PTZ相机按照所述拍摄参数调整拍摄角度与镜头焦距并抓拍所述目标；
    若所述距离估算误差大于所述景深，所述PTZ相机按照所述镜头参数调整拍摄角度与镜头焦距之后，再自动调整镜头焦距并抓拍所述目标。
  10. 根据权利要求8或9所述的快速抓拍系统，其特征在于：
    若所述立体视觉相机检测到多个目标，计算各个目标与所述立体视觉相机的距离和角度；
    所述PTZ相机还用于计算所述多个目标对应的拍摄参数，并根据优先级确定所述多个目标的抓拍顺序，按照确定的抓拍顺序调整所述第二相机抓拍所述多个目标；
    所述根据优先级确定所述多个目标的抓拍顺序所依据的参数包括目标所在的方位、角度、距离、运行轨迹或是否即将离开监控区域。
  11. 一种快速抓拍装置，其特征在于，包括第一相机、处理器和第二相机，其中：
    第一相机，包括两个镜头，用于同时从所述两个镜头获取当前的监测画面；
    处理器，用于检测所述第一相机当前画面中的目标，计算所述目标与所述第一相机之间的距离和角度，根据所述距离和角度计算第二相机的拍摄参数，并控制所述第二相机抓拍所述目标，其中，所述拍摄参数包括拍摄角度和镜头焦距；
    第二相机，用于抓拍所述目标。
  12. 根据权利要求11所述的快速抓拍装置，其特征在于，所述第一相机为立体视觉相机，所述第二相机为PTZ(Pan/Tilt/Zoom)相机。
  13. 根据权利要求12所述的快速抓拍装置，其特征在于：
    所述处理器还用于根据所述距离计算对应的距离估算误差，根据所述目标与所述PTZ相机的物距计算所述PTZ相机在所述物距出对应的景深；
    所述处理器比较所述距离估算误差与所述景深的大小关系；
    若所述距离估算误差小于等于所述景深，所述处理器按照所述拍摄参数调整所述PTZ相机并抓拍所述目标；
    若所述距离估算误差大于所述景深，所述处理器按照所述拍摄参数调整所述PTZ相机之后，还控制所述PTZ相机自动调整镜头的焦距并抓拍所述目标。
  14. 根据权利要求12或13所述的快速抓拍装置，其特征在于：
    若所述处理器检测到多个目标，则计算各个目标与所述立体视觉相机的距离和角度，以及所述多个目标对应的拍摄参数；
    所述处理器根据优先级确定所述多个目标的抓拍顺序，根据确定的抓拍顺序控制所述PTZ相机按照所述多个目标的拍摄参数调整，并抓拍所述多个目标的图像；
    所述根据优先级确定所述多个目标的抓拍顺序所依据的参数包括目标所在的方位、角度、距离、运行轨迹或是否即将离开监控区域。
PCT/CN2018/119698 2017-12-12 2018-12-07 快速抓拍的方法、装置及系统 WO2019114617A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711322098.2A CN109922251B (zh) 2017-12-12 2017-12-12 快速抓拍的方法、装置及系统
CN201711322098.2 2017-12-12

Publications (1)

Publication Number Publication Date
WO2019114617A1 true WO2019114617A1 (zh) 2019-06-20

Family

ID=66818967

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/119698 WO2019114617A1 (zh) 2017-12-12 2018-12-07 快速抓拍的方法、装置及系统

Country Status (2)

Country Link
CN (1) CN109922251B (zh)
WO (1) WO2019114617A1 (zh)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111144478A (zh) * 2019-12-25 2020-05-12 电子科技大学 一种穿帮镜头的自动检测方法
CN111652802A (zh) * 2020-05-19 2020-09-11 杭州海康威视数字技术股份有限公司 全景图制作方法、基于全景图的交互方法及装置
CN111652940A (zh) * 2020-04-30 2020-09-11 平安国际智慧城市科技股份有限公司 目标异常识别方法、装置、电子设备及存储介质
CN111986248A (zh) * 2020-08-18 2020-11-24 东软睿驰汽车技术（沈阳）有限公司 多目视觉感知方法、装置及自动驾驶汽车
CN112581547A (zh) * 2020-12-30 2021-03-30 安徽地势坤光电科技有限公司 一种调整成像镜头安装角度的快速方法
CN113038070A (zh) * 2019-12-25 2021-06-25 浙江宇视科技有限公司 一种设备调焦方法、装置和云平台
CN113452903A (zh) * 2021-06-17 2021-09-28 浙江大华技术股份有限公司 一种抓拍设备、抓拍方法及主控芯片
CN113888651A (zh) * 2021-10-21 2022-01-04 天津市计量监督检测科学研究院电子仪表实验所 一种动静态视觉检测系统

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110278382B (zh) * 2019-07-22 2020-12-08 浙江大华技术股份有限公司 一种聚焦方法、装置、电子设备及存储介质
CN113099163B (zh) * 2019-12-23 2023-04-11 中移物联网有限公司 一种监控调节方法、监控系统、电子设备及可读存储介质
CN111147840A (zh) * 2019-12-23 2020-05-12 南京工业职业技术学院 一种3d摄像摇臂的视音频采集自动控制与通信系统
CN111158107B (zh) * 2020-01-03 2021-07-06 支付宝(杭州)信息技术有限公司 一种镜头模组的对焦方法、装置及设备
CN111314602B (zh) * 2020-02-17 2021-09-17 浙江大华技术股份有限公司 目标对象聚焦方法、装置、存储介质及电子装置
WO2021184341A1 (en) * 2020-03-20 2021-09-23 SZ DJI Technology Co., Ltd. Autofocus method and camera system thereof
CN111595292A (zh) * 2020-04-29 2020-08-28 杭州电子科技大学 一种基于不等焦距的双目视觉测距方法
CN111405193B (zh) * 2020-04-30 2021-02-09 重庆紫光华山智安科技有限公司 聚焦方法、装置和摄像设备
CN112084925A (zh) * 2020-09-03 2020-12-15 厦门利德集团有限公司 一种智慧电力安全监察方法及系统
CN113422901B (zh) * 2021-05-29 2023-03-03 华为技术有限公司 一种摄像机对焦方法及相关设备
CN113569813A (zh) * 2021-09-05 2021-10-29 中国电波传播研究所（中国电子科技集团公司第二十二研究所） 一种基于服务器端的智能化图像识别系统及方法

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011031128A1 (en) * 2009-09-08 2011-03-17 Mimos Berhad Control mechanism for automated surveillance system
CN104363376A (zh) * 2014-11-28 2015-02-18 广东欧珀移动通信有限公司 一种连续对焦的方法、装置及终端
CN105407283A (zh) * 2015-11-20 2016-03-16 成都因纳伟盛科技股份有限公司 一种多目标主动识别跟踪监控方法
CN107113403A (zh) * 2015-12-09 2017-08-29 空间情报技术 利用多个立体摄像头的拍摄对象空间移动跟踪系统

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7929801B2 (en) * 2005-08-15 2011-04-19 Sony Corporation Depth information for auto focus using two pictures and two-dimensional Gaussian scale space theory
US8310554B2 (en) * 2005-09-20 2012-11-13 Sri International Method and apparatus for performing coordinated multi-PTZ camera tracking
CN100531373C (zh) * 2007-06-05 2009-08-19 西安理工大学 基于双摄像头联动结构的视频运动目标特写跟踪监视方法
CN101924923B (zh) * 2010-08-03 2013-01-09 杭州翰平电子技术有限公司 嵌入式智能自动变焦抓拍方法
US8736747B2 (en) * 2012-01-13 2014-05-27 Sony Corporation Camera autofocus adaptive blur matching model fitting
CN102867304B (zh) * 2012-09-04 2015-07-01 南京航空航天大学 双目立体视觉系统中场景立体深度与视差的关系建立方法
WO2017199556A1 (ja) * 2016-05-17 2017-11-23 富士フイルム株式会社 ステレオカメラ及びステレオカメラの制御方法
CN106251334B (zh) * 2016-07-18 2019-03-01 华为技术有限公司 一种摄像机参数调整方法、导播摄像机及系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011031128A1 (en) * 2009-09-08 2011-03-17 Mimos Berhad Control mechanism for automated surveillance system
CN104363376A (zh) * 2014-11-28 2015-02-18 广东欧珀移动通信有限公司 一种连续对焦的方法、装置及终端
CN105407283A (zh) * 2015-11-20 2016-03-16 成都因纳伟盛科技股份有限公司 一种多目标主动识别跟踪监控方法
CN107113403A (zh) * 2015-12-09 2017-08-29 空间情报技术 利用多个立体摄像头的拍摄对象空间移动跟踪系统

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111144478B (zh) * 2019-12-25 2022-06-14 电子科技大学 一种穿帮镜头的自动检测方法
CN111144478A (zh) * 2019-12-25 2020-05-12 电子科技大学 一种穿帮镜头的自动检测方法
CN113038070A (zh) * 2019-12-25 2021-06-25 浙江宇视科技有限公司 一种设备调焦方法、装置和云平台
CN113038070B (zh) * 2019-12-25 2022-10-14 浙江宇视科技有限公司 一种设备调焦方法、装置和云平台
CN111652940A (zh) * 2020-04-30 2020-09-11 平安国际智慧城市科技股份有限公司 目标异常识别方法、装置、电子设备及存储介质
CN111652802A (zh) * 2020-05-19 2020-09-11 杭州海康威视数字技术股份有限公司 全景图制作方法、基于全景图的交互方法及装置
CN111652802B (zh) * 2020-05-19 2024-03-05 杭州海康威视数字技术股份有限公司 全景图制作方法、基于全景图的交互方法及装置
CN111986248A (zh) * 2020-08-18 2020-11-24 东软睿驰汽车技术（沈阳）有限公司 多目视觉感知方法、装置及自动驾驶汽车
CN111986248B (zh) * 2020-08-18 2024-02-09 东软睿驰汽车技术（沈阳）有限公司 多目视觉感知方法、装置及自动驾驶汽车
CN112581547B (zh) * 2020-12-30 2022-11-08 安徽地势坤光电科技有限公司 一种调整成像镜头安装角度的快速方法
CN112581547A (zh) * 2020-12-30 2021-03-30 安徽地势坤光电科技有限公司 一种调整成像镜头安装角度的快速方法
CN113452903A (zh) * 2021-06-17 2021-09-28 浙江大华技术股份有限公司 一种抓拍设备、抓拍方法及主控芯片
CN113452903B (zh) * 2021-06-17 2023-07-11 浙江大华技术股份有限公司 一种抓拍设备、抓拍方法及主控芯片
CN113888651A (zh) * 2021-10-21 2022-01-04 天津市计量监督检测科学研究院电子仪表实验所 一种动静态视觉检测系统

Also Published As

Publication number Publication date
CN109922251A (zh) 2019-06-21
CN109922251B (zh) 2021-10-22

Similar Documents

Publication Publication Date Title
WO2019114617A1 (zh) 快速抓拍的方法、装置及系统
WO2019105214A1 (zh) 图像虚化方法、装置、移动终端和存储介质
WO2018201809A1 (zh) 基于双摄像头的图像处理装置及方法
KR102143456B1 (ko) 심도 정보 취득 방법 및 장치, 그리고 이미지 수집 디바이스
US8805021B2 (en) Method and apparatus for estimating face position in 3 dimensions
JP6887960B2 (ja) 自動焦点トリガのためのシステムおよび方法
WO2016049889A1 (zh) 一种自动对焦方法、装置及电子设备
WO2014153950A1 (zh) 快速自动聚焦的方法和图像采集装置
WO2015184978A1 (zh) 摄像机控制方法、装置及摄像机
WO2021000063A1 (en) Automatic focus distance extension
JP2010128820A (ja) 立体画像処理装置、方法及びプログラム並びに立体撮像装置
KR20180040336A (ko) 카메라 시스템 및 이의 객체 인식 방법
WO2020237565A1 (zh) 一种目标追踪方法、装置、可移动平台及存储介质
WO2020042581A1 (zh) 一种图像获取设备的对焦方法及装置
WO2016184131A1 (zh) 基于双摄像头拍摄图像的方法、装置及计算机存储介质
CN102629978B (zh) 一种通过自动聚焦来检测靶面平整度的高清网络半球
US10277888B2 (en) Depth triggered event feature
WO2021081909A1 (zh) 拍摄设备的对焦方法、拍摄设备、系统及存储介质
WO2022001871A1 (zh) 参数确定方法、装置、设备、系统、程序和存储介质
JP2017103564A (ja) 制御装置、制御方法、およびプログラム
WO2023072030A1 (zh) 镜头自动对焦方法及装置、电子设备和计算机可读存储介质
JP6483661B2 (ja) 撮像制御装置、撮像制御方法およびプログラム
JP6305232B2 (ja) 情報処理装置、撮像装置、撮像システム、情報処理方法およびプログラム。
JP2013098746A (ja) 撮像装置、撮像方法およびプログラム
TW201236448A (en) Auto-focusing camera and method for automatically focusing of the camera

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: 18888701

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18888701

Country of ref document: EP

Kind code of ref document: A1