WO2020151083A1 - 区域确定方法、装置、存储介质和处理器 - Google Patents

区域确定方法、装置、存储介质和处理器 Download PDF

Info

Publication number
WO2020151083A1
WO2020151083A1 PCT/CN2019/080746 CN2019080746W WO2020151083A1 WO 2020151083 A1 WO2020151083 A1 WO 2020151083A1 CN 2019080746 W CN2019080746 W CN 2019080746W WO 2020151083 A1 WO2020151083 A1 WO 2020151083A1
Authority
WO
WIPO (PCT)
Prior art keywords
area
target
image
movement
intersects
Prior art date
Application number
PCT/CN2019/080746
Other languages
English (en)
French (fr)
Inventor
臧云波
鲁邹尧
吴明辉
Original Assignee
北京明略软件系统有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京明略软件系统有限公司 filed Critical 北京明略软件系统有限公司
Priority to JP2019554376A priority Critical patent/JP6949988B2/ja
Publication of WO2020151083A1 publication Critical patent/WO2020151083A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion

Definitions

  • the prevention and treatment of the target object mainly relies on the professional experience and subjective judgment of the prevention and treatment personnel, through one-sided clues, to determine the area where the capture tool used to capture the target object is placed in the prevention area.
  • the target object is a mouse
  • the capture tool is a mouse sticky board.
  • the prevention and control personnel observe the mouse's excrement, bite marks, and infestation trails, and based on experience, along the walls, corners of the room, next to the wires and other places where the mouse may pass.
  • the mouse sticky board can be deployed in the same place, and it may be observed for several days in the same place. If there is no harvest, then the sticky mouse board will be replaced and cycled.
  • a region determination method includes: acquiring a first image obtained by photographing a monitoring area; determining a preset area indicated by the area setting instruction on the first image in response to an input area setting instruction; acquiring a monitoring area within a first target time period The first set of image data obtained by shooting; the movement track of the target object in the first target time period is determined in the monitoring area according to the first set of image data; the target that intersects the movement track is determined in the preset area Area, where the target area is used to place the target capture device, and the target capture device is used to capture the target object.
  • the method further includes at least one of the following: in a case where the target area includes the first area, a target position for placing the target capturing device Is set to include one or more positions on the part where the movement track intersects the first area; in the case that the target area includes the first area, the target position for placing the target capture device is set as the first position, where The target capture device at the first position covers at least a predetermined number of movement trajectories in the first area; when the target area includes the second area, the target position for placing the target capture device is set to include the movement trajectory intersects the second area One or more locations on the part of the; in the case that the target area includes the third area, the target location for placing the target capture device is set to the location where one or more intersections are located.
  • the method further includes: displaying a first image for identifying the movement trajectory on the first image. Three identification information.
  • acquiring the first set of image data obtained by shooting the surveillance area within the first target time period includes: acquiring a video file obtained by shooting the surveillance area by a camera device; sampling the video files to obtain a group Video frame image data, where the first set of image data includes a set of video frame image data; before determining the movement track of the target object in the first target time period in the monitoring area according to the first set of image data , The method further includes: determining a plurality of target video frame images in a set of video frame images according to the pixel values of the pixel points in a set of video frame images, wherein each target video frame image is used to indicate in the monitoring area There are moving objects; target object detection is performed on each target video frame image, and the image characteristics of each target video frame image are obtained.
  • a region determining device including one or more processors, and one or more memories storing program units, where the program units are executed by the processors,
  • the program unit includes: a first acquisition unit, configured to acquire a first image obtained by photographing a monitoring area; a response unit, configured to determine a preset indicated by the area setting instruction on the first image in response to an input area setting instruction Area; the second acquisition unit is set to acquire the first set of image data obtained by shooting the surveillance area within the first target time period; the first determination unit is set to determine in the surveillance area according to the first set of image data The movement track of the target object in the first target time period is obtained; the second determining unit is configured to determine the target area intersecting the movement track in the preset area, wherein the target area is used to place the target capture device, the target capture device Used to capture the target object.
  • the first image obtained by shooting the monitoring area is acquired; the preset area indicated by the area setting instruction is determined on the first image in response to the input area setting instruction;
  • the first set of image data obtained by shooting in the area is determined in the monitoring area according to the first set of image data; the intersecting movement trajectory is determined in the preset area
  • the target area where the target area is used to place the target capture device, and the target capture device is used to capture the target object.
  • Fig. 1 is a flowchart of a method for determining an area according to an embodiment of the present disclosure
  • Fig. 3 is a schematic diagram of an identified area where a mouse trap can be placed according to an embodiment of the present disclosure
  • Fig. 5 is a schematic diagram of a placement area of a mouse trap device according to an embodiment of the present disclosure
  • Fig. 6 is a histogram of a rat trail report according to an embodiment of the present disclosure.
  • Fig. 7 is a schematic diagram of a data processing module according to an embodiment of the present disclosure.
  • Fig. 8 is a schematic diagram of the principle of a rat infestation detection system according to an embodiment of the present disclosure.
  • Fig. 9 is a schematic diagram of a Faster-RCNN network model according to an embodiment of the present disclosure.
  • Fig. 10 is a schematic diagram of an area determining device according to an embodiment of the present disclosure.
  • FIG. 11 is a schematic structural diagram of a storage medium according to an embodiment of the present disclosure.
  • Fig. 12 is a schematic structural diagram of a processor according to an embodiment of the present disclosure.
  • the embodiment of the present disclosure provides a region determination method.
  • Fig. 1 is a flowchart of a method for determining an area according to an embodiment of the present disclosure. As shown in Figure 1, the method may include the following steps:
  • the first image obtained by shooting the monitoring area can be obtained in scenes that require prevention and treatment of target objects, such as dining scenes and factory scenes.
  • the catering scene can be a scene with high requirements for food hygiene, a catering scene in a public operating place, or a catering scene in home life, and there are no restrictions here;
  • the factory scene can be food or medicine plus There are no restrictions on production plants, food or drug storage rooms and other scenarios with high hygiene requirements.
  • the target object may be a large-sized disease vector, for example, the target object is a mouse, or it may be a small-sized disease vector, for example, a cockroach.
  • the aforementioned camera may include, but is not limited to, a camera with an infrared lighting function, for example, an infrared low-light night vision camera. Further, the camera may also include but is not limited to: motion detection function, storage function, networking function (such as wifi networking) and high-definition (such as greater than 1080p) configuration.
  • Step S104 Determine the preset area indicated by the area setting instruction on the first image in response to the input area setting instruction.
  • step S104 of the present application after acquiring the first image obtained by shooting the monitoring area, the preset area indicated by the area setting instruction is determined on the first image in response to the input area setting instruction.
  • the area setting instruction is used to determine the preset area in the monitoring area on the first image, which can be input by the user through the terminal, for example, the area is triggered according to the sliding track of the user's finger or mouse on the terminal screen
  • the area setting instructions are used to indicate a preset area
  • the preset area is a preset suitable area for placing the target capture tool, for example, the preset area is the sliding of the user's finger or mouse on the terminal screen
  • the number of areas corresponding to the area formed by the track on the first image may be multiple.
  • the area setting instruction is input, the area setting instruction is responded to, and the preset area indicated by the area setting instruction is determined on the first image.
  • the users can be related personnel such as pest control personnel, restaurant operators, etc.
  • the area setting instruction of this embodiment may be input by the user based on experience, and the preset area in the monitoring area may be determined on the first image according to the attributes and activity rules of the target object.
  • the target object is a mouse
  • the target capture tool is a sticky mouse board. Because the mouse is good at climbing and drilling holes, it is likely to move up and down along wires, water pipes, etc. during activities.
  • sticky mice can be deployed
  • the place of the board can be a flat position on the ground, a corner of a wall, a window sill, etc. Therefore, in this embodiment, a number of suitable preset areas can be set in advance in the monitoring area targeted by the video monitoring equipment, and the mouse can be placed in the corners, near the wires, etc. Priority is given to determining the preset area on the only way that is similar to conventional travel.
  • the preset area indicated by the area setting instruction can be determined on the first image in response to the input area setting instruction, and it will be used to indicate the preset area.
  • the area information is stored in the server.
  • Step S106 Obtain a first set of image data obtained by shooting the monitored area within the first target time period.
  • the first target time period may be a time period during which a predetermined target object period has passed for a period of time. It may be the previous day or a few days before the first group of images.
  • the data is used to indicate the image of the surveillance area in the first target time period, which may be video data or picture data, and may include the time for shooting the surveillance area.
  • the first set of image data is video data
  • the first set of image data can be used to indicate the continuous video images of the monitoring area in the first target time period.
  • the first set of image data is picture data
  • the first set of image data may be used to indicate a picture set of multiple pictures in the target area within the first target time.
  • the first set of image data obtained by shooting the monitored area within the first target time period is acquired through the video monitoring device.
  • Step S108 Determine the movement track of the target object in the first target time period in the monitoring area according to the first set of image data.
  • step S108 of the present application after acquiring the first set of image data obtained by photographing the monitored area within the first target time period, the target is determined in the monitored area according to the first set of image data The movement track of the object in the first target time period.
  • the movement track of the target object in the monitoring area in the first target time period is determined from the first set of image data.
  • the movement track is also the historical movement track, which can be determined by the target object in the monitoring area in the first target time period.
  • the inner passing position points are formed, and the moving track can indicate the moving direction of the target object in the monitoring area.
  • this embodiment may extract a video clip with an image of the target object from the video indicated by the first image data, and then recognize the dynamic change features in the video clip through the motion recognition technology, and use the image recognition technology to compare the dynamic For example, through artificial intelligence (AI) image recognition technology to identify the dynamic change characteristics in the video clip, it is further confirmed that the creature that appears is indeed the target object, and then combined with the dynamic change characteristics to determine the target object’s Movement track.
  • AI artificial intelligence
  • the target object in this embodiment is a mouse
  • a target video of a mouse in the surveillance area of the previous day is acquired, and the target video can be intercepted according to the time period, and a video clip including a mouse image can be extracted from it.
  • AI image recognition technology is used to identify the dynamic change features in the video clips to further determine that the target object in the monitoring area is indeed a mouse, and then combine the dynamic change characteristics to determine the mouse's movement track, and in the image of the monitoring area Shows, for example, the green line indicates the movement track of the mouse in the monitoring area.
  • this embodiment can also identify the type of the target object, the skin color of the target object, and the target object's The number, the shape of the target object, and the length of time the target object moves in the monitoring area during the first target time period and other information related to the target object.
  • the target object is a mouse
  • it in addition to identifying the movement trajectory of the mouse based on the first image data, it can also identify the number of rats, the skin color of the mouse, the shape of the mouse, the length of movement of the mouse, etc., and further identification
  • the species of mice that come out such as house mice, brown mice, yellow-breasted mice, etc.
  • the house mice usually build nests in wall foundations, warehouse stocks and insulation layers, or in broken cardboard boxes and drawers. It may also enter the room with the cargo.
  • the brown rat is very alert and can enter the room through the sewer, toilet, etc.
  • the yellow-breasted rat can climb up and down along rough walls, walk along iron wires and wires, and can invade through pipeline holes and ceilings. indoor.
  • the preset area indicated by the area setting instruction is determined on the first image in response to the input area setting instruction, and the monitoring area is determined according to the first set of image data After the movement trajectory of the target object in the first target time period, a target area that intersects the movement trajectory is determined in the preset area, where the target area is used to place the target capture device, and the target capture device is used to capture the target object.
  • the preset area can be compared with the movement track.
  • the preset area is multiple areas, and the target area that intersects the movement track among the multiple areas is determined to be used for placing the target capture device. That is, the target area is comprehensively determined by the preset area and the actual movement trajectory of the target object. It is determined as the area where the user needs to place the target capture tool in the monitoring area, so as to take into account the user’s experience and The actual haunt information of the target object is comprehensively determined to obtain the target area, thereby avoiding relying only on the professional experience and subjective judgment of the prevention and control personnel to determine the area where the capturing tool is placed, and improving the accuracy of determining the area for placing the capturing device.
  • the first image obtained by shooting the monitoring area is acquired; the preset area indicated by the area setting instruction is determined on the first image in response to the input area setting instruction;
  • the movement trajectory that intersects the target area in this embodiment meets the preset target condition, that is, not any area that intersects the movement trajectory in the preset area can be used as the target area, but the movement trajectory that intersects it needs to meet the target condition , To further improve the accuracy of determining the area where the capture device is placed.
  • the target condition can be determined according to the number of movement trajectories that intersect the preset area, the length of the movement trajectory, and the number of movement trajectories that have intersections.
  • the movement trajectory in one or more areas other than the target area does not satisfy the aforementioned target condition.
  • At least one of the following target areas that intersect the movement track is determined in one or more areas: the first area that intersects the movement track is determined in one or more areas, wherein, The target area includes a first area, and the number of movement trajectories that intersect the first area is greater than a first threshold; the second area that intersects the movement trajectory is determined in one or more areas, wherein the target area includes the second area, and The number of movement trajectories intersected by the second area is greater than the second threshold, and the length of the part where the movement trajectory intersects with the second area is greater than the third threshold; the third area intersecting the movement trajectory is determined in one or more areas, where, The target area includes a third area, and the number of movement trajectories that intersect the third area and have intersections in the third area are greater than the fourth threshold.
  • the method for determining the target area that intersects the movement track in one or more areas may include multiple methods.
  • the number of movement trajectories that intersect the target area may be different.
  • the target condition in this embodiment can be set based on the number of movement trajectories.
  • the number of movement trajectories that intersect the target area is set as the first threshold, for example, the first threshold. If a threshold is 5, the target condition may be a condition that the number of movement tracks that intersect the target area is greater than the first threshold.
  • a first area that intersects the movement track is determined from one or more areas, wherein the above-mentioned target area includes the first area, and the number of movement tracks that intersect the first area is greater than the first threshold , And then place the target capturing device in the first area, thereby improving the efficiency of capturing the target object.
  • the above-mentioned first threshold can be set according to actual application scenarios. As long as the first threshold can improve the accuracy of determining the area of the target capture device, it is within the scope of the embodiments of the present disclosure. Let me illustrate them one by one.
  • the number of movement trajectories that intersect the target area in this embodiment and the length of the movement trajectory of the part that intersects the target area are different.
  • the target condition of this embodiment can be set based on the number of movement trajectories that intersect the target area and the length of the movement trajectory of the part that intersects the target area, wherein the number of movement trajectories that intersect the target area is set as the second threshold.
  • the second threshold value is 5
  • the length of the movement track of the part that intersects with the target area is set to the third threshold value.
  • the target condition can be intersecting with the target area.
  • the condition that the number of movement trajectories is greater than the second threshold, and the length of the movement trajectory of the part where the movement trajectory intersects with the target area is greater than the third threshold.
  • a second area that intersects the movement track is determined in one or more areas, wherein the target area includes the second area, and the number of movement tracks that intersect the second area is greater than the second threshold, The length of the part where the movement track intersects the second area is greater than the third threshold, and the target capturing device is placed in the second area, thereby improving the efficiency of capturing the target object.
  • second and third thresholds can be set according to actual application scenarios, as long as the second and third thresholds that can improve the accuracy of determining the position of the target capture device are in the embodiments of the present disclosure. Within the scope of, here is no longer an example.
  • this embodiment intersects the target area and has a different number of movement trajectories where there are intersection points in the target area.
  • the target condition of this embodiment can be set based on the number of movement trajectories that intersect with the target area and have intersections in the target area, where the number of movement trajectories that intersect with the target area and that have intersections in the target area is set as The fourth threshold, for example, if the fourth threshold is 6, the target condition may be a condition that the number of movement trajectories that intersect the target area and have intersections in the target area is greater than the fourth threshold.
  • a third area that intersects the movement track is determined in one or more areas, wherein the target area includes the third area, and the third area intersects with the third area and has an intersection point in the third area.
  • the number of movement tracks is greater than the fourth threshold, and the target capture device is placed in the third area, thereby improving the efficiency of capturing the target object.
  • the foregoing fourth threshold can be set according to actual application scenarios. As long as the fourth threshold can improve the accuracy of determining the position of the target capture device, it is within the scope of the embodiments of the present disclosure. Let me illustrate them one by one.
  • the following describes the determination of the target position for placing the target capturing device in the target area in this embodiment.
  • This embodiment can determine the target area where the target object appears in the first target time period in the monitoring area, and then determine the capture device in the target area according to the movement track of the target object in the target time period and the corresponding movement time.
  • the target position achieves the purpose of determining the position of the capture tool.
  • the method further includes at least one of the following: if the target area includes the first area, it is used to place the target The target position of the capture device is set to include one or more positions on the part where the movement track intersects the first area; in the case that the target area includes the first area, the target position for placing the target capture device is set to the first Position, wherein the target capture device at the first position covers at least a predetermined number of movement tracks in the first area; in the case that the target area includes the second area, the target position for placing the target capture device is set to include the movement tracks One or more positions on the part that intersects the second area; in the case where the target area includes the third area, the target position for placing the target capturing device is set to the position where the one or more intersection points are located.
  • the target position for placing the target capturing device can be determined in the target area.
  • the number of movement tracks that intersect the first area is greater than the first threshold.
  • the target position for placing the target capture device may be determined according to the part where the movement track intersects the first area, and the target position may be set to include the movement track and the first area. One or more positions on the part where the regions intersect, thereby improving the accuracy of determining the position of the target capturing device.
  • this embodiment determines the final target position for placing the target capturing device based on the number of movement tracks that the target capturing device can cover.
  • the target area includes the first area
  • the first position of the target capturing device covering at least a predetermined number of movement tracks in the first area is determined as the final target position, and multiple movement tracks in the first area can be compared.
  • the target location is determined in dense locations, thereby improving the accuracy of determining the location of the target capturing device.
  • the number of movement trajectories that intersect the second area is greater than the second threshold, and the length of the portion of the movement trajectory that intersects the second area is greater than the third threshold.
  • the target area includes the second area, it may be determined to include the part where the movement track intersects with the second area, and one or more positions on the part where the movement track intersects the second area may be determined, and then use The target position where the target capturing device is placed is set to include one or more positions on the part where the movement track intersects the second area, thereby improving the accuracy of determining the position of the target capturing device.
  • the number of movement trajectories that intersect the third area and have intersection points in the third area are greater than the fourth threshold.
  • the target area includes the third area, determine the position of one or more intersections existing in the third area, and then set the target position for placing the target capture device to one or more intersections The location of the point, thereby improving the accuracy of determining the location of the target capture device.
  • the plurality of intersections can be selected from the plurality of intersections, which are formed by moving trajectories greater than the target number.
  • the target intersection of the target intersection, and then the location of the target intersection in the third area is determined as the target location.
  • the number of movement trajectories with intersections in the third area intersecting the movement trajectory includes A movement trajectory, B movement trajectory, C movement trajectory, and D movement trajectory. It can get the a intersection point, b intersection point, c intersection point, d intersection point formed by the movement trajectory A, B movement trajectory, C movement trajectory, and D movement trajectory.
  • the intersection a can be formed by the movement trajectory A and the movement B
  • the intersection b can be A movement trajectory, B movement trajectory, and C movement trajectory are formed
  • c intersection point can be formed by C movement trajectory and D movement trajectory
  • d intersection point can be formed by A movement trajectory and D movement trajectory. From multiple intersections, select the target intersection formed by the number of movement trajectories greater than the target.
  • the target intersection is the point on the movement trajectory that the target object has passed multiple times. It can be from the intersection a, intersection b, and intersection c. Select the target intersection b formed by the movement track greater than 2 from the intersection of d, and determine the corresponding position of the target intersection b in the third area as the target position for placing the target capture device, thereby further The accuracy of determining the position of the target capturing device is improved.
  • a priority order can be set for the multiple target positions when actually placing the target capture tool, for example, it will be easier to place the target capture tool
  • the target location is identified by the target indication information to indicate that the target capture tool can be placed first.
  • the method further includes: displaying first identification information for identifying the target area on the first image.
  • the target area can be identified.
  • the target area on the first image the target area can be identified by the first identification information.
  • the first identification information can be a striking mark such as graphics, text, and symbols, for example, the first identification information is a red circle.
  • the target area is circled in the first image by a red circle to remind the target user of the location of the target area in the entire monitoring area, and then instruct the target user to deploy the target capture tool on the target area, thereby improving the performance of the target object. The efficiency of capture.
  • displaying the first identification information for identifying the preset area on the first image includes: displaying the first identification information for identifying the range of the target area on the first image.
  • the first identification information can be used to identify the size of the range of the target area in the monitoring area to indicate the range in which the target capture tool can be placed in the monitoring area.
  • the first identification information is a red circle.
  • the size of the red circle can be used to indicate the size of the range of the target area in the monitoring area.
  • the preset area is one or more preset areas, and after the preset area indicated by the area setting instruction is determined on the first image in response to the input area setting instruction, the method further The method includes: respectively displaying second identification information for identifying one or more regions on the first image.
  • the preset area is one or more preset areas, and one or more areas can be determined on the only way through which the rodents routinely travel, such as wall corners, beside electric wires.
  • the second identification information for identifying one or more areas is displayed on the first image.
  • the information can be eye-catching marks such as graphics, text, symbols, etc., for example, a triangle, to prompt the target user of the position of the preset area in the entire monitoring area.
  • determining the target area that intersects the movement track in the preset area includes: determining the target area that intersects the movement track in one or more areas; After the target area that intersects the movement track is determined in the area, the method further includes: retaining the second identification information of the target area, and hiding the second identification information of the area except the target area in one or more areas; and/or To display the first identification information used to identify the target area.
  • the second identification information used to identify one or more regions is respectively displayed on the first image, the target region that intersects the movement track is determined in the one or more regions, and the target region is determined in the one or more regions.
  • the second identification information of the target area in one or more areas can be retained, and the second identification information of the area other than the target area in one or more areas can be hidden, such as , Only retain the triangles of the target area, and hide the triangles of one or more areas other than the target area to indicate that the area corresponding to the triangle currently displayed is the area where the target capture tool can be placed in the monitoring area.
  • this embodiment may only display the first identification information for identifying the target area in the one or more areas, for example, only display The red circle indicates that the area corresponding to the red circle currently displayed is the area where the target capture tool can be placed in the monitoring area.
  • this embodiment may retain the second identification information of the target area, and hide the areas other than the target area in one or more areas. And also display the first identification information for identifying the target area in one or more areas, for example, by displaying both a triangle and a red circle to indicate the target area in one or more areas, The area corresponding to the indicator triangle and the red circle is the area where the target capture tool can be placed in the monitoring area.
  • the method further includes: displaying on the first image The third identification information that identifies the movement track.
  • the movement trajectory of the target object in the first target time period is determined in the monitoring area according to the first set of image data
  • the movement trajectory can be identified to indicate that the target object is in the monitoring area.
  • the movement situation for example, is identified by the third identification information displayed on the first image, the third identification information may be a line, and the color and thickness of the line are not limited here.
  • determining the preset area indicated by the area setting instruction on the first image in response to the input area setting instruction includes: identifying the target type area in the monitoring area in the first image, wherein, The probability of the target object passing through the target type area is greater than the fifth threshold; the preset area indicated by the area setting instruction is determined on the target type area in response to the input area setting instruction.
  • the target type area in the monitoring area may be first identified in the first image.
  • the type area is the area determined by the attributes and activity rules of the target object.
  • the probability of the target object passing through the target type area is greater than the fifth threshold.
  • the target object is a mouse. Because the mouse is good at climbing and drilling holes, The activity is likely to move up and down along wires, water pipes, etc.
  • the place where the mouse stick can be deployed can be the ground, corners, window sills, etc., and the target type area of this embodiment can be close to wires and water pipes. The ground, corners, window sills and other flat areas.
  • the preset area indicated by the area setting instruction can be determined on the target type area in response to the input area setting instruction, and the user's finger or mouse can be displayed on the terminal screen.
  • the area setting instruction is triggered by the sliding track of the target type area, and the preset area indicated by the area setting instruction is determined on the target type area.
  • the user can input the area setting instruction based on experience to determine the area setting instruction in the target type area
  • the preset area, and then the target area intersecting the movement track is determined in the preset area, which improves the accuracy of the area for placing the target capturing device.
  • the movement track of the target object in the first target time period when the movement track of the target object in the first target time period is determined in the monitoring area according to the first set of image data, it can be first identified from the first set of image data that the target object is monitoring Multiple locations passed in the area.
  • the multiple locations can be represented by coordinate points (X, Y, Z) in the three-dimensional coordinate system, for example,
  • the multiple positions are A (X1, Y1, Z1), B (X1, Y1, Z1), C (X1, Y1, Z1), D (X1, Y1, Z1).
  • the movement trajectory can be generated from multiple locations, and multiple locations can be connected by lines, for example, location A(X1, Y1, Z1), B(X1, Y1, Z1), C(X1, Y1, Z1), D(X1, Y1, Z1) are connected by a line, thereby forming a moving track of the target object in the first target time period.
  • step S110 after the target area that intersects the movement track is determined in the preset area, the method further includes: when the target capture device is placed in the target area, capturing by the target capture device The target information of the target object is sent to the target terminal; and/or the target area is sent to the target terminal; or the second image of the target area and the monitoring area is sent to the target terminal; or the third image of the monitoring area is displayed on the target terminal An image, where the target area is displayed on the third image; or a fourth image of the monitored area is displayed on the target terminal, where the movement track and the target area are displayed on the fourth image.
  • the target information of the target object captured by the target capture device can be sent to the target when the target capture device is placed in the target area.
  • the terminal obtains the target information of the target object monitored by each monitoring device, and sends the target information of the target object to the target terminal.
  • the target information can be the type of the target object, the skin color of the target object, the number of the target object, and the target There are no restrictions on the object's shape and other information here.
  • this embodiment may also send the target area to the target terminal to instruct the target user to place the target capture tool in the monitoring area according to the target area.
  • This embodiment can also send both the target area and the second image of the monitored area to the target terminal, so that the target user can understand the specific location of the target area in the monitored area.
  • This embodiment may also display the third image including the target area on the target terminal, and may also display the fifth image of the monitored area including the movement track and the target area on the target terminal, so that the user can understand the target area and the target area in the monitored area.
  • the target capturing device is then placed on the target area, thereby improving the efficiency of capturing the target object.
  • the target terminal in this embodiment can be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a palmtop computer, and a mobile Internet device (Mobile Internet Devices, referred to as MID), PAD and other terminal devices.
  • a smart phone such as an Android phone, an iOS phone, etc.
  • a tablet computer such as a Samsung Galaxy Tabs, etc.
  • a palmtop computer such as a Samsung Galaxy Tabs, etc.
  • MID mobile Internet Devices
  • the second set of image data of the target area in each second target time period starting from the target time is obtained, and at least one set of first image data is obtained.
  • the target report of the target object in the target area can be automatically issued.
  • the second target time period may be 1 day, that is, when the target capturing device is placed at the target position, image data of the target area is acquired every day.
  • the target information of the target object entering the target area can be identified from at least one set of the second set of image data, and at least one set of target information can be obtained, and the at least one set of target information can be converted into a target report.
  • the target report may also include the target
  • the information such as the name and time of the area where the object appears can be in the form of text, table, statistical graph, etc. There are no restrictions here, and the target report is pushed to the target terminal through the server, so that the target user can use the target terminal Understand the situation of the target object in the target area, including the trend of the target object, so as to understand whether the target area is in a serious health threat, for the target user to comprehensively judge the situation on the spot, and implement the prevention and control of pests in a targeted manner Work, and can also guide whether there are loopholes in the building structure.
  • this embodiment can determine the intrusion point and hiding point of the target object in the monitoring area.
  • the first set of image data in this embodiment includes the video data of the surveillance area shot by the video surveillance equipment, and the video of the target object in the surveillance area is intercepted from the video data.
  • Obtain the first video frame in the video where the target object appears identify the position of the target object in the surveillance area from the first video frame, and determine the identified position as an intrusion of the target object in the surveillance area Point, you can use it as the entrance of the target object to invade the indoor place.
  • This embodiment can also obtain the last video frame from the video in which the target object appears, identify the position of the target object in the surveillance area from the last video frame, and determine the identified position as the hiding point of the target object. It can be used as a den for the target object, or as an exit when escaping from the surveillance area.
  • this embodiment can record the intrusion point and hiding point of the target object in the past period of time, and send information indicating the intrusion point and hiding point of the target object to the target terminal to prompt the prevention and control personnel to prevent and control the target object Further measures are taken to achieve the goal of improving the efficiency of prevention and control of target objects.
  • the target object is a mouse
  • the person in charge of pest control searches for a sewer opening with a large gap near the invasion point, or whether there is a pipeline leading to the outdoors near the invasion point, if the invasion point has a sewer opening with a large gap, or The pipes leading to the outdoors are blocked in time to sewer openings or pipes, thereby cutting off the passage for rats to invade, and improving the efficiency of prevention and control of target objects.
  • this embodiment can determine the density of the target object in the surveillance area shot by the video surveillance device, and can determine the density of the target object in different surveillance areas.
  • the ratio of the length of time for the target object in the monitoring area to the time of the entire monitoring cycle for monitoring the target object is obtained, and it is determined as the density of the target object in the monitoring area.
  • the monitoring area with the highest density of the target object is determined therefrom, and it is determined as the monitoring area where the target object frequently invades, which can be used to indicate the target object
  • the information of the frequently invaded monitoring area is sent to the target terminal to prompt the relevant personnel to take further measures to achieve the purpose of improving the efficiency of prevention and control of the target object.
  • the density of the rat in the monitoring area is also the rat density value in the monitoring area.
  • determine the monitoring area with the highest density of rats that is, record the places with higher indoor rat density values, determine them as places with frequent rat invasions, and use them for indication
  • the information of places with frequent rat intrusions is sent to the target terminal to prompt restaurant operators to further check whether the place has factors that lead to the breeding of rats and insect pests, such as whether there are residual food residues, uncleaned water marks, etc. It becomes a breeding ground for rats and insect pests. If there are factors that lead to the breeding of rats and pests in the site, the restaurant operators are instructed to further manage the site to reduce the attraction of the site to the target object, thereby improving the efficiency of the prevention and control of the target object.
  • the area determining method of this embodiment involves determining the target object, that is, determining whether there is a target object in the monitoring area, and after determining that there is a target object in the monitoring area, determine the target object in the monitoring area according to the first set of image data The movement trajectory of the target object in the first target time period, and then a target area intersecting the movement trajectory is determined in the preset area, so as to place a target capturing device for capturing the target object in the target area.
  • the algorithm for determining the target object of this embodiment will be introduced below.
  • step S106 acquiring the first set of image data obtained by shooting the surveillance area within the first target time period includes: acquiring a video file obtained by shooting the surveillance area by a camera device; Perform frame sampling to obtain a set of video frame image data, where the first set of image data includes a set of video frame image data; in step S108, the target object is determined in the monitoring area according to the first set of image data Before the movement trajectory in the first target time period, the method further includes: determining a plurality of target video frame images in a group of video frame images according to the pixel values of pixels in a group of video frame images, wherein each The target video frame image is used to indicate the presence of moving objects in the monitoring area; target object detection is performed on each target video frame image to obtain the image characteristics of each target video frame image, where the image characteristics are used to indicate the presence of moving objects Among the objects, the target image area where the similarity between the target object and the target object is greater than the sixth threshold; the motion characteristics are determined according to the image characteristics of each target video
  • the camera device may be a surveillance camera, for example, the camera device is an infrared low-light night vision camera, which is used to photograph the surveillance area to obtain a video file.
  • the monitoring area is the detected area, that is, the monitoring area is the area where the target object is detected.
  • the video file of this embodiment includes original video data obtained by shooting the monitored area, and may include a monitored video sequence of the monitored area, which is also an image video sequence.
  • the video file After acquiring the video file captured by the camera equipment in the monitoring area, the video file is preprocessed, and the video file can be sampled at the video data processing layer to obtain a set of video frame images.
  • the video file can be sampled at equal intervals to obtain a set of video frame images of the video file.
  • the video file includes 100 video frame sequences. After the frame sampling is performed, 10 video frame sequences are obtained. , The 10 video frame sequences are used as the above-mentioned set of video frame images, thereby reducing the computational complexity of the algorithm for determining the target object.
  • preprocessing the video file also includes performing dynamic detection on the video file, and determining a target video frame image indicating the presence of a moving object in the monitoring area from a set of video frame images, that is, There is a moving object in the target video frame image, and the target video frame image may be a video clip with a moving object, where the moving object may or may not be the target object.
  • a target video frame image can be determined by a dynamic detection algorithm, and multiple target video frame images can be determined in a group of video frame images according to the pixel values of pixels in a group of video frame images.
  • video frame images other than multiple target video frame images do not indicate that there are moving images in the corresponding monitoring area, and subsequent detection may not be performed.
  • the image feature is used to indicate the target image area where the moving object in the target video frame image is determined to be the target object.
  • the target object detection is performed on each target video frame image, that is, the moving object existing in the target video frame image is detected.
  • the target detection system can adopt the dynamic target detection method and the target based on neural network.
  • the detection method detects the moving objects in the target video frame image, and obtains the image characteristics of each target video frame image.
  • the dynamic target detection method has fast calculation speed and low requirements for machine configuration, while the neural network-based target The accuracy and robustness of the detection method is better.
  • the image feature can be the visual information in the rectangular frame to indicate the target image area.
  • the rectangular frame can be the detection frame to indicate that the object in motion is in line with the target.
  • the target image area where the similarity between objects is greater than the sixth threshold, that is, the object with the similarity greater than the sixth threshold may be the target object, and the target image feature is also used to indicate the target object Possible location.
  • the image characteristics of each target video frame image can be input to the motion feature extraction module, which is based on each The image features of each target video frame image determine the motion feature.
  • the motion feature is used to indicate the motion speed and direction of the moving objects in the multiple target video frame images, and further filter Remove interference images caused by the movement of non-target objects, for example, remove interference information such as the movement of mosquitoes.
  • the motion feature extraction algorithm of the motion feature extraction module can first detect the image features of each target video frame image.
  • the correlation of image features between multiple target video frame images can determine the object corresponding to the image feature with large correlation as the same object, and match the image features of each target video frame image to obtain a series of motions of the object Picture, finally you can use the 3D feature extraction network to extract the features of the motion sequence to obtain the motion features. For example, according to the detection frame of each target video frame image, calculate the correlation of the detection frame between multiple target video frame images.
  • the object corresponding to the detection frame with high correlation is determined as the same object, and the detection frame of each target video frame image is matched to obtain a series of moving pictures of the object.
  • the 3D feature extraction network is used to extract the features of the motion sequence, and get Motion characteristics, and then determine the motion speed and direction of the moving objects in multiple target video frame images.
  • this embodiment can also merge the image features of multiple target video frame images and perform feature extraction, so as to prevent the target detector of a single frame from being misjudged, and then to achieve precise screening of the target image. Determine whether the target object appears.
  • the motion features and the image features of each target video frame image can be fused and input into a pre-trained classification network, which is pre-designed Good classification network model used to determine whether there are target objects in multiple target video frames, and then determine whether there are target objects in multiple target video frames based on the motion characteristics and the image characteristics of each target video frame image , For example, to determine whether there are rats in multiple target video frames.
  • a pre-trained classification network which is pre-designed Good classification network model used to determine whether there are target objects in multiple target video frames, and then determine whether there are target objects in multiple target video frames based on the motion characteristics and the image characteristics of each target video frame image , For example, to determine whether there are rats in multiple target video frames.
  • this embodiment inputs the image characteristics of the target video frame with the target object in the multiple target video frame images to the front-end display interface, which can further display the detection frame and movement track of the target object.
  • the classification network model of this embodiment can be used to filter non-target object picture sequences, while retaining the target object picture sequence, thereby reducing the false alarm rate and ensuring the accuracy of the target object prompt information.
  • the video files in the monitoring area are sampled to obtain a group of video frame images.
  • a group of video frame images is determined to indicate that it is in the monitoring area.
  • the data of multiple target video frame images is determined in a group of video frame images according to the pixel values of pixels in a group of video frame images
  • the data of each pixel in a group of video frame images is obtained.
  • Average pixel value obtain the difference between the pixel value of each pixel in each video frame image in a group of video frame images and the corresponding average pixel value; make the difference value in a group of video frame images meet predetermined conditions
  • the video frame image of is determined as the target video frame image.
  • obtaining the difference between the pixel value of each pixel in each video frame image in a group of video frame images and the corresponding average pixel value includes: for a group of video frame images
  • determining the motion feature according to the image feature of each target video frame image includes: obtaining a target vector corresponding to the target image area represented by the image feature of each target video frame image, to obtain multiple Target vector, where each target vector is used to represent the moving speed and direction of a moving object in a corresponding target video frame image when passing through the target image area; multiple target vectors are set in accordance with each target video frame image
  • the time sequence in the video file composes the first target vector, where the motion feature includes the first target vector; or the two-dimensional optical flow diagram corresponding to the target image area represented by the image feature of each target video frame image is obtained, and the multiple A two-dimensional optical flow diagram, where each two-dimensional optical flow diagram includes the moving speed and direction of a moving object in a corresponding target video frame image when passing through the target image area; multiple two-dimensional optical flow diagrams
  • the three-dimensional second target vector is composed according to the time sequence of each target video frame image in the video file, where the motion feature includes the three-dimensional second target vector.
  • determining whether the target object appears in the multiple target video frame images includes: combining the motion feature and the image of each target video frame image
  • the features are input into a pre-trained neural network model to obtain object recognition results, where the object recognition results are used to indicate whether there are target objects in multiple target video frame images.
  • inputting the motion feature and the image feature of each target video frame image into a pre-trained neural network model to obtain the object recognition result includes: passing each image feature through a convolutional layer, The neural network layer structure of the regularization layer and the activation function layer to obtain multiple first feature vectors; fuse multiple first feature vectors with motion features to obtain a second feature vector; input the second feature vector to the fully connected layer Perform classification to obtain the first classification result.
  • the neural network model includes the neural network layer structure and the fully connected layer.
  • the object recognition result includes the first classification result. The first classification result is used to indicate whether there are multiple target video frames.
  • Target object or pass each image feature through a first neural network layer structure including a convolutional layer, a regularization layer and an activation function layer to obtain multiple first feature vectors; pass a motion feature through a convolutional layer, a regularization layer
  • the second neural network layer structure of the activation function layer is used to obtain the second feature vector; the multiple first feature vectors are merged with the second feature vector to obtain the third feature vector; the third feature vector is input to the fully connected layer to perform Classification to obtain a second classification result, where the neural network model includes a first neural network layer structure, a second neural network layer structure, and a fully connected layer, and the object recognition result includes a second classification result, and the second classification result is used to represent multiple Whether the target object appears in the target video frame image.
  • multiple first feature vectors and motion features can be spliced (or called a combination) to obtain a second feature vector.
  • first feature vectors and second feature vectors can be spliced (or called a combination) to obtain a third feature vector.
  • inputting the motion feature and the image feature of each target video frame image into a pre-trained neural network model to obtain the object recognition result includes: passing each image feature through multiple blocks in turn, Obtain a plurality of first feature vectors, where in each block, the input of the block is sequentially performed on the convolution operation on the convolution layer, the regularization operation on the regularization layer, and the activation operation on the activation function layer; The first feature vector is spliced with the motion feature to obtain the second feature vector; the second feature vector is input to the fully connected layer, and the first classification result is obtained through the output of the fully connected layer.
  • the neural network model includes multiple blocks and full In the connection layer, the object recognition result includes the first classification result.
  • the first classification result is used to indicate whether the target object appears in the multiple target video frame images; or each image feature passes through multiple first blocks in turn to obtain multiple first blocks.
  • the feature passes through multiple second blocks in turn to obtain a second feature vector. In each second block, the input of the second block is sequentially performed on the convolution layer and the regularization operation on the regularization layer.
  • performing frame sampling on a video file to obtain a group of video frame images includes: sampling a video sequence in the video file at equal intervals to obtain a group of video frame images.
  • acquiring a video file captured by a camera device on a monitored area includes: the acquired video file includes: acquiring a video file captured by an infrared low-light night vision camera on the monitored area, where in the video file The video frame image is an image captured by an infrared low-light night vision camera.
  • the method further includes: in the case where it is determined that the target object appears in the multiple target video frame images, determining the target The position of the object in multiple target video frames; the position is displayed in multiple target video frames.
  • the method for determining the target object is executed by a server set locally.
  • the scene video sequence is collected by an infrared low-light night vision camera, and the data processing module receives the video sequence and detects whether there is a mouse in the video. If a mouse is detected, a series of information such as the position of the mouse is output To the front-end display interface, the front-end display interface displays the location, appearance time, and active area of the mouse, and can immediately alarm the mouse.
  • this embodiment realizes the automatic determination of the placement area of the mouse trap in the scenes of restaurants, factories, etc., by acquiring the first image obtained by shooting the monitoring area, and responding to the input area setting instruction on the first image Determine the preset area indicated by the area setting instruction, obtain the first set of image data obtained by shooting the surveillance area within the first target time period, and determine the target object in the surveillance area according to the first set of image data
  • the target area that intersects the movement trajectory is determined in the preset area, which achieves the purpose of determining the area for placing the target capture tool, and improves the
  • the accuracy of determining the area of the target capture tool that is, the use of a computer to complete the automatic judgment of the placement area of the target capture tool, instead of manual judgment and empirical judgment, can be used to assist the target in a clean and hygienic environment such as restaurants and factories.
  • the prevention and control of the object thereby guiding the effective development of the prevention and control of the target object, ensuring that key places and facilities in the catering industry are not attacked by the target object, and eliminating the need to manually mark the placement area of the target capture tool in each monitoring area Process, thereby saving labor costs.
  • the target object is a mouse
  • the target capture tool is a mousetrap
  • This embodiment applies digital technology and proposes a method for automatically determining the placement area of the mouse trap in the restaurant, factory, etc. scenes.
  • the computer can be used to complete the automatic determination of the placement area of the mouse trap, thereby avoiding manual determination.
  • the placement area and empirical determination of placement area can be used in restaurants, factories and other places that pay attention to cleaning and sanitation to assist rodent prevention and rodent control, thereby improving the prevention and control of rodents and other harmful organisms, and ensuring the catering industry Key places and facilities are not attacked by rodents, and the process of manually marking the area where the rodent trap is placed in each monitoring area is eliminated, thereby saving labor costs.
  • Fig. 2 is a flowchart of a method for determining the placement position of a rodent killer according to an embodiment of the present disclosure. As shown in Figure 2, the method includes the following steps:
  • Step S201 Identify the placeable area of the mouse trap in the monitoring area.
  • Step S202 Identify the historical movement track of the mouse in the monitoring area.
  • Step S203 Determine the target area for placing the mouse trap according to the placeable area of the mouse trap and the historical movement track of the mouse.
  • the area where the deployment device can be placed can be flat areas such as the ground, corners, windowsills, etc. . Therefore, in this embodiment, a number of suitable placement areas are preset in the static area where the video surveillance equipment is aligned, and corners, next to electric wires, and other areas where rats normally travel can be optimized.
  • the video monitoring device may be a camera.
  • Fig. 3 is a schematic diagram of an identified area where a mouse trap can be placed according to an embodiment of the present disclosure.
  • the static image of the surveillance area is acquired by the camera equipment, and the ground, corners and other areas can be identified from the image of the surveillance area through image recognition technology, and the ground, the corner of the wall can also be identified manually from the image of the surveillance area. Corners, windowsills and other areas.
  • the triangle in the image of the monitoring area in FIG. 3 is used to indicate the area where the mouse trap can be placed, and the area includes multiple, for example, five.
  • video surveillance equipment can be installed in different surveillance areas to identify areas where the mouse trap can be placed in different surveillance areas.
  • the identification of the area where the mouse trap can be placed is completed, and the identified area where the mouse trap can be placed can be stored in advance.
  • the server is waiting to be used.
  • the historical movement track of the mouse can be acquired through the video monitoring device, and the target video of the mouse in the surveillance area of the previous day can be acquired, the target video is intercepted according to the time period, and the video segment including the mouse image can be extracted. Then use motion recognition technology to identify the dynamic change features in the video clip, and then use the image recognition technology to identify the dynamic change feature, for example, through the AI image recognition technology to identify the dynamic change feature in the video clip, and further determine the occurrence
  • the creature is indeed a mouse, and then combined with the dynamic change characteristics to determine the mouse's moving track, which is shown in the image of the monitoring area.
  • Fig. 4 is a schematic diagram of an identified rat track according to an embodiment of the present disclosure.
  • AI image recognition technology and motion recognition technology are used to extract the movement track of the mouse in the surveillance video of the surveillance area. Marked in the image of the monitoring area, the lines in it are the movement track of the mouse in the monitoring area.
  • This embodiment identifies the placeable area of the mouse trap in the monitoring area. After identifying the historical movement track of the mouse in the monitored area, the placeable area of the mouse trap is compared with the historical movement track of the mouse in the monitoring area. The area where the overlapping part is located is determined as the target area for placing the mouse trap.
  • Fig. 5 is a schematic diagram of a placement area of a mouse trap device according to an embodiment of the present disclosure.
  • the target area corresponding to the part where the mouse trap can be placed in the monitoring area ( Figure 3) and the historical movement track of the mouse in the monitoring area ( Figure 4) overlap with a circle
  • a red circle or other eye-catching color it is the area where the control personnel need to place the mouse trap in the monitoring area that day.
  • this embodiment can also be used to determine the intrusion point and hiding point of the pest in the surveillance area. For example, if the pest is a rat, calculate the invasion point and hiding point of the rat in the surveillance area. This embodiment can intercept the video of the rat in the surveillance area from the video indicated by the video data. Obtain the first video frame of the rat-infested video, identify the position of the rat in the surveillance area from the first video frame, and determine the position as the rat’s intrusion point in the surveillance area, which can be used as The rat invaded the entrance of the indoor place.
  • This embodiment can also obtain the last video frame from the rat-infested video, identify the position of the rat in the surveillance area from the last video frame, and determine the position as the rat’s hiding spot, which can be used as the rat den , Or the exit when escaping indoors.
  • this embodiment can record the intrusion points and hiding points of the pests in the past period of time, and send information indicating the intrusion points and hiding points of the pests to the terminal to prompt the person in charge of pest control to treat the pests Take further measures for the prevention and control of the country.
  • the pest is a mouse
  • this embodiment can also be used to determine the density of pests in the monitoring area, and the density of pests in different monitoring areas can be determined, and the length of time the pests infested in the monitoring area can be accounted for by the pests.
  • the time ratio of the entire monitoring cycle of monitoring is determined as the density of pests in the monitoring area. For example, if the pest is a mouse, the density of rats in the surveillance area captured by the video surveillance equipment can be calculated, and the ratio of the time of the rat in the surveillance area to the time of the entire monitoring cycle of the surveillance of rats is determined as the surveillance area Rat density.
  • the monitoring area with the highest density of pests is determined therefrom, and it is determined as an area with frequent pest invasion, which can be used to indicate pest invasion Frequent area information is sent to the terminal to prompt relevant personnel to take further measures.
  • the pest is a rat.
  • determine the monitoring area with the highest density of rats that is, record the places with high rat density values in the room, and determine it as frequent rat invasions It also sends information indicating places with frequent rat invasions to the terminal to prompt restaurant operators to further check whether there are residual food residues, uncleaned water marks, etc., making it a breeding place for rats and insect pests. If there are residual food residues, uncleaned water marks, etc., instruct the restaurant operators to further do the sanitation and cleaning work in the place to reduce the attraction of the place to harmful organisms.
  • a mouse disease video monitoring device may include several components: an infrared low-light night vision camera, a data processing module, and a front-end display component.
  • the working principle of the above device is as follows: the infrared low-light night vision camera is responsible for Collect the scene video sequence, the data processing module receives the video sequence and detects whether there is a mouse in the video. If a mouse is detected, a series of information such as the position of the mouse is output to the front-end display interface.
  • the front-end display interface displays the mouse's position, appearance time, and activity Area and can immediately alarm for rat infestation.
  • Fig. 7 is a schematic diagram of a data processing module according to an embodiment of the present disclosure.
  • the data processing module includes: a video acquisition module 702, a video processing module 704, and a storage module 706.
  • the video acquisition module 702 includes an ARM board 7022 and a video preprocessing module 7024.
  • the video processing module 704 includes: Embedded GPU processor 7042.
  • the video acquisition module 702 collects video data through the ARM board 7022 and performs preprocessing.
  • the video processing module 704 reads in the trained model and performs video processing in the embedded GPU processor 7042 according to the deep learning algorithm. If the deep learning network detects a certain When there is a mouse in a segment, the segment and the corresponding detection result are stored in the storage module 706, and the storage module 706 outputs the series of information to the front end.
  • each frame of the pre-processed video sequence is detected, and image features (such as the visual information in the corresponding detection frame) are acquired at the position where rats may exist, and the motion feature extraction module is used to extract each The information between the video image frames is fused and feature extracted to prevent the target detector of a single frame from being misjudged. Then the extracted motion features and image features are input into the classification network, and the classification network determines whether it is a mouse. If it is a mouse, Then the rectangular detection frame of the mouse at each frame is transmitted to the front-end display interface.
  • image features such as the visual information in the corresponding detection frame
  • the dynamic target detection algorithm includes background difference and frame difference methods, using the following formula (1) to calculate the difference between the current frame and the background or the previous frame:
  • (x, y) is used to indicate the coordinates of the pixel in the coordinate system established with the upper left corner of the image as the origin, the width direction as the X axis, and the height direction as the Y axis.
  • K is the index of the current frame
  • f represents the current Frame
  • b represents the background or the previous frame.
  • RPN is used to represent the region proposal network, and a series of candidate frames will be proposed.
  • the pooling layer maps the area of the feature map mentioned by the convolutional layer under the coordinates of the RPN output into a fixed size (w, h)
  • the rectangular frame is sent to the classifier and frame regression composed of fully connected layers.
  • the frame regression outputs the possible coordinate position of the mouse, and the output of the classifier is the confidence level of the mouse at that position.
  • the motion feature extraction algorithm first calculates the correlation of the detection frame between frames according to the detection frame obtained in each frame, and the detection frame with a large correlation is considered the same object , Match the detection frame of each frame to obtain a series of moving pictures of the object, and finally use the 3D feature extraction network to extract the features of the motion sequence.
  • the first acquiring unit 10 is configured to acquire a first image obtained by photographing the monitored area.
  • the second determining unit 50 is configured to determine a target area that intersects the movement track in a preset area, where the target area is used to place the target capture device, and the target capture device is used to capture the target object.
  • the preset area indicated by the area setting instruction is determined on the first image obtained by shooting the monitoring area, and the first image obtained by shooting the monitoring area within the first target time period is determined.
  • the group of image data determines the movement trajectory of the target object in the first target time period in the monitoring area, and then determines the target area that intersects the movement trajectory in the preset area, avoiding relying on the professional experience and subjectivity of the prevention and control personnel.
  • the judgment to determine the area where the capture tool is placed solves the technical problem of low accuracy in determining the area where the capture device is placed, and achieves the technical effect of improving the accuracy of determining the area where the capture device is placed.
  • a computer program is stored in the aforementioned storage medium, where the computer program can be used to execute the area determination method when the computer program is set to run.
  • the preset area is one or more preset areas
  • the program code when the computer program is executed by the processor, the program code further implements the following steps:
  • a target area that intersects the movement track is determined in one or more areas, where the movement track that intersects the target area meets a preset target condition, and the target condition is determined according to at least one of the following: Movement that intersects the target area The number of trajectories, the length of the intersection of the movement trajectory and the target area, and the number of movement trajectories that intersect the target area and have an intersection point in the target area.
  • a second area that intersects the movement track is determined in one or more areas, where the target area includes the second area, the number of movement tracks intersecting the second area is greater than the second threshold, and the movement track intersects the second area.
  • the length of the part is greater than the third threshold
  • a third area that intersects with the movement trajectory is determined in one or more areas, where the target area includes a third area, and the number of movement trajectories that intersect with the third area and have intersections in the third area is greater than the fourth threshold .
  • the storage medium may also be set to determine the program code of various preferred or optional method steps provided by the area determination method.
  • the computer-readable storage medium may include a data signal propagated in baseband or as a part of a carrier wave, and readable program code is carried therein.
  • This propagated data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing.
  • the computer-readable storage medium can send, propagate, or transmit the program for use by or in combination with the instruction execution system, apparatus, or device.
  • the program code contained in the computer-readable storage medium can be transmitted by any suitable medium, including but not limited to wireless, wired, optical cable, radio frequency, etc., or any suitable combination of the foregoing.
  • an embodiment of the present disclosure further provides a processor.
  • Fig. 12 is a schematic structural diagram of a processor according to an embodiment of the present disclosure. As shown in FIG. 12, the processor 120 is used to run a program, where the method for determining a region in any one of the embodiments of the present disclosure is executed when the program is running.
  • the aforementioned processor 120 may execute an operating program of the area determination method.
  • the processor 120 may be configured to perform the following steps:
  • a target area that intersects with the movement track is determined in the preset area, where the target area is used to place the target capture device, and the target capture device is used to capture the target object.
  • the preset area is one or more preset areas
  • the processor 120 may also be configured to perform the following steps:
  • a target area that intersects the movement track is determined in one or more areas, where the movement track that intersects the target area meets a preset target condition, and the target condition is determined according to at least one of the following: Movement that intersects the target area The number of trajectories, the length of the intersection of the movement trajectory and the target area, and the number of movement trajectories that intersect the target area and have an intersection point in the target area.
  • the processor 120 may also be configured to perform the following steps:
  • a second area that intersects the movement track is determined in one or more areas, where the target area includes the second area, the number of movement tracks intersecting the second area is greater than the second threshold, and the movement track intersects the second area.
  • the length of the part is greater than the third threshold
  • a third area that intersects with the movement trajectory is determined in one or more areas, where the target area includes a third area, and the number of movement trajectories that intersect with the third area and have intersections in the third area is greater than the fourth threshold .
  • the above-mentioned processor 120 may execute various functional applications and data processing by running software programs and modules stored in the memory, that is, realize the above-mentioned area determination method.
  • the storage medium may include: flash disk, read-only memory (Read-Only Memory, ROM for short), random access memory (Random Access Memory, RAM for short), magnetic disk or optical disk, etc.
  • modules or steps of the present disclosure can be implemented by a general computing device, and they can be concentrated on a single computing device or distributed in a network composed of multiple computing devices.
  • they can be implemented with program codes executable by a computing device, so that they can be stored in a storage device for execution by the computing device, or they can be made into individual integrated circuit modules, or they can be Multiple modules or steps are made into a single integrated circuit module to achieve.
  • the present disclosure is not limited to any specific hardware and software combination.
  • the first image obtained by photographing the monitored area determines the preset area indicated by the area setting instruction on the first image in response to the input area setting instruction; obtain the obtained by photographing the monitored area within the first target time period
  • the first set of image data according to the first set of image data, determine the movement track of the target object in the first target time period in the monitoring area; determine the target area that intersects the movement track in the preset area, where ,
  • the target area is used to place the target capture device, and the target capture device is used to capture the target object.
  • the preset area indicated by the area setting instruction is determined on the first image obtained by shooting the monitored area, and the preset area indicated by the area setting instruction is determined according to the first image obtained by shooting the monitored area within the first target time period.
  • a set of image data determines the movement trajectory of the target object in the first target time period in the monitoring area, and then determines the target area that intersects the movement trajectory in the preset area, avoiding relying on the professional experience and
  • the subjective judgment to determine the area where the capture tool is placed solves the technical problem of low accuracy in determining the area where the capture device is placed, and achieves the technical effect of improving the accuracy of determining the area where the capture device is placed.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)
  • Catching Or Destruction (AREA)
  • Closed-Circuit Television Systems (AREA)

Abstract

本公开实施例公开了一种区域确定方法、装置、存储介质和处理器。该方法包括：获取对监控区域进行拍摄得到的第一图像；响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域；获取在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据；根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹；在预设区域中确定出与移动轨迹相交的目标区域，其中，目标区域用于放置目标捕捉装置，目标捕捉装置用于捕捉目标对象。

Description

区域确定方法、装置、存储介质和处理器
本申请要求于2019年01月24日提交中国专利局、优先权号为201910068842.3、发明名称为“区域确定方法、装置、存储介质和处理器”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
本公开实施例涉及虫鼠害防治领域，具体而言，涉及一种区域确定方法、装置、存储介质和处理器。
背景技术
目前，在对目标对象进行防治时，主要依赖于防治人员的从业经验和主观判断，通过片面的线索，来确定在防治区域中放置用于对目标对象进行捕捉的捕捉工具的区域。比如，目标对象为老鼠，捕捉工具为粘鼠板，防治人员通过观察老鼠的排泄物、啃咬的痕迹、出没的踪迹，依据经验，沿墙边、房间角落、电线旁等老鼠可能经过的场所的位置来部署粘鼠板，在同一个地方可能持续观察若干天，如果没有收获，则再更换粘鼠板的放置位置，以此循环。
但是，当防治人员经验不足，或者防治人员与防治区域的管理人员沟通不充分时，就难以保证捕捉工具在防治区域放置的区域的准确性。比如，防治人员对餐厅建筑环境没有进行全面地了解，对下水道和墙洞、屋顶空隙等老鼠能钻过的通道可能漏放粘鼠板，则有老鼠再次侵入的可能，即使在运气好的情况下抓到了老鼠，也难以保证是否还有存留的其它老鼠。
针对现有技术中对用于放置捕捉装置的区域进行确定的准确性低的问题，目前尚未提出有效的解决方案。
发明内容
本公开至少部分实施例提供了一种区域确定方法、装置、存储介质和处理器，以至少解决对用于放置捕捉装置的区域进行确定的准确性低的技术问题。
为了实现上述目的，根据本公开的一个方面，提供了一种区域确定方法。该方法包括：获取对监控区域进行拍摄得到的第一图像；响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域；获取在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据；根据第一组图像数据在监控区域中确定出目标对象在 第一目标时间段内的移动轨迹；在预设区域中确定出与移动轨迹相交的目标区域，其中，目标区域用于放置目标捕捉装置，目标捕捉装置用于捕捉目标对象。
可选地，预设区域为预设的一个或多个区域，在预设区域中确定出与移动轨迹相交的目标区域包括：在一个或多个区域中确定出与移动轨迹相交的目标区域，其中，与目标区域相交的移动轨迹满足预设的目标条件，其中，目标条件根据以下至少之一确定：与目标区域相交的移动轨迹的数量，移动轨迹与目标区域相交的长度，与目标区域相交且在目标区域中存在交叉点的移动轨迹的数量。
可选地，在一个或多个区域中确定出与移动轨迹相交的目标区域以下至少之一：在一个或多个区域中确定出与移动轨迹相交的第一区域，其中，目标区域包括第一区域，与第一区域相交的移动轨迹的数量大于第一阈值；在一个或多个区域中确定出与移动轨迹相交的第二区域，其中，目标区域包括第二区域，与第二区域相交的移动轨迹的数量大于第二阈值，移动轨迹与第二区域相交的部分的长度大于第三阈值；在一个或多个区域中确定出与移动轨迹相交的第三区域，其中，目标区域包括第三区域，与第三区域相交、在第三区域中存在交叉点的移动轨迹的数量大于第四阈值。
可选地，在预设区域中确定出与移动轨迹相交的目标区域之后，该方法还包括以下至少之一：在目标区域包括第一区域的情况下，将用于放置目标捕捉装置的目标位置设置为包括移动轨迹与第一区域相交的部分上的一个或多个位置；在目标区域包括第一区域的情况下，将用于放置目标捕捉装置的目标位置设置为第一位置，其中，位于第一位置的目标捕捉装置覆盖第一区域中至少预定数量的移动轨迹；在目标区域包括第二区域的情况下，将用于放置目标捕捉装置的目标位置设置为包括移动轨迹与第二区域相交的部分上的一个或多个位置；在目标区域包括第三区域的情况下，将用于放置目标捕捉装置的目标位置设置为一个或多个交叉点所在的位置。
可选地，在预设区域中确定出与移动轨迹相交的目标区域之后，该方法还包括：在第一图像上显示用于标识目标区域的第一标识信息。
可选地，在第一图像上显示用于标识预设区域的第一标识信息包括：在第一图像上显示用于标识目标区域的范围的第一标识信息。
可选地，预设区域为预设的一个或多个区域，在响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域之后，该方法还包括：在第一图像上分别显示用于标识一个或多个区域的第二标识信息。
可选地，在预设区域中确定出与移动轨迹相交的目标区域包括：在一个或多个区域中确定出与移动轨迹相交的目标区域；在一个或多个区域中确定出与移动轨迹相交的目标区域之后，该方法还包括：保留目标区域的第二标识信息，隐藏一个或多个区 域中除目标区域之外的区域的第二标识信息；和/或，显示用于标识目标区域的第一标识信息。
可选地，在根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹之后，该方法还包括：在第一图像上显示用于标识移动轨迹的第三标识信息。
可选地，响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域包括：在第一图像中识别出监控区域中的目标类型区域，其中，目标对象经过目标类型区域的概率大于第五阈值；响应输入的区域设置指令在目标类型区域上确定出区域设置指令所指示的预设区域。
可选地，根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹包括：从第一组图像数据中识别出目标对象在监控区域中经过的多个位置；通过多个位置生成移动轨迹，其中，多个位置位于移动轨迹上。
可选地，在预设区域中确定出与移动轨迹相交的目标区域之后，该方法还包括：在目标捕捉装置放置在目标区域的情况下，将由目标捕捉装置捕捉到的目标对象的目标信息发送至目标终端；和/或将目标区域发送至目标终端；或者将目标区域和监控区域的第二图像发送至目标终端；或者在目标终端上显示监控区域的第三图像，其中，第三图像上显示有目标区域；或者在目标终端上显示监控区域的第四图像，其中，第四图像上显示有移动轨迹和目标区域。
可选地，获取在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据包括：获取摄像设备对监控区域拍摄得到的视频文件；对视频文件进行抽帧采样，得到一组视频帧图像的数据，其中，第一组图像数据包括一组视频帧图像的数据；在根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹之前，该方法还包括：根据一组视频帧图像中的像素点的像素值在一组视频帧图像中确定出多个目标视频帧图像，其中，每个目标视频帧图像用于指示在监控区域中存在运动的对象；对每个目标视频帧图像进行目标对象检测，得到每个目标视频帧图像的图像特征，其中，图像特征用于表示在存在运动的对象中，与目标对象之间的相似度大于第六阈值的对象所在的目标图像区域；根据每个目标视频帧图像的图像特征确定出运动特征，其中，运动特征用于表示多个目标视频帧图像中存在运动的对象的运动速度和运动方向；根据运动特征和每个目标视频帧图像的图像特征，确定多个目标视频帧图像中是否出现有目标对象。
为了实现上述目的，根据本公开的另一方面，提供了一种区域确定装置，包括一个或多个处理器，以及一个或多个存储程序单元的存储器，其中，程序单元由处理器 执行，所述程序单元包括：第一获取单元，设置为获取对监控区域进行拍摄得到的第一图像；响应单元，设置为响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域；第二获取单元，设置为获取在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据；第一确定单元，设置为根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹；第二确定单元，设置为在预设区域中确定出与移动轨迹相交的目标区域，其中，目标区域用于放置目标捕捉装置，目标捕捉装置用于捕捉目标对象。
通过本公开，采用获取对监控区域进行拍摄得到的第一图像；响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域；获取在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据；根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹；在预设区域中确定出与移动轨迹相交的目标区域，其中，目标区域用于放置目标捕捉装置，目标捕捉装置用于捕捉目标对象。也就是说，响应输入的区域设置指令在对监控区域进行拍摄得到的第一图像上确定出区域设置指令所指示的预设区域，根据在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹，进而在预设区域中确定出与移动轨迹相交的目标区域，避免了依赖于防治人员的从业经验和主观判断来确定放置捕捉工具的区域，解决了对用于放置捕捉装置的区域进行确定的准确性低的技术问题，达到了提高对用于放置捕捉装置的区域进行确定的准确性的技术效果。
附图说明
构成本申请的一部分的附图用来提供对本公开实施例的进一步理解，本公开的示意性实施例及其说明用于解释本公开，并不构成对本公开的不当限定。在附图中：
图1是根据本公开实施例的一种区域确定方法的流程图；
图2是根据本公开实施例的一种确定灭鼠器的放置位置的方法的流程图；
图3是根据本公开实施例的一种识别出的可放置捕鼠装置区域的示意图；
图4是根据本公开实施例的一种识别出的鼠迹的示意图；
图5是根据本公开实施例的一种捕鼠装置的放置区域的示意图；
图6是根据本公开实施例的一种鼠迹报告的直方图；
图7是根据本公开实施例的一种数据处理模块的示意图；
图8是根据本公开实施例的一种鼠患检测系统的原理示意图；
图9是本公开实施例的一种Faster-RCNN网络模型的示意图；
图10是根据本公开实施例的一种区域确定装置的示意图；
图11是根据本公开实施例的一种存储介质的结构示意图；以及
图12是根据本公开实施例的一种处理器的结构示意图。
具体实施方式
需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互组合。下面将参考附图并结合实施例来详细说明本公开。
为了使本技术领域的人员更好地理解本申请实施例的方案，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的实施例，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，都应当属于本申请保护的范围。
需要说明的是，本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便这里描述的本申请的实施例。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
本公开实施例提供了一种区域确定方法。
图1是根据本公开实施例的一种区域确定方法的流程图。如图1所示，该方法可以包括以下步骤：
步骤S102，获取对监控区域进行拍摄得到的第一图像。
在本申请上述步骤S102提供的技术方案中，可以在餐饮场景、工厂场景等需要对目标对象进行防治的场景下，获取对监控区域进行拍摄得到的第一图像。其中，餐饮场景可以为对饮食卫生要求较高的场景，可以为公共运营场所中的餐饮场景，也可以为居家生活中的餐饮场景，此处不做任何限制；工厂场景可以为食品或药品加生产厂房、食品或药品储存间等对卫生要求较高的场景，此处不做任何限制。其中，目标对象可以为体型较大的病媒生物，比如，目标对象为老鼠，也可以为体型较小的病媒生物，比如，为蟑螂。
该实施例的第一图像可以通过视频监控设备对监控区域进行拍摄得到，比如，该视频监控设备为摄像头，其数量可以为多个，分别设置在不同的监控区域中，该监控区域为待放置用于捕捉目标对象的目标捕捉装置的区域，可以为静态区域，也即，该监控区域中不存在其它动态的干扰因素，以方便进行观察。可选地，该实施例的监控区域可以为预先设定的餐饮场景中的某一可视区域，该监控区域可以为目标对象活动频繁且需要提高卫生质量的区域，比如，为餐厅、厨房、烧烤间、水果台等食品操作区，该监控区域还可以为预先设定的工厂场景中的某一可视区域，比如，该监控区域为食品粗加工制作区、食品库房等关键区域，此处不做任何限制。
可选地，上述摄像头可以包括但不限于：带有红外照明功能的摄像头，例如，红外微光夜视摄像头。进一步，该摄像头还可以包括但不限于：移动侦测功能、存储功能、联网功能(如wifi联网)及高清晰度(如大于1080p)配置。
步骤S104，响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域。
在本申请上述步骤S104提供的技术方案中，在获取对监控区域进行拍摄得到的第一图像之后，响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域。
在该实施例中，区域设置指令用于在第一图像上确定出监控区域中的预设区域，可以由用户通过终端输入，比如，按照用户手指或者鼠标在终端屏幕上的滑动轨迹来触发区域设置指令，该区域设置指令用于指示预设区域，该预设区域为预设的用于放置目标捕捉工具的合适区域，比如，该预设区域为由用户手指或者鼠标在终端屏幕上的滑动轨迹所形成的区域在第一图像上所对应的区域，其数量可以为多个。在输入区域设置指令之后，响应该区域设置指令，进而在第一图像上确定出由区域设置指令所指示的预设区域。其中，用户可以为虫害防治人员、餐厅运营人员等相关人员。
可选地，该实施例的区域设置指令可以由用户根据经验输入，可以根据目标对象的属性、活动规律来在第一图像上确定出监控区域中的预设区域。比如，目标对象为老鼠，目标捕捉工具为粘鼠板，由于老鼠具有善于攀爬、钻洞的特性，其在活动时很有可能是沿着电线、水管等上下移动，另外，能够部署粘鼠板的地方可以为地面、墙角、窗台等平整的位置，因而该实施例可以在视频监控设备所对准的监控区域中，预先设置若干个适合的预设区域，可以在墙角、电线旁等鼠类常规行进的必经之路上优先确定预设区域。
可选地，该实施例在每个视频监控设备安装完毕之后，就可以响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域，并将用于指示该预设区 域的信息存储在服务器中。
步骤S106，获取在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据。
在本申请上述步骤S106提供的技术方案中，第一目标时间段可以为预先设定的目标对象周期过去一段时间出没的时间段，可以为前一天，或者为前几天，第一组图像数据用于指示监控区域在第一目标时间段内的图像，可以为视频数据，也可以为图片数据，可以包括对监控区域进行拍摄的时间。在第一组图像数据为视频数据的情况下，第一组图像数据可以用于指示监控区域在第一目标时间段内的连续视频图像，在第一组图像数据为图片数据的情况下，第一组图像数据可以用于指示目标区域在第一目标时间内的包括多张图片的图片集。
可选地，该实施例通过视频监控设备获取在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据。
步骤S108，根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹。
在本申请上述步骤S108提供的技术方案中，在获取在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据之后，根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹。
从第一组图像数据中确定出目标对象在监控区域中在第一目标时间段内的移动轨迹，该移动轨迹也即历史移动轨迹，可以由目标对象在监控区域中在第一目标时间段内经过的位置点形成，该移动轨迹可以指示出目标对象在监控区域中的移动方向。
可选地，该实施例可以从第一图像数据所指示的视频中提取出有目标对象的图像的视频片段，再通过动作识别技术识别出视频片段中动态的变化特征，通过图像识别技术对动态的变化特征进行进一步识别，比如，通过人工智能(AI)图像识别技术对视频片段中动态的变化特征进行识别，进一步确定出现的生物确实为目标对象，进而结合动态的变化特征确定出目标对象的移动轨迹。
举例而言，该实施例的目标对象为老鼠，获取前一天的监控区域中有老鼠出没的目标视频，可以对该目标视频按照时间段进行截取，从中提取出包括有老鼠图像的视频片段。再通过AI图像识别技术对视频片段中动态的变化特征进行识别，进一步确定监控区域中出现的目标对象确实为老鼠，进而结合动态的变化特征确定出老鼠的移动轨迹，并在监控区域的图像中表示出，比如，通过绿色线条指示出老鼠在监控区域中的移动轨迹。
可选地，该实施例除了根据第一组图像数据可以确定目标对象的移动轨迹之外，还可以根据第一组图像数据识别出目标对象的种类、目标对象的皮肤颜色、目标对象的数量、目标对象的形态、目标对象在第一目标时间段内在监控区域的移动时长等与目标对象相关的信息。比如，目标对象为老鼠，除了根据第一图像数据识别出老鼠的移动轨迹之外，还可以识别出老鼠的数量、老鼠的皮肤颜色、老鼠的形态、老鼠的移动时长等信息，还可以进一步识别出老鼠的品种，比如，小家鼠、褐家鼠、黄胸鼠等，其中，小家鼠筑巢多在墙基、库房货堆中和保温层内打洞或在破纸箱、抽屉中，也可能会随货物一起进入室内，褐家鼠警觉性强，可以通过下水道、马桶等侵入室内，黄胸鼠可以沿粗糙墙直上直下攀爬，沿铁丝、电线行走，可以通过管线孔洞，天花板入侵室内。
步骤S110，在预设区域中确定出与移动轨迹相交的目标区域。
在本申请上述步骤S110提供的技术方案中，在响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域，且根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹之后，在预设区域中确定出与移动轨迹相交的目标区域，其中，目标区域用于放置目标捕捉装置，目标捕捉装置用于捕捉目标对象。
在该实施例中，可以将预设区域和移动轨迹进行比较，可选地，该预设区域为多个区域，将多个区域中与移动轨迹相交的目标区域确定为用于放置目标捕捉装置的区域，也即，该目标区域是由预设区域和目标对象实际的移动轨迹综合确定得到的，将其确定为用户需要将目标捕捉工具在监控区域中放置的区域，从而兼顾用户的经验和目标对象的实际出没信息综合确定得到目标区域，从而避免仅依赖于防治人员的从业经验和主观判断来确定放置捕捉工具的区域，提高了对用于放置捕捉装置的区域进行确定的准确性。
通过本申请上述步骤S102至步骤S110，采用获取对监控区域进行拍摄得到的第一图像；响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域；获取在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据；根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹；在预设区域中确定出与移动轨迹相交的目标区域，其中，目标区域用于放置目标捕捉装置，目标捕捉装置用于捕捉目标对象。也就是说，响应输入的区域设置指令在对监控区域进行拍摄得到的第一图像上确定出区域设置指令所指示的预设区域，根据在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹，进而在预设区域中确定出与移动轨迹相交的目标区域，避免了依赖于防治人员的从业经验和主观判断来确定放置捕捉工具的区域，解决了对 用于放置捕捉装置的区域进行确定的准确性低的技术问题，达到了提高对用于放置捕捉装置的区域进行确定的准确性的技术效果。
作为一种可选的实施方式，预设区域为预设的一个或多个区域，步骤S110，在预设区域中确定出与移动轨迹相交的目标区域包括：在一个或多个区域中确定出与移动轨迹相交的目标区域，其中，与目标区域相交的移动轨迹满足预设的目标条件，其中，目标条件根据以下至少之一确定：与目标区域相交的移动轨迹的数量，移动轨迹与目标区域相交的长度，与目标区域相交且在目标区域中存在交叉点的移动轨迹的数量。
在该实施例中，预设区域为响应输入的区域设置指令在第一图像上确定出的区域，可以为多个区域，比如，为地面、墙角、窗台等平整的区域，此处不做任何限制。在一个或多个区域中确定出与移动轨迹相交的目标区域，该目标区域也可以为多个，比如，预设区域包括a区域、b根据、c区域、d区域等，在a区域、b区域、c区域、d区域中确定出与移动轨迹相交的a区域、b区域，将a区域、b区域确定为目标区域。
该实施例的与目标区域相交的移动轨迹满足预设的目标条件，也即，不是预设区域中与移动轨迹相交的任何区域都可以作为目标区域，而是与其相交的移动轨迹需要满足目标条件，以进一步提高对用于放置捕捉装置的区域进行确定的准确性。该目标条件可以根据与预设区域相交的移动轨迹的数量、移动轨迹的长度、存在交叉点的移动轨迹的数量进行确定，比如，根据与目标区域相交的移动轨迹的数量、移动轨迹与目标区域相交的长度、与目标区域相交且在目标区域中存在交叉点的移动轨迹的数量中的至少之一来确定目标条件，从而进一步提高对用于放置捕捉装置的区域进行确定的准确性。
可选地，在一个或多个区域中除目标区域之外的其它区域中的移动轨迹不满足上述目标条件。
在在一个或多个区域中确定出与移动轨迹相交的目标区域之后，目标区域中放置目标捕捉装置，进而提高了对目标对象进行捕捉的效率。
下面对该实施例的在一个或多个区域中确定出与移动轨迹相交的目标区域进行介绍。
作为一种可选的实施方式，在一个或多个区域中确定出与移动轨迹相交的目标区域以下至少之一：在一个或多个区域中确定出与移动轨迹相交的第一区域，其中，目标区域包括第一区域，与第一区域相交的移动轨迹的数量大于第一阈值；在一个或多个区域中确定出与移动轨迹相交的第二区域，其中，目标区域包括第二区域，与第二区域相交的移动轨迹的数量大于第二阈值，移动轨迹与第二区域相交的部分的长度大于第三阈值；在一个或多个区域中确定出与移动轨迹相交的第三区域，其中，目标区 域包括第三区域，与第三区域相交、在第三区域中存在交叉点的移动轨迹的数量大于第四阈值。
在该实施例中，在一个或多个区域中确定出与移动轨迹相交的目标区域的方法可以包括多种。与目标区域相交的移动轨迹的数量可以不同，该实施例的目标条件可以基于移动轨迹的数量来进行设置，其中，将与目标区域相交的移动轨迹的数量设置为第一阈值，比如，该第一阈值为5条，则该目标条件可以为与目标区域相交的移动轨迹的数量大于第一阈值的条件。通过该目标条件，从一个或多个区域中确定出与移动轨迹相交的第一区域，其中，上述的目标区域包括该第一区域，与该第一区域相交的移动轨迹的数量大于第一阈值，进而在第一区域中放置目标捕捉装置，从而提高了对目标对象进行捕捉的效率。
需要说明的是，上述第一阈值可以根据实际应用场景进行设置，只要可以提高对目标捕捉装置的区域进行确定的准确性的第一阈值，都在本公开实施例的范围之内，此处不再一一举例说明。
可选地，该实施例与目标区域相交的移动轨迹的数量，以及与目标区域相交的部分的移动轨迹的长度不同。该实施例的目标条件可以基于与目标区域相交的移动轨迹的数量，以及与目标区域相交的部分的移动轨迹的长度进行设置，其中，将与目标区域相交的移动轨迹的数量设置为第二阈值，比如，该第二阈值为5条，将与目标区域相交的部分的移动轨迹的长度设置为第三阈值，比如，该第三阈值为0.5米，则该目标条件可以为与目标区域相交的移动轨迹的数量大于第二阈值，移动轨迹与目标区域相交的部分的移动轨迹的长度大于第三阈值的条件。通过该目标条件，在一个或多个区域中确定出与移动轨迹相交的第二区域，其中，上述目标区域包括该第二区域，与该第二区域相交的移动轨迹的数量大于第二阈值，移动轨迹与该第二区域相交的部分的长度大于第三阈值，进而在第二区域中放置目标捕捉装置，从而提高了对目标对象进行捕捉的效率。
需要说明的是，上述第二阈值、第三阈值可以根据实际应用场景进行设置，只要可以提高对目标捕捉装置的位置进行确定的准确性的第二阈值、第三阈值，都在本公开实施例的范围之内，此处不再一一举例说明。
可选地，该实施例与目标区域相交、在目标区域中存在交叉点的移动轨迹的数量不同。该实施例的目标条件可以基于与目标区域相交、在目标区域中存在交叉点的移动轨迹的数量进行设置，其中，将与目标区域相交、在目标区域中存在交叉点的移动轨迹的数量设置为第四阈值，比如，该第四阈值为6条，则该目标条件可以为与目标区域相交、在目标区域中存在交叉点的移动轨迹的数量大于第四阈值的条件。通过该目标条件，在一个或多个区域中确定出与移动轨迹相交的第三区域，其中，上述目标 区域包括该第三区域，与该第三区域相交、在第三区域中存在交叉点的移动轨迹的数量大于第四阈值，进而在第三区域中放置目标捕捉装置，从而提高了对目标对象进行捕捉的效率。
需要说明的是，上述第四阈值可以根据实际应用场景进行设置，只要可以提高对目标捕捉装置的位置进行确定的准确性的第四阈值，都在本公开实施例的范围之内，此处不再一一举例说明。
下面对该实施例的在目标区域中确定用于放置目标捕捉装置的目标位置进行介绍。
该实施例可以在监控区域中确定在第一目标时间段内有目标对象出现的目标区域，进而根据目标对象在目标时间段内的移动轨迹和对应的移动时长，在目标区域中确定捕捉装置的目标位置，达到了对捕捉工具的位置进行确定的目的。
作为一种可选的实施方式，在预设区域中确定出与移动轨迹相交的目标区域之后，该方法还包括以下至少之一：在目标区域包括第一区域的情况下，将用于放置目标捕捉装置的目标位置设置为包括移动轨迹与第一区域相交的部分上的一个或多个位置；在目标区域包括第一区域的情况下，将用于放置目标捕捉装置的目标位置设置为第一位置，其中，位于第一位置的目标捕捉装置覆盖第一区域中至少预定数量的移动轨迹；在目标区域包括第二区域的情况下，将用于放置目标捕捉装置的目标位置设置为包括移动轨迹与第二区域相交的部分上的一个或多个位置；在目标区域包括第三区域的情况下，将用于放置目标捕捉装置的目标位置设置为一个或多个交叉点所在的位置。
在该实施例中，在预设区域中确定出与移动轨迹相交的目标区域之后，可以在目标区域中确定用于放置目标捕捉装置的目标位置。该实施例的与第一区域相交的移动轨迹的数量大于第一阈值。可选地，在目标区域包括第一区域的情况下，可以根据移动轨迹与第一区域相交的部分来确定用于放置目标捕捉装置的目标位置，可以将目标位置设置为包括移动轨迹与第一区域相交的部分上的一个或多个位置，从而提高了对目标捕捉装置的位置进行确定的准确性。
可选地，该实施例基于目标捕捉装置可以覆盖的移动轨迹的数量来确定最终用于放置目标捕捉装置的目标位置。在目标区域包括第一区域的情况下，将目标捕捉装置覆盖第一区域中至少预定数量的移动轨迹的第一位置，确定为最终的目标位置，可以在第一区域中的多条移动轨迹较为密集的位置上确定目标位置，从而提高了对目标捕捉装置的位置进行确定的准确性。
该实施例的与第二区域相交的移动轨迹的数量大于第二阈值，移动轨迹与第二区域相交的部分的长度大于第三阈值。可选地，在目标区域包括第二区域的情况下，可 以确定包括移动轨迹与第二区域相交的部分，确定包括移动轨迹与第二区域相交的部分上的一个或多个位置，进而将用于放置目标捕捉装置的目标位置设置为包括移动轨迹与第二区域相交的部分上的一个或多个位置，从而提高了对目标捕捉装置的位置进行确定的准确性。
该实施例的与第三区域相交、在第三区域中存在交叉点的移动轨迹的数量大于第四阈值。可选地，在目标区域包括第三区域的情况下，确定第三区域中存在的一个或多个交叉点所在的位置，进而将用于放置目标捕捉装置的目标位置设置为一个或多个交叉点所在的位置，从而提高了对目标捕捉装置的位置进行确定的准确性。
可选地，该实施例在将用于放置目标捕捉装置的目标位置设置为一个或多个交叉点所在的位置时，可以从多个交叉点中，选择出由大于目标数量的移动轨迹所形成的目标交叉点，进而将目标交叉点在第三区域中所在的位置，确定为目标位置。
举例而言，与移动轨迹相交的第三区域中存在交叉点的移动轨迹的数量包括A移动轨迹、B移动轨迹、C移动轨迹、D移动轨迹。可以获取A移动轨迹、B移动轨迹、C移动轨迹、D移动轨迹形成的a交点、b交点、c交点、d交点，其中，a交点可以由A移动轨迹、B移动轨迹形成，b交点可以由A移动轨迹、B移动轨迹、C移动轨迹形成，c交点可以由C移动轨迹、D移动轨迹形成，d交点可以由A移动轨迹、D移动轨迹形成。从多个交叉点中，选择出由大于目标数量的移动轨迹所形成的目标交叉点，该目标交叉点为目标对象多次经过的移动轨迹上的点，可以从a交点、b交点、c交点、d交点中选择出由大于2的移动轨迹所形成的目标交叉点b，可以将目标交叉点b在第三区域中所对应的位置，确定为用于放置目标捕捉装置的目标位置，从而进一步提高了对目标捕捉装置的位置进行确定的准确性。
可选地，如果该实施例的用于放置目标捕捉工具的位置包括多个目标位置，则可以对多个目标位置在实际放置目标捕捉工具时设置优先级顺序，比如，将容易放置目标捕捉工具的目标位置通过目标指示信息标识出来，以指示可以优先放置目标捕捉工具。
作为一种可选的实施方式，在预设区域中确定出与移动轨迹相交的目标区域之后，该方法还包括：在第一图像上显示用于标识目标区域的第一标识信息。
在该实施例中，在预设区域中确定出与移动轨迹相交的目标区域之后，可以对目标区域进行标识。可以在第一图像上的目标区域中，通过第一标识信息将目标区域标识出来，该第一标识信息可以为图形、文字、符号等醒目的标记，比如，第一标识信息为红色的圆圈，通过红色的圆圈将目标区域在第一图像中圈出来，以向目标用户提示目标区域在整个监控区域中的位置，进而指示目标用户将目标捕捉工具部署在目标 区域上，从而提高对目标对象进行捕捉的效率。
作为一种可选的实施方式，在第一图像上显示用于标识预设区域的第一标识信息包括：在第一图像上显示用于标识目标区域的范围的第一标识信息。
在该实施例中，第一标识信息可以用于标识目标区域在监控区域中的范围大小，以指示目标捕捉工具在监控区域中可以进行放置的范围，比如，第一标识信息为红色圆圈，该红色圆圈的大小可以用于指示目标区域在监控区域中的范围的大小。
作为一种可选的实施方式，预设区域为预设的一个或多个区域，在响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域之后，该方法还包括：在第一图像上分别显示用于标识一个或多个区域的第二标识信息。
在该实施例中，预设区域为预设的一个或多个区域，可以在墙角、电线旁等鼠类常规行进的必经之路上确定的一个或多个区域。在响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域之后，在第一图像上分别显示用于标识一个或多个区域的第二标识信息，该第二标识信息可以为图形、文字、符号等醒目的标记，比如，为三角形，以向目标用户提示预设区域在整个监控区域中的位置。
作为一种可选的实施方式，在步骤S110，预设区域中确定出与移动轨迹相交的目标区域包括：在一个或多个区域中确定出与移动轨迹相交的目标区域；在一个或多个区域中确定出与移动轨迹相交的目标区域之后，该方法还包括：保留目标区域的第二标识信息，隐藏一个或多个区域中除目标区域之外的区域的第二标识信息；和/或，显示用于标识目标区域的第一标识信息。
在该实施例中，第一图像上分别显示用于标识一个或多个区域的第二标识信息，在一个或多个区域中确定出与移动轨迹相交的目标区域，并且在一个或多个区域中确定出与移动轨迹相交的目标区域之后，可以保留一个或多个区域中的目标区域的第二标识信息，隐藏一个或多个区域中除目标区域之外的区域的第二标识信息，比如，仅保留目标区域的三角形，而隐藏一个或多个区域中除目标区域之外的区域的三角形，以指示目前显示的三角形所对应的区域为监控区域中可以放置目标捕捉工具的区域。
可选地，该实施例在一个或多个区域中确定出与移动轨迹相交的目标区域之后，可以仅显示用于标识一个或多个区域中的目标区域的第一标识信息，比如，仅显示红色圆圈，以指示目前显示的红色圆圈所对应的区域为监控区域中可以放置目标捕捉工具的区域。
可选地，该实施例在一个或多个区域中确定出与移动轨迹相交的目标区域之后，可以既保留目标区域的第二标识信息，隐藏一个或多个区域中除目标区域之外的区域的第二标识信息，并且还显示用于标识一个或多个区域中的目标区域的第一标识信息， 比如，通过既显示三角形，又显示红色圆圈来指示一个或多个区域中的目标区域，以指示三角形和红色圆圈所对应的区域为监控区域中可以放置目标捕捉工具的区域。
作为一种可选的实施方式，在根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹之后，该方法还包括：在第一图像上显示用于标识移动轨迹的第三标识信息。
在该实施例中，在根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹之后，可以对移动轨迹进行标识，以指示目标对象在监控区域中的移动情况，比如，通过第一图像上显示的第三标识信息进行标识，该第三标识信息可以为线条，此处对线条的颜色和粗细不做任何限定。
作为一种可选的实施方式，响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域包括：在第一图像中识别出监控区域中的目标类型区域，其中，目标对象经过目标类型区域的概率大于第五阈值；响应输入的区域设置指令在目标类型区域上确定出区域设置指令所指示的预设区域。
在该实施例中，在响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域时，可以首先在第一图像中识别出监控区域中的目标类型区域，该目标类型区域是通过目标对象的属性和活动规律确定的区域，目标对象经过该目标类型区域的概率大于第五阈值，比如，目标对象为老鼠，由于老鼠具有善于攀爬、钻洞的特性，其在活动时很有可能是沿着电线、水管等上下移动，另外，能够部署粘鼠板的地方可以为地面、墙角、窗台等平整的位置，则该实施例的目标类型区域可以为靠近电线、水管的地面、墙角、窗台等平整的区域。在第一图像中识别出监控区域中的目标类型区域之后，可以进一步响应输入的区域设置指令在目标类型区域上确定出区域设置指令所指示的预设区域，按照用户手指或者鼠标在终端屏幕上的滑动轨迹来触发区域设置指令，在目标类型区域上确定出区域设置指令所指示的预设区域，可以由用户根据经验输入区域设置指令，从而在目标类型区域上确定出区域设置指令所指示的预设区域，进而在预设区域中确定出与移动轨迹相交的目标区域，提高了对用于放置目标捕捉装置的区域的准确性。
作为一种可选的实施方式，步骤S108，根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹包括：从第一组图像数据中识别出目标对象在监控区域中经过的多个位置；通过多个位置生成移动轨迹，其中，多个位置位于移动轨迹上。
在该实施例中，在根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹时，可以先从第一组图像数据中识别出目标对象在监控区域中经 过的多个位置。可选地，先从第一组图像数据中识别出目标特征，在该目标特征为目标对象的特征的情况下，确定监控区域中出现了目标对象，可以通过目标特征在第一组图像中的多个位置来确定目标对象在第一目标时间段内在监控区域中经过的多个位置，该多个位置可以通过三维坐标系下的坐标点(X，Y，Z)进行表示，比如，多个位置分别为A(X1，Y1，Z1)、B(X1，Y1，Z1)、C(X1，Y1，Z1)、D(X1，Y1，Z1)。在从第一组图像数据中识别出目标对象在监控区域中经过的多个位置之后，可以通过多个位置生成移动轨迹，可以将多个位置通过线连接，比如，将位置A(X1，Y1，Z1)、B(X1，Y1，Z1)、C(X1，Y1，Z1)、D(X1，Y1，Z1)通过线连接，从而形成目标对象在第一目标时间段内的移动轨迹。
作为一种可选的实施方式，在步骤S110，预设区域中确定出与移动轨迹相交的目标区域之后，该方法还包括：在目标捕捉装置放置在目标区域的情况下，将由目标捕捉装置捕捉到的目标对象的目标信息发送至目标终端；和/或将目标区域发送至目标终端；或者将目标区域和监控区域的第二图像发送至目标终端；或者在目标终端上显示监控区域的第三图像，其中，第三图像上显示有目标区域；或者在目标终端上显示监控区域的第四图像，其中，第四图像上显示有移动轨迹和目标区域。
在该实施例中，在预设区域中确定出与移动轨迹相交的目标区域之后，在目标捕捉装置放置在目标区域的情况下，可以将由目标捕捉装置捕捉到的目标对象的目标信息发送至目标终端，比如，获取各个监控设备监控到的目标对象的目标信息，将目标对象的目标信息发送至目标终端，该目标信息可以为目标对象的种类、目标对象的皮肤颜色、目标对象的数量、目标对象的形态等信息，此处不做任何限制。
可选地，该实施例还可以将目标区域发送至目标终端，以指示目标用户按照目标区域在监控区域中放置目标捕捉工具。该实施例还可以将目标区域以及监控区域的第二图像均发送至目标终端，使得目标用户可以了解目标区域在监控区域中的具体位置。该实施例还可以在目标终端显示包括目标区域的第三图像，还可以在目标终端上显示包括有移动轨迹和目标区域的监控区域的第五图像，从而使得用户了解监控区域中的目标区域和进入的目标对象的情况，进而将目标捕捉装置放置在目标区域上，进而提高了对目标对象进行捕捉的效率。
需要说明的是，该实施例的目标终端可以为智能手机(如Android手机、iOS手机等)、平板电脑、掌上电脑以及移动互联网设备(Mobile Internet Devices，简称为MID)、PAD等终端设备。
作为一种可选的实施方式，在目标捕捉装置放置在目标区域上的情况下，获取目标区域在从目标时刻开始每第二目标时间段内的第二组图像数据，得到至少一组第二组图像数据，其中，目标时刻为第一目标时间段之后的时刻；分别从至少一组第二组 图像数据中识别出进入目标区域的目标对象的第二目标信息，得到至少一组第二目标信息；将至少一组第二目标信息转化为目标报告，其中，目标报告包括以下至少之一形式：文本形式、表格形式、统计图形式；通过服务器将目标报告推送至目标终端。
在该实施例中，在目标捕捉装置放置在目标位置上的情况下，可以自动出具目标区域的目标对象的目标报告。可选地，获取目标区域在从目标时刻开始每第二目标时间段内的第二组图像数据，得到至少一组第二组图像数据，该目标时刻为第一目标时间段之后的时刻，第二目标时间段可以为1天，也即，在目标捕捉装置放置在目标位置上的情况下，获取目标区域每天的图像数据。可以分别从至少一组第二组图像数据中识别出进入目标区域的目标对象的目标信息，得到至少一组目标信息，将至少一组目标信息转化为目标报告，该目标报告还可以包括目标对象出现的区域的名称、时间等信息，形式可以为文本形式、表格形式、统计图形式等，此处不做任何限制，进而通过服务器将目标报告推送至目标终端，使得目标用户通过目标终端可以了解目标区域的目标对象的情况，包括目标对象的变化趋势等，从而了解目标区域是否处于严重的卫生威胁之中，供目标用户综合判断现场的情况，并有针对性地实施对有害生物的防治工作，并且还可以指导建筑结构是否存在漏洞。
作为另一种可选的实施方式，该实施例可以确定目标对象在监控区域中的入侵点和藏匿点。该实施例的第一组图像数据包括视频监控设备所拍摄的监控区域的视频数据，从该视频数据中截取目标对象在监控区域中出没的视频。获取该目标对象出没的视频中的第一个视频帧，从该第一个视频帧中识别出目标对象在监控区域中的位置，将识别出的该位置确定为目标对象在监控区域中的入侵点，可以将其作为目标对象入侵室内场所的入口。该实施例还可以从目标对象出没的视频中获取最后一个视频帧，从该最后一个视频帧中识别出目标对象在监控区域中的位置，将识别出的该位置确定为目标对象的藏匿点，可以将其作为目标对象的窝点、或者是在逃离监控区域时的出口。
可选地，该实施例可以记录目标对象在过去一段时间的入侵点和藏匿点，将用于指示目标对象的入侵点和藏匿点的信息发送至目标终端，以提示防治人员对目标对象的防治进一步采取措施，从而达到了提高对目标对象进行防治的效率的目的。
举例而言，目标对象为老鼠，记录老鼠在过去三天的入侵点和藏匿点，将用于指示老鼠的入侵点和藏匿点的信息发送至目标终端，以提示有害生物防治负责人对老鼠的防治进一步采取措施，比如，有害生物防治负责人寻找入侵点附近是否有较大缝隙的下水道口，或者寻找入侵点附近是否有通往室外的管道，如果入侵点有较大缝隙的下水道口，或者通往室外的管道，则及时地封堵下水道口或者管道，从而断绝老鼠入侵的通道，提高了对目标对象进行防治的效率。
作为一种可选的实施方式，该实施例可以确定目标对象在视频监控设备所拍摄的监控区域中的密度，可以确定目标对象在不同监控区域中的密度。可选地，获取监控区域中目标对象出没的时长和对目标对象进行监测的整个监测周期的时间之比，将其确定为监控区域的目标对象的密度。
可选地，该实施例在获取目标对象在不同监控区域中的密度之后，从中确定目标对象的密度最高的监控区域，将其确定为目标对象入侵频繁的监控区域，可以将用于指示目标对象入侵频繁的监控区域的信息发送至目标终端，以提示相关人员进一步采取措施，达到提高对目标对象进行防治的效率的目的。
举例而言，目标对象为老鼠，老鼠在监控区域中的密度也即监控区域中的鼠密度值。在获取老鼠在不同监控区域中的密度之后，从中确定老鼠的密度最高的监控区域，也即，记录室内鼠密度值较高的场所，将其确定为老鼠入侵频繁的场所，并且将用于指示老鼠入侵频繁的场所的信息发送至目标终端，以提示餐厅运营人员进一步检查该场所是否有导致老鼠和虫害滋生的因素，比如，该场所是否有残留的食物残渣、未清理的水迹等，使其成为老鼠和虫害滋生的场所。如果该场所有导致老鼠和虫害滋生的因素，则指示餐厅运营人员进一步做好该场所的管理工作，以减少该场所对目标对象的吸引力，从而提高了对目标对象进行防治的效率。
该实施例的区域确定方法涉及对目标对象的确定，也即，确定监控区域中是否有目标对象，在确定出监控区域中有目标对象之后，根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹，进而在预设区域中确定出与移动轨迹相交的目标区域，以在目标区域放置用于捕捉目标对象的目标捕捉装置。下面对该实施例的对目标对象的确定的算法进行介绍。
作为一种可选的实施方式，步骤S106，获取在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据包括：获取摄像设备对监控区域拍摄得到的视频文件；对视频文件进行抽帧采样，得到一组视频帧图像的数据，其中，第一组图像数据包括一组视频帧图像的数据；在步骤S108，根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹之前，该方法还包括：根据一组视频帧图像中的像素点的像素值在一组视频帧图像中确定出多个目标视频帧图像，其中，每个目标视频帧图像用于指示在监控区域中存在运动的对象；对每个目标视频帧图像进行目标对象检测，得到每个目标视频帧图像的图像特征，其中，图像特征用于表示在存在运动的对象中，与目标对象之间的相似度大于第六阈值的对象所在的目标图像区域；根据每个目标视频帧图像的图像特征确定出运动特征，其中，运动特征用于表示多个目标视频帧图像中存在运动的对象的运动速度和运动方向；根据运动特征和每个目标视频帧图像的图像特征，确定多个目标视频帧图像中是否出现有目标对象，也即，确定监控 区域中是否有目标对象。
在该实施例中，摄像设备可以为监控摄像头，比如，该摄像设备为红外微光夜视摄像头，用于对监控区域进行拍摄，得到视频文件。其中，监控区域为被检测区域，也即，该监控区域为检测是否有目标对象出现的区域。该实施例的视频文件包括对监控区域进行拍摄得到的原始视频数据，可以包括监控区域的监控视频序列，该监控视频序列也即图像视频序列。
在获取摄像设备对监控区域拍摄得到的视频文件之后，对视频文件进行预处理，可以在视频数据处理层对视频文件进行抽帧采样，得到一组视频帧图像。该实施例可以对视频文件进行等间隔的抽帧采样，从而得到视频文件的一组视频帧图像，比如，视频文件包括100个视频帧序列，在进行抽帧采样之后，得到10个视频帧序列，则将这10个视频帧序列作为上述一组视频帧图像，从而减少对目标对象进行确定的算法的运算量。
在该实施例中，对视频文件进行预处理，还包括对视频文件进行动态检测，从一组视频帧图像中确定用于指示在监控区域中存在运动的对象的目标视频帧图像，也即，在该目标视频帧图像中存在运动的对象，该目标视频帧图像可以为存在运动的对象的视频片段，其中，存在运动的对象可能是目标对象，也可能不是。该实施例可以通过动态检测算法确定目标视频帧图像，根据一组视频帧图像中的像素点的像素值在一组视频帧图像中确定出多个目标视频帧图像。可选地，在一组视频帧图像中，除多个目标视频帧图像之外的视频帧图像未指示出在对应的监控区域中有运动的图像，可以不进行后续的检测。
在根据一组视频帧图像中的像素点的像素值在一组视频帧图像中确定出多个目标视频帧图像之后，对每个目标视频帧图像进行目标对象检测，得到每个目标视频帧图像的图像特征，其中，图像特征针对每个目标视频帧图像而言，用于表示目标视频帧图像中存在运动的对象被判定为目标对象时存在运动的对象所在的目标图像区域。
在该实施例中，对每个目标视频帧图像进行目标对象检测，也即，对目标视频帧图像中存在的运动对象进行检测，可以通过目标检测系统采用动态目标检测方法和基于神经网络的目标检测方法对目标视频帧图像中存在的运动对象进行检测，得到每个目标视频帧图像的图像特征，其中，动态目标检测方法的运算速度快、对机器配置要求较低，而基于神经网络的目标检测方法的准确性和鲁棒性更好，图像特征可以为矩形框中的视觉信息，用于表示目标图像区域，该矩形框可以为检测框，用于表示在存在运动的对象中，与目标对象之间的相似度大于第六阈值的对象所在的目标图像区域，也即，与目标对象之间的相似度大于第六阈值的对象可能为目标对象，目标图像特征也是用于指示目标对象的可能位置。
在对每个目标视频帧图像进行目标对象检测，得到每个目标视频帧图像的图像特征之后，可以将每个目标视频帧图像的图像特征输入至运动特征提取模块，该运动特征提取模块根据每个目标视频帧图像的图像特征确定出运动特征，该运动特征针对多个目标视频帧图像而言，用于表示多个目标视频帧图像中存在运动的对象的运动速度和运动方向，同时进一步过滤掉非目标对象的移动所造成的干扰图像，比如，删除掉蚊虫的移动等干扰信息。
可选地，在该实施例中，由于每个目标视频帧图像中存在运动的对象的运动是连续的，运动特征提取模块的运动特征提取算法可以先根据每个目标视频帧图像的图像特征检测多个目标视频帧图像之间的图像特征的相关性，可以将相关性大的图像特征对应的对象确定为同一对象，对每一目标视频帧图像的图像特征进行匹配，得到对象的一系列运动图片，最后可以使用3D的特征提取网络提取运动序列的特征，从而得到运动特征，比如，根据每个目标视频帧图像的检测框，计算多个目标视频帧图像之间检测框的相关性，可以将相关性大的检测框对应的对象确定为同一对象，对每个目标视频帧图像的检测框进行匹配，得到对象的一系列运动图片，最后使用3D的特征提取网络提取运动序列的特征，得到运动特征，进而确定多个目标视频帧图像中存在运动的对象的运动速度和运动方向。
可选地，该实施例也可以将多个目标视频帧图像的图像特征进行融合且进行特征提取，从而防止单帧的目标检测器出现误判的情况，进而实现对目标图像进行精筛以准确确定出是否出现目标对象。
在根据每个目标视频帧图像的图像特征确定出运动特征之后，可以将运动特征和每个目标视频帧图像的图像特征进行融合，输入至预先训练好的分类网络中，该分类网络为预先设计好的用于确定多个目标视频帧图像中是否出现有目标对象的分类网络模型，进而根据运动特征和每个目标视频帧图像的图像特征，确定多个目标视频帧图像中是否出现有目标对象，比如，确定多个目标视频帧图像中是否出现有老鼠。
可选地，该实施例将多个目标视频帧图像中有目标对象的目标视频帧的图像特征输入至前端显示界面，该前端显示界面可以进而显示出目标对象的检测框和移动轨迹。
可选地，该实施例的分类网络模型可以用于过滤非目标对象的图片序列，而保留目标对象的图片序列，从而降低虚警率，保证目标对象提示信息的准确性。
该实施例对监控区域的视频文件进行抽帧采样，得到一组视频帧图像，根据一组视频帧图像中的像素点的像素值在一组视频帧图像中确定出用于指示在监控区域中存在运动的对象的多个目标视频帧图像，再根据每个目标视频帧图像的图像特征确定出运动特征，进而根据运动特征和每个目标视频帧图像的图像特征，达到自动确定多个 目标视频帧图像中是否出现有目标对象的目的，不仅大大减少了确定目标对象的人力成本，而且提高了确定目标对象的准确率，解决了对目标对象进行确定的效率低的问题，进而达到了提高鼠患检测准确度的效果。
可选地，在根据一组视频帧图像中的像素点的像素值在一组视频帧图像中确定出多个目标视频帧图像的数据时，获取一组视频帧图像中的每个像素点的平均像素值；获取一组视频帧图像中的每个视频帧图像中的每个像素点的像素值与对应的平均像素值之间的差值；将一组视频帧图像中差值满足预定条件的视频帧图像确定为目标视频帧图像。
作为一种可选的实施方式，获取一组视频帧图像中的每个视频帧图像中的每个像素点的像素值与对应的平均像素值之间的差值包括：对于一组视频帧图像中的每个视频帧图像中的每个像素点执行以下操作，其中，在执行以下操作时将每个视频帧图像视为当前视频帧图像，将每个像素点视为当前像素点：D(x,y)＝|f(x,y)-b(x,y)|，其中，(x,y)为当前像素点在当前视频帧图像中的坐标，f(x,y)表示当前像素点的像素值，b(x,y)表示当前像素点的平均像素值，D(x,y)表示当前像素点的像素值与对应的平均像素值之间的差值。
作为一种可选的实施方式，将一组视频帧图像中差值满足预定条件的视频帧图像确定为目标视频帧图像包括：对于一组视频帧图像中的每个视频帧图像中的每个像素点执行以下操作，其中，在执行以下操作时将每个视频帧图像视为当前视频帧图像，将每个像素点视为当前像素点：
Figure PCTCN2019080746-appb-000001
其中，D(x,y)表示为当前像素点的像素值与对应的平均像素值之间的差值，T为第一预设阈值；其中，预定条件包括：目标视频帧图像中M(x,y)＝1的像素点的个数超过第二预设阈值。
作为一种可选的实施方式，根据每个目标视频帧图像的图像特征确定出运动特征包括：获取与每个目标视频帧图像的图像特征所表示的目标图像区域对应的目标矢量，得到多个目标矢量，其中，每个目标矢量用于表示对应的一个目标视频帧图像中存在运动的对象在经过目标图像区域时的运动速度和运动方向；将多个目标矢量按照每个目标视频帧图像在视频文件中的时间顺序组成第一目标向量，其中，运动特征包括第一目标向量；或者获取与每个目标视频帧图像的图像特征所表示的目标图像区域对应的二维光流图，得到多个二维光流图，其中，每个二维光流图包括对应的一个目标视频帧图像中存在运动的对象在经过目标图像区域时的运动速度和运动方向；将多个二维光流图按照每个目标视频帧图像在视频文件中的时间顺序组成三维第二目标向量，其中，运动特征包括三维第二目标向量。
作为一种可选的实施方式，根据运动特征和每个目标视频帧图像的图像特征，确定多个目标视频帧图像中是否出现有目标对象包括：将运动特征和每个目标视频帧图像的图像特征输入到预先训练好的神经网络模型中，得到对象识别结果，其中，对象识别结果用于表示多个目标视频帧图像中是否出现有目标对象。
作为一种可选的实施方式，将运动特征和每个目标视频帧图像的图像特征输入到预先训练好的神经网络模型中，得到对象识别结果包括：将每个图像特征经过包括卷积层、正则化层和激活函数层的神经网络层结构，得到多个第一特征向量；将多个第一特征向量与运动特征进行融合，得到第二特征向量；将第二特征向量输入到全连接层进行分类，得到第一分类结果，其中，神经网络模型包括神经网络层结构和全连接层，对象识别结果包括第一分类结果，第一分类结果用于表示多个目标视频帧图像中是否出现有目标对象；或者将每个图像特征经过包括卷积层、正则化层和激活函数层的第一神经网络层结构，得到多个第一特征向量；将运动特征经过包括卷积层、正则化层、激活函数层的第二神经网络层结构，得到第二特征向量；将多个第一特征向量与第二特征向量进行融合，得到第三特征向量；将第三特征向量输入到全连接层进行分类，得到第二分类结果，其中，神经网络模型包括第一神经网络层结构、第二神经网络层结构和全连接层，对象识别结果包括第二分类结果，第二分类结果用于表示多个目标视频帧图像中是否出现有目标对象。
作为一种可选的融合方式，可以将多个第一特征向量与运动特征进行拼接(或称为组合)，得到第二特征向量。
作为一种可选的融合方式，可以将多个第一特征向量与第二特征向量进行拼接(或称为组合)，得到第三特征向量。
作为另一种可选的示例，将运动特征和每个目标视频帧图像的图像特征输入到预先训练好的神经网络模型中，得到对象识别结果包括：将每个图像特征依次经过多个块，得到多个第一特征向量，其中，在每个块中会对块的输入依次执行卷积层上的卷积操作、正则化层上的正则化操作、激活函数层上的激活操作；将多个第一特征向量与运动特征进行拼接，得到第二特征向量；将第二特征向量输入到全连接层，通过全连接层输出得到第一分类结果，其中，神经网络模型包括多个块和全连接层，对象识别结果包括第一分类结果，第一分类结果用于表示多个目标视频帧图像中是否出现有目标对象；或者将每个图像特征依次经过多个第一块，得到多个第一特征向量，其中，在每个第一块中会对第一块的输入依次执行卷积层上的卷积操作、正则化层上的正则化操作、激活函数层上的激活操作；将运动特征依次经过多个第二块，得到第二特征向量，其中，在每个第二块中会对第二块的输入依次执行卷积层上的卷积操作、正则化层上的正则化操作、激活函数层上的激活操作；将多个第一特征向量与第二特征向 量进行拼接，得到第三特征向量；将第三特征向量输入到全连接层，通过全连接层输出得到第二分类结果，其中，神经网络模型包括多个第一块、多个第二块和全连接层，对象识别结果包括第二分类结果，第二分类结果用于表示多个目标视频帧图像中是否出现有目标对象。
作为一种可选的实施方式，对视频文件进行抽帧采样，得到一组视频帧图像包括：对视频文件中的视频序列进行等间隔的抽帧采样，得到一组视频帧图像。
作为一种可选的实施方式，获取摄像设备对监控区域拍摄得到的视频文件包括：获取的视频文件包括：获取红外微光夜视摄像头对监控区域拍摄得到的视频文件，其中，视频文件中的视频帧图像为通过红外微光夜视摄像头拍摄到的图像。
作为一种可选的实施方式，在确定多个目标视频帧图像中是否出现有目标对象之后，该方法还包括：在确定出多个目标视频帧图像中出现有目标对象的情况下，确定目标对象在多个目标视频帧图像中的位置；将位置显示在多个目标视频帧图像中。
作为一种可选的实施方式，目标对象的确定方法由设置在本地的服务器执行。
可选地，在该实施例中，通过红外微光夜视摄像头采集场景视频序列，数据处理模块接收视频序列并且检测视频中有无老鼠，若检测到老鼠，将老鼠的位置等一系列信息输出至前端显示界面，前端显示界面显示老鼠的位置、出现时间、活动区域并且可以即时进行鼠患的报警。
该实施例基于上述方法，实现了在餐厅、工厂等场景下的自动确定捕鼠工具的放置区域，通过获取对监控区域进行拍摄得到的第一图像，响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域，获取在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据，根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹，在预设区域中确定出与移动轨迹相交的目标区域，达到了对用于放置目标捕捉工具的区域进行确定的目的，提高了对用于放置目标捕捉装置的区域进行确定的准确性，也即，使用计算机完成对目标捕捉工具的放置区域的自动判别，替代人工判别和凭经验判别，在餐厅、工厂等注重清洁卫生的环境中可以用于辅助对目标对象的防治工作，从而指导对目标对象的防治工作的有效开展，保障了餐饮业关键场所和设施不受目标对象的侵袭，并且免去人工标记目标捕捉工具在每处监控区域中的放置区域的工序，从而节省了人力成本。
需要说明的是，在附图的流程图示出的步骤可以在诸如一组计算机可执行指令的计算机系统中执行，并且，虽然在流程图中示出了逻辑顺序，但是在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤。
下面结合优选的实施例对本公开的技术方案进行举例说明。具体以目标对象为老 鼠、目标捕捉工具为捕鼠器进行举例说明。
该实施例应用数字化技术，提出了一种在餐厅、工厂等场景下的自动确定捕鼠工具的放置区域的方法，可以使用计算机完成对捕鼠装置的放置区域进行自动确定，从而避免了人工确定放置区域和凭经验确定放置区域，可以用于在餐厅、工厂等注重清洁卫生的场所中辅助防鼠、灭鼠的工作，从而改进了对老鼠等有害生物的防治的效果，保障了餐饮业的关键场所和设施不受鼠害侵袭，并且免去人工标记捕鼠装置在每个监控区域中放置的区域的工序，从而节省了人力成本。
图2是根据本公开实施例的一种确定灭鼠器的放置位置的方法的流程图。如图2所示，该方法包括以下步骤：
步骤S201，在监控区域中识别捕鼠装置的可放置区域。
步骤S202，在监控区域中识别老鼠的历史移动轨迹。
步骤S203，根据捕鼠装置的可放置区域和老鼠的历史移动轨迹，确定用于放置捕鼠装置的目标区域。
下面对该实施例的在监控区域中识别捕鼠装置的可放置区域的方法进行介绍。
在该实施例中，考虑到老鼠善于攀爬、钻洞的特性，其很可能是沿着电线、水管等进行上下移动，而可以放置部署装置的区域可以是地面、墙角、窗台等平整的区域。因而该实施例在视频监控设备对准的静态区域中，预先设置若干个适合的放置区域，可以优选墙角、电线旁等老鼠常规行进的必经区域。其中，视频监控设备可以为摄像头。
图3是根据本公开实施例的一种识别出的可放置捕鼠装置区域的示意图。如图3所示，通过摄像设备获取静态的监控区域的图像，可以通过图像识别技术从监控区域的图像中可以识别出地面、墙角等区域，也可以人工从监控区域的图像中识别出地面、墙角、窗台等区域。图3中监控区域的图像中的三角形用于指示可放置捕鼠装置的区域，该区域包括多个，比如，为5个。
该实施例可以在不同的监控区域中安装视频监控设备，以在不同的监控区域中识别出捕鼠装置的可放置区域。
可选地，该实施例在每个视频监控设备在对应的监控区域中安装完毕之后，可放置捕鼠装置区域的识别工作也就完成，将识别出的可放置捕鼠装置区域可以预先存储在服务器中以待取用。
下面对该实施例的在监控区域中识别老鼠的历史移动轨迹的方法进行介绍。
该实施例可以通过视频监控设备获取老鼠的历史移动轨迹，可以获取前一天的监控区域中有老鼠出没的目标视频，对该目标视频按照时间段进行截取，提取出包括老鼠图像的视频片段。再通过动作识别技术识别出视频片段中动态的变化特征，再通过图像识别技术对动态的变化特征进行识别，比如，通过AI图像识别技术对视频片段中动态的变化特征进行识别，进一步确定出现的生物确实为老鼠，进而结合动态的变化特征确定出老鼠的移动轨迹，在监控区域的图像中表示出。
图4是根据本公开实施例的一种识别出的鼠迹的示意图。如图4所示，应用AI图像识别技术、动作识别技术，提取出监控区域的监控视频中的老鼠的移动轨迹。在监控区域的图像中标示出，其中的线条即为老鼠在监控区域中的移动轨迹。
下面对该实施例的根据捕鼠装置的可放置区域和老鼠的历史移动轨迹，确定用于放置捕鼠装置的目标区域的方法进行介绍。
该实施例在监控区域中识别捕鼠装置的可放置区域，在监控区域中识别老鼠的历史移动轨迹之后，捕鼠装置的可放置区域和监控区域中老鼠的历史移动轨迹进行比较，可以在相重合的部分所在的区域确定为用于放置捕鼠装置的目标区域。
图5是根据本公开实施例的一种捕鼠装置的放置区域的示意图。如图5所示，可以将捕鼠装置在监控区域中的可放置区域(如图3)和监控区域中老鼠的历史移动轨迹(如图4)相重合的部分所对应的目标区域用圆圈标出，比如，用红色等醒目颜色的圆圈标出，即为当天防治人员需将捕鼠装置在监控区域中放置的区域。
可选地，该实施例还可以用于确定有害生物在监控区域中的入侵点和藏匿点。比如，有害生物为老鼠，计算老鼠在监控区域中的入侵点和藏匿点。该实施例可以从视频数据所指示的视频中截取老鼠在监控区域中出没的视频。获取该老鼠出没的视频中的第一个视频帧，从该第一个视频帧中识别出老鼠在监控区域中的位置，将该位置确定为老鼠在监控区域中的入侵点，可以将其作为老鼠入侵室内场所的入口。该实施例还可以从老鼠出没的视频中获取最后一个视频帧，从该最后一个视频帧中识别出老鼠在监控区域中的位置，将该位置确定为老鼠的藏匿点，可以将其作为老鼠窝点、或者是在逃离室内时的出口。
可选地，该实施例可以记录有害生物在过去一段时间的入侵点和藏匿点，将用于指示有害生物的入侵点和藏匿点的信息发送至终端，以提示有害生物防治负责人对有害生物的防治进一步采取措施。比如，有害生物为老鼠，记录老鼠在过去三天的入侵点和藏匿点，将用于指示老鼠的入侵点和藏匿点的信息发送至终端，以提示有害生物防治负责人对老鼠的防治进一步采取措施，比如，寻找入侵点附近是否有较大缝隙的下水道口，或者寻找入侵点附近是否有通往室外的管道，如果有，则及时地封堵水道 口和管道，从而断绝老鼠入侵的通道。
可选地，该实施例还可以用于确定有害生物在监控区域中的密度，可以确定有害生物在不同监控区域中的密度，可以将在该监控区域中有害生物出没的时长占对有害生物进行监测的整个监测周期的时间之比，确定为监控区域的有害生物的密度。比如，有害生物为老鼠，可以计算视频监控设备所拍摄的监控区域的鼠密度，将在该监控区域中老鼠出没的时长占对老鼠进行监测的整个监测周期的时间之比，确定为监控区域的鼠密度。
可选地，该实施例在计算有害生物在不同监控区域中的密度之后，从中确定有害生物的密度最高的监控区域，将其确定为有害生物入侵频繁的区域，可以将用于指示有害生物入侵频繁的区域的信息发送至终端，以提示相关人员进一步采取措施。比如，有害生物为老鼠，在计算老鼠在不同监控区域中的密度之后，从中确定老鼠的密度最高的监控区域，也即，记录室内鼠密度值较高的场所，将其确定为老鼠入侵频繁的场所，并且将用于指示老鼠入侵频繁的场所的信息发送至终端，以提示餐厅运营人员进一步检查是否有残留的食物残渣、未清理的水迹等，使其成为老鼠和虫害滋生的场所。如果有残留的食物残渣、未清理的水迹等，则指示餐厅运营人员进一步做好该场所中的卫生清洁工作，以减少该场所对有害生物的吸引力。
该实施例可以汇总待部署捕鼠装置的目标区域，并且在将捕捉装置放置在目标区域之后，可以汇总当天各个视频监控设备收集的老鼠的信息，通过客户端(APP)呈现给餐厅运营人员以及防治人员，可以自动按天出具报告；也可通过微信公众号、即时信息、短信等可选的方式，推送给餐厅运营的相关人员。
图6是根据本公开实施例的一种鼠迹报告的直方图。如图6所示，可以以天为周期，检测监控区域在目标时间段内的老鼠的活跃指数，可以检测日期11/29到日期12/12时间段内的老鼠的活跃指数，通过摄像头获取监控区域的图像数据，根据识别出的老鼠的信息，确定出监控区域在每一天的老鼠活跃指数，其中，老鼠活跃指数可以通过老鼠的活跃时长、老鼠的数量等信息确定，从而使得给餐厅运营人员以及虫害防治人员可以了解目标区域的老鼠的情况，进而采取措施进行防治。
根据本公开实施例的一种鼠患视频监测装置可以包括分为几个部件：红外微光夜视摄像头、数据处理模块和前端显示部件，上述装置工作时原理如下：红外微光夜视摄像头负责采集场景视频序列，数据处理模块接收视频序列并且检测视频中有无老鼠，若检测到老鼠，将老鼠的位置等一系列信息输出至前端显示界面，前端显示界面显示老鼠的位置、出现时间、活动区域并且可以即时进行鼠患的报警。
图7是根据本公开实施例的一种数据处理模块的示意图。如图7所示，该数据处 理模块包括：视频采集模块702、视频处理模块704和存储模块706，其中，视频采集模块702包括：ARM板7022和视频预处理模块7024，视频处理模块704包括：嵌入式GPU处理器7042。
视频采集模块702通过ARM板7022采集视频数据并进行预处理，视频处理模块704读入以训练好的模型在嵌入式GPU处理器7042中根据深度学习算法进行视频处理，若深度学习网络检测到某一个片段时间有老鼠，则将该片段以及相应的检测结果存储至存储模块706，存储模块706将这一系列信息输出至前端。
图8是根据本公开实施例的一种鼠患检测系统的原理示意图。如图8所示，该算法包括以下几个模块：预处理、目标检测、运动特征提取和分类网络，系统的输入为原始的视频序列，预处理包含两个步骤：抽帧和动态检测，先是对原始视频序列进行等间隔的抽帧采样，减少算法的运算量，然后利用目标检测算法进行目标检测，判断图像中是否有运动物体，若无运动物体，则不进行后续的检测，若有运动物体，则将有运动物体的视频片段送入后续模块。在目标检测过程中，对预处理后的视频序列的每一帧进行检测，在可能存在老鼠的位置获取图像特征(如该对应的检测框内的视觉信息)并通过运动特征提取模块，将各个视频图像帧之间的信息进行融合和特征提取，防止单帧的目标检测器出现误判的情况，随后将提取的运动特征与图像特征输入分类网络，由分类网络判别是否是老鼠，若是老鼠，则将老鼠在每一帧所在位置的矩形检测框传给前端显示界面。
需要说明的是，在本实施例中，上述目标检测过程是根据具体的机器计算资源分配了两种算法：动态目标检测算法和基于神经网络的目标检测算法，前者运算速度快、对机器配置要求低，后者准确性和鲁棒性。
1)动态目标检测算法包含背景差和帧差法，利用下述公式(1)，计算当前帧和背景或者前一帧的差值：
D k(x，y)＝|f k(x，y)-b k(x，y)|     (1)
上式中，(x,y)用于表示以图像左上角为原点，宽方向为X轴，高方向为Y轴建立的坐标系中像素点的坐标，k为当前帧的索引，f代表当前帧，b代表背景或者上一帧。利用公式(2)判断是否存在运动目标：
Figure PCTCN2019080746-appb-000002
M(x,y)用于表示运动图像，T为阈值，若M(x,y)为1表示有运动目标，所有X(x,y)的像素组成了运动目标图像，经过形态学运算合并像素点可得出所有运动的 目标，作为该模块的输出。
2)基于神经网络的目标检测将图片输入预先训练好的网络模型，得出所有可能的目标和其置信度，大于某个置信度阈值的检测框作为该模块的输出。使用的网络模型包含但不限于SSD、Faster-RCNN、FPN等。图9是本公开实施例的一种Faster-RCNN网络模型的示意图。如图9所示，其中conv是卷积层，包括卷积层1至卷积层5，构成残差网络101，该实施例由卷积核(是一个矩阵)在输入上进行划窗，对每个输入的划窗位置都和矩阵根据公式(3)相点乘，结果F作为该划窗位置的特征输出。
F＝Σ 0≤i，j≤nk(i，j)*I(i，j)    (3)
RPN用于表示区域提出网络，会提出一系列的候选框，池化层(ROI pooling)将卷积层提到的特征图在RPN输出的坐标下的区域映射成大小(w，h)固定的矩形框，送入由全连接层构成的分类器和边框回归器，边框回归输出老鼠的可能坐标位置，分类器输出是该位置老鼠的置信度。
对于上述运动特征提取：因为物体的运动是连续的，运动特征提取算法先根据每一帧得到的检测框，计算帧与帧之间检测框的相关性，相关性大的检测框认为是同一物体，对每一帧的检测框进行匹配，得到物体的一系列运动图片，最后使用3D的特征提取网络提取运动序列的特征。
对于上述分类网络：将目标检测框中的视觉信息和运动特征融合，送入设计好的分类的网络模型，以筛除非老鼠的图片序列，从而降低虚警率，进而将结果送入前端显示界面，显示老鼠的检测框和轨迹。
在本公开实施例中，对于整体的框架，还可以包括但不限于通过目标检测和分类网络来达到检测识别的目的，以节省框架布局成本。
本公开实施例提出了利用图像识别算法，自动识别监控视频中的老鼠，无需放置鼠夹鼠笼，也无需花费人力进行观测，将监测鼠害变为高效全自动的流程工作，不仅大大减少了监测鼠害的人力成本，同时准确率高，方便对后厨鼠害卫生的监管，同时，还可以提供老鼠活动的轨迹，便于人员选择灭鼠工具放置位置，方便了进一步的除害工作。
该实施例通过在监控区域中自动标记用于放置捕鼠装置的区域，可以为餐厅、工厂等室内运营场所自动确定用于放置捕鼠装置的区域，增加老鼠被捕捉的概率，通过虫鼠害的汇总情况，了解当前餐厅是否处于严重的卫生威胁之中，以供专业的虫鼠害防治人员综合判断现场情况，并有针对性地采取防治措施实施防治工作，从而指导防鼠工作的有效开展。
本公开实施例还提供了一种区域确定装置。需要说明的是，该实施例的区域确定装置可以用于执行本公开实施例的区域确定方法。
图10是根据本公开实施例的一种区域确定装置的示意图。如图10所示，该装置包括一个或多个处理器，以及一个或多个存储程序单元的存储器，其中，程序单元由处理器执行，程序单元包括：第一获取单元10、响应单元20、第二获取单元30、第一确定单元40和第二确定单元50。
第一获取单元10，设置为获取对监控区域进行拍摄得到的第一图像。
响应单元20，设置为响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域。
第二获取单元30，设置为获取在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据。
第一确定单元40，设置为根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹。
第二确定单元50，设置为在预设区域中确定出与移动轨迹相交的目标区域，其中，目标区域用于放置目标捕捉装置，目标捕捉装置用于捕捉目标对象。
此处需要说明的是，上述第一获取单元10、响应单元20、第二获取单元30、第一确定单元40和第二确定单元50可以作为装置的一部分运行在终端中，可以通过终端中的处理器来执行上述单元实现的功能，终端也可以是智能手机(如Android手机、iOS手机等)、平板电脑、掌声电脑以及移动互联网设备(Mobile Internet Devices，MID)、PAD等终端设备。
该实施例响应输入的区域设置指令在对监控区域进行拍摄得到的第一图像上确定出区域设置指令所指示的预设区域，根据在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹，进而在预设区域中确定出与移动轨迹相交的目标区域，避免了依赖于防治人员的从业经验和主观判断来确定放置捕捉工具的区域，解决了对用于放置捕捉装置的区域进行确定的准确性低的技术问题，达到了提高对用于放置捕捉装置的区域进行确定的准确性的技术效果。
可选地，本公开实施例还提供了一种存储介质。该存储介质包括存储的程序，其中，在程序运行时控制存储介质所在设备执行本公开实施例中任意一项的区域确定方法。
本申请实施例所提供的各个功能模块可以在区域确定装置或者类似的运算装置中运行，也可以作为存储介质的一部分进行存储。
可选地，在本实施例中，上述存储介质中存储有计算机程序，其中，所述计算机程序被设置为运行时可以用于执行区域确定方法。
图11是根据本公开实施例的一种存储介质的结构示意图。如图11所示，描述了根据本公开的实施方式的程序产品110，其上存储有计算机程序，所述计算机程序被处理器执行时实现如下步骤的程序代码：
获取对监控区域进行拍摄得到的第一图像；响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域；
获取在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据；
根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹；
在预设区域中确定出与移动轨迹相交的目标区域，其中，目标区域用于放置目标捕捉装置，目标捕捉装置用于捕捉目标对象。
可选地，在本实施例中，预设区域为预设的一个或多个区域，所述计算机程序被处理器执行时还实现如下步骤的程序代码：
在一个或多个区域中确定出与移动轨迹相交的目标区域，其中，与目标区域相交的移动轨迹满足预设的目标条件，其中，目标条件根据以下至少之一确定：与目标区域相交的移动轨迹的数量，移动轨迹与目标区域相交的长度，与目标区域相交且在目标区域中存在交叉点的移动轨迹的数量。
可选地，在本实施例中，所述计算机程序被处理器执行时实现如下至少之一步骤的程序代码：
在一个或多个区域中确定出与移动轨迹相交的第一区域，其中，目标区域包括第一区域，与第一区域相交的移动轨迹的数量大于第一阈值；
在一个或多个区域中确定出与移动轨迹相交的第二区域，其中，目标区域包括第二区域，与第二区域相交的移动轨迹的数量大于第二阈值，移动轨迹与第二区域相交的部分的长度大于第三阈值；
在一个或多个区域中确定出与移动轨迹相交的第三区域，其中，目标区域包括第三区域，与第三区域相交、在第三区域中存在交叉点的移动轨迹的数量大于第四阈值。
可选地，在本实施例中，存储介质还可以被设置为确定区域确定方法提供的各种优选地或可选的方法步骤的程序代码。
可选地，本实施例中的具体示例可以参考上述实施例中所描述的示例，本实施例在此不再赘述。
计算机可读存储介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了可读程序代码。这种传播的数据信号可以采用多种形式，包括但不限于电磁信号、光信号或上述的任意合适的组合。计算机可读存储介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。
计算机可读存储介质中包含的程序代码可以用任何适当的介质传输，包括但不限于无线、有线、光缆、射频等等，或者上述的任意合适的组合。
为了实现上述目的，根据本公开的另一方面，本公开实施例还提供了一种处理器。
图12是根据本公开实施例的一种处理器的结构示意图。如图12所示，该处理器120用于运行程序，其中，程序运行时执行权利本公开实施例中任意一项的区域确定方法。
在发明本实施例中，上述处理器120可以执行区域确定方法的运行程序。
可选地，在本实施例中，处理器120可以被设置为执行下述步骤：
获取对监控区域进行拍摄得到的第一图像；
响应输入的区域设置指令在第一图像上确定出区域设置指令所指示的预设区域；
获取在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据；
根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹；
在预设区域中确定出与移动轨迹相交的目标区域，其中，目标区域用于放置目标捕捉装置，目标捕捉装置用于捕捉目标对象。
可选地，在本实施例中，预设区域为预设的一个或多个区域，处理器120还可以被设置为执行下述步骤：
在一个或多个区域中确定出与移动轨迹相交的目标区域，其中，与目标区域相交的移动轨迹满足预设的目标条件，其中，目标条件根据以下至少之一确定：与目标区域相交的移动轨迹的数量，移动轨迹与目标区域相交的长度，与目标区域相交且在目标区域中存在交叉点的移动轨迹的数量。
可选地，在本实施例中，处理器120还可以被设置为执行下述步骤：
在一个或多个区域中确定出与移动轨迹相交的第一区域，其中，目标区域包括第一区域，与第一区域相交的移动轨迹的数量大于第一阈值；
在一个或多个区域中确定出与移动轨迹相交的第二区域，其中，目标区域包括第二区域，与第二区域相交的移动轨迹的数量大于第二阈值，移动轨迹与第二区域相交的部分的长度大于第三阈值；
在一个或多个区域中确定出与移动轨迹相交的第三区域，其中，目标区域包括第三区域，与第三区域相交、在第三区域中存在交叉点的移动轨迹的数量大于第四阈值。
上述处理器120可以通过运行存储在存储器内的软件程序以及模块，从而执行各种功能应用以及数据处理，即实现上述的区域确定方法。
本领域普通技术人员可以理解上述实施例的各种方法中的全部或部分步骤是可以通过程序来指示确定区域确定的装置相关的硬件来完成，该程序可以存储于一确定区域确定装置可读存储介质中，存储介质可以包括：闪存盘、只读存储器(Read-Only Memory，简称为ROM)、随机存取器(Random Access Memory，简称为RAM)、磁盘或光盘等。
如上参照附图以示例的方式描述了根据本公开的区域确定方法、装置、存储介质和处理器。但是，本领域技术人员应当理解，对于上述本公开所提出的区域确定方法、装置、存储介质和处理器，还可以在不脱离本公开内容的基础上做出各种改进。因此，本公开的保护范围应当由所附的权利要求书的内容确定。
显然，本领域的技术人员应该明白，上述的本公开的各模块或各步骤可以用通用的计算装置来实现，它们可以集中在单个的计算装置上，或者分布在多个计算装置所组成的网络上，可选地，它们可以用计算装置可执行的程序代码来实现，从而，可以将它们存储在存储装置中由计算装置来执行，或者将它们分别制作成各个集成电路模块，或者将它们中的多个模块或步骤制作成单个集成电路模块来实现。这样，本公开不限制于任何特定的硬件和软件结合。
以上所述仅为本公开的优选实施例而已，并不用于限制本公开，对于本领域的技术人员来说，本公开可以有各种更改和变化。凡在本公开的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本公开的保护范围之内。
工业实用性
采用获取对监控区域进行拍摄得到的第一图像；响应输入的区域设置指令在第一 图像上确定出区域设置指令所指示的预设区域；获取在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据；根据第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹；在预设区域中确定出与移动轨迹相交的目标区域，其中，目标区域用于放置目标捕捉装置，目标捕捉装置用于捕捉目标对象。也就是说，响应输入的区域设置指令在对监控区域进行拍摄得到的第一图像上确定出区域设置指令所指示的预设区域，根据在第一目标时间段内对监控区域进行拍摄得到的第一组图像数据在监控区域中确定出目标对象在第一目标时间段内的移动轨迹，进而在预设区域中确定出与移动轨迹相交的目标区域，避免了依赖于防治人员的从业经验和主观判断来确定放置捕捉工具的区域，解决了对用于放置捕捉装置的区域进行确定的准确性低的技术问题，达到了提高对用于放置捕捉装置的区域进行确定的准确性的技术效果。

Claims (16)

  1. 一种区域确定方法，包括：
    获取对监控区域进行拍摄得到的第一图像；
    响应输入的区域设置指令在所述第一图像上确定出所述区域设置指令所指示的预设区域；
    获取在第一目标时间段内对所述监控区域进行拍摄得到的第一组图像数据；
    根据所述第一组图像数据在所述监控区域中确定出目标对象在所述第一目标时间段内的移动轨迹；
    在所述预设区域中确定出与所述移动轨迹相交的目标区域，其中，所述目标区域用于放置目标捕捉装置，所述目标捕捉装置用于捕捉所述目标对象。
  2. 根据权利要求1所述的方法，其中，所述预设区域为预设的一个或多个区域，在所述预设区域中确定出与所述移动轨迹相交的目标区域包括：
    在所述一个或多个区域中确定出与所述移动轨迹相交的目标区域，其中，与所述目标区域相交的移动轨迹满足预设的目标条件，其中，所述目标条件根据以下至少之一确定：与所述目标区域相交的移动轨迹的数量，移动轨迹与所述目标区域相交的长度，与所述目标区域相交且在所述目标区域中存在交叉点的移动轨迹的数量。
  3. 根据权利要求2所述的方法，其中，在所述一个或多个区域中确定出与所述移动轨迹相交的所述目标区域以下至少之一：
    在所述一个或多个区域中确定出与所述移动轨迹相交的第一区域，其中，所述目标区域包括所述第一区域，与所述第一区域相交的移动轨迹的数量大于第一阈值；
    在所述一个或多个区域中确定出与所述移动轨迹相交的第二区域，其中，所述目标区域包括所述第二区域，与所述第二区域相交的移动轨迹的数量大于第二阈值，所述移动轨迹与所述第二区域相交的部分的长度大于第三阈值；
    在所述一个或多个区域中确定出与所述移动轨迹相交的第三区域，其中，所述目标区域包括所述第三区域，与所述第三区域相交、在所述第三区域中存在交叉点的移动轨迹的数量大于第四阈值。
  4. 根据权利要求3所述的方法，其中，在所述预设区域中确定出与所述移动轨迹相交的目标区域之后，所述方法还包括以下至少之一：
    在所述目标区域包括所述第一区域的情况下，将用于放置所述目标捕捉装置的目标位置设置为包括所述移动轨迹与所述第一区域相交的部分上的一个或多个位置；
    在所述目标区域包括所述第一区域的情况下，将用于放置所述目标捕捉装置的目标位置设置为第一位置，其中，位于所述第一位置的所述目标捕捉装置覆盖所述第一区域中至少预定数量的移动轨迹；
    在所述目标区域包括所述第二区域的情况下，将用于放置所述目标捕捉装置的目标位置设置为包括所述移动轨迹与所述第二区域相交的部分上的一个或多个位置；
    在所述目标区域包括所述第三区域的情况下，将用于放置所述目标捕捉装置的目标位置设置为一个或多个所述交叉点所在的位置。
  5. 根据权利要求1所述的方法，其中，在所述预设区域中确定出与所述移动轨迹相交的目标区域之后，所述方法还包括：
    在所述第一图像上显示用于标识所述目标区域的第一标识信息。
  6. 权利要求5所述的方法，其中，在所述第一图像上显示用于标识所述预设区域的第一标识信息包括：
    在所述第一图像上显示用于标识所述目标区域的范围的所述第一标识信息。
  7. 根据权利要求1所述的方法，其中，所述预设区域为预设的一个或多个区域，在响应输入的区域设置指令在所述第一图像上确定出所述区域设置指令所指示的预设区域之后，所述方法还包括：
    在所述第一图像上分别显示用于标识所述一个或多个区域的第二标识信息。
  8. 根据权利要求7所述的方法，其中，
    在所述预设区域中确定出与所述移动轨迹相交的目标区域包括：在所述一个或多个区域中确定出与所述移动轨迹相交的目标区域；
    在所述一个或多个区域中确定出与所述移动轨迹相交的目标区域之后，所述方法还包括：保留所述目标区域的所述第二标识信息，隐藏所述一个或多个区域中除所述目标区域之外的区域的所述第二标识信息；和/或，显示用于标识所述目标区域的第一标识信息。
  9. 根据权利要求1所述的方法，其中，在根据所述第一组图像数据在所述监控区域中确定出所述目标对象在所述第一目标时间段内的移动轨迹之后，所述方法还包括：
    在所述第一图像上显示用于标识所述移动轨迹的第三标识信息。
  10. 根据权利要求1至9中任意一项所述的方法，其中，响应输入的区域设置指令在所述第一图像上确定出所述区域设置指令所指示的预设区域包括：
    在所述第一图像中识别出所述监控区域中的目标类型区域，其中，所述目标对象经过所述目标类型区域的概率大于第五阈值；
    响应输入的所述区域设置指令在所述目标类型区域上确定出所述区域设置指令所指示的所述预设区域。
  11. 根据权利要求1至9中任意一项所述的方法，其中，根据所述第一组图像数据在所述监控区域中确定出所述目标对象在所述第一目标时间段内的移动轨迹包括：
    从所述第一组图像数据中识别出所述目标对象在所述监控区域中经过的多个位置；
    通过所述多个位置生成所述移动轨迹，其中，所述多个位置位于所述移动轨迹上。
  12. 根据权利要求1至9中任意一项所述的方法，其中，在所述预设区域中确定出与所述移动轨迹相交的目标区域之后，所述方法还包括：
    在所述目标捕捉装置放置在所述目标区域的情况下，将由所述目标捕捉装置捕捉到的所述目标对象的目标信息发送至目标终端；和/或
    将所述目标区域发送至目标终端；或者
    将所述目标区域和所述监控区域的第二图像发送至目标终端；或者
    在目标终端上显示所述监控区域的第三图像，其中，所述第三图像上显示有所述目标区域；或者
    在目标终端上显示所述监控区域的第四图像，其中，所述第四图像上显示有所述移动轨迹和所述目标区域。
  13. 根据权利要求1至9中任意一项所述的方法，其中，
    获取在第一目标时间段内对所述监控区域进行拍摄得到的第一组图像数据包括：获取摄像设备对所述监控区域拍摄得到的视频文件；对所述视频文件进行抽 帧采样，得到一组视频帧图像的数据，其中，所述第一组图像数据包括所述一组视频帧图像的数据；
    在根据所述第一组图像数据在所述监控区域中确定出目标对象在所述第一目标时间段内的移动轨迹之前，所述方法还包括：根据所述一组视频帧图像中的像素点的像素值在所述一组视频帧图像中确定出多个目标视频帧图像，其中，每个所述目标视频帧图像用于指示在所述监控区域中存在运动的对象；对每个所述目标视频帧图像进行目标对象检测，得到每个所述目标视频帧图像的图像特征，其中，所述图像特征用于表示在所述存在运动的对象中，与所述目标对象之间的相似度大于第六阈值的对象所在的目标图像区域；根据每个所述目标视频帧图像的图像特征确定出运动特征，其中，所述运动特征用于表示所述多个目标视频帧图像中所述存在运动的对象的运动速度和运动方向；根据所述运动特征和每个所述目标视频帧图像的图像特征，确定所述多个目标视频帧图像中是否出现有所述目标对象。
  14. 一种区域确定装置，包括一个或多个处理器，以及一个或多个存储程序单元的存储器，其中，所述程序单元由所述处理器执行，所述程序单元包括：
    第一获取单元，设置为获取对监控区域进行拍摄得到的第一图像；
    响应单元，设置为响应输入的区域设置指令在所述第一图像上确定出所述区域设置指令所指示的预设区域；
    第二获取单元，设置为获取在第一目标时间段内对所述监控区域进行拍摄得到的第一组图像数据；
    第一确定单元，设置为根据所述第一组图像数据在所述监控区域中确定出目标对象在所述第一目标时间段内的移动轨迹；
    第二确定单元，设置为在所述预设区域中确定出与所述移动轨迹相交的目标区域，其中，所述目标区域用于放置目标捕捉装置，所述目标捕捉装置用于捕捉所述目标对象。
  15. 一种存储介质，其特征在于，所述存储介质包括存储的程序，其中，在所述程序运行时控制所述存储介质所在设备执行权利要求1至13中任意一项所述的方法。
  16. 一种处理器，其特征在于，所述处理器用于运行程序，其中，所述程序运行时执行权利要求1至13中任意一项所述的方法。
PCT/CN2019/080746 2019-01-24 2019-04-01 区域确定方法、装置、存储介质和处理器 WO2020151083A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2019554376A JP6949988B2 (ja) 2019-01-24 2019-04-01 領域特定方法、装置、記憶媒体及びプロセッサ

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910068842.3A CN109919966A (zh) 2019-01-24 2019-01-24 区域确定方法、装置、存储介质和处理器
CN201910068842.3 2019-01-24

Publications (1)

Publication Number Publication Date
WO2020151083A1 true WO2020151083A1 (zh) 2020-07-30

Family

ID=66960682

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/080746 WO2020151083A1 (zh) 2019-01-24 2019-04-01 区域确定方法、装置、存储介质和处理器

Country Status (3)

Country Link
JP (1) JP6949988B2 (zh)
CN (1) CN109919966A (zh)
WO (1) WO2020151083A1 (zh)

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111930992A (zh) * 2020-08-14 2020-11-13 腾讯科技（深圳）有限公司 神经网络训练方法、装置及电子设备
CN112040186A (zh) * 2020-08-28 2020-12-04 北京市商汤科技开发有限公司 目标对象的活动区域确定方法、装置、设备及存储介质
CN113505681A (zh) * 2021-07-02 2021-10-15 中标慧安信息技术股份有限公司 用于监测市场内部地面卫生情况的方法
CN113610688A (zh) * 2021-06-25 2021-11-05 格讯科技(深圳)有限公司 一种基于大数据分析的食安监管方法及存储介质
CN113762027A (zh) * 2021-03-15 2021-12-07 北京京东振世信息技术有限公司 一种异常行为的识别方法、装置、设备及存储介质
CN114143712A (zh) * 2021-12-09 2022-03-04 中国地质调查局水文地质环境地质调查中心 一种监测方法、装置及存储介质
CN114821392A (zh) * 2022-03-23 2022-07-29 深圳市鸿彩云实业发展有限公司 一种智能四害监控及防治系统
CN115214430A (zh) * 2022-03-23 2022-10-21 广州汽车集团股份有限公司 车辆座椅调节方法及车辆
CN115457717A (zh) * 2022-09-05 2022-12-09 温州市铁路与轨道交通投资集团有限公司 一种光缆周界安防报警系统
CN115633248A (zh) * 2022-12-22 2023-01-20 浙江宇视科技有限公司 多场景协同检测方法与系统
CN116313018A (zh) * 2023-05-18 2023-06-23 北京大学第三医院（北京大学第三临床医学院） 提高滑雪场与近地医院的急救效率的系统及方法

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110490931A (zh) * 2019-08-20 2019-11-22 上海秒针网络科技有限公司 轨迹生成方法和装置、存储介质及电子装置
CN110662106B (zh) * 2019-09-18 2021-08-27 浙江大华技术股份有限公司 一种录像回放的方法及装置
CN110598670B (zh) * 2019-09-20 2022-03-25 腾讯科技（深圳）有限公司 设置监控区域禁区的方法、装置、存储介质和计算机设备
CN111131770A (zh) * 2019-12-10 2020-05-08 上海秒针网络科技有限公司 目标放置区域的确定方法及装置
CN111127440A (zh) * 2019-12-22 2020-05-08 上海眼控科技股份有限公司 轮胎规格的检测方法、装置、电子设备及可读存储介质
JPWO2021171855A1 (zh) * 2020-02-27 2021-09-02
CN112149561B (zh) * 2020-09-23 2024-04-16 杭州睿琪软件有限公司 图像处理方法和装置、电子设备和存储介质
CN114513625A (zh) * 2020-11-16 2022-05-17 中国石油化工股份有限公司 存储器、罐区泄漏的安全联锁方法、装置及设备
CN112866631B (zh) * 2020-12-30 2022-09-02 杭州海康威视数字技术股份有限公司 一种区域确定方法、系统、装置及电子设备

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130230295A1 (en) * 2001-07-13 2013-09-05 Canon Kabushiki Kaisha Information recording apparatus
CN104954738A (zh) * 2015-04-30 2015-09-30 广州视声光电有限公司 一种移动侦测方法及装置
CN108259830A (zh) * 2018-01-25 2018-07-06 深圳冠思大数据服务有限公司 基于云服务器的鼠患智能监控系统和方法

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20090095083A (ko) * 2008-03-04 2009-09-09 주식회사 세스코 바퀴벌레 추적장치 및 이를 이용한 바퀴벌레 방제방법
JP6159179B2 (ja) * 2013-07-09 2017-07-05 キヤノン株式会社 画像処理装置、画像処理方法
CN106303403B (zh) * 2015-06-12 2019-03-12 中国人民公安大学 监控装置预置位设置、切换方法及系统
CN105279898A (zh) * 2015-10-28 2016-01-27 小米科技有限责任公司 报警方法及装置
CN106651916B (zh) * 2016-12-29 2019-09-03 深圳市深网视界科技有限公司 一种目标的定位跟踪方法及装置
CN108665700B (zh) * 2017-03-31 2020-11-27 杭州海康威视数字技术股份有限公司 侧方停车的状态检测方法及装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130230295A1 (en) * 2001-07-13 2013-09-05 Canon Kabushiki Kaisha Information recording apparatus
CN104954738A (zh) * 2015-04-30 2015-09-30 广州视声光电有限公司 一种移动侦测方法及装置
CN108259830A (zh) * 2018-01-25 2018-07-06 深圳冠思大数据服务有限公司 基于云服务器的鼠患智能监控系统和方法

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111930992A (zh) * 2020-08-14 2020-11-13 腾讯科技（深圳）有限公司 神经网络训练方法、装置及电子设备
CN112040186A (zh) * 2020-08-28 2020-12-04 北京市商汤科技开发有限公司 目标对象的活动区域确定方法、装置、设备及存储介质
CN113762027B (zh) * 2021-03-15 2023-09-08 北京京东振世信息技术有限公司 一种异常行为的识别方法、装置、设备及存储介质
CN113762027A (zh) * 2021-03-15 2021-12-07 北京京东振世信息技术有限公司 一种异常行为的识别方法、装置、设备及存储介质
CN113610688A (zh) * 2021-06-25 2021-11-05 格讯科技(深圳)有限公司 一种基于大数据分析的食安监管方法及存储介质
CN113610688B (zh) * 2021-06-25 2024-03-12 格讯科技(深圳)有限公司 一种基于大数据分析的食安监管方法及存储介质
CN113505681A (zh) * 2021-07-02 2021-10-15 中标慧安信息技术股份有限公司 用于监测市场内部地面卫生情况的方法
CN114143712B (zh) * 2021-12-09 2023-06-16 中国地质调查局水文地质环境地质调查中心 一种监测方法、装置及存储介质
CN114143712A (zh) * 2021-12-09 2022-03-04 中国地质调查局水文地质环境地质调查中心 一种监测方法、装置及存储介质
CN114821392A (zh) * 2022-03-23 2022-07-29 深圳市鸿彩云实业发展有限公司 一种智能四害监控及防治系统
CN115214430B (zh) * 2022-03-23 2023-11-17 广州汽车集团股份有限公司 车辆座椅调节方法及车辆
CN115214430A (zh) * 2022-03-23 2022-10-21 广州汽车集团股份有限公司 车辆座椅调节方法及车辆
CN115457717A (zh) * 2022-09-05 2022-12-09 温州市铁路与轨道交通投资集团有限公司 一种光缆周界安防报警系统
CN115457717B (zh) * 2022-09-05 2024-04-02 温州市铁路与轨道交通投资集团有限公司 一种光缆周界安防报警系统
CN115633248A (zh) * 2022-12-22 2023-01-20 浙江宇视科技有限公司 多场景协同检测方法与系统
CN116313018A (zh) * 2023-05-18 2023-06-23 北京大学第三医院（北京大学第三临床医学院） 提高滑雪场与近地医院的急救效率的系统及方法
CN116313018B (zh) * 2023-05-18 2023-09-15 北京大学第三医院（北京大学第三临床医学院） 用于滑雪场与近地医院的急救系统及方法

Also Published As

Publication number Publication date
JP2021515919A (ja) 2021-06-24
JP6949988B2 (ja) 2021-10-13
CN109919966A (zh) 2019-06-21

Similar Documents

Publication Publication Date Title
WO2020151083A1 (zh) 区域确定方法、装置、存储介质和处理器
CN109886999B (zh) 位置确定方法、装置、存储介质和处理器
WO2020151084A1 (zh) 目标对象的监控方法、装置及系统
CN109922310B (zh) 目标对象的监控方法、装置及系统
CN109886129B (zh) 提示信息生成方法和装置，存储介质及电子装置
KR101168760B1 (ko) 화염 검출 방법 및 장치
WO2021063046A1 (zh) 一种分布式目标监测系统和方法
CN109886555A (zh) 食品安全的监测方法及装置
CN112733690A (zh) 一种高空抛物检测方法、装置及电子设备
JP2011138310A (ja) 有害鳥類の数及び停留位置の確定方法及びその確定装置並びに有害鳥類の防除方法
CN110928282A (zh) 清扫机器人的控制方法及装置
CN109255360A (zh) 一种目标分类方法、装置及系统
Schiano et al. Autonomous detection and deterrence of pigeons on buildings by drones
KR20200009530A (ko) 이상 개체 검출 시스템 및 방법
Amarasinghe et al. Suppressing dengue via a drone system
Bhusal et al. Bird deterrence in a vineyard using an unmanned aerial system (uas)
CN111814665A (zh) 基于宠物情绪识别的陪护方法、装置、服务器及存储介质
JP2012088861A (ja) 侵入物体検出装置
Liu et al. Towards continuous surveillance of fruit flies using sensor networks and machine vision
Dastidar et al. Tracking human intrusion through a CCTV
CN109272535A (zh) 一种基于图像识别的配电房安全区预警方法
WO2021022427A1 (zh) 一种社区流浪动物监测系统及监测方法
Ram et al. A self induced warning system for wild animal trespassing using machine vision system
CN112464714A (zh) 基于视频监控的有害动物检测方法、装置和电子装置
CN103699907A (zh) 基于机器学习的农药喷洒检测方法

Legal Events

Date Code Title Description
ENP Entry into the national phase

Ref document number: 2019554376

Country of ref document: JP

Kind code of ref document: A

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

Ref document number: 19911735

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 11.11.2021)

122 Ep: pct application non-entry in european phase

Ref document number: 19911735

Country of ref document: EP

Kind code of ref document: A1