CN113452903B - Snapshot equipment, snap method and main control chip - Google Patents

Snapshot equipment, snap method and main control chip Download PDF

Info

Publication number
CN113452903B
CN113452903B CN202110669350.7A CN202110669350A CN113452903B CN 113452903 B CN113452903 B CN 113452903B CN 202110669350 A CN202110669350 A CN 202110669350A CN 113452903 B CN113452903 B CN 113452903B
Authority
CN
China
Prior art keywords
snapshot
image
target
detail
camera
Prior art date
Legal status (The legal status 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 status listed.)
Active
Application number
CN202110669350.7A
Other languages
Chinese (zh)
Other versions
CN113452903A (en
Inventor
贾明昊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang Dahua Technology Co Ltd
Original Assignee
Zhejiang Dahua Technology Co Ltd
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 Zhejiang Dahua Technology Co Ltd filed Critical Zhejiang Dahua Technology Co Ltd
Priority to CN202110669350.7A priority Critical patent/CN113452903B/en
Publication of CN113452903A publication Critical patent/CN113452903A/en
Application granted granted Critical
Publication of CN113452903B publication Critical patent/CN113452903B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/80Camera processing pipelines; Components thereof

Landscapes

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

Abstract

The embodiment of the invention provides snapshot equipment, a snapshot method and a main control chip, which can solve the problem that a snapshot image with better image quality cannot be obtained due to randomness of a snapshot process in the prior art. Wherein, the candid photograph equipment includes: the main control chip is used for detecting whether a first snapshot target appears or not based on the panoramic image; when the first snapshot target is determined to appear, controlling a first camera to snapshot the first snapshot target, and obtaining a first snapshot image; when the confidence coefficient of the face detection result of the first snapshot image is lower than a first preset threshold value, controlling at least one camera of the N-1 detail cameras to take a snapshot of the first snapshot target again to obtain a second snapshot image; and comparing the confidence degrees of the face detection results of the second snapshot image and the first snapshot image, and transmitting the snapshot image with the highest confidence degree of the face detection result to the server.

Description

Snapshot equipment, snap method and main control chip
Technical Field
The invention relates to the technical field of video monitoring, in particular to snapshot equipment, a snapshot method and a main control chip.
Background
Video surveillance, an important technical means for modern security management, has been widely applied to various scenes, such as urban traffic management and social security management. In the prior art, a panoramic camera and a variable-magnification rotatable detail camera are generally configured for a video camera, so that a target detected by the panoramic camera is captured by the detail camera, but the capturing time is random, so that it is difficult to ensure that a captured image with better image quality can be obtained.
As can be seen, in the prior art, a captured image with better image quality may not be obtained due to the randomness of the capturing process.
Disclosure of Invention
The embodiment of the invention provides snapshot equipment, a snapshot method and a main control chip, which can solve the problem that a snapshot image with better image quality cannot be obtained due to randomness of a snapshot process in the prior art.
In a first aspect, an embodiment of the present invention provides a capturing device, where the capturing device includes a panoramic camera unit, a first detail camera unit, and a main control chip, where the main control chip is connected to the panoramic camera unit and the first detail camera unit, and the first detail camera unit includes N detail cameras, and a viewing range of the N detail cameras is 360 degrees in total; wherein:
The panoramic camera unit is used for acquiring panoramic images in the corresponding view range;
the main control chip is used for dividing the panoramic image into N sub-areas, establishing corresponding relations between the N sub-areas and the N detail cameras, and detecting whether a first snapshot target appears or not based on the panoramic image; when the first snapshot target is determined to appear, controlling the first detail camera to snapshot the first snapshot target based on the corresponding relation between the N detail cameras and the N sub-areas to obtain a first snapshot image; performing face detection on the first snapshot image, and controlling at least one detail camera in the remaining N-1 detail cameras to perform secondary snapshot on the first snapshot target after a preset time interval when the confidence level of the face detection result of the first snapshot image is lower than a first preset threshold value, so as to obtain at least one second snapshot image; performing face detection on the at least one second captured image; comparing the face detection result of the at least one second snap-shot image with the confidence of the face detection result of each first snap-shot image; and determining an image corresponding to a detection result with highest confidence as a first target snapshot image, and transmitting the first target snapshot image to a server so that the server determines the state of the first snapshot target according to the first target snapshot image, wherein the first detail camera is a detail camera corresponding to a subarea where the first snapshot target is located.
Optionally, the main control chip is specifically configured to:
controlling the N-1 detail cameras to take a candid photograph of the first candid photograph target to obtain N-1 second candid photograph images;
and determining the image with the highest confidence coefficient of the face detection result in the N-1 second snap shot images and the first snap shot images as the first target snap shot image.
Optionally, the main control chip is further configured to:
before at least one of the remaining N-1 detail cameras is controlled to capture the first capture object again, determining whether the first capture object moves from a current subarea to other subareas;
when the first snapshot target is determined to move from the current subarea to other subareas, controlling a second detail camera corresponding to the other subareas to snapshot the first snapshot target, and obtaining a second snapshot image;
and determining the image with the highest confidence coefficient of the face detection result in the second snapshot image and the first snapshot image as the first target snapshot image.
Optionally, the snapshot device further includes: the second detail camera unit is a detail camera with a view range of 360 degrees, and the main control chip is connected with the second detail camera unit; the main control chip is also used for:
When the confidence level of the first target snapshot image is lower than a first preset threshold value, controlling the second detail camera unit to snapshot the first snapshot target again after the preset duration is spaced, and obtaining a third snapshot image;
and determining an image with high confidence of the face detection result in the first target snapshot image and the third snapshot image as a second target snapshot image, and transmitting the second target snapshot image to the server.
Optionally, the main control chip is further configured to:
before controlling a first detail camera to capture the first capture object, determining whether the first detail camera is capturing an image;
and if the first detail camera is capturing images, controlling the second detail camera unit to capture the first capture object.
Optionally, the main control chip is further configured to:
determining whether the second detail camera unit is capturing an image before at least one of the remaining N-1 detail cameras of the control performs a second capture of the first capture object;
and if the second detail camera unit does not capture the image, controlling the second detail camera unit to capture the first capture object.
In a second aspect, an embodiment of the present invention provides a capturing method, where the capturing method is applied to capturing equipment, where the capturing equipment includes a panoramic camera unit, a first detail camera unit, and a main control chip, where the main control chip is connected to the panoramic camera unit and the first detail camera unit, and the first detail camera unit includes N detail cameras, and a view range of the N detail cameras is 360 degrees in total, and the method includes:
acquiring a panoramic image in a view range corresponding to the panoramic camera unit;
dividing the panoramic image into N sub-areas, establishing corresponding relations between the N sub-areas and the N detail cameras, and detecting whether a first snapshot target appears or not based on the panoramic image;
when the first snapshot target is determined to appear, based on the corresponding relation between the N detail cameras and the N sub-areas, controlling the first detail camera to snapshot the first snapshot target to obtain a first snapshot image, wherein the first detail camera is a detail camera corresponding to the sub-area where the first snapshot target is located;
performing face detection on the first snapshot image, and controlling at least one detail camera in the remaining N-1 detail cameras to perform secondary snapshot on the first snapshot target after a preset time interval when the confidence level of the face detection result of the first snapshot image is lower than a first preset threshold value, so as to obtain at least one second snapshot image;
Performing face detection on the at least one second snapshot image, comparing the face detection result of the at least one second snapshot image with the confidence level of the face detection result of each first snapshot image, and determining an image corresponding to the detection result with the highest confidence level as a first target snapshot image;
and transmitting the first target snapshot image to a server so that the server can determine the state of the first snapshot target according to the first target snapshot image.
In a third aspect, an embodiment of the present invention provides a main control chip, where the main control chip is applied to a snapshot device, where the snapshot device includes a panoramic camera unit, a first detail camera unit, and a main control chip, where the main control chip is connected to the panoramic camera unit and the first detail camera unit, the first detail camera unit includes N detail cameras, and a viewing area range of the N detail cameras is 360 degrees, and the main control chip includes:
an acquisition unit configured to acquire a panoramic image in a view range corresponding to the panoramic camera unit;
the detection unit is used for dividing the panoramic image into N sub-areas, establishing corresponding relations between the N sub-areas and the N detail cameras, and detecting whether a first snapshot target appears or not based on the panoramic image;
The snapshot unit is used for controlling the first detail camera to snapshot the first snapshot target based on the corresponding relation between the N detail cameras and the N sub-areas when the first snapshot target exists, so as to obtain a first snapshot image, wherein the first detail camera is a detail camera corresponding to the sub-area where the first snapshot target is located;
the snapshot unit is further configured to perform face detection on the first snapshot image, and when the confidence level of the face detection result of the first snapshot image is lower than a first preset threshold, after a preset time interval, control at least one detail camera of the remaining N-1 detail cameras to perform snapshot again on the first snapshot target, so as to obtain at least one second snapshot image;
the comparison unit is used for carrying out face detection on the at least one second snapshot image, comparing the face detection result of the at least one second snapshot image with the confidence level of the face detection result of each first snapshot image, and determining an image corresponding to the detection result with the highest confidence level as a first target snapshot image;
and the transmission unit is used for transmitting the first target snapshot image to a server so that the server can determine the state of the first snapshot target according to the first target snapshot image.
In a fourth aspect, an embodiment of the present invention provides a snapshot device, where the snapshot device includes a processor and a memory, and the processor is configured to execute a computer program stored in the memory, where the computer program is configured to implement the steps of the method according to the embodiment of the second aspect.
In a sixth aspect, embodiments of the present invention provide a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method according to the embodiments of the second aspect.
In this embodiment of the present application, the snapshot device is configured with panorama camera unit, first detail camera unit and main control chip, first detail camera unit includes N detail camera, the range of vision of N detail camera is 360 degrees altogether. When the snapshot task is executed, the panoramic camera unit can be used for acquiring panoramic images within the view range of the panoramic camera unit, and on the basis, the main control chip can divide the panoramic images into N sub-areas and establish the corresponding relation between N detail cameras and the N sub-areas. If the main control chip detects that the first snapshot target exists in the panoramic image, the first camera corresponding to the subarea where the first snapshot target is currently located can be controlled to snapshot the first snapshot target based on the established corresponding relation between the N detail cameras and the N subareas, so that the first snapshot image is obtained. Performing face detection on the first snap-shot image, and if the confidence coefficient of the detection result is lower than a first preset threshold value, indicating that the image quality of the first snap-shot image is poor; at this time, the main control chip may control at least one of the remaining N-1 detail cameras to take a snapshot again of the first snapshot target, for example, may take a snapshot again after a preset time interval from the time when the first snapshot image is obtained, and obtain at least one second snapshot image. And then comparing the confidence degrees of the face detection results of the first snapshot image and at least one second snapshot image, and transmitting the snapshot image with the highest confidence degree of the face detection result to a server. When the snapshot device determines that the quality of the snapshot image formed by the first camera is poor, at least one camera of the remaining N-1 detail cameras can be controlled to take the same snapshot target again, and the snapshot image with the best image quality is selected from the snapshot images obtained by the two snapshots to transmit to the server, so that the server can more accurately determine the state of the snapshot target based on the snapshot image with the best image quality.
Drawings
Fig. 1 is a schematic side view of a first snapshot device according to an embodiment of the present invention;
fig. 2 is a schematic top view of a first snapshot device according to an embodiment of the present invention;
fig. 3 is a schematic side structural diagram of a second snapshot device according to an embodiment of the present invention;
fig. 4 is a schematic top view of a second snapshot device according to an embodiment of the present invention;
fig. 5 is a schematic side structural diagram of a third snapshot device according to an embodiment of the present invention;
fig. 6 is a schematic top view of a third snapshot device according to an embodiment of the present invention;
fig. 7 is a schematic flow chart of a snapshot method according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of a main control chip according to an embodiment of the present invention;
fig. 9 is a schematic structural diagram of a snapshot device according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent.
Currently, a video camera for capturing images is configured with a panoramic camera unit and a variable magnification, rotatable detail camera, and captures images of an object to be captured detected by the panoramic camera unit using the detail camera. However, because the object to be captured is in a continuously moving state, many factors such as temporary shielding and turning can cause poor image quality of the obtained captured image. And limited by the self snapshot mechanism of the camera, namely, a detail camera can only snapshot the same target to be snapshot once, even if the images of the snapshot images are poor, the snapshot images with poor image quality can only be transmitted to the server, so that the server is not beneficial to determining the state of the target to be snapshot based on the snapshot images.
In view of this, in the embodiment of the present application, the capturing device may simultaneously configure N detail cameras with a view range of 360 degrees. After a first camera of the N detail cameras captures a first captured target, if the acquired captured images are poor in image quality, at least one camera of the remaining N-1 detail cameras can be controlled to capture the first captured target again, and captured images with the best image quality are selected from captured images obtained by capturing twice and transmitted to the server, so that the server can accurately determine the state of the captured target based on the captured images with the best image quality.
The following describes the snapshot device provided by the embodiment of the invention in detail with reference to the drawings.
Example 1
Fig. 1 and fig. 2 are schematic structural diagrams of a first snapshot device according to an embodiment of the present invention. The snapshot device (for example, a video camera) includes a panoramic camera unit 101, a first detail camera unit 102, and a main control chip (not shown in the figure), where the main control chip is respectively connected to the panoramic camera unit 101 and the first detail camera unit 102, and the first detail camera unit 102 includes N detail cameras, and the view ranges of the N detail cameras are 360 degrees in total. In the embodiment of the present application, n=3 is taken as an example, when n=3, the included angle of each detail camera is 120 degrees, and the viewing range is 120 degrees; wherein:
A panoramic camera unit 101 for acquiring a panoramic image within a corresponding view range;
the main control chip is used for dividing the panoramic image into three sub-areas, establishing corresponding relations between the three sub-areas and the three first detail cameras, and detecting whether a first snapshot target appears or not based on the panoramic image; when the first snapshot target is determined to appear, based on the corresponding relation between the three first detail camera units and the three sub-areas in the panoramic image, controlling a first camera corresponding to the sub-area where the first snapshot target is currently located to snapshot the first snapshot target, and obtaining a first snapshot image; when the confidence coefficient of the face detection result of the first snapshot image is lower than a first preset threshold value, after a preset time interval from the time when the first snapshot image is obtained, controlling at least one of the remaining two detail cameras to snapshot the first snapshot target again to obtain at least one second snapshot image; and comparing the confidence degrees of the face detection results of the at least one second snapshot image and the first snapshot image, and transmitting the snapshot image with the highest confidence degree of the face detection result to the server so that the server determines the state of the first snapshot target according to the snapshot image with the highest confidence degree.
In the embodiment of the present invention, the panoramic camera unit 101 may acquire a panoramic image within the view range of itself under the control of the main control chip. It should be understood that a panoramic image is an image with a field of view of 360 degrees. Therefore, the panoramic camera unit 101 may be a camera having one view range of 360 degrees, or may be a camera having two view ranges of 180 degrees each (for example, A1 and A2 in fig. 2 each correspond to a camera having one view range of 180 degrees), and the number of panoramic camera units 101 for acquiring panoramic images is not particularly limited here.
Considering that, in the first detail camera unit 102, any one of the detail cameras can only capture a first capture object that appears in the own view range. Therefore, in order to determine which camera should be controlled to take a snapshot according to the specific position of the first snapshot target in the panoramic image, in the embodiment of the present invention, after the panoramic image is acquired, the main control chip may divide the panoramic image into three sub-areas, and establish the correspondence between the three sub-areas and the three detail cameras of the first detail camera unit. For example, the panoramic image is divided into a sub-area 1, a sub-area 2, and a sub-area 3, the sub-area 1 corresponding to B1 in the first detail camera unit 102, the sub-area 2 corresponding to B2 in the first detail camera unit 102, and the sub-area 3 corresponding to B3 in the first detail camera unit 102. If the main control chip determines that the first snapshot target exists based on the panoramic image, it can further determine which sub-region of the sub-regions 1-3 the first snapshot target is located in, so that based on the established correspondence between the three detail cameras and the three sub-regions, the first camera corresponding to the sub-region where the first snapshot target is currently located can be controlled to snapshot the first snapshot target, and thus the first snapshot image is obtained. It should be appreciated that the first snap shot object, while appearing within the field of view of the panoramic camera unit 101 during movement, may be farther from the panoramic camera unit 101, and may be smaller in the panoramic image acquired by the panoramic camera unit 101. In order to clearly capture the first capture object, the main control chip can be considered to pertinently adjust the magnification of the first camera before controlling the first camera to capture the first capture object, so that the capture requirement of the first capture object when the first capture object appears at any position in the panoramic image is met.
After the first snapshot image is obtained, the main control chip can perform face detection on the first snapshot image based on the face detection model stored by the main control chip, so that a first detection result and a first confidence of the first detection result can be obtained. The first detection result is used for indicating whether a face appears in the first snapshot image, and the first confidence coefficient is used for indicating the probability that the face appearing in the first snapshot image is an actual face. If the first confidence coefficient of the face detection model output the first detection result is higher, the likelihood that the face exists in the first snap image is indicated to be higher. It should be appreciated that in general, the likelihood that a face is detected when there is no occlusion or frontal orientation is high, and the quality of the snap-shot image may be considered to be good; otherwise, the possibility that the face is detected when the face is temporarily blocked or the face is sideways is low, and the quality of the snap-shot image can be considered to be poor. Therefore, the higher the confidence of the face detection result, the better the image quality of the snap-shot image can be considered.
If the main control chip determines that the confidence (for example, the first confidence) of the first detection result of the first captured image is lower than the first preset threshold, the main control chip indicates that the image quality of the first captured image is poor. For example, due to an unreasonable snapshot angle. Considering that the first snapshot target is in a continuously moving state, after a certain distance, the face may face to one of the remaining two detail cameras, and the possibility of capturing the complete face is high. Therefore, in the embodiment of the invention, after a preset time interval from the time when the first snapshot image is obtained, the control chip can control at least one of the remaining two detail cameras to take a snapshot again on the first snapshot target, so as to obtain at least one second snapshot image. For example, the preset time period may be 30s or 60s, and the value of the preset time period is not particularly limited here.
After obtaining at least one second captured image, the main control chip can also perform face detection on the at least one second captured image based on the face detection model stored by the main control chip, and obtain at least one second detection result and a second confidence coefficient corresponding to the at least one second detection result. The main control chip can compare the first confidence coefficient with at least one second confidence coefficient, so that the snap shot image with the highest confidence coefficient is transmitted to the server. That is, the best quality snapshot of the at least two snapshots is transmitted to the server so that the server can determine the status of the first snapshot target based on the best quality snapshot.
The following specifically describes a case of controlling one detail camera and controlling two detail cameras to capture the first capturing object again, respectively.
First case: under the condition that the image quality of the first snapshot image is poor, because the first snapshot target is in a moving state, the sub-region where the first snapshot target is located can be considered to be changed after a preset time interval by taking the time for acquiring the first snapshot image as a reference. At this time, the main control chip can further determine the subarea where the first snapshot target is actually located based on the panoramic image acquired again by the panoramic camera unit, and control the second camera corresponding to the subarea to take the first snapshot target again, so as to obtain a second snapshot image. The confidence of the face detection result of the first snapshot image and the confidence of the face detection result of the second snapshot image can be compared, and then the snapshot image with high confidence of the face detection result can be determined to be the target snapshot image, and the target snapshot image is transmitted to the server.
For example, at the moment of obtaining the first snapshot image, the first snapshot target is located in the sub-area 1, and after a preset time interval, the first snapshot target is located in the sub-area 2, then a second camera corresponding to the sub-area 2 can be controlled to take a snapshot of the first snapshot target again, and a second snapshot image can be obtained. On the basis, the confidence of the face detection result of the first snapshot image and the confidence of the face detection result of the second snapshot image can be compared, so that the snapshot image (for example, the second snapshot image) with high confidence of the face detection result can be determined to be the target snapshot image, and the second snapshot image is transmitted to the server.
Second case: under the condition that the image quality of the first snapshot image is poor, because the first snapshot target is in a moving state, the sub-region where the first snapshot target is located can be considered to be changed after a preset time interval by taking the time for acquiring the first snapshot image as a reference. At the moment, the main control chip can directly control the two remaining detail cameras to simultaneously snapshot the first snapshot target without determining the subarea where the first terrible target is located according to the panoramic image, so that two second snapshot images are obtained. The first snapshot target with the changed position necessarily appears in a certain second snapshot image in the two second snapshot images, so that the confidence degree of the face detection results of the first snapshot image and the two second snapshot images can be compared, and further the snapshot image with the highest confidence degree of the face detection results can be determined and transmitted to the server.
In some embodiments, referring to fig. 3 and 4, the video camera further includes a second detail camera unit 103, where the second detail camera unit 103 is a detail camera with a view range of 360 degrees. If the image quality of the first captured image is poor, and in the moving process of the first captured object, the image quality of at least one second captured image obtained by the remaining two detail cameras is also poor, at this time, the second detail camera unit 103 may be controlled to capture the first captured object again, and obtain a third captured image.
After the third snapshot image is obtained, the main control chip can perform face detection on the third snapshot image based on the face detection model stored by the main control chip, and obtain a detection result and a confidence coefficient corresponding to the detection result, which is not described herein. The main control chip can compare the confidence coefficient of the first target snapshot image with the confidence coefficient of the third snapshot image, so that the snapshot image with high confidence coefficient of the face detection result can be determined to be the second target snapshot image, and the second target snapshot image is transmitted to the server. The first target snapshot image is the second target snapshot image, and the determined snapshot image with high confidence is the second target snapshot image after the first snapshot image is compared with at least one second snapshot image, so that only the confidence of the first target snapshot image and the third snapshot image can be compared in the process.
In other embodiments, the main control chip may further compare the confidence degrees of the first captured image, the at least one second captured image, and the third captured image, determine the captured image with the highest confidence degree as the second target captured image, and transmit the second target captured image to the server.
In one possible implementation, when the main control chip controls the first camera to capture the first capturing object, the first camera is performing a capturing task, for example, capturing the second capturing object, at this time, the first camera cannot perform the task of capturing the first capturing object, so the main control chip may further determine whether the first camera is capturing an image before controlling the first camera to capture the first capturing object, and if the first camera is capturing an image, the main control chip controls the second camera unit 103 to capture the first capturing object.
In one possible implementation, before controlling at least one of the remaining N-1 detail cameras to capture the first capturing object, the main control chip may further determine whether the second detail camera unit 103 is capturing an image, and if the second detail camera unit 103 does not capture an image, may further directly control the second detail camera unit 103 to capture the first capturing object.
In a specific implementation process, the second detail camera unit is used as a replacement camera unit, so that the problem that when two snapshot targets appear in a view range corresponding to the same subarea, one of the snapshot targets cannot be snapshot is solved.
Example two
Fig. 5 and fig. 6 are schematic structural diagrams of a third snapshot device according to an embodiment of the present invention. The video camera includes a panoramic camera unit 101, a third detail camera unit 104, and a main control chip (not shown in the figure). The third detail camera unit 104 includes a first detail camera 1041 and a second detail camera 1042, each of which has a view range of 360 degrees. The main control chip is respectively connected to the panoramic camera unit 101, the first detail camera 1041, and the second detail camera 1042.
A panoramic camera unit 101 for acquiring a panoramic image within a corresponding view range;
the main control chip is used for detecting whether a first snapshot target appears or not based on the panoramic image; when the first snapshot target is determined to appear, the first detail camera 1041 is controlled to snapshot the first snapshot target, so as to obtain a first snapshot image; when the confidence coefficient of the face detection result of the first snapshot image is lower than a first preset threshold value, controlling a second detail camera 1042 to snapshot the first snapshot target again after a preset time interval from the time when the first snapshot image is obtained to obtain a second snapshot image; and comparing the confidence coefficient of the face detection result of the second snapshot image with the confidence coefficient of the face detection result of the first snapshot image, and transmitting the snapshot image with high confidence coefficient of the face detection result in the two snapshot images to the server so that the server determines the state of the first snapshot target according to the snapshot image with high confidence coefficient.
In the embodiment of the present invention, the panoramic camera unit 101 may acquire a panoramic image within the view range of itself under the control of the main control chip. As in the first embodiment, the number of panoramic camera units 101 for acquiring panoramic images is not particularly limited in this embodiment, and will not be described here again.
When the main control chip determines that the first snapshot target exists based on the panoramic image, the first detail camera 1041 can be controlled to snapshot the first snapshot target, and a corresponding first snapshot image is obtained. The first detail camera 1041 and the second detail camera 1042 in this embodiment also support the function of amplifying the snapshot target in the same manner as in the first embodiment, and are not described here again.
After the first snapshot image is obtained, the main control chip can perform face detection on the first snapshot image based on the face detection model stored by the main control chip, so that a first detection result and a first confidence of the first detection result can be obtained.
If the main control chip determines that the first confidence coefficient of the first detection result of the first snapshot image is lower than a first preset threshold value, the image quality of the first snapshot image is poor. For example, due to the temporary occlusion of the face by an obstacle. Considering that the first snapshot target is in a continuously moving state, after a certain distance is moved, the face can be considered not to be blocked, and if the snapshot is performed again, the possibility of capturing the face is high. Therefore, in the embodiment of the present invention, after a preset time interval from the time when the first snapshot image is obtained, the control chip may control the second detail camera 1042 to take a snapshot of the first snapshot target again, and obtain the second snapshot image. The setting of the preset duration may refer to the first embodiment, and will not be described herein.
After the second snapshot image is obtained, the main control chip can also perform face detection on the second snapshot image based on the face detection model stored by the main control chip, and a second detection result and a second confidence coefficient corresponding to the second detection result are obtained. The main control chip can compare the first confidence coefficient with the second confidence coefficient, if the second confidence coefficient is higher than the first confidence coefficient, the main control chip indicates that the image quality of the second snapshot image is better, and the first snapshot image is transmitted to the server at the moment; otherwise, if the second confidence is lower than the first confidence, the image quality of the first captured image is better, and the first captured image is transmitted to the server. That is, the better quality of the two captured images is transmitted to the server, so that the server can determine the state of the first captured object based on the better quality of the captured images.
Referring to fig. 7, based on the same inventive concept, an embodiment of the present invention provides a capturing method, which is applied to a capturing device as shown in fig. 1-2, where the capturing device includes a panoramic camera unit 101, a first detail camera unit 102, and a main control chip (not shown in the figure), the main control chip is respectively connected to the panoramic camera unit 101 and the first detail camera unit 102, the first detail camera unit 102 includes N detail cameras, and the view ranges of the N detail cameras are 360 degrees, and the flow of the method is described as follows:
Step 201: a panoramic image within a field of view corresponding to the panoramic camera unit is acquired.
Step 202: dividing the panoramic image into N sub-areas, establishing corresponding relations between the N sub-areas and the N detail cameras, and detecting whether a first snapshot target appears or not based on the panoramic image.
Step 203: when the first snapshot target is determined to appear, based on the corresponding relation between the N detail cameras and the N sub-areas, controlling the first camera to snapshot the first snapshot target, and obtaining a first snapshot image, wherein the first camera is a detail camera corresponding to the sub-area where the first snapshot target is located.
Step 204: and performing face detection on the first snapshot image, and controlling at least one camera of the remaining N-1 detail cameras to perform secondary snapshot on the first snapshot target after a preset time interval when the confidence of the face detection result of the first snapshot image is lower than a first preset threshold value, so as to obtain at least one second snapshot image.
Step 205: and carrying out face detection on the at least one second snapshot image, comparing the face detection result of the at least one second snapshot image with the confidence of the face detection result of each first snapshot image, and determining the image corresponding to the detection result with the highest confidence as the first target snapshot image.
Step 206: and transmitting the first target snapshot image to a server so that the server can determine the state of the first snapshot target according to the first target snapshot image.
Referring to fig. 8, based on the same inventive concept, an embodiment of the present invention provides a main control chip, where the main control chip is applied to a capturing device as in fig. 1-2, the capturing device includes a panoramic camera unit 101, a first detail camera unit 102, and a main control chip (not shown in the figure), the main control chip is respectively connected to the panoramic camera unit 101 and the first detail camera unit 102, the first detail camera unit 102 includes N detail cameras, and a view range of the N detail cameras is 360 degrees, where the main control chip includes: an acquisition unit 301, a detection unit 302, a snapshot unit 303, a comparison unit 304, and a transmission unit 305.
An acquisition unit 301 for acquiring a panoramic image in a view range corresponding to the panoramic camera unit 101;
the detecting unit 302 is configured to divide the panoramic image into N sub-areas, establish correspondence between the N sub-areas and the N detail cameras, and detect whether a first capturing object appears based on the panoramic image.
And the snapshot unit 303 is configured to, when determining that a first snapshot target exists, control a first camera to snapshot the first snapshot target based on the correspondence between the N detail cameras and the N sub-areas, so as to obtain a first snapshot image, where the first camera is a detail camera corresponding to the sub-area where the first snapshot target is located.
The snapshot unit 303 is further configured to perform face detection on the first snapshot image, and when the confidence level of the face detection result of the first snapshot image is lower than a first preset threshold, after a preset time interval, control at least one of the remaining N-1 detail cameras to perform secondary snapshot on the first snapshot target, so as to obtain at least one second snapshot image.
Comparison unit 304: the face detection method is used for carrying out face detection on the at least one second snapshot image, comparing the face detection result of the at least one second snapshot image with the confidence level of the face detection result of each first snapshot image, and determining the image corresponding to the detection result with the highest confidence level as a first target snapshot image.
And the transmission unit 305 is configured to transmit the first target snapshot image to a server, so that the server determines a state of the first snapshot target according to the first target snapshot image.
Referring to fig. 9, based on the same inventive concept, an embodiment of the present invention provides a capturing apparatus, where the capturing apparatus includes at least one processor 401, and the processor 401 is configured to execute a computer program stored in a memory, to implement steps of a capturing method according to the embodiment of the present invention, where the steps are shown in fig. 7.
Alternatively, the processor 401 may be a central processing unit, a specific ASIC, or may be one or more integrated circuits for controlling the execution of a program.
Optionally, the main control chip may further include a memory 402 connected to the at least one processor 401, and the memory 402 may include ROM, RAM, and disk memory. The memory 402 is used for storing data required by the processor 401 when running, i.e. instructions executable by at least one processor 401, the at least one processor 401 executing the method as shown in fig. 7 by executing the instructions stored by the memory 402. Wherein the number of memories 402 is one or more. The memory 402 is also shown in fig. 9, but it should be noted that the memory 402 is not an essential functional block, and is therefore shown in fig. 9 by a broken line.
The physical devices corresponding to the acquiring unit 301, the detecting unit 302, the capturing unit 303, the comparing unit 304, and the transmitting unit 305 may be the aforementioned processor 401. The snapshot device may be used to perform the method provided by the embodiment shown in fig. 7. Therefore, regarding the functions that can be implemented by each functional module in the snapshot device, reference may be made to the corresponding description in the embodiment shown in fig. 7, which is not repeated.
Embodiments of the present invention also provide a computer storage medium storing computer instructions that, when executed on a computer, cause the computer to perform a method as described in fig. 7.
It will be apparent to those skilled in the art that various modifications and variations can be made in the present application without departing from the spirit or scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims and the equivalents thereof, the present application is intended to cover such modifications and variations.

Claims (9)

1. The snapshot equipment is characterized by comprising a panoramic camera unit, a first detail camera unit, a second detail camera unit and a main control chip, wherein the main control chip is respectively connected with the panoramic camera unit, the first detail camera unit and the second detail camera unit, the first detail camera unit comprises N detail cameras, the view ranges of the N detail cameras are 360 degrees, and the second detail camera unit is a detail camera with the view range of 360 degrees; wherein:
the panoramic camera unit is used for acquiring panoramic images in the corresponding view range;
The main control chip is used for dividing the panoramic image into N sub-areas, establishing corresponding relations between the N sub-areas and the N detail cameras, and detecting whether a first snapshot target appears or not based on the panoramic image; when the first snapshot target is determined to appear, controlling a first camera to snapshot the first snapshot target based on the corresponding relation between the N detail cameras and the N sub-areas, so as to obtain a first snapshot image; performing face detection on the first snapshot image, and controlling at least one camera of the remaining N-1 detail cameras to perform secondary snapshot on the first snapshot target after a preset time interval when the confidence level of the face detection result of the first snapshot image is lower than a first preset threshold value, so as to obtain at least one second snapshot image; performing face detection on the at least one second captured image; comparing the face detection result of the at least one second snap-shot image with the confidence of the face detection result of each first snap-shot image; determining an image corresponding to a detection result with highest confidence as a first target snapshot image, and transmitting the first target snapshot image to a server so that the server determines the state of the first snapshot target according to the first target snapshot image, wherein the first camera is a detail camera corresponding to a subarea where the first snapshot target is located;
The main control chip is further used for determining whether the first camera is capturing images before controlling the first camera to capture the first capturing target; and if the first camera is capturing images, controlling the second detail camera unit to capture the first capturing target.
2. The snapshot device of claim 1, wherein the main control chip is specifically configured to:
controlling the N-1 detail cameras to take a candid photograph of the first candid photograph target to obtain N-1 second candid photograph images;
and determining the image with the highest confidence coefficient of the face detection result in the N-1 second snap shot images and the first snap shot images as the first target snap shot image.
3. A snapshot device as recited in claim 1, wherein said main control chip is further configured to:
determining whether the first snapshot target moves from a current subarea to other subareas before at least one camera of the remaining N-1 detail cameras is controlled to snapshot the first snapshot target again;
when the first snapshot target is determined to move from the current subarea to other subareas, controlling a second camera corresponding to the other subareas to snapshot the first snapshot target, and obtaining a second snapshot image;
And determining the image with high confidence of the face detection result in the second snapshot image and the first snapshot image as the first target snapshot image.
4. A snapshot device as recited in claim 1, wherein said main control chip is further configured to:
when the confidence level of the first target snapshot image is lower than a first preset threshold value, controlling the second detail camera unit to snapshot the first snapshot target again after the preset duration is spaced, and obtaining a third snapshot image;
and determining an image with the highest confidence coefficient of the face detection result in the first target snapshot image and the third snapshot image as a second target snapshot image, and transmitting the second target snapshot image to the server.
5. A snapshot device as recited in claim 1, wherein said main control chip is further configured to:
determining whether the second detail camera unit is capturing an image before at least one of the remaining N-1 detail cameras of the control takes a second capture of the first capture object;
and if the second detail camera unit does not capture the image, controlling the second detail camera unit to capture the first capture object.
6. The utility model provides a snap shot method which characterized in that is applied to snap shot equipment, snap shot equipment includes panorama camera unit, first detail camera unit and main control chip, main control chip connects respectively panorama camera unit first detail camera unit, first detail camera unit includes N detail cameras, the field of vision scope of N detail cameras is 360 degrees altogether, the method includes:
acquiring a panoramic image in a view range corresponding to the panoramic camera unit;
dividing the panoramic image into N sub-areas, establishing corresponding relations between the N sub-areas and the N detail cameras, and detecting whether a first snapshot target appears or not based on the panoramic image;
when the first snapshot target is determined to appear, based on the corresponding relation between the N detail cameras and the N sub-areas, controlling a first camera to snapshot the first snapshot target to obtain a first snapshot image, wherein the first camera is a detail camera corresponding to the sub-area where the first snapshot target is located;
performing face detection on the first snapshot image, and controlling at least one camera of the remaining N-1 detail cameras to perform secondary snapshot on the first snapshot target after a preset time interval when the confidence level of the face detection result of the first snapshot image is lower than a first preset threshold value, so as to obtain at least one second snapshot image;
Performing face detection on the at least one second snapshot image, comparing the face detection result of the at least one second snapshot image with the confidence level of the face detection result of each first snapshot image, and determining an image corresponding to the detection result with the highest confidence level as a first target snapshot image;
and transmitting the first target snapshot image to a server so that the server can determine the state of the first snapshot target according to the first target snapshot image.
7. The utility model provides a master control chip, its characterized in that is applied to snapshot equipment, snapshot equipment includes panorama camera unit, first detail camera unit, second detail camera unit and master control chip, master control chip connects respectively panorama camera unit first detail camera unit with second detail camera unit, first detail camera unit includes N detail camera, the view scope of N detail camera is 360 degrees altogether, and second detail camera unit is 360 degrees detail camera of view scope, master control chip includes:
an acquisition unit configured to acquire a panoramic image in a view range corresponding to the panoramic camera unit;
The detection unit is used for dividing the panoramic image into N sub-areas, establishing corresponding relations between the N sub-areas and the N detail cameras, and detecting whether a first snapshot target appears or not based on the panoramic image;
the snapshot unit is used for controlling a first camera to snapshot the first snapshot target based on the corresponding relation between the N detail cameras and the N sub-areas when the first snapshot target exists, so as to obtain a first snapshot image, wherein the first camera is a detail camera corresponding to the sub-area where the first snapshot target is located;
the snapshot unit is further configured to perform face detection on the first snapshot image, and when the confidence level of the face detection result of the first snapshot image is lower than a first preset threshold, after a preset time interval, control at least one of the remaining N-1 detail cameras to perform secondary snapshot on the first snapshot target, so as to obtain at least one second snapshot image;
the comparison unit is used for carrying out face detection on the at least one second snapshot image, comparing the face detection result of the at least one second snapshot image with the confidence level of the face detection result of each first snapshot image, and determining an image corresponding to the detection result with the highest confidence level as a first target snapshot image;
The transmission unit is used for transmitting the first target snapshot image to a server so that the server can determine the state of the first snapshot target according to the first target snapshot image;
the snapshot unit is further used for determining whether the first camera is capturing images before controlling the first camera to capture the first snapshot target; and if the first camera is capturing images, controlling the second detail camera unit to capture the first capturing target.
8. A snapshot device comprising at least one processor and a memory coupled to the at least one processor, the at least one processor for implementing the steps of the method of claim 6 when executing a computer program stored in the memory.
9. A computer readable storage medium having stored thereon a computer program, the computer program being executed by a processor to perform the steps of the method of claim 6.
CN202110669350.7A 2021-06-17 2021-06-17 Snapshot equipment, snap method and main control chip Active CN113452903B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110669350.7A CN113452903B (en) 2021-06-17 2021-06-17 Snapshot equipment, snap method and main control chip

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110669350.7A CN113452903B (en) 2021-06-17 2021-06-17 Snapshot equipment, snap method and main control chip

Publications (2)

Publication Number Publication Date
CN113452903A CN113452903A (en) 2021-09-28
CN113452903B true CN113452903B (en) 2023-07-11

Family

ID=77811853

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110669350.7A Active CN113452903B (en) 2021-06-17 2021-06-17 Snapshot equipment, snap method and main control chip

Country Status (1)

Country Link
CN (1) CN113452903B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115361504B (en) * 2022-10-24 2023-04-28 中国铁塔股份有限公司 Monitoring video processing method and device and electronic equipment

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2018186514A (en) * 2018-06-13 2018-11-22 パナソニックIpマネジメント株式会社 Image processing device, imaging system having the same and image processing method
CN109151375A (en) * 2017-06-16 2019-01-04 杭州海康威视数字技术股份有限公司 A kind of target object grasp shoot method, device and video monitoring equipment
WO2019114617A1 (en) * 2017-12-12 2019-06-20 华为技术有限公司 Method, device, and system for fast capturing of still frame
CN110490041A (en) * 2019-05-31 2019-11-22 杭州海康威视数字技术股份有限公司 Human face image collecting device and method
CN111372037A (en) * 2018-12-25 2020-07-03 杭州海康威视数字技术股份有限公司 Target snapshot system and method

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003284053A (en) * 2002-03-27 2003-10-03 Minolta Co Ltd Monitoring camera system and monitoring camera control device
CN102025900A (en) * 2009-09-18 2011-04-20 鸿富锦精密工业(深圳)有限公司 Panoramic camera and target panoramic image monitoring method thereof
CN202443476U (en) * 2012-02-20 2012-09-19 华焦宝 Multiple camera human face recognition system
CN107016322B (en) * 2016-01-28 2020-01-14 浙江宇视科技有限公司 Method and device for analyzing followed person
CN108419014B (en) * 2018-03-20 2020-02-21 北京天睿空间科技股份有限公司 Method for capturing human face by linkage of panoramic camera and multiple capturing cameras
US11553129B2 (en) * 2020-06-22 2023-01-10 Intel Corporation Camera orchestration technology to improve the automated identification of individuals
CN111914781B (en) * 2020-08-10 2024-03-19 杭州海康威视数字技术股份有限公司 Face image processing method and device
CN112911249B (en) * 2021-03-23 2023-03-24 浙江大华技术股份有限公司 Target object tracking method and device, storage medium and electronic device

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109151375A (en) * 2017-06-16 2019-01-04 杭州海康威视数字技术股份有限公司 A kind of target object grasp shoot method, device and video monitoring equipment
WO2019114617A1 (en) * 2017-12-12 2019-06-20 华为技术有限公司 Method, device, and system for fast capturing of still frame
JP2018186514A (en) * 2018-06-13 2018-11-22 パナソニックIpマネジメント株式会社 Image processing device, imaging system having the same and image processing method
CN111372037A (en) * 2018-12-25 2020-07-03 杭州海康威视数字技术股份有限公司 Target snapshot system and method
CN110490041A (en) * 2019-05-31 2019-11-22 杭州海康威视数字技术股份有限公司 Human face image collecting device and method

Also Published As

Publication number Publication date
CN113452903A (en) 2021-09-28

Similar Documents

Publication Publication Date Title
CN110163885B (en) Target tracking method and device
JP4241742B2 (en) Automatic tracking device and automatic tracking method
US11107246B2 (en) Method and device for capturing target object and video monitoring device
CN113870550B (en) Regional anomaly detection method and system based on edge calculation
CN105898107B (en) A kind of target object grasp shoot method and system
CN112770056B (en) Shooting method, shooting device and electronic equipment
CN105979143B (en) Method and device for adjusting shooting parameters of dome camera
CN111277760B (en) Shooting composition method, terminal and storage medium
CN111083358A (en) Target snapshot method, device and system
JP2017228082A (en) Tracking device, tracking method, and program
CN112487888B (en) Image acquisition method and device based on target object
CN113452903B (en) Snapshot equipment, snap method and main control chip
CN107395972B (en) Shooting method and terminal for fast moving object
CN110248148B (en) Method and device for determining positioning parameters
WO2019196240A1 (en) Photographing method, apparatus, computer device, and storage medium
JP6799325B2 (en) Image correction device, image correction method, attention point recognition device, attention point recognition method and abnormality detection system
JP5360403B2 (en) Mobile imaging device
JP5127692B2 (en) Imaging apparatus and tracking method thereof
US20120026292A1 (en) Monitor computer and method for monitoring a specified scene using the same
US10748397B2 (en) Monitoring system
CN115550555B (en) Holder calibration method and related device, camera device and storage medium
CN106851052A (en) A kind of control method and electronic equipment
CN115601738B (en) Parking information acquisition method, device, equipment, storage medium and program product
CN110248090A (en) Control method and control device
CN112887605B (en) Image anti-shake method and device and electronic equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant