Flexible wake-up method and system based on multi-focal-length multi-view camera group

Abstract

The invention discloses a flexible wake-up method based on a multi-focal-length multi-view camera group, the multi-focal-length multi-view camera group comprises a first detection camera, a first snapshot camera and a second snapshot camera, the focal length of the first snapshot camera is smaller than that of the second snapshot camera, and the method comprises the following steps: S101, shooting a preset area by using the first detection camera, when it is identified that the target exists in the preset area, executing a step S102 or a step S103; s102, when the pixel number of all the targets is larger than or equal to a preset pixel number threshold value, the first snapshot camera is awakened for snapshot; when the pixel number of all the targets is smaller than a preset pixel number threshold value, a second snapshot camera is awakened for snapshot; otherwise, awakening the first snapshot camera and the second snapshot camera for snapshot; s103, a short-distance area and a long-distance area are divided, and when the positions of all the targets are in the short-distance area, the first snapshot camera is awakened for snapshot; when the positions of all the targets are in the long-distance area, waking up a second snapshot camera for snapshot; otherwise, the first snapshot camera and the second snapshot camera are awakened for snapshot.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring technology, and in particular to a flexible wake-up method and system based on a multi-focal-segment multi-eye camera group. Background Art

[0002] As we all know, with the widespread application of intelligent monitoring systems in security, traffic management and other fields, target detection and ranging technology based on camera visual recognition has become the core means to achieve intelligent perception, especially in high-voltage power transmission grid scenarios. Cameras are often used to intelligently monitor target scenes, and after determining that a target appears in the target scene, the camera is used to capture and analyze it to determine whether the target has a risk of invading the high-voltage transmission grid and avoid safety problems.

[0003] The inventors have found that in the current existing real-time monitoring system for power transmission lines, the following two schemes are mainly used for monitoring: Solution 1: Use high-resolution cameras to directly perform full-time monitoring tasks and simultaneously complete target capture and analysis functions. However, when this solution is actually applied, high-resolution cameras need to continuously run image processing algorithms, and their hardware power consumption is significantly higher than that of low-resolution devices. Especially in multi-camera deployment scenarios, the overall energy consumption of the system increases exponentially. In addition, the camera in this solution is always in a high-load state, resulting in a shortened device life and increased maintenance costs. In addition, when collecting multi-focal length images, all high-resolution cameras will capture the target synchronously after it appears, and each focal length will be measured once, resulting in excessive power consumption.

[0004] Solution 2: Use a low-resolution short-focus camera for continuous monitoring, and wake up all high-resolution cameras for capture and analysis when a target is detected. Although this solution reduces the power consumption of normal monitoring, it still has significant defects: First, the limited field of view and resolution of the short-focus camera result in insufficient detection capabilities for long-distance or small-sized targets, which can easily lead to missed detections and problems where targets are too far away to be detected; second, all high-resolution cameras are forced to wake up after the target is triggered. Regardless of the actual spatial position of the target, multiple cameras are required to perform ranging calculations synchronously, resulting in redundant power consumption. For example, when the target is only within the effective range of a certain high-resolution camera, the activation of the remaining cameras will result in ineffective energy consumption.

[0005] It can be seen that the key contradiction of the existing technology is the failure to achieve dynamic coordination between target position perception and camera resource scheduling. The multi-focal-length and multi-eye camera wake-up strategy does not match the actual target needs, resulting in low hardware resource utilization. Its specific manifestation is: there is a lack of real-time judgment mechanism for the spatial distribution of the target, and it is impossible to adaptively select the snapshot camera with the optimal focal length for snapshot analysis according to the target. Summary of the invention

[0006] The technical problem to be solved by the present invention is: the present invention discloses a flexible wake-up method and system based on a multi-focal-length multi-eye camera group to solve the problem that the optimal focal-length camera cannot be selected for snapshot analysis due to the lack of a real-time judgment mechanism for the target spatial distribution in the existing camera wake-up strategy.

[0007] In order to solve the above technical problems, the technical solution adopted by the present invention is: a flexible wake-up method based on a multi-focal-segment multi-eye camera group, wherein the multi-focal-segment multi-eye camera group includes a first detection camera, a first snapshot camera and a second snapshot camera, wherein the focal length of the first snapshot camera is smaller than that of the second snapshot camera, and the flexible wake-up method includes the steps of: S101: photographing a preset area using the first detection camera, and analyzing the photographed image information, so as to execute step S102 or step S103 when a target is identified in the preset area; S102: Obtain the number of pixels of all targets in the image information, and when the number of pixels of all targets is ≥ a preset pixel number threshold, wake up the first snapshot camera to capture; when the number of pixels of all targets is < the preset pixel number threshold, wake up the second snapshot camera to capture; otherwise, wake up the first snapshot camera and the second snapshot camera at the same time to capture; S103: Divide the image information into a short-distance area and a long-distance area, and obtain the positions of all targets in the image information. When the positions of all targets are in the short-distance area, wake up the first snapshot camera to capture; when the positions of all targets are in the long-distance area, wake up the second snapshot camera to capture; otherwise, wake up the first snapshot camera and the second snapshot camera at the same time to capture.

[0008] The study found that the existing camera wake-up strategy lacks a real-time judgment mechanism for the target's spatial distribution, which makes it impossible to select a high-resolution snapshot camera with the optimal focal length in a multi-focal length multi-eye camera group for snapshot analysis. As a result, all high-resolution snapshot cameras are often forced to wake up after the target is triggered. Regardless of the target's actual spatial position, multiple cameras are required to perform snapshot analysis synchronously, resulting in redundant power consumption.

[0009] To this end, in the above-mentioned technical scheme of the present invention, the present invention discloses a new flexible wake-up method based on a multi-focal-length multi-eye camera group. The flexible wake-up method can shoot a preset area through a first detection camera, and analyze the captured image information, so as to adaptively select a first capture camera and / or a second capture camera with a suitable focal length for capture through preliminary judgment.

[0010] Specifically, in actual application, the above-mentioned first detection camera is a low-power camera, which can run all day and shoot the preset area, perform target analysis and preliminary judgment in real time, and select the appropriate first capture camera and / or second capture camera for capturing according to the preliminary judgment, so as to avoid the high-power consumption first capture camera and / or second capture camera working all day and reduce power consumption.

[0011] Correspondingly, when the above-mentioned first detection camera performs preliminary judgment, it can select the corresponding snapshot camera for snapshot according to the difference in the number of target pixels and the target position in the acquired image information, so as to flexibly wake up the above-mentioned first snapshot camera and / or the second snapshot camera, which can effectively reduce resource waste and reduce power consumption, and has good promotion prospects and application value.

[0012] Furthermore, in the flexible wake-up method based on a multi-focal-segment multi-camera group according to the present invention, in step S101, specifically: Using the first detection camera to shoot the power transmission scene, setting a warning area in the power transmission scene, and using the warning area as the preset area; When it is identified that there is a target in the preset area, the dynamics of the target are continuously tracked; and when it is detected that the target is moving in the direction of the power transmission cable in the power transmission scene, step S102 or step S103 is executed.

[0013] In the above technical solution of the present invention, in actual application, in order to further reduce power consumption and reduce the snapshot situations of the first snapshot camera and / or the second snapshot camera under unnecessary circumstances, when the above flexible wake-up method is actually applied to the power transmission scenario, even if the first detection camera recognizes the target in the preset area, that is, recognizes the existence of the target in the warning distinction, it will not immediately execute the above step S102 or step S103 to wake up the snapshot camera for snapshot.

[0014] In some preferred embodiments, the first detection camera will continuously track the dynamics of the target in a preset area in the power transmission scene for a period of time, and only execute the process of waking up the snapshot camera in the above step S102 or step S103 when it detects that the target is moving in the direction of the transmission cable in the power transmission scene; when it is detected that the target is moving in the preset area in the direction away from the transmission cable, the process in the above step S102 or step S103 is not executed.

[0015] Furthermore, in the flexible wake-up method based on the multi-focal-segment multi-camera group described in the present invention, in step S102 or step S103, the first snapshot camera and / or the second snapshot camera captures and measures the distance between the target and the transmission cable.

[0016] In the above technical solution of the present invention, when the above-mentioned first snapshot camera and / or second snapshot camera is actually used for snapshot, the snapshot image can be effectively obtained, and the target can be detected from the snapshot image, and the distance between the target and the transmission cable can be obtained at the same time, so as to provide early warning to remind power grid workers.

[0017] Furthermore, in the flexible wake-up method based on the multi-focal-segment multi-eye camera group described in the present invention, the multi-focal-segment multi-eye camera group also includes a second detection camera, and the focal lengths of the first detection camera and the second detection camera are different. The flexible wake-up method also includes the steps of: S201: Establishing a matching relationship between the first snapshot camera, the second snapshot camera and the first detection camera, the second detection camera; S202: photographing a preset area using the first detection camera and the second detection camera, and analyzing the photographed image information; S203: When the first detection camera or the second detection camera recognizes that there is a target in the preset area, the corresponding first capture camera or the second capture camera is awakened to capture.

[0018] In the above technical solution of the present invention, the inventors have found that when using a single low-resolution short-focus camera for monitoring and shooting, there may be a problem that the target cannot be detected because it is too far away; in order to further improve the detection capability of the multi-focal-length multi-eye camera, in actual application, the multi-focal-length multi-eye camera group can also be set to also include a second detection camera, and the above-mentioned first detection camera and second detection camera are used to simultaneously shoot and monitor the preset area, so as to improve its own detection capability and improve the accuracy of detection.

[0019] To this end, when actually constructing the above-mentioned flexible wake-up method of the present invention, a matching relationship between the first snapshot camera, the second snapshot camera and the first detection camera, the second detection camera can also be established, and the first detection camera and the second detection camera can be used to shoot the preset area at the same time. When any one of the two detection cameras recognizes the target in the preset area and the other camera does not detect the target, the first snapshot camera or the second snapshot camera corresponding to the detection camera that recognizes the target is awakened to take the snapshot, ensuring that a snapshot camera with a suitable focal length can be selected to complete the capture of the target.

[0020] Furthermore, in the flexible wake-up method based on a multi-focal-segment multi-camera group according to the present invention, in step S202, the following steps are further included: Calibrate the position relationship between the first detection camera and the second detection camera to obtain the coordinate mapping relationship of the pixel points on the two cameras.

[0021] Furthermore, in the flexible wake-up method based on a multi-focal-segment multi-camera group according to the present invention, in step S203, the following steps are further included: When the first detection camera and the second detection camera simultaneously identify that there is a target in the preset area, obtain masks of each target, and match the targets detected by the first detection camera and the second detection camera based on the coordinate mapping relationship; When there is a one-to-one correspondence between the objects detected by the first detection camera and the objects detected by the second detection camera, executing steps S101-S103; When there is no one-to-one correspondence between the objects detected by the first detection camera and the objects detected by the second detection camera, the first capture camera and the second capture camera are simultaneously awakened to capture.

[0022] In the above technical solution of the present invention, when the first detection camera and the second detection camera are used to monitor the preset area at the same time, it is inevitable that the first detection camera and the second detection camera will detect the target at the same time. At this time, the targets detected in the first detection camera and the second detection camera can be matched according to the coordinate mapping relationship of the pixel points on the two cameras, so as to execute a suitable wake-up capture strategy according to the matching results to reduce power consumption.

[0023] Furthermore, in the flexible wake-up method based on a multi-focal-segment multi-camera group described in the present invention, the focal length of the first detection camera is smaller than that of the second detection camera.

[0024] In the above technical solution of the present invention, in order to achieve better monitoring effects, in actual applications, the first detection camera and the second detection camera can also select lenses with different focal lengths to obtain a more comprehensive monitoring effect and improve the accuracy and reliability of the above flexible wake-up method.

[0025] Furthermore, in the flexible wake-up method based on a multi-focal-segment multi-eye camera group described in the present invention, in step S201, the focal length, sensor target surface size and field of view angle of the first detection camera, the second detection camera, the first snapshot camera and the second snapshot camera are obtained, so that for the same field of view angle when the sensor ratio is consistent, the relationship between the focal length and the sensor target surface size is f1 / D1=f2 / D2, and the matching relationship between the first snapshot camera, the second snapshot camera and the first detection camera, the second detection camera is established; wherein f1 and D1 are the focal length and sensor target surface size of the first detection camera or the second detection camera, respectively; f2 and D2 are the focal length and sensor target surface size of the first snapshot camera or the second snapshot camera, respectively.

[0026] Accordingly, another object of the present invention is to disclose a flexible wake-up system based on a multi-focal-segment multi-eye camera group, which can simply execute the above-mentioned flexible wake-up method of the present invention to realize flexible wake-up of a first snapshot camera and a second snapshot camera with different focal lengths, which specifically includes: A multi-focal-length multi-eye camera group, a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, each step of a flexible wake-up method based on a multi-focal-length multi-eye camera group as described above in the present invention is implemented.

[0027] Further, in the flexible wake-up system based on the multi-focal-segment multi-eye camera group described in the present invention, the multi-focal-segment multi-eye camera group includes a main device and a sub-device, the first detection camera, the first snapshot camera, and the second snapshot camera are arranged in the main device, and the sub-device includes a plurality of sub-device snapshot cameras having the same focal length as each snapshot camera in the main device; Among them, the processor controls the main device to execute each step of a flexible wake-up method based on a multi-focal-length multi-eye camera group as described in any one of claims 1-9, and when controlling the first snapshot camera and / or the second snapshot camera of the main device to capture, synchronously wakes up the secondary device snapshot camera with the same focal length in the secondary device to capture.

[0028] The beneficial effects of the present invention are as follows: the present invention discloses a flexible wake-up method based on a multi-focal-segment multi-eye camera group, which can shoot a preset area through a first detection camera, and analyze the captured image information, so as to adaptively select a snapshot camera with a suitable focal length for snapshot through preliminary judgment according to the difference in the number of target pixels and the target position in the image information, so as to flexibly wake up the above-mentioned first snapshot camera and / or the second snapshot camera, which can effectively reduce resource waste and reduce power consumption, and has good promotion prospects and application value.

[0029] Correspondingly, a flexible wake-up system based on a multi-focal-segment multi-eye camera group disclosed in the present invention can effectively execute the above-mentioned flexible wake-up method of the present invention, and it also has the above-mentioned advantages and beneficial effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 The present invention is a flowchart of the steps of a flexible wake-up method based on a multi-focal-segment multi-camera group in one embodiment. DETAILED DESCRIPTION

[0031] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in conjunction with the implementation modes and the accompanying drawings.

[0032] See also Figure 1As shown, the present invention specifically discloses a new flexible wake-up method based on a multi-focal-length multi-eye camera group; and, in this embodiment, the multi-focal-length multi-eye camera group specifically includes a first detection camera, a second detection camera, a first snapshot camera and a second snapshot camera; wherein, the focal length of the first snapshot camera is smaller than the focal length of the second snapshot camera, and the focal length of the first detection camera is smaller than the second detection camera.

[0033] The reason for setting up two detection cameras with different focal lengths is that when using a single low-resolution short-focus camera for monitoring and shooting, there may be a problem that the target is too far away to be detected due to the single focal length.

[0034] Therefore, in this embodiment, in order to further improve the detection capability of the multi-focal-length multi-eye camera, in actual application, the first detection camera and the second detection camera can also select lenses with different focal lengths, and the multi-focal-length multi-eye camera group is set to simultaneously use the first detection camera and the second detection camera to shoot and monitor the preset area, so as to improve its own detection capability, obtain a more comprehensive monitoring effect, and improve the accuracy and reliability of detection.

[0035] To this end, based on the camera settings of the multi-focal-segment multi-eye camera group, when actually constructing the flexible wake-up method of the present invention, the following steps may be specifically included: S201: Establishing a matching relationship between a first snapshot camera, a second snapshot camera, and a first detection camera, and a second detection camera.

[0036] In the above step S201, the focal length, sensor target surface size and field of view angle of the above-mentioned first detection camera, second detection camera, first snapshot camera and second snapshot camera can be specifically obtained, so as to establish a matching relationship between the above-mentioned first snapshot camera, second snapshot camera and the first detection camera, second detection camera according to the relationship between the focal length and the sensor target surface size of f1 / D1=f2 / D2 for the same field of view when the sensor ratio is consistent; wherein f1 and D1 are the focal length and sensor target surface size of the first detection camera or the second detection camera respectively; f2 and D2 are the focal length and sensor target surface size of the first snapshot camera or the second snapshot camera respectively.

[0037] Based on the above relationship, in the present embodiment, the actual application scenario is a power transmission scenario, the first detection camera can be specifically selected as a 4mm focal length camera, the second detection camera can be specifically selected as a 12mm focal length camera, and the first detection camera and the second detection camera are both 200w low-resolution cameras with a target surface of 2 / 3 inches; the first snapshot camera can be specifically selected as an 8mm focal length camera, the second snapshot camera can be specifically selected as a 25mm focal length camera, and the first snapshot camera and the second snapshot camera are both 1200w high-resolution cameras with a target surface of 1 / 2.8 inches; wherein, in the power transmission scenario of the present embodiment, the first detection camera establishes a matching relationship with the first snapshot camera, and the second detection camera establishes a matching relationship with the second snapshot camera.

[0038] Among them, the sensor target size refers to the physical size of the camera's photosensitive element (i.e., image sensor); the photosensitive element is the part inside the camera used to capture light and convert it into image signals. Common types of photosensitive elements include CCD (charge-coupled device) and CMOS (complementary metal oxide semiconductor).

[0039] S202: Using the first detection camera and the second detection camera to simultaneously photograph the preset area, and analyzing the photographed image information.

[0040] S203: When any one of the two detection cameras recognizes the target in the preset area and the other camera does not detect the target, the first capture camera or the second capture camera corresponding to the detection camera that recognizes the target is awakened to capture the target, thereby ensuring that a capture camera with a suitable focal length can be flexibly selected to capture the target.

[0041] It should be pointed out that in actual applications, when the first detection camera and the second detection camera are used to monitor the preset area at the same time, in some specific circumstances, the first detection camera and the second detection camera will inevitably detect the target at the same time. At this time, the targets detected in the first detection camera and the second detection camera can be matched according to the coordinate mapping relationship of the pixel points on the two cameras, so as to execute the appropriate wake-up capture strategy according to the matching results to reduce power consumption.

[0042] To this end, in the flexible wake-up method designed in the present invention, in the above step S202, it also specifically includes the step of calibrating the posture relationship between the above first detection camera and the above second detection camera to obtain the coordinate mapping relationship between the pixel points on the two cameras.

[0043] In this embodiment, each camera can be calibrated by a feature point matching method or other means to obtain the relationship between the fields of view of each camera, and the posture relationship between the first detection camera and the second detection camera can be determined by camera calibration to obtain the coordinate mapping relationship between the pixel points on the first detection camera and the second detection camera.

[0044] At the same time, in the flexible wake-up method designed by the present invention, in the above step S203, the following steps are specifically included: when the first detection camera and the second detection camera simultaneously identify the existence of a target in the preset area, a mask of each target is obtained, and based on the coordinate mapping relationship obtained in the above step S202, coordinate matching is performed on the targets detected by the first detection camera and the second detection camera: When there is no one-to-one correspondence between the targets detected by the first detection camera and the second detection camera, the first capture camera and the second capture camera are simultaneously awakened to capture; When there is a one-to-one correspondence between the objects detected by the first detection camera and the objects detected by the second detection camera, the following steps S101-S103 are performed: S101: The first detection camera analyzes the captured image information, and when a target is identified in a preset area, executes step S102 or step S103; S102: Obtain the number of pixels of all targets in the image information of the first detection camera. When the number of pixels of all targets is ≥ a preset pixel number threshold, wake up the first snapshot camera to take a snapshot; when the number of pixels of all targets is < the preset pixel number threshold, wake up the second snapshot camera to take a snapshot; otherwise, when the number of pixels of both targets is ≥ the preset pixel number threshold and the number of pixels of both targets is < the preset pixel number threshold, wake up the first snapshot camera and the second snapshot camera to take a snapshot at the same time; S103: Divide the image information of the first detection camera into a short-range area and a long-range area, and obtain the positions of all targets in the image information. When the positions of all targets are in the short-range area, wake up the first snapshot camera to capture; when the positions of all targets are in the long-range area, wake up the second snapshot camera to capture; otherwise, when there are targets in both the short-range area and the long-range area, wake up the first snapshot camera and the second snapshot camera at the same time to capture.

[0045] In the above technical solution of the present invention, in actual application, the above-mentioned first detection camera is a low-power camera, which can run all day and shoot a preset area, perform target analysis and preliminary judgment in real time, and make judgments based on the number of target pixels and target position, so as to select a suitable first capture camera and / or second capture camera for capture based on the preliminary judgment, thereby avoiding the high-power consumption first capture camera and / or second capture camera working all day and reducing power consumption.

[0046] In addition, it should be noted that when the flexible wake-up method based on the multi-focal-length multi-camera group designed by the present invention is actually applied, the above-mentioned flexible wake-up method can be specifically applied to power transmission scenarios to detect whether there is a target close to the transmission cable in the power transmission scenario and there is a risk of intrusion into the transmission cable.

[0047] To this end, in actual application, under certain preferred embodiments, in order to further reduce power consumption and reduce the snapshot situations of the first snapshot camera and / or the second snapshot camera in unnecessary situations, when the above-mentioned flexible wake-up method is actually applied to the power transmission scenario, a warning area is set in advance in the power transmission scenario, and the warning area is used as the preset area; in actual operation, even if the above-mentioned first detection camera and / or the second detection camera recognizes the target in the preset area, that is, when it recognizes the existence of the target in the warning area, it will not immediately wake up the snapshot camera to capture.

[0048] When the first detection camera and / or the second detection camera recognizes the existence of a target in the preset area, it will continue to track the dynamics of the target; and only when it is detected that the target is moving in the direction of the transmission cable in the power transmission scene, it will report the detection of the target and flexibly wake up the snapshot camera to take a snapshot; when it is detected that the target is moving in the preset area in the direction away from the transmission cable, no reporting will be performed.

[0049] Accordingly, when the first snapshot camera and / or the second snapshot camera are actually used for snapshot, the snapshot image can be effectively obtained, and the target detected in the snapshot image and the distance between the target and the transmission cable can be obtained at the same time, so as to provide early warning to the power grid staff and facilitate the power grid staff to intuitively and accurately obtain whether the distance between the target and the transmission cable is within a safe distance.

[0050] Correspondingly, in the present invention, a flexible wake-up system based on a multi-focal-segment multi-eye camera group is also disclosed. The flexible wake-up system can simply execute the above-mentioned flexible wake-up method of the present invention to realize the flexible wake-up capture analysis of the multi-focal-segment multi-eye camera group, which specifically includes: The multi-focal-length multi-eye camera group, the memory, the processor and the computer program stored in the memory and executable on the processor. When the processor executes the computer program, each step of the flexible wake-up method based on the multi-focal-length multi-eye camera group of the present invention is implemented, that is, the above steps are executed.

[0051] Of course, in practical applications, considering that multiple devices may be used for synchronous shooting, in order to facilitate meeting practical application requirements, in some embodiments, the multi-focal-segment multi-eye camera group used in the flexible wake-up system may specifically include a main device and a sub-device, and the first detection camera, the first capture camera, and the second capture camera are set in the main device; at the same time, multiple sub-device capture cameras with the same focal length as each capture camera in the main device are set in the sub-device; In actual application, the above-mentioned processor can control the first detection camera, the first snapshot camera and the second snapshot camera in the main device to actually execute the various steps in the flexible wake-up method of the present invention, and when controlling the first snapshot camera and / or the second snapshot camera of the main device to capture, synchronously wake up the secondary device snapshot camera with the same focal length in the secondary device to capture.

[0052] For ease of understanding, for example: the above-mentioned secondary device is provided with a first secondary device snapshot camera, whose focal length is the same as that of the first snapshot camera of the main device; the above-mentioned secondary device is provided with a second secondary device snapshot camera, whose focal length is the same as that of the second snapshot camera of the main device; When the master device executes the flexible wake-up method of the present invention and finally determines that the first snapshot camera is taking a snapshot, the processor will also synchronously wake up the first snapshot camera in the slave device to take a snapshot; When the master device executes the flexible wake-up method of the present invention and finally determines that the second snapshot camera is taking a snapshot, the processor will also synchronously wake up the second slave device snapshot camera in the slave device to take a snapshot; When the first snapshot camera and the second snapshot camera of the main device are synchronously capturing, the secondary processor will also synchronously wake up the first secondary device snapshot camera and the second secondary device snapshot camera in the secondary device to capture.

[0053] From the above description, it can be seen that the flexible awakening method and system based on the multi-focal-segment multi-eye camera group designed by the present invention can shoot a preset area through a detection camera, and analyze the captured image information, so as to adaptively select a capture camera with a suitable focal length for capture based on the difference in the number of target pixels and the target position in the image information through preliminary judgment, so as to flexibly awaken the above-mentioned first capture camera and / or second capture camera. At the same time, in actual application, multiple detection cameras with different focal lengths can also be set to match the above-mentioned first capture camera and second capture camera respectively, so as to flexibly awaken the first capture camera and / or second capture camera for capture analysis according to the actual detection situation under the premise of improving the detection accuracy, so as to call the capture camera with a suitable focal length for capture analysis, which can effectively reduce power consumption while ensuring the capture effect, has low cost, is easy to implement, and has good promotion prospects and application value.

[0054] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A flexible wake-up method based on a multi-focal-segment multi-camera group, characterized in that: The multi-focal-segment multi-camera group includes a first detection camera, a first snapshot camera, and a second snapshot camera, wherein the focal length of the first snapshot camera is smaller than that of the second snapshot camera, and the flexible wake-up method includes the steps of: S101: photographing a preset area using the first detection camera, and analyzing the photographed image information, so as to execute step S102 or step S103 when a target is identified in the preset area; S102: Obtain the number of pixels of all targets in the image information, and when the number of pixels of all targets is ≥ a preset pixel number threshold, wake up the first snapshot camera to capture; when the number of pixels of all targets is < the preset pixel number threshold, wake up the second snapshot camera to capture; otherwise, wake up the first snapshot camera and the second snapshot camera at the same time to capture; S103: Divide the image information into a short-distance area and a long-distance area, and obtain the positions of all targets in the image information. When the positions of all targets are in the short-distance area, wake up the first snapshot camera to capture; when the positions of all targets are in the long-distance area, wake up the second snapshot camera to capture; otherwise, wake up the first snapshot camera and the second snapshot camera at the same time to capture.

2. The flexible wake-up method based on a multi-focal-segment multi-camera group according to claim 1, characterized in that: In step S101, specifically: Using the first detection camera to shoot the power transmission scene, setting a warning area in the power transmission scene, and using the warning area as the preset area; When it is identified that there is a target in the preset area, the dynamics of the target are continuously tracked; and when it is detected that the target is moving in the direction of the power transmission cable in the power transmission scene, step S102 or step S103 is executed.

3. The flexible wake-up method based on a multi-focal-segment multi-camera group according to claim 2, characterized in that: In step S102 or step S103, the first snapshot camera and / or the second snapshot camera captures and measures the distance between the target and the transmission cable.

4. The flexible wake-up method based on a multi-focal-segment multi-camera group according to claim 1, characterized in that: The multi-focal-segment multi-camera group further includes a second detection camera, and the focal lengths of the first detection camera and the second detection camera are different. The flexible wake-up method further includes the steps of: S201: Establishing a matching relationship between the first snapshot camera, the second snapshot camera and the first detection camera, the second detection camera; S202: photographing a preset area using the first detection camera and the second detection camera, and analyzing the photographed image information; S203: When the first detection camera or the second detection camera recognizes that there is a target in the preset area, the corresponding first capture camera or the second capture camera is awakened to capture.

5. The flexible wake-up method based on a multi-focal-segment multi-camera group according to claim 4, characterized in that: In step S202, the method further includes the following steps: Calibrate the position relationship between the first detection camera and the second detection camera to obtain the coordinate mapping relationship of the pixel points on the two cameras.

6. The flexible wake-up method based on a multi-focal-segment multi-camera group according to claim 5, characterized in that: In step S203, the method further includes the following steps: When the first detection camera and the second detection camera simultaneously identify that there is a target in the preset area, obtain masks of each target, and match the targets detected by the first detection camera and the second detection camera based on the coordinate mapping relationship; When there is a one-to-one correspondence between the objects detected by the first detection camera and the objects detected by the second detection camera, executing steps S101-S103; When there is no one-to-one correspondence between the objects detected by the first detection camera and the objects detected by the second detection camera, the first capture camera and the second capture camera are simultaneously awakened to capture.

7. The flexible wake-up method based on a multi-focal-segment multi-camera group according to claim 4, characterized in that: The focal length of the first detection camera is smaller than that of the second detection camera.

8. The flexible wake-up method based on a multi-focal-segment multi-camera group according to claim 4, characterized in that: In step S201, the focal length, sensor target surface size and field of view angle of the first detection camera, the second detection camera, the first snapshot camera and the second snapshot camera are obtained, so as to establish a matching relationship between the first snapshot camera, the second snapshot camera and the first detection camera, the second detection camera according to the relationship between the focal length and the sensor target surface size of f1 / D1=f2 / D2 for the same field of view when the sensor ratio is consistent; wherein f1 and D1 are the focal length and sensor target surface size of the first detection camera or the second detection camera, respectively; and f2 and D2 are the focal length and sensor target surface size of the first snapshot camera or the second snapshot camera, respectively.

9. A flexible wake-up system based on a multi-focal-length multi-eye camera group, comprising a multi-focal-length multi-eye camera group, a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, each step of a flexible wake-up method based on a multi-focal-segment multi-eye camera group as described in any one of claims 1 to 8 is implemented.

10. The flexible wake-up system according to claim 9, characterized in that: The multi-focal-segment multi-eye camera group includes a main device and a sub-device, wherein the first detection camera, the first snapshot camera and the second snapshot camera are arranged in the main device, and the sub-device includes a plurality of sub-device snapshot cameras having the same focal length as each snapshot camera in the main device; Among them, the processor controls the main device to execute each step of a flexible wake-up method based on a multi-focal-length multi-eye camera group as described in any one of claims 1-8, and when controlling the first snapshot camera and / or the second snapshot camera of the main device to capture, synchronously wakes up the secondary device snapshot camera with the same focal length in the secondary device to capture.

Images