Control method, device and storage medium of multi-channel camera equipment

Abstract

The present application relates to a control method, device and storage medium for a multi-channel camera device. The control method for the multi-channel camera device includes: obtaining a single-channel image captured by each camera device under a device channel; one device channel corresponds to one camera device; detecting the single-channel image and calculating the number of valid targets under each device channel; based on the number of valid targets, determining the target channel in each device channel, and switching the image processing mode of the target channel to the target intelligent mode. The present application solves the problem of low intelligent control of multi-channel camera devices.

Description

Technical Field

[0001] The present application relates to the field of camera equipment, and in particular to a control method, device and storage medium for multi-channel camera equipment. Background Art

[0002] The channels of multi-channel camera devices are the key paths or interfaces used to connect and display the images of multiple surveillance cameras in the monitoring system. By rationally configuring and managing the channels of camera devices, efficient video monitoring and security management can be achieved. In the related art, in order to improve the efficiency and accuracy of image data processing, intelligent functions can be enabled for the image processing process under each device channel. However, usually only one set of intelligent image processing functions can be enabled under one device channel, and the channel will remain in the enabled state after the intelligent function is enabled, resulting in a low degree of intelligence in the control of multi-channel camera devices.

[0003] Currently, no effective solution has been proposed for the problem of low intelligence level in the control of multi-channel camera equipment in related technologies. Summary of the invention

[0004] The embodiments of the present application provide a control method, device and storage medium for a multi-channel camera device, so as to at least solve the problem of low intelligent control level of the multi-channel camera device in the related art.

[0005] In a first aspect, an embodiment of the present application provides a method for controlling a multi-channel camera device, the method comprising:

[0006] Obtaining a single channel of images collected by each camera device under a device channel; one device channel corresponds to one camera device;

[0007] Detecting the single-channel image and calculating the number of valid targets under each device channel;

[0008] Based on the number of valid targets, a target channel in each of the device channels is determined, and the image processing mode of the target channel is switched to a target intelligent mode.

[0009] In some embodiments, determining a target channel in each of the device channels based on the number of valid targets includes:

[0010] Obtaining priority information assigned to each of the device channels;

[0011] The target channel is determined according to the number of valid targets and the priority information.

[0012] In some embodiments, the obtaining priority information assigned to each of the device channels includes:

[0013] Obtaining channel detection information of each of the device channels;

[0014] Weight values ​​are respectively assigned to the number of valid targets and the channel detection information, and based on the weight values, weighted calculation is performed on the number of valid targets and the channel detection information to obtain the priority information.

[0015] In some embodiments, determining the target channel according to the number of valid targets and the priority information includes:

[0016] From each of the device channels, searching for a primary inspection channel whose number of valid targets is greater than or equal to a preset target number threshold;

[0017] When the number of the preliminary inspection channels is less than or equal to the preset maximum computing power specification number of channels, the preliminary inspection channels are determined as the target channels; when the number of the preliminary inspection channels is greater than the maximum computing power specification number of channels, the target channels are screened out from the preliminary inspection channels based on the priority information.

[0018] In some embodiments, determining a target channel in each of the device channels based on the number of valid targets includes:

[0019] Based on the single-channel image, determine the valid channels that trigger the valid targets in each of the device channels, and calculate the number of valid channels;

[0020] The target channel is determined according to the number of valid targets and the number of valid channels.

[0021] In some embodiments, determining the target channel according to the number of valid targets and the number of channels includes:

[0022] When it is detected that the effective target number of each device channel is less than the preset target number threshold, and the effective channel number is greater than or equal to the preset intelligent path number threshold, the effective channel number is compared with the preset maximum computing power specification path number;

[0023] If the comparison result is that the number of valid channels is less than or equal to the number of channels of the maximum computing power specification, the number of valid channels is determined as the target channels;

[0024] If the comparison result is that the number of valid channels is greater than the number of channels with the maximum computing power specification, the single-channel images corresponding to the valid channels are stitched to obtain a first stitched image, and the stitched channel for the first stitched image is determined as the target channel.

[0025] In some embodiments, determining the target channel according to the number of valid targets and the number of channels includes:

[0026] When searching for a preliminary inspection channel whose valid target number is greater than or equal to a preset target number threshold from each of the device channels, and the valid channel number is greater than the preset maximum computing power specification number of channels, comparing the valid channel number with the preset intelligent channel number threshold;

[0027] If the comparison result is that the number of valid channels is less than the intelligent path number threshold, the target channel is screened out from the initial inspection channels based on the priority information assigned to each of the device channels;

[0028] If the comparison result is that the number of valid channels is greater than or equal to the intelligent path threshold, the target channel in the initial inspection channel is determined based on the maximum computing power specification path and the priority information.

[0029] In some embodiments, determining the target channel in the initial inspection channel based on the maximum computing power specification path and the priority information includes:

[0030] Based on the priority information, the target channels are selected from the initially inspected channels, and the number of the target channels matches the number of channels with the maximum computing power specification; or,

[0031] Based on the priority information, a preset number of priority channels are screened out from the primary inspection channel, and the single-channel images corresponding to other channels are spliced ​​to obtain a second spliced ​​image; the other channels are used to represent the remaining channels in the primary inspection channel except the priority channels;

[0032] The priority channel and a stitching channel for the second stitched image are determined as the target channels.

[0033] In some embodiments, switching the image processing mode of the target channel to the target intelligent mode includes:

[0034] The single-channel images corresponding to the target channel are stitched together to obtain a target stitched image, and image processing is performed on the target stitched image in the target intelligent mode.

[0035] In some embodiments, the step of stitching the single-channel images corresponding to the target channels to obtain the target stitched image includes:

[0036] Determine the cutout area in the single-channel image corresponding to each of the target channels;

[0037] Based on the number of valid targets, a stitching ratio is determined, and the cutout area is stitched according to the stitching ratio to obtain the target stitching image.

[0038] In some embodiments, determining the cutout region in the single-channel image corresponding to each of the target channels includes:

[0039] Obtain the cutout area set by the user; or,

[0040] The cutout area is calculated according to the target detection results in each of the single-channel images.

[0041] In a second aspect, an embodiment of the present application provides a control device for a multi-channel camera device, comprising:

[0042] An acquisition module is used to acquire a single channel of images collected by each camera device under a device channel; one device channel corresponds to one camera device;

[0043] A detection module, used to detect the single-channel image and calculate the number of valid targets under each device channel;

[0044] A mode switching module is used to determine a target channel in each of the device channels based on the number of valid targets, and switch the image processing mode of the target channel to a target intelligent mode.

[0045] In a fourth aspect, an embodiment of the present application provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the control method for a multi-channel camera device as described in the first aspect above.

[0046] Compared with the related art, the control method, device and storage medium of the multi-channel camera equipment provided in the embodiment of the present application obtains the single-channel image collected by each camera equipment under the device channel; one device channel corresponds to one camera equipment; the single-channel image is detected and the number of valid targets under each device channel is calculated; based on the number of valid targets, the target channel in each device channel is determined, and the image processing mode of the target channel is switched to the target intelligent mode. Based on this, the adaptive dynamic switching of the intelligent mode of multiple channels is realized. By adopting the intelligent switching method, the channels that really need intelligent analysis can be dynamically opened, so that the number of channels that finally open the intelligent mode can be consistent with the actual situation, thereby making full use of the intelligent analysis resource scheduling of the system, avoiding the problem that the opened intelligent channel is fixed, resulting in the inability to effectively use the equipment resources to perform intelligent algorithm analysis on the data of other channels, and effectively improving the control intelligence of the multi-channel camera equipment.

[0047] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0049] Figure 1 It is a hardware structure block diagram of a terminal of a control method for a multi-channel camera device according to an embodiment of the present application;

[0050] Figure 2 is a flow chart of a control method of a multi-channel camera device according to an embodiment of the present application;

[0051] Figure 3A is a schematic diagram of manual cutout according to an embodiment of the present application;

[0052] Figure 3B is a schematic diagram of automatic cutout according to an embodiment of the present application;

[0053] Figure 4 is a schematic diagram of image stitching according to an embodiment of the present application;

[0054] Figure 5 is a flow chart of another method for controlling a multi-channel camera device according to an embodiment of the present application;

[0055] Fig. 6A is a schematic diagram of a processing flow of an intelligent analysis algorithm according to an embodiment of the present application;

[0056] Figure 6B is a schematic diagram of another processing flow of feeding into an intelligent analysis algorithm according to an embodiment of the present application;

[0057] Figure 7 It is a structural block diagram of a control device for a multi-channel camera device according to an embodiment of the present application. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application. In addition, it can also be understood that although the efforts made in this development process may be complex and lengthy, for ordinary technicians in the field related to the contents disclosed in the present application, some changes such as design, manufacturing or production based on the technical contents disclosed in the present application are only conventional technical means, and should not be understood as insufficient contents disclosed in the present application.

[0059] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0060] Unless otherwise defined, the technical terms or scientific terms involved in this application should be understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the" and the like involved in this application do not indicate a quantity limitation, and may indicate the singular or plural. The terms "include", "comprise", "have" and any of their variations involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" involved in this application means greater than or equal to two. "And / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, "A and / or B" can mean: A exists alone, A and B exist at the same time, and B exists alone. The terms "first", "second", "third" and the like involved in the present application are merely used to distinguish similar objects and do not represent a specific ordering of the objects.

[0061] The method embodiment provided in this embodiment can be executed in a terminal, a computer or a similar computing device. Taking running on a terminal as an example, Figure 1 1 is a hardware structure block diagram of a terminal of a method for controlling a multi-channel camera device according to an embodiment of the present application. Figure 1 As shown, the terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data. Optionally, the terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is for illustration only and does not limit the structure of the above terminal. Figure 1more or fewer components shown, or having a configuration different from that Figure 1 shown.

[0062] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the control method of the multi-channel camera device in the embodiments of the present application. The processor 102 executes various functional applications and image processing by running the computer program stored in the memory 104, that is, implements the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some instances, the memory 104 may further include a memory remotely disposed relative to the processor 102, and these remote memories can be connected to the terminal through a network. Examples of the above network include but are not limited to the Internet, enterprise intranet, local area network, mobile communication network, and combinations thereof.

[0063] The transmission device 106 is used to receive or send data via a network. Specific examples of the above network may include a wireless network provided by a communication provider of the terminal. In one instance, the transmission device 106 includes a network adapter (abbreviated as NIC), which can be connected to other network devices through a base station and thus communicate with the Internet. In one instance, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0064] The present embodiment provides a control method for a multi-channel camera device, Figure 2 is a flowchart of a control method for a multi-channel camera device according to an embodiment of the present application, as Figure 2 shown, and the process includes the following steps:

[0065] Step S210, obtain single-channel images collected by each multi-channel camera device under the device channel; one device channel corresponds to one multi-channel camera device.

[0066] For the above multi-channel camera device, usually multiple camera devices deployed on-site are linked to take pictures of a certain scene. For example, the multi-channel camera device can be applied in a security monitoring system. By installing multiple cameras, comprehensive coverage of the monitoring area can be achieved, including in public places such as shopping malls, parks, banks, and traffic intersections, where multiple cameras are set up to monitor the flow of people, etc. Another example is that the multi-channel camera device can be applied in a vehicle-mounted camera system. By multiple camera devices surrounding and monitoring the surrounding environment of the vehicle, driving safety can be improved.

[0067] The device channel can be understood as the connecting bridge between the camera device and the control system. It is responsible for transmitting the video signal or data stream of the camera device to the system for processing and storage. In a multi-channel camera device control system, each camera device corresponds to a device channel, which is independent and configurable in the system. In actual application, the system can obtain a single-channel image collected by the corresponding camera device from each device channel to provide input for subsequent detection and processing.

[0068] Step S220, detecting a single channel image and calculating the number of valid targets in each device channel.

[0069] In this step, target detection can be performed for the single-channel image transmitted by each device channel. Target detection is an image processing process used to identify specific objects (such as people, cars, animals, etc.) in the image based on the current intelligent business. Then, the valid target of the corresponding single-channel image can be determined based on the target detection result. Among them, the valid target refers to the target object that has been processed by the algorithm, correctly identified and meets the business requirements in the image processing task; for example, for the intelligent transportation system, the target objects such as vehicles and pedestrians are identified as valid targets, for security monitoring, abnormal behaviors, intruders, etc. are detected as valid targets, and for animal protection business, wild animal populations are monitored as valid targets. The method of judging the valid target can be: according to the actual business needs, the judgment conditions of the valid target (such as target type, movement speed, appearance time, etc.) are set, and the targets that meet these judgment conditions are regarded as valid targets. Finally, for each channel, the number of valid targets in the single-channel image is counted. In other words, the number of valid targets is a quantitative representation of the target detection results. During the image processing process, the algorithm will identify multiple potential target objects in the image. However, not all identified targets are required by the business; the number of valid targets is the total number of targets that meet business needs after screening and confirmation among these identified targets.

[0070] Specifically, the above detection method can be implemented through the system's built-in smart motion detection (Smart Motion Detection, SMD) module or dynamic detection algorithm. It should be understood that the SMD module currently only detects and counts valid targets such as people and cars, so the resource consumption of the device in this process is negligible compared to intelligent algorithms such as face, perimeter, and behavior analysis. The dynamic detection algorithm does not consume the chip's intelligent computing power for the step of calculating the target coordinates of the screen area. Therefore, through the above step S220, low-computing-power intelligent analysis of a single-channel image is achieved without occupying more computing power of the system.

[0071] Step S230, based on the number of valid targets, determine the target channel in each device channel, and switch the image processing mode of the target channel to the target intelligent mode.

[0072] After obtaining the number of valid targets corresponding to each of the above-mentioned device channels, by analyzing these data, determine which device channels (target channels) need to automatically switch to the target intelligent mode. The target intelligent mode generally refers to a more intelligent image processing method, for example, it may include more complex image analysis, target tracking, anomaly detection and other functions. Among them, the method of determining the target channel based on the number of valid targets can be: respectively compare the number of valid targets of each device channel with the preset target number threshold, and determine the target channel based on the channels whose number of valid targets is greater than or equal to the target number threshold. The target number threshold can be pre-set by the user based on actual conditions. For example, the target number threshold can be 10; when it is detected that the number of valid targets under a certain device channel is greater than or equal to 10, it means that the current target flow under the device channel is large, and it is necessary to perform intelligent analysis on the centralized computing power of the image data collected under the device channel, so the device channel is determined as the target channel.

[0073] More specifically, each device channel is provided with an image processing module, which is used to process the image data detected by the corresponding camera device through the respective algorithm units. When a target channel is detected, the target intelligent mode of the image processing module of the target channel can be turned on, so as to perform in-depth analysis on the image data of the target channel using the intelligent algorithm.

[0074] Through the above step S230, according to the target detection results, the device channels that need to turn on the smart mode and the number of device channels that need to turn on the smart mode can be dynamically determined to achieve dynamic adjustment of resource allocation and enable the smart mode only when needed. For example, if a large number of targets are detected under one or some device channels, the channel can be automatically switched to the smart mode for more in-depth analysis. During the entire video detection process, if the number of valid targets under one or some device channels drops to within a preset target number threshold, and the number of valid targets of one of the remaining channels in the device channel rises to exceed the preset target number threshold, the device channel with the reduced number of valid targets can be switched back to the normal image processing mode, and the device channel with the increased number of valid targets can be switched to the target smart mode.

[0075] It should also be noted that the target intelligent mode can also include multiple intelligent modes with different intelligence levels. For example, for device channel 1 and device channel 2, the number of valid targets detected is relatively large, and the number of valid targets of device channel 1 is greater than the number of valid targets of device channel 2, then the image processing mode corresponding to device channel 1 can be switched to intelligent mode A with a relatively higher algorithm intelligence level in the target intelligent mode, and the image processing mode corresponding to device channel 2 can be switched to intelligent mode B with a relatively lower algorithm intelligence level in the target intelligent mode, while the image processing modes of the remaining device channels remain in the normal processing mode.

[0076] In the control method of the above-mentioned multi-channel camera equipment, the target channel that needs to be switched to the target intelligent mode is determined by the number of valid targets under each device channel detected by a single-channel image, thereby realizing adaptive dynamic switching of the intelligent modes of multiple channels. By adopting the intelligent switching method, the channels that really need intelligent analysis can be dynamically opened, so that the number of channels that finally turn on the intelligent mode can be consistent with the actual situation. In addition, the above-mentioned method can also achieve the effect of expanding the number of intelligent channels of the system, thereby making full use of the system's intelligent analysis resource scheduling, avoiding the problem of fixed opened intelligent channels, resulting in the inability to effectively utilize device resources to perform intelligent algorithm analysis on the data of other channels, and effectively improving the control intelligence level of the multi-channel camera equipment.

[0077] In some embodiments, the step of determining the target channel in each device channel based on the number of valid targets may further include the following steps:

[0078] Step S231, obtaining priority information assigned to each device channel, wherein the priority information refers to level identification information of relative importance or urgency set for the device channel of each camera device and requiring the target intelligent mode to be activated.

[0079] The above priority information can be set by the user in advance according to the actual situation. For example, the device channel priority corresponding to the camera equipment arranged at the main channel in the park can be set to the highest level, and the device channel priority corresponding to the camera equipment arranged at a more remote location can be set to the lowest level. Alternatively, the priority information can also be set according to the number of valid targets calculated in the above steps, that is, the more valid targets there are, the higher the priority of the corresponding device channel.

[0080] Alternatively, in another embodiment, in order to improve the accuracy of the intelligent mode switching control, the calculation process of the priority information of each device channel may further include the following steps:

[0081] The channel detection information of each device channel is obtained; weight values ​​are assigned to the number of valid targets and the channel detection information respectively, and based on the weight values, the number of valid targets and the channel detection information are weightedly calculated to obtain priority information.

[0082] Among them, the channel detection information specifically refers to the detection value corresponding to each device channel detected based on the preset influencing factors that will affect the intelligent computing power analysis of each channel. For example, the preset influencing factor can be the maximum value of the target that can be detected by the image processing algorithm corresponding to the device channel; in other words, the larger the maximum value of the target that can be detected by a device channel, the higher the priority of setting the device channel to turn on the intelligent mode. At this time, the channel detection information corresponding to the preset influencing factor is the maximum value of the detectable target of the device channel, and a weighted calculation is performed on the maximum value of the target and the number of valid targets; in this weighted calculation, the weight values ​​assigned to each item can be set according to the actual business needs and other application situations. For this example, one of the priority calculation methods is provided, as shown in the following formula:

[0083] S i =b×A+cO i

[0084] In the above formula, S i Indicates the priority score corresponding to the i-th device channel, that is, the above priority information; i is a positive integer. A is a constant whose value is less than or equal to the maximum value of the target that can be detected by the device channel. i is the number of valid targets detected under the i-th device channel. b and c represent the maximum value of targets that can be detected by the channel and the weight value assigned to the number of valid targets, respectively. Combined with the above formula, at the same time, the number of valid targets detected by device channel 1 is less than that of device channel 2, but at this time, due to some reasons, it may be more necessary to pay attention to the target detection of device channel 1. In this case, the corresponding calculated priority score of device channel 1 is higher than that of device channel 2.

[0085] For another example, the preset influencing factor may be the deployment position of the camera device corresponding to the device channel in the actual application scenario; that is, based on the deployment position, the closer the detected device channel of the camera device is to the center range of the monitoring area, the higher the value of the channel detection information. For example, if the camera device of device channel 1 is deployed at the center of the monitoring area, the channel detection information value is set to 10; if the camera device of device channel 9 is deployed at the edge of the monitoring area, the channel detection information value is set to 1; and for the device channels corresponding to a series of deployed cameras from the center to the edge, the channel detection information values ​​increase or decrease with the position information.

[0086] For another example, the above-mentioned preset influencing factors can also be the actual device performance parameters or other factors of the camera device corresponding to the device channel; the value of the channel detection information can be set based on the detected device performance parameters such as optical performance or video performance, which will not be repeated here.

[0087] It should be understood that there may be one or more types of the above-mentioned channel detection information; in other words, the above-mentioned multiple influencing factors may be combined to calculate the value of the channel detection information corresponding to each influencing factor, and weighted calculation may be performed on multiple channel detection information combined with the number of valid targets.

[0088] It can be seen that through the above-mentioned embodiments, weighted calculation is performed in combination with the number of valid targets and channel detection information to determine the priority information corresponding to each device channel, so that the calculation of the priority of each channel is not only determined by the number of valid targets of the channel, thereby fully considering the richness and complexity of the environment under the monitoring field of the camera equipment, improving the accuracy and comprehensiveness of the priority calculation of each device channel, and thus also improving the accuracy of the control of multi-channel camera equipment.

[0089] Step S232, determining the target channel according to the number of valid targets and the priority information.

[0090] Among them, channels whose effective target number is greater than or equal to the preset target number threshold can be searched from the above-mentioned device channels, and the channels can be sorted in descending order according to the priority information corresponding to the searched channels, and the channels ranked in the first N are set as the target channels that need to turn on the target smart mode. N is a positive integer, and the value of N can be pre-set by the user according to actual business needs, or can be determined according to the maximum number of channels that the system can allow to turn on the smart mode, for example, it can be set to 2.

[0091] In another embodiment, determining the target channel according to the number of valid targets and the priority information in the above step S232 may further include the following steps:

[0092] From each device channel, search for a preliminary inspection channel whose number of valid targets is greater than or equal to a preset target number threshold; when the number of preliminary inspection channels is less than or equal to the preset maximum computing power specification number of channels, determine the preliminary inspection channel as the target channel; when the number of preliminary inspection channels is greater than the maximum computing power specification number of channels, filter out the target channel from the preliminary inspection channels based on the priority information.

[0093] The above-mentioned maximum computing power specification number refers to the maximum value of the number of device channels that are allowed to enable the target intelligent mode in the entire multi-channel camera control system. For example, for a 16-channel multi-channel camera device, under the current system architecture, a maximum of 3 device channels are supported to enable the target intelligent mode, and the maximum computing power specification number is 3. Therefore, if the number of initial inspection channels is less than the maximum computing power specification number, it means that the current monitoring scene of the multi-channel camera device is a scene with a large number of targets and relatively concentrated targets, such as crowds, and the target intelligent mode can be directly enabled for the initial inspection channel to perform intelligent analysis. On the contrary, it can be determined according to the priority information which channel or channels of the channel device to enable the intelligent mode.

[0094] Through the above embodiments, an intelligent mode switching method is provided for scenes with a large number of targets and relatively concentrated targets, and the accuracy of controlling multiple camera devices is effectively improved by setting priorities.

[0095] In some embodiments, the step of determining the target channel in each device channel based on the number of valid targets may further include the following steps:

[0096] Based on a single-channel image, determine the valid channels that trigger valid targets in each device channel, and calculate the number of valid channels; determine the target channel based on the number of valid targets and the number of valid channels.

[0097] The above-mentioned effective channel specifically refers to a channel where effective targets are detected after preliminary motion detection processing is performed on the single-channel images collected under each device channel. The number of all channels determined to be effective is counted to obtain the above-mentioned effective channel number.

[0098] If the number of valid channels currently counted is large, it means that the distribution of various detection targets such as people flow is relatively scattered; otherwise, it means that the distribution of detection targets is relatively concentrated. Based on this analysis, the number of valid targets and the number of valid channels can be comprehensively considered to further distinguish between the situation where the number of detection targets is large and concentrated, the situation where the number of detection targets is small and scattered, and the situation where the number of targets is large and scattered, so as to adaptively determine the target channel for different situations, thereby helping to improve the accuracy of intelligent mode switching control of each device channel.

[0099] In some embodiments, the step of determining the target channel according to the number of valid targets and the number of channels may further include the following steps:

[0100] When it is detected that the effective target number of each device channel is less than the preset target number threshold, and the effective channel number is greater than or equal to the preset intelligent path threshold, the effective channel number is compared with the preset maximum computing power specification path; if the comparison result is that the effective channel number is less than or equal to the maximum computing power specification path, the effective channel number is determined as the target channel; if the comparison result is that the effective channel number is greater than the maximum computing power specification path, the single-channel image corresponding to the effective channel is stitched to obtain a first stitched image, and the stitched channel for the first stitched image is determined as the target channel.

[0101] The above-mentioned intelligent channel number threshold specifically refers to a value set by the user in advance based on actual business needs, which is used to characterize whether to trigger the subsequent image stitching and fusion steps. For example, if there are 16 channels of multi-channel camera equipment at present, the maximum number of device channels supported by the system for turning on the intelligent mode is 2, and the intelligent channel number threshold set by the user is 11. Then, when it is detected that the number of valid channels is greater than 11, that is, the intelligent channel number threshold, it means that the detection target is dispersed at this time, and the number of valid channels that trigger the target in this scenario has reached the preset threshold that needs to trigger the subsequent stitching process; in order to improve the accuracy and efficiency of image processing, you can choose to stitch the images of each device channel, and enable the intelligent algorithm for the stitched images to process the image data.

[0102] The above calculation process is explained below.

[0103] First, the number of valid targets of each device channel is detected. This number is compared with the preset target number threshold. If the number of valid targets is less than the target number threshold, and the number of valid channels is greater than or equal to another preset value, namely the intelligent channel number threshold, then the multi-channel camera device control system needs to further determine whether splicing processing is required, which leads to a comparison with the maximum computing power specification channel number.

[0104] Next, the number of valid channels is compared with the maximum computing power specification. If the number of valid channels is less than or equal to the maximum computing power specification, it means that the system can support intelligent computing power analysis for all valid channels that have triggered valid targets. Therefore, each detected valid channel can be determined as a target channel that needs to turn on the target intelligent mode, so that the single-channel image under the target channel can be sent to the intelligent algorithm for in-depth analysis.

[0105] If the number of valid channels is greater than the maximum computing power specification, this means that the current system cannot directly process the data of all valid channels. In this embodiment, it has been determined that the number of valid targets has reached the intelligent channel threshold, so the system will choose to stitch the single-channel images corresponding to these channels to obtain a larger image-the first stitched image. Then, the first stitched channel corresponding to the stitched image will be determined as the target channel, and then the stitched image under the target channel will be sent to the intelligent algorithm for in-depth analysis. Among them, the first stitched channel here refers to the channel required to process the first stitched image, which can be a virtual channel or a channel composed of multiple physical channels.

[0106] If the number of valid channels is less than the intelligent path threshold, it may mean that the current number of channels is not enough to trigger the need for splicing processing, or the system can directly process the data of these channels without splicing. Specifically, if it is detected that the target number of at least one channel is greater than or equal to the target number threshold, and the number of valid channels is less than the intelligent path threshold, then when it is detected that the number of valid channels is greater than the maximum computing power specification path, the number of channels with the maximum computing power specification path can be screened out from each valid channel, or a preset number of channels can be used as the target channels. Among them, this screening method can be random screening, or it can also compare the priority information corresponding to each valid channel to decide which one or several intelligent paths to turn on.

[0107] Therefore, the intelligent path number threshold plays a key triggering role here. It is set according to the actual needs of the user (such as processing efficiency, image quality, system resources, etc.) to ensure that the system can operate and process data in the expected manner.

[0108] In summary, the intelligent path number threshold is a user-defined numerical standard used to determine whether the system needs to perform image stitching and fusion. In practical applications, it plays an important role in balancing system performance, processing efficiency and user needs.

[0109] In some embodiments, the step of determining the target channel according to the number of valid targets and the number of channels may further include the following steps:

[0110] When searching for initial inspection channels with valid target numbers greater than or equal to the preset target number threshold from each device channel, and the number of valid channels is greater than the preset maximum computing power specification number of channels, the number of valid channels is compared with the preset intelligent channel number threshold. In this case, it means that in the current monitoring scenario of multiple camera devices, each device channel detects a large number of targets, and almost every channel wants to apply for resources to perform intelligent analysis and processing on image data.

[0111] If the comparison result shows that the number of valid channels is less than the intelligent channel threshold, the target channel is selected from the initial inspection channel based on the priority information assigned to each device channel. In this case, when the computing power resources are relatively abundant, the priority of the channel is given priority. In other words, when the number of valid channels is less than the intelligent channel threshold, it means that the threshold for image stitching has not been triggered at this time. Therefore, it is directly determined based on the priority information corresponding to each initial inspection channel to decide which channel to enable the intelligent mode.

[0112] If the comparison result shows that the number of valid channels is greater than or equal to the intelligent path number threshold, the target channel in the initial inspection channel is determined based on the maximum computing power specification path number and priority information. This step means that in the case of tight computing power resources, it is necessary to comprehensively consider computing power requirements and channel priorities to determine the target channel.

[0113] Through the above steps, the target channels that meet specific conditions can be efficiently screened out from multiple device channels, thereby optimizing the processing efficiency and computing power allocation of the system or device.

[0114] In another embodiment, in order to achieve flexible processing of intelligent mode switching, two different target channel determination methods can also be provided. That is, for the above-mentioned determination of the target channel in the initial inspection channel based on the maximum computing power specification number and priority information, the following steps can also be included:

[0115] Based on the priority information, a preset number of priority channels are selected from the primary inspection channels, and the single-channel images corresponding to other channels are spliced ​​to obtain a second spliced ​​image; the other channels are used to characterize the remaining channels in the primary inspection channels except the priority channels; the priority channels and the spliced ​​channels for the second spliced ​​image are determined as target channels. The second spliced ​​channel refers to the channel required for processing the second spliced ​​image, which can be a virtual channel or a channel composed of multiple physical channels.

[0116] Specifically, taking the case where there are two channels with the maximum computing power specification as an example, at this time, based on the priority information, the single channel with the highest priority score can be selected to start the target intelligent mode, and then the fusion and stitching method can be selected in the remaining channels to obtain the second stitched image, and the second stitched image can be sent to another computing power for intelligent analysis.

[0117] Alternatively, based on the priority information, the target channels are selected from the initial inspection channels, and the number of target channels matches the maximum computing power specification. In the case of limited computing power resources, the system will give priority to channels with high priority and that can meet computing power requirements as target channels.

[0118] Specifically, based on the priority information, each initial inspection channel is sorted in descending order according to the priority score; if the priority score difference between the initial inspection channels ranked in the first M (M is the number of channels with the maximum computing power specification) is less than the preset difference threshold (the difference threshold can be set in advance based on the actual situation), it means that the priority score difference between the high-priority channels is not large, then it is directly based on the priority information. At this time, if the above method is still used to directly turn on the intelligent mode of some channels and splice the image data of other channels, it may be difficult to distinguish the device channels that need to directly turn on the intelligent mode. Based on this, in this embodiment, another method is adopted, that is, based on the priority information, the target channels that match the number of channels with the maximum computing power specification are selected and the target intelligent mode is turned on.

[0119] Still taking the example of 2 channels with maximum computing power specification, determine the difference between the highest priority score S1 and the second highest priority score S2 in each initial inspection channel; if the difference is not large, and the highest priority score S1 and the second highest priority score S2 far exceed the priority scores of other initial inspection channels, then the initial inspection channels corresponding to S1 and S2 are determined as target channels, and the intelligent mode of these two target channels is enabled first.

[0120] Through the above process design, the system can flexibly determine the target channel based on computing resources and priority information, thereby optimizing resource allocation and improving processing efficiency.

[0121] In some embodiments, switching the image processing mode of the target channel to the target intelligent mode may further include the following steps:

[0122] The single-channel images corresponding to the target channel are stitched together to obtain a target stitched image, and image processing is performed on the target stitched image in a target intelligent mode.

[0123] Among them, when it is detected that single-channel images of multiple channels need to be stitched, for example, when the number of target channels is greater than the channel threshold preset according to actual conditions or business needs, the single-channel images under each channel are stitched and the stitched images are sent to the algorithm for analysis.

[0124] Through the above-mentioned embodiments, the system can flexibly process the data of multiple target channels and select the most appropriate processing method and algorithm according to the amount and characteristics of the data, thereby improving the efficiency and accuracy of image processing.

[0125] In some embodiments, the above-mentioned stitching process of the single-channel images corresponding to the target channel to obtain the target stitched image may also include the following steps:

[0126] Determine the cutout area in the single-channel image corresponding to each target channel.

[0127] The method for determining the cutout area may be: obtaining the cutout area set by the user. Since the installation location of each camera device at the monitoring site is set according to the user's actual application scenario, the user is more familiar with which areas will frequently have detection events in this scenario. Therefore, in this embodiment, the user can manually set the area to be cutout under each device channel. For example, a multi-channel camera device includes 4 device channels, please refer to Figure 3A For device channels 1 to 4, a manual cutout area is set by the user for different areas of each channel.

[0128] Through the above method of determining the cutout area, the user sets the cutout area for each device channel, which ensures the stability of the area setting, so that the algorithm detection in the subsequent process and the target location information returned by the algorithm are more accurate and timely. In addition, this method can also support the planning of the cutout area according to time, and different cutout areas are effective in different time periods, which can also be used to coordinate the update of the above parameters such as the number of valid targets.

[0129] Alternatively, the above-mentioned cutout area can also be determined by: calculating the cutout area according to the target detection results in each single-channel image. Specifically, the cutout area is automatically cut out according to the target position information detected by the dynamic detection or SMD module. Figure 3B , under device channel 1 to device channel 4, the detection targets of each channel are different, so at least one target frame can be automatically determined according to the detected target (such as a target human body, a target vehicle or other intelligent target), and then an automatic cutout area is generated. By implementing the cutout area in this way, the target information to be detected can be automatically cut out, and the problem of easy omission of targets due to manual cutout area fixation can be effectively avoided. It has a high degree of intelligence and a strong fault tolerance rate.

[0130] Next, based on the number of valid targets, the stitching ratio is determined, and the cutout area is stitched according to the stitching ratio to obtain a target stitching image.

[0131] In this step, the stitching image to be fed into the algorithm can be stitched in a manner of multiple rectangular rows and columns arranged side by side. Specifically, it can be divided into different equal parts, such as 2×2, 3×3, 4×4 and other different proportions. The specific proportion of stitching can be determined by the situation of detecting targets in different device channels. For example, if there are more valid channels for triggering targets, but fewer valid targets in each channel, it means that the image content of each device channel is relatively sparse, and the number of stitching areas can be increased (such as 4x4 or a larger ratio), so that the number of divided areas fed into the algorithm at one time is relatively large. If a channel triggers a large number of valid targets, and the priority score of the channel is still high, it means that the image content is dense and the target is important, and the number of stitching areas should be reduced (such as 2×2 or a smaller ratio), that is, fewer stitching areas can be divided.

[0132] See also Figure 4 , provides a stitching diagram of four equal stitching ratios. Among them, each stitching area fed into the algorithm adopts the image equivalent filling method, that is, it is filled with the pixels around the cutout area itself, rather than stretching it to the equally divided stitching area. In this way, the detection effect of the algorithm will be better than the image deformation after stretching.

[0133] Through the above scheme, the image stitching method can be flexibly adjusted dynamically according to the number and priority of targets, thereby improving the performance and effect of the target detection algorithm.

[0134] It should be understood that the first stitched image and / or the second stitched image may also be processed using the stitching method described in any of the above embodiments.

[0135] The following describes the invention in conjunction with specific embodiments. Figure 5 : is a flow chart of another method for controlling a multi-channel camera device according to an embodiment of the present application, the flow chart comprising the following steps:

[0136] Step S501: collect a single channel of images shot by a corresponding camera device through each device channel.

[0137] Step S502: Perform dynamic inspection on each single-channel image or perform inspection and analysis through an SMD module.

[0138] Step S503: The user sets parameters.

[0139] Step S504, through the detection results of each single-channel image in the above step S502, the number of effective targets Oi and the number of effective channels C of the i-th channel are obtained, and the target number threshold ThrO and the intelligent channel number threshold ThrC are obtained through the user parameter setting of the above step S503, and the priority score S of each device channel is calculated. i , which is the priority information mentioned above.

[0140] Specifically, information on motor vehicles, non-motor vehicles, people, etc. in each channel is counted through an intelligent algorithm with low computational complexity, and the number of valid targets Oi in each channel and the total number of channels C that have triggered valid targets are counted according to the intelligent type corresponding to the channel, where i is the channel number.

[0141] According to the usage scenario, set the target threshold ThrO, intelligent path threshold ThrC and channel priority score threshold S in each channel. Among them, the target threshold ThrO represents the number threshold of the channel target. If the number exceeds this threshold, it means that the number of targets is large, and if it is lower than this threshold, it means that the number of targets is small; the intelligent path threshold ThrC indicates the threshold of the number of intelligent paths that you want to enable when multiple channels have targets at the same time. The channel priority score threshold S indicates the priority of enabling intelligent functions separately for this channel.

[0142] Step S505, selecting a smart mode activation plan through statistics and calculations of various parameters.

[0143] Step S506: If the target number O of a channel is detected through the above step S505, i If the number of targets is greater than or equal to the set target number threshold ThrO, it is judged to be case 1, which means that there are too many targets to be detected and the intelligent computing power needs to be fully turned on for intelligent analysis. If the number of targets in only one channel is greater than the threshold ThrO, the intelligence of the channel is directly turned on; if multiple channels are greater than the set target number threshold ThrO at the same time, select several channels of intelligence to be turned on according to the priority score S.

[0144] Step S507: if the target number of each device channel is detected through the above step S505, i If it is lower than the set target number threshold ThrO, and the number of valid channels C is greater than the intelligent path threshold ThrC, it is determined that this belongs to situation 2, which means that the number of targets in each device channel is small, but there are more channels that need to open the intelligent path. At this time, directly turn on a channel for full intelligent analysis or stitch the areas of multiple channels and send them to the intelligent analysis. More specifically, if C to be turned on is less than or equal to the maximum computing power specification path, turn it on directly. If it is greater than the maximum computing power specification path, the images of these channels are spliced, and the intelligent algorithm is turned on to process these fused images, and the detection results of each channel are returned. At this time, only the computing power of one intelligent path of the device is consumed, which can reduce the operating pressure of the device. If a detection target appears in a certain channel at this time, other intelligent computing powers can be directly turned on to achieve the purpose of timely response.

[0145] Step S508: If the target number O of a channel is detected through the above step S505, iIf it is greater than or equal to the set target number threshold ThrO, and the number of valid channels C is greater than the intelligent path threshold ThrC, then it is determined that this belongs to situation 3, which means that each channel detects many targets in this scenario, and almost every channel wants to apply for resources for intelligent analysis. However, due to the limitation of the intelligent specifications of the equipment, assuming that there are only two channels with the maximum computing power specification, there will be two processing methods: Method 1, according to the priority score S, the single channel with the highest score will be selected to open, and then the fusion splicing method will be selected from the remaining channels to send to another computing power for analysis. Method 2, determine the difference between the highest priority score S1 and the second highest priority score S2. If the difference is not large and far exceeds the score S of other channels, the channel intelligence corresponding to S1 and S2 will be enabled first.

[0146] In the above steps S506 to S508, the processing flow for sending the image data to intelligent analysis includes the following steps:

[0147] See also Fig. 6A If the number of smart channels to be enabled in case 1 and case 2 is less than the number of smart channels of the device, the smart channels are directly enabled and the image data collected by the channels are directly sent to the algorithm for analysis. Figure 6B If multiple channels are needed to send intelligence together in case 2 or 3, each channel needs to be spliced ​​and then sent to the algorithm.

[0148] In order to better understand the present solution, a list of enabling the smart mode in various situations is also provided, as shown in the following Table 1:

[0149] Table 1 Smart mode activation schemes in various situations

[0150]

[0151] Through the above embodiment, the intelligent switching method can be used to dynamically open the channel that really needs intelligent analysis, and the effect of expanding the number of intelligent paths of the device can be achieved. This solves the problem that the number of intelligent paths that can be opened in the prior art is far less than the number of channels actually supported by the device for collection and display, the one-to-one correspondence of the opened intelligent data stream is very fixed, the intelligence of the channel can only detect the data collected by the channel, and the opened intelligent channel is fixed, and the device resources cannot be effectively used to make other channels intelligent.

[0152] It should be noted that the steps shown in the above process or the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0153] This embodiment also provides a control device for a multi-channel camera device, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. As used below, the terms "module", "unit", "subunit", etc. can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0154] Figure 7 is a structural block diagram of a control device for a multi-channel camera device according to an embodiment of the present application, such as Figure 7 As shown, the device comprises:

[0155] The acquisition module 71 acquires the single-channel image collected by each camera device under the device channel; one device channel corresponds to one camera device; the detection module 72 is used to detect the single-channel image and calculate the number of valid targets under each device channel; the mode switching module 73 is used to determine the target channel in each device channel based on the number of valid targets, and switch the image processing mode of the target channel to the target intelligent mode.

[0156] In some of the embodiments, the mode switching module 73 is further used to obtain priority information assigned to each device channel; the mode switching module 73 is further used to determine the target channel according to the number of valid targets and the priority information.

[0157] In some of the embodiments, the mode switching module 73 is also used to obtain channel detection information of each device channel; the mode switching module 73 is also used to assign weight values ​​to the number of valid targets and the channel detection information, respectively, and based on the weight values, perform weighted calculations on the number of valid targets and the channel detection information to obtain priority information.

[0158] In some of the embodiments, the mode switching module 73 is also used to search for a preliminary inspection channel from each device channel whose number of valid targets is greater than or equal to a preset target number threshold; the mode switching module 73 is also used to determine the preliminary inspection channel as the target channel when the number of preliminary inspection channels is less than or equal to the preset maximum computing power specification number of channels; when the number of preliminary inspection channels is greater than the maximum computing power specification number of channels, the target channel is screened out from the preliminary inspection channel based on the priority information.

[0159] In some of the embodiments, the mode switching module 73 is also used to determine the effective channels that trigger effective targets in each device channel based on a single-channel image, and calculate the number of effective channels; the mode switching module 73 is also used to determine the target channel based on the number of effective targets and the number of effective channels.

[0160] In some of the embodiments, the mode switching module 73 is further used to compare the number of valid channels with a preset maximum computing power specification number of channels when it is detected that the valid target number of each device channel is less than a preset target number threshold, and the number of valid channels is greater than or equal to a preset intelligent number threshold; if the comparison result of the mode switching module 73 is that the number of valid channels is less than or equal to the maximum computing power specification number of channels, the valid channel number is determined as the target channel; if the comparison result of the mode switching module 73 is that the number of valid channels is greater than the maximum computing power specification number of channels, the single-channel image corresponding to the valid channel is stitched to obtain a first stitched image, and the stitched channel for the first stitched image is determined as the target channel.

[0161] In some of the embodiments, the mode switching module 73 is also used to compare the number of valid channels with a preset intelligent path threshold when searching for a preliminary inspection channel from each device channel whose valid target number is greater than or equal to a preset target number threshold, and the number of valid channels is greater than a preset maximum computing power specification path; if the comparison result of the mode switching module 73 is that the number of valid channels is less than the intelligent path threshold, then based on the priority information assigned to each device channel, the target channel is screened out from the preliminary inspection channel; if the comparison result of the mode switching module 73 is that the number of valid channels is greater than or equal to the intelligent path threshold, then based on the maximum computing power specification path and the priority information, the target channel in the preliminary inspection channel is determined.

[0162] In some of the embodiments, the mode switching module 73 is further used to stitch the single-channel images corresponding to the target channel to obtain a target stitched image, and perform image processing on the target stitched image in the target intelligent mode.

[0163] In some of the embodiments, the mode switching module 73 is also used to determine the cutout area in the single-channel image corresponding to each target channel; the mode switching module 73 determines the stitching ratio based on the number of valid targets, and stitches the cutout area according to the stitching ratio to obtain the target stitching image.

[0164] In some embodiments, the mode switching module 73 is further configured to obtain a cutout region set by a user; or, the mode switching module 73 calculates the cutout region according to the target detection results in each single-channel image.

[0165] It should be noted that the above modules can be functional modules or program modules, and can be implemented by software or hardware. For modules implemented by hardware, the above modules can be located in the same processor; or the above modules can be located in different processors in any combination. For specific examples in this embodiment, reference can be made to the examples described in the above embodiments and optional implementations, and will not be repeated in this embodiment.

[0166] This embodiment further provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0167] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0168] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0169] S1, obtaining a single channel image collected by each camera device under the device channel; one device channel corresponds to one camera device.

[0170] S2, detects a single-channel image and calculates the number of valid targets under each device channel.

[0171] S3, based on the number of valid targets, determining the target channel in each device channel, and switching the image processing mode of the target channel to the target intelligent mode.

[0172] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

[0173] In addition, in combination with the control method of the multi-channel camera device in the above embodiment, the embodiment of the present application can provide a storage medium for implementation. The storage medium stores a computer program; when the computer program is executed by a processor, any of the control methods of the multi-channel camera device in the above embodiment is implemented.

[0174] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0175] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, an image processing logic device based on quantum computing, etc., but are not limited to this.

[0176] Those skilled in the art should understand that the technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0177] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for controlling a multi-channel camera device, characterized in that: The method comprises: Obtaining a single channel of images collected by each camera device under a device channel; one device channel corresponds to one camera device; Detecting the single-channel image and calculating the number of valid targets under each device channel; Determining a target channel in each of the device channels based on the valid target quantity includes: Based on the single-channel image, determine the valid channels that trigger the valid targets in each of the device channels, and calculate the number of valid channels; Determining the target channel according to the number of valid targets and the number of valid channels includes: When it is detected that the number of valid targets of each of the device channels is less than a preset target number threshold, and the number of valid channels is greater than or equal to a preset intelligent path number threshold, the number of valid channels is compared with a preset maximum computing power specification path number; the intelligent path number threshold is a value used to indicate whether to trigger a subsequent image stitching and fusion step; If the comparison result is that the number of valid channels is less than or equal to the number of channels of the maximum computing power specification, the number of valid channels is determined as the target channels; If the comparison result is that the number of valid channels is greater than the number of channels of the maximum computing power specification, the single-channel images corresponding to the valid channels are stitched to obtain a first stitched image, and the stitched channel for the first stitched image is determined as the target channel; The image processing mode of the target channel is switched to the target intelligent mode.

2. The control method according to claim 1, characterized in that: The determining of the target channel in each of the device channels based on the valid target quantity includes: Obtaining priority information assigned to each of the device channels; The target channel is determined according to the number of valid targets and the priority information.

3. The control method according to claim 2, characterized in that: The obtaining of priority information assigned to each of the device channels includes: Obtaining channel detection information of each of the device channels; Weight values ​​are respectively assigned to the number of valid targets and the channel detection information, and based on the weight values, weighted calculation is performed on the number of valid targets and the channel detection information to obtain the priority information.

4. The control method according to claim 2, characterized in that: The step of determining the target channel according to the number of valid targets and the priority information includes: From each of the device channels, searching for a primary inspection channel whose number of valid targets is greater than or equal to a preset target number threshold; When the number of the preliminary inspection channels is less than or equal to the preset maximum computing power specification number of channels, the preliminary inspection channels are determined as the target channels; when the number of the preliminary inspection channels is greater than the maximum computing power specification number of channels, the target channels are screened out from the preliminary inspection channels based on the priority information.

5. The control method according to claim 1, characterized in that: The step of determining the target channel according to the number of valid targets and the number of channels includes: When searching for a preliminary inspection channel whose valid target number is greater than or equal to a preset target number threshold from each of the device channels, and the valid channel number is greater than the preset maximum computing power specification number of channels, comparing the valid channel number with the preset intelligent channel number threshold; If the comparison result is that the number of valid channels is less than the intelligent path number threshold, the target channel is screened out from the initial inspection channels based on the priority information assigned to each of the device channels; If the comparison result is that the number of valid channels is greater than or equal to the intelligent path threshold, the target channel in the initial inspection channel is determined based on the maximum computing power specification path and the priority information.

6. The control method according to claim 5, characterized in that: The determining the target channel in the initial inspection channel based on the maximum computing power specification path number and the priority information includes: Based on the priority information, the target channels are selected from the initially inspected channels, and the number of the target channels matches the number of channels with the maximum computing power specification; or, Based on the priority information, a preset number of priority channels are screened out from the primary inspection channel, and the single-channel images corresponding to other channels are spliced ​​to obtain a second spliced ​​image; the other channels are used to represent the remaining channels in the primary inspection channel except the priority channels; The priority channel and a stitching channel for the second stitched image are determined as the target channels.

7. The control method according to any one of claims 1 to 4, characterized in that: The step of switching the image processing mode of the target channel to the target intelligent mode includes: The single-channel images corresponding to the target channel are stitched together to obtain a target stitched image, and image processing is performed on the target stitched image in the target intelligent mode.

8. The control method according to claim 7, characterized in that: The step of performing stitching processing on the single-channel images corresponding to the target channel to obtain a target stitched image includes: Determine the cutout area in the single-channel image corresponding to each of the target channels; Based on the number of valid targets, a stitching ratio is determined, and the cutout area is stitched according to the stitching ratio to obtain the target stitching image.

9. The control method according to claim 8, characterized in that: The determining of the cutout area in the single-channel image corresponding to each of the target channels includes: Obtain the cutout area set by the user; or, The cutout area is calculated according to the target detection results in each of the single-channel images.

10. A control device for a multi-channel camera device, characterized in that: include: An acquisition module is used to acquire a single channel image captured by each camera device under a device channel; One of the device channels corresponds to one of the camera devices; A detection module, used to detect the single-channel image and calculate the number of valid targets under each device channel; A mode switching module, used for determining a target channel in each of the device channels based on the number of valid targets, and switching the image processing mode of the target channel to a target intelligent mode; The mode switching module is also used to determine the valid channel that triggers the valid target in each of the device channels based on the single-channel image, and calculate the number of valid channels; The mode switching module is further used to compare the number of valid channels with the preset maximum computing power specification number of channels when it is detected that the number of valid targets of each device channel is less than a preset target number threshold and the number of valid channels is greater than or equal to a preset intelligent number threshold; the intelligent number threshold is a value used to indicate whether to trigger a subsequent image stitching and fusion step; The mode switching module is further configured to determine the number of valid channels as the target channels if the comparison result shows that the number of valid channels is less than or equal to the number of channels of the maximum computing power specification; The mode switching module is also used to stitch the single-channel images corresponding to the valid channels to obtain a first stitched image if the comparison result is that the number of valid channels is greater than the maximum computing power specification number, and determine the stitching channel for the first stitched image as the target channel.

11. A storage medium, characterized in that: The storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the control method of the multi-channel camera device according to any one of claims 1 to 9.

Images