Multi-split air conditioner control method and device and storage medium

Abstract

The invention provides a multi-split air conditioner control method and device and a storage medium, and relates to the data processing technology, a first target image in a target area is obtained based on an image collecting device, first human body recognition processing is conducted on the first target image, and a plurality of first human body recognition targets are obtained; extracting a second target image meeting a preset requirement in a previous time sequence in the target area, and verifying the first human body recognition target based on a second human body recognition target in the second target image to obtain a third human body recognition target; obtaining a corresponding point of each third human body recognition target in the first target image, and performing division based on the position relation of all the corresponding points to obtain a corresponding division point group; and a coverage control chart of the multi-split air conditioners in the target area is obtained, the air conditioners to be subjected to joint control and the joint control mode of each air conditioner to be subjected to joint control are obtained based on the division point groups and the coverage control chart, customized regulation and control are conducted on the multi-split air conditioners, and the energy-saving effect is achieved.

Description

Technical Field

[0001] The present invention relates to data technology, and in particular to a multi-split air conditioner control method, device and storage medium. Background Art

[0002] Multi-split air conditioner is a type of user central air conditioner, commonly known as "one to many", which means that one outdoor unit is connected to two or more indoor units through piping. In modern buildings, such as large conference centers, libraries, shopping malls and other public places, intelligent management of air conditioning systems has become increasingly important. These places are usually huge in area, with frequent personnel flow and uneven density. Therefore, how to effectively save energy while meeting comfort has become a major challenge for air conditioning management.

[0003] In the prior art, when regulating the air conditioner, the air conditioner is often turned on and off according to a preset schedule or regional temperature control. However, this method cannot respond to changes in the distribution of people in the area in real time, resulting in energy waste and reduced comfort. In addition, the traditional system cannot distinguish between people who are passing by briefly and those who stay for a long time, resulting in the air conditioner being turned on unnecessary, thereby wasting energy.

[0004] Therefore, how to customize the multi-split air conditioner according to the actual number of people in the air-conditioning temperature control area to achieve energy-saving effects has become an urgent problem that needs to be solved. Summary of the invention

[0005] The present invention provides a multi-split air conditioner control method, device and storage medium, which can perform customized regulation and control on the multi-split air conditioner according to the actual number of people in the air-conditioning temperature control area to achieve energy-saving effect.

[0006] A first aspect of the present invention provides a multi-split air conditioner control method, comprising: Acquire a first target image in the target area based on the image acquisition device, perform a first human body recognition process on the first target image, and obtain a plurality of first human body recognition targets; Extracting a second target image of the target area that meets the preset requirements at a previous time sequence, verifying the first human recognition target based on the second human recognition target in the second target image, and obtaining a third human recognition target; Obtain corresponding points of each third human recognition target in the first target image, and divide the corresponding points based on the positional relationship of all corresponding points to obtain corresponding divided point groups; A coverage control diagram of the multi-split air conditioner in the target area is obtained, and based on the division point group and the coverage control diagram, the air conditioners to be jointly controlled and the joint control mode of each of the air conditioners to be jointly controlled are obtained.

[0007] Optionally, in a possible implementation manner of the first aspect, the image acquisition device acquires a first target image in the target area, and performs a first human body recognition process on the first target image to obtain multiple first human body recognition targets, including: After determining that the image acquisition device acquires the first field of view image of the target area, the target area is divided into effective acquisition sub-areas based on each image acquisition device, and each image acquisition device has a pre-configured effective acquisition sub-area; Taking the image in the first field of view image corresponding to the divided effective acquisition sub-area as the first target image; Based on the open source OpenCV human body recognition model, human body contour recognition processing is performed on the first target image to obtain a first human body recognition target.

[0008] Optionally, in a possible implementation manner of the first aspect, extracting a second target image in which the target area meets a preset requirement at a previous time sequence, verifying the first human recognition target based on a second human recognition target in the second target image to obtain a third human recognition target includes: If the first human recognition target is not empty, extracting a second target image of the target area that is not empty and meets the preset requirement under the previous time sequence, wherein the second target image is at least one, and the preset requirement is at least a preset time length; Comparing a second target attribute of the second human recognition target with a first target attribute of the first human recognition target to obtain a target comparison result, wherein the second target attribute includes at least position information and basic feature information; If the target comparison result passes, the corresponding first human recognition target is used as the third human recognition target. If the target comparison result fails, the first human recognition target is not used as the third human recognition target.

[0009] Optionally, in a possible implementation manner of the first aspect, the comparing the second target attribute of the second human recognition target with the first target attribute of the first human recognition target to obtain a target comparison result, wherein the second target attribute includes at least position information and basic feature information, includes: Coordinate processing is performed on the first target image and the second target image respectively; Extracting the coordinates of the outline pixel points of the first human recognition target in the first target image to obtain a first coordinate point set, extracting the coordinates of the outline pixel points of the second human recognition target in the second target image to obtain a second coordinate point set, wherein the position information includes the coordinate point set; Based on the coordinate relationship between the first coordinate point set and the second coordinate point set, the contours of the first human body recognition target and the second human body recognition target are respectively divided into a plurality of first segmented contours and a second segmented contour according to a preset segmentation strategy, and a pixel value of each segmented contour is extracted to obtain pixel feature information, wherein the basic feature information includes pixel feature information; The first segmentation contour is compared with the second segmentation contour to obtain a target comparison result.

[0010] Optionally, in a possible implementation manner of the first aspect, based on the coordinate relationship between the first coordinate point set and the second coordinate point set, the contours of the first human body recognition target and the second human body recognition target are respectively divided according to a preset division strategy to obtain a plurality of first segmented contours and a second segmented contour, and a pixel value of each segmented contour is extracted to obtain pixel feature information, and the basic feature information includes pixel feature information, including: A first ordinate interval value is obtained based on a difference between a maximum ordinate value and a minimum ordinate value in the first coordinate point set, and a second ordinate interval value is obtained based on a difference between a maximum ordinate value and a minimum ordinate value in the second coordinate point set; Calculate the ratio of the first ordinate interval value to the second ordinate interval value to obtain an interval ratio, obtain a standard coefficient based on the interval ratio, and the interval ratio of each numerical interval corresponds to a preset standard coefficient; According to the number of divisions in the preset division strategy, the first ordinate interval value and the second ordinate interval value are respectively divided into contours to obtain multiple first segmentation contours and second segmentation contours. After the first segmentation contours and the second segmentation contours are grouped, divided and screened, the pixel value of each segmentation contour is extracted to obtain pixel feature information.

[0011] Optionally, in a possible implementation manner of the first aspect, extracting a pixel value of each segmentation contour to obtain pixel feature information after grouping, dividing and screening the first segmentation contour and the second segmentation contour includes: Labeling each first segmented contour in sequence according to the positive direction of the ordinate to obtain a first label, and labeling each second segmented contour in sequence according to the positive direction of the ordinate to obtain a second label; The first segmentation contour and the second segmentation contour are grouped and divided based on the coordinates of the first coordinate point set and the second coordinate point set, and the grouped first segmentation contour and the second segmentation contour are used as contours to be grouped; Filtering the first segmentation contours and the second segmentation contours corresponding to the first and second labels of the preset values, deleting the first and second segmentation contours, and treating the contours to be grouped with the same first and second labels after filtering as a contour group; Pixel feature information corresponding to each first segmented contour and each second segmented contour in each contour group is extracted.

[0012] Optionally, in a possible implementation manner of the first aspect, the grouping and dividing the first segmentation contour and the second segmentation contour based on the coordinates of the first coordinate point set and the second coordinate point set, and using the grouped first segmentation contour and the second segmentation contour as contours to be grouped, includes: Determine the coordinates of the center point based on the extreme value coordinates of all coordinate points in the first coordinate point set of the first segmented contour, generate a movable range according to a preset radius based on the coordinates of the center point, and extract a range coordinate set of the movable range; A second segmentation contour intersecting with the range coordinate set is determined, and the corresponding first segmentation contour and the second segmentation contour are grouped and divided to obtain contours to be grouped.

[0013] Optionally, in a possible implementation manner of the first aspect, comparing the first segmentation contour and the second segmentation contour to obtain a target comparison result includes: Calculating a first pixel mean value of first pixel feature information in a first segmented contour of each contour group, and calculating a second pixel mean value of second pixel feature information in a second segmented contour; Calculate the difference between the first pixel mean and the second pixel mean of each contour group to obtain a comparison difference; The comparison coefficient value is obtained by statistically calculating the weighted sum of the comparison differences between all the first segmented contours and the second segmented contours under the two complete contours. If the comparison coefficient value is greater than the standard coefficient, the target comparison result is judged to be unsuccessful. If the comparison coefficient value is less than or equal to the standard coefficient, the target comparison result is judged to be passed.

[0014] Optionally, in a possible implementation manner of the first aspect, obtaining corresponding points of each third human recognition target in the first target image, and dividing the corresponding points based on positional relationships of all corresponding points to obtain corresponding divided point groups includes: Determine the center point of the third human recognition target as the corresponding point in the first target image; All corresponding points in the same preset air conditioning processing division interval are counted to obtain a corresponding division point group of the corresponding preset air conditioning processing division interval.

[0015] Optionally, in a possible implementation of the first aspect, the acquiring a coverage control map of the multi-split air conditioner in the target area, and obtaining the air conditioners to be jointly controlled and the joint control mode of each of the air conditioners to be jointly controlled based on the division point group and the coverage control map, include: Extracting a coverage control diagram of the multi-split air conditioner in the target area, wherein the coverage control diagram has a corresponding relationship between each air conditioner and an air outlet; Determine the air outlet and air conditioner corresponding to each division point group, and use the determined air conditioner as the air conditioner to be jointly controlled; The joint control mode of each air conditioner to be joint controlled is determined based on the required number of third-party human recognition targets, and each required number has a preset joint control mode.

[0016] A second aspect of the present invention provides a multi-split air conditioner control device, comprising: A recognition module, configured to acquire a first target image in a target area based on an image acquisition device, and perform a first human body recognition process on the first target image to obtain a plurality of first human body recognition targets; A verification module is used to extract a second target image of the target area that meets the preset requirements at a previous time sequence, and verify the first human recognition target based on the second human recognition target in the second target image to obtain a third human recognition target; A division module, used for obtaining corresponding points of each third human recognition target in the first target image, and performing division based on the positional relationship of all corresponding points to obtain corresponding division point groups; The acquisition module is used to acquire the coverage control diagram of the multi-split air conditioner in the target area, and obtain the air conditioner to be jointly controlled and the joint control mode of each air conditioner to be jointly controlled based on the division point group and the coverage control diagram.

[0017] According to a third aspect of the present invention, a storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, it is used to implement the method of the first aspect of the present invention and various possible designs of the first aspect.

[0018] The beneficial effects of the present invention are as follows: 1. The present invention can perform customized regulation and control on multi-split air conditioners according to the actual number of people in the air-conditioning temperature control area to achieve energy-saving effects. First, the present invention can obtain the first target image at the current moment in real time to perform personnel recognition processing, and obtain the first human recognition target at the current moment in each area, that is, the personnel situation corresponding to each area at the current moment, and obtain the second target image corresponding to the historical moment, so as to extract the second human recognition target in the second target image to verify the first human recognition target, so as to determine the third human recognition target corresponding to each area, that is, the staying personnel. Secondly, by grouping and dividing the corresponding points corresponding to the third human recognition target, a division point group is obtained, and the distribution of personnel in the area is quickly identified, so as to subsequently determine the multi-split air conditioners and joint control methods that need to be regulated in combination with the coverage control diagram of the multi-split air conditioner, and realize customized regulation and control of the multi-split air conditioner to achieve energy-saving effects.

[0019] 2. The present invention can extract previous time-series images to compare and verify the positions and features of people in the area, thereby realizing dynamic judgment of the stay of people. By comparing the personnel identification targets in the current and historical images, the people who stay for a long time are confirmed, thereby avoiding the false start-up of the air conditioner caused by people who pass by briefly, and effectively reducing energy consumption. At the same time, the present invention can utilize the open source OpenCV human body recognition model, which can identify the outline of people in the image, determine the center point position of each identification target, obtain corresponding points, and group and divide them according to the positional relationship of the corresponding points to obtain a divided point group. In this way, the distribution of people in the area can be quickly and accurately identified, so that the air conditioner that needs to be regulated and the joint control method can be located according to the distribution of people and the temperature control area of ​​the air conditioner, so as to reduce resource waste and achieve energy-saving effects.

[0020] 3. The present invention can be combined with the coverage control diagram of the multi-split air conditioner to realize the intelligent control of multiple air conditioners in the area. Among them, the present invention can determine the air conditioner to be controlled and its specific joint control method based on the identified personnel distribution information and the control area of ​​each air conditioner in the area, determine the corresponding air conditioner to be adjusted according to the location area where the division point group is located, and at the same time, automatically select the appropriate joint control method according to the number of people in the corresponding area of ​​the air conditioner, so as to ensure the comfort experience of the personnel while achieving energy saving effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A flow chart of a multi-split air conditioner control method provided by the present invention; Figure 2 A schematic diagram of an effective acquisition sub-area provided by the present invention; Figure 3 A schematic structural diagram of a multi-split air conditioning control device provided by the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] The terms "first", "second", "third", "fourth", etc. (if any) in the description and claims of the present invention and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in sequences other than those illustrated or described herein.

[0024] It should be understood that in various embodiments of the present invention, the size of the sequence number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0025] It should be understood that in the present invention, "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0026] It should be understood that in the present invention, "plurality" refers to two or more than two. "And / or" is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "Contains A, B and C", "Contains A, B, C" means that A, B, and C are all included, "Contains A, B or C" means that one of A, B, and C is included, and "Contains A, B and / or C" means that any one, any two, or any three of A, B, and C are included.

[0027] It should be understood that in the present invention, "B corresponding to A", "B corresponding to A", "A corresponds to B" or "B corresponds to A" means that B is associated with A and B can be determined based on A. Determining B based on A does not mean determining B based only on A, but B can also be determined based on A and / or other information. A and B match when the similarity between A and B is greater than or equal to a preset threshold.

[0028] Depending on the context, "if" as used herein may be interpreted as "when" or "when" or "in response to determining" or "in response to detecting."

[0029] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0030] like Figure 1 As shown, the present invention provides a flow chart of a multi-split air conditioner control method, the multi-split air conditioner control method comprising: S1, acquiring a first target image in a target area based on an image acquisition device, performing a first human body recognition process on the first target image, and obtaining a plurality of first human body recognition targets.

[0031] It should be noted that in a large area space, in order to maintain the temperature of the entire area space, multiple air conditioners or air conditioning outlets will be installed in the target area. Therefore, when there are people staying in the area, the air conditioner in the corresponding area can be turned on according to the location of the people, avoiding starting all the air conditioners in the target area at the same time, reducing resource waste, and achieving energy-saving effects. Therefore, the first target image can be obtained according to the image acquisition device in the target area, so as to subsequently identify the people in the image, so as to facilitate the subsequent judgment of the location and method of turning on the air conditioner according to the personnel information, so as to save resources.

[0032] Among them, the image acquisition device is a device for collecting image information, such as a camera, the target area is an area for monitoring and processing, such as a conference hall, a library, etc., the first target image is the image acquired by the image acquisition device, and the space of the target area is usually large. Therefore, there are multiple image acquisition devices at different positions in the target area, and each image acquisition device has a corresponding first target image, so that multiple first target images corresponding to the target area can be obtained.

[0033] At the same time, the first human body is an image of a person in the first target image, and the first human body recognition target is the person in the first target image.

[0034] It is not difficult to understand that an image acquisition device (such as a camera) is used to obtain a first target image in the target area. The first target image contains spatial information at the current moment, and preliminary human recognition processing is performed on the first target image to identify multiple first human recognition targets in the image. This step ensures that the system can detect the presence of people in the current area and their approximate location.

[0035] In some embodiments, the specific implementation of step S1 (obtaining a first target image in the target area based on an image acquisition device, performing a first human body recognition process on the first target image, and obtaining a plurality of first human body recognition targets) includes: S11, after determining that the image acquisition device acquires the first field of view image of the target area, the target area is divided into effective acquisition sub-areas based on each image acquisition device, and each image acquisition device has a pre-configured effective acquisition sub-area.

[0036] It is understandable that, due to the different angles of the image acquisition device, the corresponding image acquisition field of view is also different. Figure 2 As shown, there are image acquisition devices in the southeast, northwest and northeast directions of the target area, and the field of view of each image acquisition device is inconsistent. At the same time, each image acquisition device has a pre-configured effective acquisition sub-area. Since the area that can be actually photographed by the image acquisition device is relatively large, in order to ensure the clarity of the image for subsequent recognition and analysis, the target area can be divided according to the effective acquisition sub-area, so as to subsequently determine the first target image corresponding to each image acquisition device.

[0037] The first field of view image is a field of view image that can be captured by the image acquisition device in the target area, and the effective acquisition sub-area is an effective area for corresponding image acquisition by each image acquisition device.

[0038] S12: Use the image in the first field of view image corresponding to the divided effective acquisition sub-area as the first target image.

[0039] It can be understood that the first target image is an image corresponding to the effectively captured sub-region in the first field of view image, that is, an image that can be effectively and clearly captured in a corresponding image capture device.

[0040] Through the above implementation, the present invention can acquire the first target image so as to subsequently perform human body recognition on the first target image, so as to determine the air conditioner that needs to be regulated, intelligently adjust the operating status of the air conditioner, and improve the energy-saving effect.

[0041] S13, performing human body contour recognition processing on the first target image based on the open source OpenCV human body recognition model to obtain a first human body recognition target.

[0042] It can be understood that through the existing image recognition technology OpenCV human recognition model, the human body contour is identified in the first target image, thereby obtaining the first human body recognition target, so as to facilitate subsequent position judgment, thereby determining the air conditioner that needs to be adjusted, avoiding the turning on of all air conditioners, and achieving energy saving effects.

[0043] The human body contour is the human body contour in the first target image.

[0044] S2, extracting a second target image of the target area that meets the preset requirements at a previous time sequence, verifying the first human recognition target based on the second human recognition target in the second target image, and obtaining a third human recognition target.

[0045] It should be noted that when performing human body recognition on the first target image, a person may be recognized in the target area, but the recognized person may just be a passerby. In order to avoid wasting resources by automatically turning on the air conditioner when a person passes by, it is necessary to judge the position of the corresponding person, so that the second target image can be extracted for recognition, so as to obtain the second human body recognition target to verify the first human body recognition target, and thus determine the person with the same position as the first human body recognition target as the third human body recognition target, and then determine whether the conditions for turning on the air conditioner are met, so that the corresponding air conditioner can be controlled to turn on automatically.

[0046] Among them, the preset requirements are set manually in advance, such as the requirement of selecting images at an interval of 5 seconds. The second target image is the captured image corresponding to the first target image at a different time, for example, it can be an image 5 seconds before the corresponding time of the first target image. The second human recognition target is the person appearing in the second target image, and the third human recognition target is the person staying there who meets the requirement of turning on the air conditioner.

[0047] It is not difficult to understand that the second target images in the previous time series are extracted from the historical data of the target area. These images are selected because they meet certain preset requirements (such as a preset time interval). By comparing the second human recognition target in the second target image, the multiple first human recognition targets in the first target image are verified, and the third human recognition target with a stable position is identified. This step enhances the accuracy of recognition and reduces the risk of misjudging the automatic start of the air conditioner.

[0048] In some embodiments, the specific implementation of step S2 (the extraction of a second target image of the target area that meets the preset requirements at the previous time sequence, and verification of the first human recognition target based on the second human recognition target in the second target image to obtain a third human recognition target) includes: S21, if the first human recognition target is not empty, extracting a second target image of the target area that is not empty and meets the preset requirements in the previous time sequence, the second target image is at least 1, and the preset requirements are at least a preset time length.

[0049] It can be understood that when it is determined that the first human recognition target is not empty, it can be explained that there is a person in the current area, and the second target image that meets the preset requirements in the previous time sequence in the non-empty area can be extracted. For example, when the preset requirement is an interval of 5 seconds and there is a person in area A and it is not empty, then the image corresponding to 5 seconds before the first target image and the image corresponding to 10 seconds before the first target image in the image acquisition device corresponding to area A can be extracted, that is, there is at least one second target image, and it can be multiple.

[0050] The preset time length is a preset time interval length, for example, 5 seconds.

[0051] Through the above implementation, the present invention can acquire the second target image, so as to subsequently identify the second human recognition target according to the second target image and perform position verification on the first human recognition target.

[0052] It is worth mentioning that when the first human recognition target is not identified in the corresponding first target image, it can be said that there is no person in the corresponding area, and there is no need to adjust the air conditioning in the corresponding area. Therefore, there is no need to perform image recognition, that is, the implementation of subsequent steps can be determined by whether there is the first human recognition target in the first target image, which can reduce the amount of data processing.

[0053] S22, comparing the second target attribute of the second human recognition target with the first target attribute of the first human recognition target to obtain a target comparison result, wherein the second target attribute at least includes position information and basic feature information.

[0054] It can be understood that the first target attribute is the image information attribute corresponding to the first human recognition target, and the second target attribute is the image information attribute corresponding to the second human recognition target, wherein the first target attribute and the second target attribute both include at least position information and basic feature information, so that the information with the same attributes can be compared later, the position information is the position of the corresponding person in the image, and the basic feature information is the basic image information corresponding to the person, including pixel feature information, such as the color of clothes in the image corresponding to the first human recognition target or the second human recognition target, etc., so that the movement of people or the number of people can be judged through attribute comparison later, so that the corresponding air-conditioning operation status can be dynamically adjusted.

[0055] The target comparison result is a comparison result after comparing the second target attribute of the second human recognition target with the first target attribute of the first human recognition target, for example, it may be a result of passing the comparison or failing the comparison.

[0056] In some embodiments, a specific implementation of step S22 (comparing the second target attribute of the second human recognition target with the first target attribute of the first human recognition target to obtain a target comparison result, wherein the second target attribute includes at least location information and basic feature information) includes: S221, coordinate processing is performed on the first target image and the second target image respectively.

[0057] It can be understood that in order to subsequently realize the verification of the first human recognition target by the second human recognition target and obtain the corresponding position information, the first target image and the second target image can be coordinate-processed, that is, coordinate systems can be constructed in the first target image and the second target image respectively. Since the first target image and the second target image are images extracted at different times in the same image acquisition device, the specifications of the first target image and the second target image are consistent. Furthermore, the coordinate points in the subsequent images after the coordinate processing are one-to-one corresponding, which is convenient for the subsequent comparison results of the first human recognition target and the second human recognition target.

[0058] S222, extracting the contour pixel point coordinates of the first human recognition target in the first target image to obtain a first coordinate point set, extracting the contour pixel point coordinates of the second human recognition target in the second target image to obtain a second coordinate point set, the position information includes the coordinate point set.

[0059] It can be understood that the contour pixel point coordinates are the pixel point coordinates of the contour of the person image in the image, the first coordinate point set is the set of contour pixel point coordinates corresponding to the first human recognition target, the second coordinate point set is the set of contour pixel point coordinates corresponding to the second human recognition target, and the coordinate point set is a set including multiple pixel point coordinates.

[0060] Through the above-mentioned implementation mode, the present invention can obtain a first coordinate point set and a second coordinate point set, so that the contours of the first human body recognition target and the second human body recognition target can be compared and divided respectively according to the first coordinate point set and the second coordinate point set to obtain a first segmented contour and a second segmented contour, which is convenient for the subsequent verification of the first human body recognition target and the second human body recognition target.

[0061] S223, based on the coordinate relationship between the first coordinate point set and the second coordinate point set, the contours of the first human body recognition target and the second human body recognition target are respectively divided into a plurality of first segmented contours and a second segmented contour according to a preset segmentation strategy, and the pixel value of each segmented contour is extracted to obtain pixel feature information, wherein the basic feature information includes pixel feature information.

[0062] It should be noted that in order to improve the verification accuracy, the contours corresponding to the first human recognition target and the second human recognition target can be segmented respectively, so as to subsequently compare the first segmented contour with the corresponding second segmented contour to obtain a target comparison result.

[0063] It can be understood that the first segmented contour is the contour obtained by segmenting the contour of the first human recognition target, the second segmented contour is the contour obtained by segmenting the contour of the second human recognition target, and the pixel feature information is the feature of the pixel value.

[0064] In some embodiments, the specific implementation of step S223 (based on the coordinate relationship between the first coordinate point set and the second coordinate point set, dividing the contours of the first human body recognition target and the second human body recognition target respectively according to a preset division strategy to obtain a plurality of first segmented contours and second segmented contours, extracting the pixel value of each segmented contour to obtain pixel feature information, and the basic feature information includes pixel feature information) includes: S2231, obtaining a first ordinate interval value based on the difference between the maximum ordinate value and the minimum ordinate value in the first coordinate point set, and obtaining a second ordinate interval value based on the difference between the maximum ordinate value and the minimum ordinate value in the second coordinate point set.

[0065] It can be understood that in order to improve the accuracy of subsequent comparison between the first human recognition target and the second human recognition target, the interval of the vertical coordinate corresponding to the outline of the person can be obtained based on the outline of the person in the captured image, and the vertical coordinate intervals corresponding to different postures of the person sitting and standing in the target area are different. At the same time, when the person leans forward or stands straight, the corresponding body shape is different, which will also affect the vertical coordinate interval. Therefore, in order to determine the changes of the person in the first target image and the second target image, the first vertical coordinate interval value and the second vertical coordinate interval value can be obtained, so as to subsequently obtain the degree of change of the person at different times.

[0066] Among them, the maximum value of the ordinate is the maximum value of the ordinate of the pixel point corresponding to the personnel contour, the minimum value of the ordinate is the minimum value of the ordinate of the pixel point corresponding to the personnel contour, the first ordinate interval value is the numerical interval of the ordinate in the first coordinate point set, that is, the difference between the maximum value and the minimum value of the ordinate in the first coordinate point set, and the second ordinate interval value is the numerical interval of the ordinate in the second coordinate point set, that is, the difference between the maximum value and the minimum value of the ordinate in the second coordinate point set.

[0067] S2232, calculating the ratio of the first ordinate interval value to the second ordinate interval value to obtain an interval ratio, and obtaining a standard coefficient based on the interval ratio, wherein the interval ratio of each numerical interval corresponds to a preset standard coefficient.

[0068] It is understandable that when a person changes, the corresponding contour coordinates of the person will also change, and the pixel feature information will also change. For example, when a person leans to one side, the corresponding person's contour changes, and thus the coordinates of the corresponding contour pixel points also change. Moreover, when a person is wearing red clothes, the number of corresponding red pixels in the image may decrease as the body tilts. Therefore, the first ordinate interval value and the second ordinate interval value can be determined by the pixel coordinates of the person's contour in the image, and compared so as to subsequently determine the third human recognition target, and then determine the air conditioning in the corresponding area for adjustment.

[0069] Among them, the interval ratio is the ratio of different interval values, that is, the ratio of the first ordinate interval value to the second ordinate interval value, and the standard coefficient is the coefficient value of the standard degree to which the person is judged to have not changed in the previous and subsequent time periods, which can be manually set in advance.

[0070] It is not difficult to understand that each interval ratio has a corresponding standard coefficient, and the larger the interval ratio, the greater the degree of personnel movement, and the smaller the corresponding standard coefficient.

[0071] S2233, contour division is performed on the first ordinate interval value and the second ordinate interval value respectively according to the number of divisions in the preset division strategy to obtain a plurality of first segmentation contours and second segmentation contours, and the pixel value of each segmentation contour is extracted after grouping, division and screening the first segmentation contours and the second segmentation contours to obtain pixel feature information.

[0072] It can be understood that the number of divisions is the number of divisions of the contour, for example, it can be 5, and then the contour of the person in the image can be divided into 5 parts vertically.

[0073] It is not difficult to understand that when the corresponding human body contour is divided according to the number of divisions and the first ordinate interval value, the first ordinate interval value can be compared with the number of divisions to obtain the height difference corresponding to each part, so that the minimum ordinate value can be selected as the starting ordinate, and the contour of the first human body recognition target in the first target image can be segmented by accumulating a height difference in sequence, thereby obtaining multiple first segmented contours. Similarly, the contour of the second human body recognition target can be segmented to obtain multiple second segmented contours, so that the first segmented contours and the second segmented contours can be subsequently grouped and divided to obtain corresponding pixel feature information for comparison.

[0074] In some embodiments, a specific implementation of step S2233 (the step of extracting pixel values ​​of each segmentation contour to obtain pixel feature information after grouping, dividing and screening the first segmentation contour and the second segmentation contour) includes: S22331, label each first segmented contour in turn according to the positive direction of the longitudinal coordinate to obtain a first label, and label each second segmented contour in turn according to the positive direction of the longitudinal coordinate to obtain a second label.

[0075] It can be understood that the first label is the label corresponding to the first segmentation contour, for example, multiple first segmentation contours are marked as 1, 2, 3, 4, 5 in sequence, the second label is the label corresponding to the second segmentation contour, for example, multiple second segmentation contours are marked as 1, 2, 3, 4, 5 in sequence, and the positive direction of the ordinate is the positive direction corresponding to the ordinate axis, for example, it can be a vertical upward pointing direction.

[0076] It is not difficult to understand that the multiple first segmentation contours corresponding to the first human recognition target can be numbered in sequence according to the vertical coordinate direction. At the same time, the multiple second segmentation contours corresponding to the second human recognition target can be numbered according to the same direction, so that the segmentation contours with the same number can be determined for comparison later. Dividing the contour into multiple parts for comparison can improve the accuracy of the comparison results.

[0077] S22332, grouping and dividing the first segmentation contour and the second segmentation contour based on the coordinates of the first coordinate point set and the second coordinate point set, and using the grouped first segmentation contour and the second segmentation contour as contours to be grouped.

[0078] It can be understood that the contours to be grouped are segmentation contours that are grouped and combined. For example, the first segmentation contour labeled 1 in the first coordinate point set and the second segmentation contour labeled 1 in the second coordinate point set are divided into the same group, and the first segmentation contour and the second segmentation contour with corresponding labels are used as the contours to be grouped, the first segmentation contour labeled 2 in the first coordinate point set and the second segmentation contour labeled 2 in the second coordinate point set are divided into the same group, the first segmentation contour labeled 3 in the first coordinate point set and the second segmentation contour labeled 3 in the second coordinate point set are divided into the same group, and so on.

[0079] In some embodiments, a specific implementation of step S22332 (grouping and dividing the first segmentation contour and the second segmentation contour based on the coordinates of the first coordinate point set and the second coordinate point set, and using the grouped first segmentation contour and the second segmentation contour as the contour to be grouped) includes: S223321, determining the center point coordinates based on the extreme value coordinates of all coordinate points in the first coordinate point set of the first segmentation contour, generating a movable range according to a preset radius based on the center point coordinates, and extracting a range coordinate set of the movable range.

[0080] It is understandable that since people may swing a small distance at the corresponding position, there will be certain deviations in the contour coordinate point set in the corresponding images at different times. Therefore, in order to determine the second segmentation contour corresponding to the first segmentation contour and facilitate subsequent information comparison, the center point coordinates corresponding to the first segmentation contour area can be determined based on the extreme values ​​of the coordinate points corresponding to the first segmentation contour in the first coordinate point set, and the movable range can be determined together with the preset radius to obtain a range coordinate set, which is convenient for the subsequent determination of the second segmentation contour corresponding to the corresponding first segmentation contour in the second target image, so as to facilitate the subsequent extraction of corresponding information for comparison.

[0081] Among them, the center point coordinates are the coordinates of the center point position of the area corresponding to the first segmentation contour, the preset radius is the radius of the movable distance in a preset short time, and the movable range is the area range in which personnel can move. For example, taking the center point coordinates as the circle coordinates and the preset radius as 1 cm, the movable range of the circle can be determined, and the range coordinate set is the set of coordinates of the position points of the area corresponding to the movable range.

[0082] Through the above implementation, the present invention can obtain a range coordinate set corresponding to the first segmentation contour, so that a second segmentation contour having a position intersection with the movable range can be determined subsequently, thereby improving the accuracy of subsequent comparison results.

[0083] S223322, determine a second segmentation contour that intersects with the range coordinate set, and group and divide the corresponding first segmentation contour and second segmentation contour to obtain contours to be grouped.

[0084] It can be understood that when the second segmentation contour and the range coordinate set have an intersection, it can be said that the second segmentation contour corresponds to the corresponding first segmentation contour. Therefore, the corresponding first segmentation contour and the second segmentation contour can be grouped into one group to obtain the contour to be grouped.

[0085] S22333, filter out the first segmentation contours and the second segmentation contours corresponding to the first and second labels of the preset values, and delete them, and treat the contours to be grouped with the same first and second labels after filtering as a contour group.

[0086] It should be noted that after segmenting the contour of the same human recognition target, the segmented contour corresponding to the largest number usually corresponds to the head of the person, and the features of the human head are roughly the same. Therefore, the topmost segmented contour can be deleted, that is, there is no need to compare the head part in the first target image and the second target image. At the same time, since the area image corresponding to the bottommost segmented contour may be the chair where the person is sitting or the shooting angle, the bottom image is unclear or incomplete. Therefore, the segmented contour with the corresponding number can also be deleted to ensure that the image shot for comparison can reduce the amount of data processing without distortion.

[0087] It can be understood that the preset value is a pre-set value, which may be 1 or 5, so as to delete the segmentation contour corresponding to the value preset by the personnel and reduce the amount of data processing.

[0088] Among them, the contour group is the combination of contours corresponding to the same first and second numbers after screening among the contours to be grouped, that is, the contour combination corresponding to the first number 2 and the second number 2 is a contour group, and the contour combination corresponding to the first number 3 and the second number 3 is a contour group.

[0089] S22334, extracting pixel feature information corresponding to each first segmented contour and each second segmented contour in each contour group.

[0090] It can be understood that the pixel feature information in the corresponding first segmentation contour in the contour group is extracted. For example, the pixel feature information may include the pixel value and the number of pixel points corresponding to the pixel value. At the same time, the pixel feature information in the second segmentation contour is extracted for subsequent comparison to determine whether the person has stayed in the corresponding area for a long time, so as to subsequently adjust the operating status of the air conditioner in the corresponding area.

[0091] S224, comparing the first segmentation contour and the second segmentation contour to obtain a target comparison result.

[0092] It can be understood that the first segmentation contour and the second segmentation contour are compared to determine a target comparison result, so as to provide a data basis for subsequent implementation operations.

[0093] In some embodiments, a specific implementation of step S224 (comparing the first segmentation contour with the second segmentation contour to obtain a target comparison result) includes: S2241, calculating a first pixel mean value of first pixel feature information in a first segmented contour of each contour group, and calculating a second pixel mean value of second pixel feature information in a second segmented contour.

[0094] It can be understood that the first pixel mean is the average value of the first pixel feature information in the first segmentation contour, and the second pixel mean is the average value of the second pixel feature information in the second segmentation contour, which can be the average value of the pixel values ​​corresponding to the pixel points in the contour.

[0095] Through the above implementation, the present invention can obtain the first pixel mean value and the second pixel mean value for subsequent comparison.

[0096] S2242, calculating the difference between the first pixel mean and the second pixel mean of each contour group to obtain a comparison difference.

[0097] It can be understood that the comparison difference is the difference between the first pixel mean and the second pixel mean corresponding to each contour group.

[0098] S2243, a comparison coefficient value is obtained by weighted summing up the comparison differences between all the first segmented contours and the second segmented contours under the two complete contours. If the comparison coefficient value is greater than the standard coefficient, the target comparison result is judged to be unsuccessful; if the comparison coefficient value is less than or equal to the standard coefficient, the target comparison result is judged to be successful.

[0099] It can be understood that the complete contour is the contour corresponding to the person in the image, including the contour corresponding to the first human recognition target in the first target image and the contour corresponding to the second human recognition target in the second target image, so that the comparison difference between the multiple first segmented contours and the second segmented contours after contour segmentation can be weightedly summed to obtain the comparison coefficient value, so as to determine the target comparison result by comparing the comparison coefficient value with the standard coefficient.

[0100] Among them, the comparison coefficient value is the coefficient value of the comparison result of the person image in the first target image and the second target image, that is, the value obtained by weighted summing up the comparison differences of all the first segmented contours and the second segmented contours under the two complete contours, and the standard coefficient is the standard value of the preset comparison result.

[0101] It is not difficult to understand that when the comparison coefficient value is greater than the standard coefficient, it can be explained that the differences between the human body contour information of the corresponding chain are large, and the conditions for the person to stay and turn on the air conditioner are not met. The person may be a passerby, so the target comparison result fails. When the comparison coefficient value is less than or equal to the standard coefficient, it can be explained that the comparison results of the person information in the first target image and the second target image corresponding to different times are consistent, indicating that the person stays in the corresponding target area, and thus, the target comparison result is passed.

[0102] S23, if the target comparison result passes, the corresponding first human recognition target is used as the third human recognition target; if the target comparison result fails, the first human recognition target is not used as the third human recognition target.

[0103] It can be understood that when it is determined that the target comparison result passes, it can be explained that the position of the current first human recognition target has not changed and has been maintained for a certain period of time, and may stay in the target area for a long time. In order to maintain the temperature experience of the personnel, the corresponding first human recognition target can be used as the third human recognition target for subsequent identification statistics, so as to facilitate the determination of the corresponding air conditioner to start. When it is determined that the target comparison result fails, it means that the first human recognition target may be a passerby and will not stay in the corresponding target area for a long time. There is no need to adjust the air conditioner according to the passers-by. Therefore, there is no need to use the first human recognition target as the third human recognition target, thereby improving the accuracy of the subsequent determination of the air conditioner startup result.

[0104] S3, obtaining corresponding points of each third human recognition target in the first target image, and dividing the corresponding points based on the positional relationship of all corresponding points to obtain corresponding divided point groups.

[0105] It can be understood that all corresponding points of all third-party human recognition targets with similar distances can be divided into the same group, among which 3 people are relatively concentrated and correspond to air conditioner No. 1, and the other 5 people are in close positions and correspond to air conditioner No. 2. The air conditioner in the corresponding area can be determined according to the position of the corresponding division point group, so that it can be turned on in a targeted manner later, saving resources while maintaining the comfortable temperature of the people.

[0106] Among them, the corresponding points are the position points of the third human body recognition target in the first target image, which represent the specific position of each human body target. The division point group is the combination of position points for dividing and integrating the corresponding points. The spatial position relationship of all corresponding points is analyzed and reasonably divided to form different division point groups, so as to determine the air conditioner in the corresponding area to be turned on.

[0107] In some embodiments, a specific implementation of step S3 (obtaining corresponding points of each third human recognition target in the first target image, dividing the corresponding points based on the positional relationship of all corresponding points to obtain corresponding divided point groups) includes: S31, determining the center point of the third human recognition target as the corresponding point in the first target image.

[0108] It can be understood that the corresponding point in the first target image is the center point of the third human recognition target.

[0109] It is not difficult to understand that by taking the center point of the third human recognition target as the corresponding point in the first target image, the position information of the corresponding person in the image can be intuitively reflected through the position of the corresponding point, which is convenient for subsequent grouping according to the corresponding points, so as to subsequently determine the air conditioner that needs to be adjusted.

[0110] S32, counting all corresponding points in the same preset air conditioning processing division interval to obtain a corresponding division point group of the corresponding preset air conditioning processing division interval.

[0111] It can be understood that each air conditioner has a regional interval where temperature can be controlled. Thus, multiple corresponding points within the target area can be divided, and all corresponding points in the same preset air conditioning processing division interval can be divided into a group to obtain a corresponding division point group corresponding to the preset air conditioning processing division interval, so as to subsequently adjust the air conditioner corresponding to the preset air conditioning processing division interval.

[0112] The preset air conditioning processing division interval is an area where the air conditioner can perform temperature control, which is artificially pre-divided, and the division point group is a combination of corresponding points in the same preset air conditioning processing division interval.

[0113] S4, obtaining a coverage control diagram of the multi-split air conditioners in the target area, and obtaining the air conditioners to be jointly controlled and the joint control mode of each of the air conditioners to be jointly controlled based on the division point group and the coverage control diagram.

[0114] It can be understood that the coverage control diagram is the coverage area diagram of the corresponding temperature control of the multi-split air conditioner, the air conditioner to be controlled is the air conditioner that needs to be turned on, and the joint control method is the air conditioner start-up method, for example, first gear, second gear, etc., that is, the more people there are, the higher the gear number of the air conditioner.

[0115] It is not difficult to understand that the coverage control diagram of the multi-split air-conditioning system in the target area is obtained, which shows the coverage range of each air conditioner. Combined with the division point group and the coverage control diagram, the joint control method of each air conditioner to be controlled is determined, and it is clear which air conditioners need to be started or adjusted to achieve precise control. This step ensures the optimization of energy efficiency and improvement of comfort of the air conditioner.

[0116] In some embodiments, the specific implementation of step S4 (obtaining the coverage control map of the multi-split air conditioner in the target area, obtaining the air conditioner to be jointly controlled and the joint control mode of each air conditioner to be jointly controlled based on the division point group and the coverage control map) includes: S41, extracting a coverage control diagram of the multi-split air conditioner in the target area, wherein the coverage control diagram has a corresponding relationship between each air conditioner and an air outlet.

[0117] It is understandable that the coverage control diagram of the multi-split air conditioner is obtained to determine the position of each air conditioner and the air outlet, so that the corresponding air conditioner can be controlled to adjust the temperature according to the divided point group.

[0118] S42, determining the air outlet and air conditioner corresponding to each division point group, and taking the determined air conditioner as the air conditioner to be jointly controlled.

[0119] It can be understood that, according to the regional location of the division point group, the air conditioners in the corresponding coverage area are determined as the air conditioners to be jointly controlled, so that the air conditioners to be jointly controlled can be adjusted later.

[0120] Among them, the air conditioner to be controlled is the air conditioner waiting to be turned on for control.

[0121] It is not difficult to understand that when there is no corresponding point in the preset air conditioning processing division interval corresponding to the air conditioner, it means that there is no person in the area, so the air conditioner corresponding to the corresponding area can be turned off to achieve energy saving effect.

[0122] S43, determining a joint control mode for each air conditioner to be joint controlled based on the required number of the third human recognition target, each required number having a preset joint control mode.

[0123] It can be understood that the demand quantity is the number of third-party human recognition targets within the preset air-conditioning processing interval, and the joint control method is the method of adjusting the air-conditioning, for example, it can be the first gear, the second gear, etc., which is pre-set according to the demand quantity.

[0124] like Figure 3 FIG. 1 is a structural flow chart of a multi-split air conditioner control device provided by the present invention, and the multi-split air conditioner control device comprises: The recognition module is used to obtain a first target image in a target area based on an image acquisition device, perform a first human body recognition process on the first target image, and obtain multiple first human body recognition targets.

[0125] The verification module is used to extract a second target image of the target area that meets the preset requirements at a previous time sequence, and verify the first human recognition target based on the second human recognition target in the second target image to obtain a third human recognition target.

[0126] The division module is used to obtain corresponding points of each third human recognition target in the first target image, and divide them based on the positional relationship of all corresponding points to obtain corresponding division point groups.

[0127] The acquisition module is used to acquire the coverage control diagram of the multi-split air conditioner in the target area, and obtain the air conditioner to be jointly controlled and the joint control mode of each air conditioner to be jointly controlled based on the division point group and the coverage control diagram.

[0128] The present invention also provides a storage medium, in which a computer program is stored. When the computer program is executed by a processor, it is used to implement the methods provided by the various embodiments described above.

[0129] Among them, the storage medium can be a computer storage medium or a communication medium. The communication medium includes any medium that facilitates the transmission of a computer program from one place to another. The computer storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer. For example, the storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an application-specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). In addition, the ASIC can be located in a user device. Of course, the processor and the storage medium can also exist in a communication device as discrete components. The storage medium can be a read-only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0130] The present invention also provides a program product, which includes an execution instruction, which is stored in a storage medium. At least one processor of a device can read the execution instruction from the storage medium, and at least one processor executes the execution instruction so that the device implements the methods provided in the above various embodiments.

[0131] In the above-mentioned terminal or server embodiments, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0132] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-split air conditioner control method, characterized in that: include: Acquire a first target image in the target area based on an image acquisition device, perform a first human body recognition process on the first target image, and obtain a plurality of first human body recognition targets; Extracting a second target image of the target area that meets the preset requirements at a previous time sequence, verifying the first human recognition target based on the second human recognition target in the second target image, and obtaining a third human recognition target; Obtain corresponding points of each third human recognition target in the first target image, and divide the corresponding points based on the positional relationship of all corresponding points to obtain corresponding divided point groups; A coverage control diagram of the multi-split air conditioner in the target area is obtained, and based on the division point group and the coverage control diagram, the air conditioners to be jointly controlled and the joint control mode of each of the air conditioners to be jointly controlled are obtained.

2. The multi-split air conditioner control method according to claim 1, characterized in that: The first target image in the target area is acquired based on the image acquisition device, and the first human body recognition processing is performed on the first target image to obtain multiple first human body recognition targets, including: After determining that the image acquisition device acquires the first field of view image of the target area, the target area is divided into effective acquisition sub-areas based on each image acquisition device, and each image acquisition device has a pre-configured effective acquisition sub-area; Taking the image in the first field of view image corresponding to the divided effective acquisition sub-area as the first target image; Based on the open source OpenCV human body recognition model, human body contour recognition processing is performed on the first target image to obtain a first human body recognition target.

3. The multi-split air conditioner control method according to claim 1, characterized in that: The extracting of a second target image of the target area that meets the preset requirements at a previous time sequence, verifying the first human recognition target based on the second human recognition target in the second target image, and obtaining a third human recognition target, comprises: If the first human recognition target is not empty, extracting a second target image of the target area that is not empty and meets the preset requirement under the previous time sequence, wherein the second target image is at least one, and the preset requirement is at least a preset time length; Comparing a second target attribute of the second human recognition target with a first target attribute of the first human recognition target to obtain a target comparison result, wherein the second target attribute includes at least position information and basic feature information; If the target comparison result passes, the corresponding first human recognition target is used as the third human recognition target. If the target comparison result fails, the first human recognition target is not used as the third human recognition target.

4. The multi-split air conditioner control method according to claim 3, characterized in that: The comparing the second target attribute of the second human recognition target with the first target attribute of the first human recognition target to obtain a target comparison result, wherein the second target attribute at least includes position information and basic feature information, includes: Coordinate processing is performed on the first target image and the second target image respectively; Extracting the coordinates of the outline pixel points of the first human recognition target in the first target image to obtain a first coordinate point set, extracting the coordinates of the outline pixel points of the second human recognition target in the second target image to obtain a second coordinate point set, wherein the position information includes the coordinate point set; Based on the coordinate relationship between the first coordinate point set and the second coordinate point set, the contours of the first human body recognition target and the second human body recognition target are respectively divided into a plurality of first segmented contours and a second segmented contour according to a preset segmentation strategy, and a pixel value of each segmented contour is extracted to obtain pixel feature information, wherein the basic feature information includes pixel feature information; The first segmentation contour is compared with the second segmentation contour to obtain a target comparison result.

5. The multi-split air conditioner control method according to claim 4, characterized in that: Based on the coordinate relationship between the first coordinate point set and the second coordinate point set, the contours of the first human body recognition target and the second human body recognition target are respectively divided into a plurality of first segmented contours and a second segmented contour according to a preset segmentation strategy, and a pixel value of each segmented contour is extracted to obtain pixel feature information, wherein the basic feature information includes pixel feature information, including: A first ordinate interval value is obtained based on a difference between a maximum ordinate value and a minimum ordinate value in the first coordinate point set, and a second ordinate interval value is obtained based on a difference between a maximum ordinate value and a minimum ordinate value in the second coordinate point set; Calculate the ratio of the first ordinate interval value to the second ordinate interval value to obtain an interval ratio, obtain a standard coefficient based on the interval ratio, and the interval ratio of each numerical interval corresponds to a preset standard coefficient; According to the number of divisions in the preset division strategy, the first ordinate interval value and the second ordinate interval value are respectively divided into contours to obtain multiple first segmentation contours and second segmentation contours. After the first segmentation contours and the second segmentation contours are grouped, divided and screened, the pixel value of each segmentation contour is extracted to obtain pixel feature information.

6. The multi-split air conditioner control method according to claim 5, characterized in that: The step of extracting pixel values ​​of each segmentation contour to obtain pixel feature information after grouping, dividing and screening the first segmentation contour and the second segmentation contour comprises: Labeling each first segmented contour in sequence according to the positive direction of the ordinate to obtain a first label, and labeling each second segmented contour in sequence according to the positive direction of the ordinate to obtain a second label; The first segmentation contour and the second segmentation contour are grouped and divided based on the coordinates of the first coordinate point set and the second coordinate point set, and the grouped first segmentation contour and the second segmentation contour are used as contours to be grouped; Filtering the first segmentation contours and the second segmentation contours corresponding to the first and second labels of the preset values, deleting the first and second segmentation contours, and treating the contours to be grouped with the same first and second labels after filtering as a contour group; Extract pixel feature information corresponding to each first segmented contour and the second segmented contour in each contour group; The method of grouping and dividing the first segmentation contour and the second segmentation contour based on the coordinates of the first coordinate point set and the second coordinate point set, and using the grouped first segmentation contour and the second segmentation contour as contours to be grouped, comprises: Determine the coordinates of the center point based on the extreme value coordinates of all coordinate points in the first coordinate point set of the first segmented contour, generate a movable range according to a preset radius based on the coordinates of the center point, and extract a range coordinate set of the movable range; A second segmentation contour intersecting with the range coordinate set is determined, and the corresponding first segmentation contour and the second segmentation contour are grouped and divided to obtain contours to be grouped.

7. The multi-split air conditioner control method according to claim 4, characterized in that: The comparing the first segmentation contour and the second segmentation contour to obtain a target comparison result includes: Calculating a first pixel mean value of first pixel feature information in a first segmented contour of each contour group, and calculating a second pixel mean value of second pixel feature information in a second segmented contour; Calculate the difference between the first pixel mean and the second pixel mean of each contour group to obtain a comparison difference; The comparison coefficient value is obtained by statistically calculating the weighted sum of the comparison differences between all the first segmented contours and the second segmented contours under the two complete contours. If the comparison coefficient value is greater than the standard coefficient, the target comparison result is judged to be unsuccessful. If the comparison coefficient value is less than or equal to the standard coefficient, the target comparison result is judged to be passed.

8. The multi-split air conditioner control method according to claim 1, characterized in that: The step of obtaining corresponding points of each third human recognition target in the first target image and dividing the corresponding points based on the positional relationship of all corresponding points to obtain corresponding divided point groups includes: Determine the center point of the third human recognition target as the corresponding point in the first target image; Counting all corresponding points in the same preset air conditioning processing division interval to obtain a corresponding division point group of the corresponding preset air conditioning processing division interval; The method of obtaining a coverage control diagram of the multi-split air conditioner in the target area, and obtaining the air conditioner to be jointly controlled and the joint control mode of each air conditioner to be jointly controlled based on the division point group and the coverage control diagram, includes: Extracting a coverage control diagram of the multi-split air conditioner in the target area, wherein the coverage control diagram has a corresponding relationship between each air conditioner and an air outlet; Determine the air outlet and air conditioner corresponding to each division point group, and use the determined air conditioner as the air conditioner to be jointly controlled; The joint control mode of each air conditioner to be joint controlled is determined based on the required number of third-party human recognition targets, and each required number has a preset joint control mode.

9. A multi-connected air conditioner control device, characterized in that: include: A recognition module, configured to acquire a first target image in a target area based on an image acquisition device, and perform a first human body recognition process on the first target image to obtain a plurality of first human body recognition targets; A verification module is used to extract a second target image of the target area that meets the preset requirements at a previous time sequence, and verify the first human recognition target based on the second human recognition target in the second target image to obtain a third human recognition target; A division module, used for obtaining corresponding points of each third human recognition target in the first target image, and performing division based on the positional relationship of all corresponding points to obtain corresponding division point groups; The acquisition module is used to acquire the coverage control diagram of the multi-split air conditioner in the target area, and obtain the air conditioner to be jointly controlled and the joint control mode of each air conditioner to be jointly controlled based on the division point group and the coverage control diagram.

10. A storage medium, characterized in that: The storage medium stores a computer program, which is used to implement the method according to any one of claims 1 to 8 when executed by a processor.

Images