Device installation detection method and device, medium, product and electronic device

Abstract

The present disclosure provides a device installation detection method, device, medium, product and electronic device, and relates to the technical field of image processing. The method comprises the following steps: obtaining an installation image corresponding to a target device, wherein the installation image comprises a device region corresponding to the target device and a background region; determining a target region in the installation image, wherein the target region comprises the device region and a set region around the device region, and the set region is smaller than the background region; and detecting whether the target device meets a set installation specification according to the target region. The present disclosure can reduce the interference of irrelevant factors around the device on the device installation detection by reducing the detection area, effectively improve the detection accuracy, reduce the data processing amount, and improve the detection efficiency.

Description

Technical Field

[0001] This disclosure relates to the field of image processing technology, and in particular to a method, apparatus, medium, product, and electronic device for detecting device installation. Background Technology

[0002] In related technologies, the installation of some equipment has specific installation requirements. For example, after the indoor unit of an air conditioner is installed, the distance between the top of the indoor unit and the ceiling of the room must exceed a certain distance, and the indoor unit must be installed horizontally and stably to ensure air intake space and safety.

[0003] After the installation engineer completes the equipment installation, he uploads photos of the installed equipment to the review platform. Review staff can judge whether the equipment meets the installation specifications by viewing the photos. However, this method is very labor-intensive and is neither efficient nor accurate. Summary of the Invention

[0004] To overcome the problems existing in related technologies, this disclosure provides a method, apparatus, medium, product, and electronic equipment for testing equipment installation.

[0005] According to a first aspect of the present disclosure, a method for detecting device installation is provided, comprising: Obtain an installation image corresponding to the target device, wherein the installation image includes a device area and a background area corresponding to the target device; Determine a target region in the installation image, the target region including the device region and a defined region surrounding the device region, the defined region being smaller than the background region; Based on the target area, detect whether the target device conforms to the set installation specifications.

[0006] Optionally, determining the target region in the installation image includes: Image recognition is performed on the installation image to determine the original detection box corresponding to the device area. The original detection box is used to determine the position of the device area in the installation image. The original detection frame is expanded. The target region is determined based on the expanded detection bounding box.

[0007] Optionally, the expansion process of the original detection box includes: Determine the first size information corresponding to the original detection frame; Based on the first size information, determine the expansion distance corresponding to the boundary to be expanded of the original detection frame; The original detection box is expanded according to the expansion distance corresponding to the boundary to be expanded; The step of determining the expansion distance corresponding to the expansion boundary of the original detection frame based on the first size information includes: Based on the first size information, the extension distance corresponding to the set boundary of the original detection frame is determined, and the set boundary is determined according to the relative positional relationship between the reference object and the target device.

[0008] Optionally, determining the expansion distance corresponding to the expansion boundary of the original detection box based on the first size information further includes: Based on the first size information and the set scaling factor, the expansion distances corresponding to the other boundaries of the original detection frame are determined respectively. The other boundaries are at least one boundary among all the boundaries of the original detection frame other than the set boundary.

[0009] Optionally, expanding the original detection box according to the expansion distance corresponding to the boundary to be expanded further includes: Based on the second size information of the installation image and the original detection box, determine the maximum expansion distance corresponding to the boundary to be expanded; Based on the numerical relationship between the expansion distance corresponding to the boundary to be expanded and the maximum expansion distance, the target expansion distance corresponding to the boundary to be expanded is determined; The original detection box is expanded based on the target expansion distance corresponding to the boundary to be expanded.

[0010] Optionally, determining the target region based on the expanded detection bounding box includes: Based on the expanded detection box, the installation image is cropped to determine the target region.

[0011] Optionally, the step of performing image recognition on the installation image to determine the original detection box corresponding to the device area includes: The installation image is subjected to image recognition using a target detection model to determine the original detection box corresponding to the device area.

[0012] Optionally, obtaining the installation image corresponding to the target device includes: In response to acquiring multiple images, image recognition is performed on the multiple images using an object classification model to determine the installation image from the multiple images; And / or, The step of detecting whether the target device conforms to the set installation specifications based on the target area includes: By installing a classification model based on the target area, it is determined whether the target device conforms to the set installation specifications.

[0013] Optionally, the target device conforms to the set installation specifications, including: The distance between the target device and the relative surfaces of the reference object is greater than a set distance; The angle between the target device and the relative surfaces of the reference object is less than a set angle.

[0014] Optionally, the target device is the indoor unit of an air conditioning unit, and the reference object is the ceiling or wall of the room; or, The target device is the outdoor unit of an air conditioning unit, and the reference object is the obstruction or wall facing the air outlet of the outdoor unit.

[0015] According to a second aspect of the present disclosure, a device for detecting equipment installation is provided, comprising: The acquisition module is configured to acquire an installation image corresponding to the target device, wherein the installation image includes a device area and a background area corresponding to the target device; A determination module is configured to determine a target region in the installation image, the target region including the device region and a set region surrounding the device region, the set region being smaller than the background region; The detection module is configured to detect whether the target device conforms to the set installation specifications based on the target area.

[0016] According to a third aspect of the present disclosure, a computer-readable storage medium is provided having computer program instructions stored thereon, which, when executed by a processor, implement the device installation detection method described in the first aspect of the present disclosure.

[0017] According to a fourth aspect of the present disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the device installation detection method described in the first aspect of the present disclosure.

[0018] According to a fifth aspect of the present disclosure, an electronic device is provided, comprising: Storage device for storing computer programs; An execution device is used to execute the computer program to implement the device installation detection method described in the first aspect of the present disclosure.

[0019] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects: This disclosure identifies a target area within an installation image corresponding to the target device and, based on this target area, detects whether the target device conforms to a set installation specification. The installation image includes the device area corresponding to the target device and a background area; the target area includes the device area and a surrounding set area, which is smaller than the background area. Thus, by reducing the detection area, interference from irrelevant factors around the device on the device installation detection can be reduced, effectively improving detection accuracy, reducing data processing volume, and increasing detection efficiency.

[0020] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0021] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0022] Figure 1 This is a flowchart illustrating a device installation detection method according to an exemplary embodiment.

[0023] Figure 2 This is a schematic diagram of a raw detection frame according to an exemplary embodiment.

[0024] Figure 3 This is a schematic diagram of another original detection frame according to an exemplary embodiment.

[0025] Figure 4 This is a schematic diagram illustrating the principle of determining a target area according to an exemplary embodiment.

[0026] Figure 5 This is a schematic diagram illustrating a MobileNet network structure according to an exemplary embodiment.

[0027] Figure 6 This is a schematic diagram illustrating a ResNet network structure according to an exemplary embodiment.

[0028] Figure 7 This is a flowchart illustrating a method for identifying an installation image of a target device according to an exemplary embodiment.

[0029] Figure 8 This is a flowchart illustrating a device installation detection method according to an exemplary embodiment.

[0030] Figure 9 This is a block diagram illustrating a detection device installed in an apparatus according to an exemplary embodiment.

[0031] Figure 10This is a block diagram illustrating an electronic device according to an exemplary embodiment.

[0032] Figure 11 This is a block diagram illustrating an apparatus for device installation detection according to an exemplary embodiment. Detailed Implementation

[0033] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0034] In related technologies, the installation of some equipment has specific requirements. For example, most wall-mounted air conditioners use top air intake, and the air intake space directly affects the airflow and the heat exchange efficiency of the evaporator, thus affecting the cooling or heating performance. Therefore, the distance between the top of the indoor unit and the ceiling of the room should not be less than a certain distance, such as 15cm. Similarly, to ensure that the heat generated by the stove during use is not directly conducted to the wall, thereby preventing damage to the wall materials due to prolonged heat exposure, the net distance between the gas stove and the wall should not be less than 10cm.

[0035] In addition, the installed equipment may have horizontal or vertical installation requirements relative to the reference object. That is, the equipment should be horizontal or vertical relative to the reference object, and the angle between the opposite surfaces of the equipment and the reference object should be less than the set angle threshold.

[0036] In some inspection methods, after the installation engineer completes the equipment installation, he uploads photos of the installed equipment to the auditing platform. Auditing staff can then judge whether the equipment meets the installation specifications by viewing the photos. However, this method is very labor-intensive, inefficient, and inaccurate. Furthermore, the inspection is not accurate when the staff has limited experience.

[0037] In other inspection methods, the operating status of electrical equipment can be remotely monitored, and the installation status can be determined based on the operating data. However, due to the multi-dimensionality and multivariate coupling of the operating data, detecting whether the installation distance is too close by detecting the operating status has a relatively low accuracy rate.

[0038] Therefore, a method for detecting equipment installation is proposed, which can use image recognition models or algorithms to detect whether the equipment installation complies with specifications. However, directly using images taken by the installation engineer for model judgment to obtain the final detection result, without considering the influence of irrelevant factors in the images, may lead to inaccurate detection results.

[0039] Reference Figure 1 , Figure 1 This is a flowchart illustrating a device installation detection method according to an exemplary embodiment, such as... Figure 1 As shown, the detection method for the installed equipment includes the following steps.

[0040] In step S101, an installation image corresponding to the target device is obtained, the installation image including the device area and background area corresponding to the target device.

[0041] In step S102, a target area in the installation image is determined. The target area includes the device area and a set area surrounding the device area, and the set area is smaller than the background area.

[0042] In step S103, based on the target area, it is detected whether the target device conforms to the set installation specifications.

[0043] For example, the target equipment is equipment that has already been installed. The target equipment could be the outdoor or indoor unit of an air conditioning unit, or it could be a stove, some light fixtures, switches, etc. The installation specifications are set according to the design requirements or relevant installation standards of the target equipment. Setting installation specifications may include distance conditions and / or relative angle conditions. For example, conditions such as maintaining a safe distance or a fixed relative angle relationship between the equipment and its surrounding environment or other equipment during installation.

[0044] For example, the installation image is an image that includes the target device. The installation image includes a device area corresponding to the target device and a background area. The background area includes a large amount of surrounding environmental information about the target device. The target area includes the device area of ​​the target device in the installation image, and a surrounding defined area. The defined area is smaller than the background area, meaning that the environmental information within the defined area is less than the environmental information in the background area.

[0045] For example, the device area is the area occupied by the target device in the installation image. Environmental information refers to various information about the environment surrounding the target device, such as the location and status of walls, floors, ceilings, obstacles, other equipment or facilities, etc.

[0046] For example, the setting area and installation specifications can be set according to the actual situation of the target device and the testing requirements. For instance, if the target device is an indoor air conditioner unit, and the installation specifications stipulate that the distance between the installation position of the indoor unit and the ceiling of the room is not less than 15cm, the setting area can be the area after extending the original detection frame above the device area by its height, or the area after extending it upwards by half the height of the original detection frame, etc. The distance between the setting area and the device area can also be set to a preset value. For example, the original detection frame can be extended upwards by 50 pixels, etc. This is only an example and is not intended to limit this implementation method.

[0047] For example, after the installation engineer completes the installation of the target device, they can use a mobile phone camera or other image acquisition device to capture images of the target device's installation location. These photos can then be uploaded to the acceptance platform to obtain the corresponding installation image of the target device.

[0048] For example, there are often other objects unrelated to the installation inspection near the installation location of the target equipment. For instance, near the installation location of the indoor unit of an air conditioner, there may be objects such as curtains, windows, and wall decorations, which can interfere with the inspection of whether the indoor unit's installation complies with regulations.

[0049] Before detecting the installation location of the target device, the target area in the installation image can be determined. The target area includes the device area where the target device is located, as well as a surrounding defined area, which is smaller than the background area of ​​the original installation image. The background area typically contains a large amount of information unrelated to the device installation, which may interfere with the detection algorithm's judgment. Whether using manual identification, image recognition algorithms, or image recognition models, by narrowing down the actual detection area, the focus of detection can be concentrated on the device and its surrounding key areas, rather than the entire background area, thus eliminating interference from other irrelevant areas.

[0050] For example, image recognition algorithms, image recognition models, etc., can be used to detect whether the target device conforms to the set installation specifications based on the target area.

[0051] For example, image recognition algorithms such as convolutional neural networks (CNNs) can be used to extract and analyze features of the target area to determine whether the target device meets the set installation specifications.

[0052] For example, target detection algorithms, such as the R-CNN series, YOLO (You Only Look Once) series, and SSD (Single Shot MultiBox Detector), can be used to locate and identify key points of the target device and its corresponding reference object within the target area. Furthermore, by combining this information with the actual dimensions of the target device and possible perspective data, it can be determined whether the device conforms to the established installation specifications.

[0053] For example, convolutional neural network models such as AlexNet, VGG, and ResNet can be used to extract features and classify target regions to determine whether the device meets the set installation specifications.

[0054] Alternatively, a binary classification model can be used to determine whether the equipment conforms to the set installation specifications. During training, experts can label the installation images of the target equipment; images of the target equipment conforming to the set installation specifications are marked as qualified, while images that do not are marked as unqualified. Alternatively, images where the target equipment is installed far from a reference object are marked as qualified, while images where the target equipment is installed close to the reference object are marked as unqualified. The binary classification model learns the phase distance and / or relative angle information between the target equipment's installation position and the reference object to determine whether the target equipment's installation conforms to the set installation specifications.

[0055] It should be noted that in the installation image, if the distance between the installation location of the target device and the reference object is far, the reference object may not exist in the target area. However, since the distance between the target device and the reference object has been determined to meet the distance requirements even when the target device is outside the defined area, it can still be determined that the target device meets the distance requirements corresponding to the defined installation specifications.

[0056] This disclosure identifies a target area within an installation image corresponding to the target device and, based on this target area, detects whether the target device conforms to a set installation specification. The installation image includes the device area corresponding to the target device and a background area; the target area includes the device area and a surrounding set area, which is smaller than the background area. Thus, by reducing the detection area, interference from irrelevant factors around the device on the device installation detection can be reduced, effectively improving detection accuracy, reducing data processing volume, and increasing detection efficiency.

[0057] As an optional embodiment, determining the target region in the installation image includes: Image recognition is performed on the installation image to determine the original detection box corresponding to the device area. The original detection box is used to determine the position of the device area in the installation image. The original detection frame is expanded. The target region is determined based on the expanded detection bounding box.

[0058] For example, the original detection box is a rectangular box used to locate the device area in the installation image. The position and extent of the target device in the image can be determined based on the original detection box. The original detection box marks the position and extent of the device area by using coordinate information and size parameters in the image coordinate system.

[0059] The original detection bounding box can be described in two ways. For example... Figure 2 As shown, the original detection box can mark the location and extent of the device area using the coordinates (x, y) of its center point, its length w, and its height h. Figure 3 As shown, the original detection box can mark the location and extent of the device area using the coordinates (x1, y1) and (x2, y2) of its diagonal points. The diagonal points of the original detection box can be either the top-left and bottom-right corners, or the bottom-left and bottom-right corners; there are no restrictions here.

[0060] For example, the target area is the region marked on the installation image by the expanded detection frame after the original detection frame has been expanded. Based on the determined target area, it is possible to detect whether the target device conforms to the set installation specifications. The target area includes the device area corresponding to the target device and the defined area surrounding the device area.

[0061] In this context, corresponding to the original detection frame, when the original detection frame represents the position and extent of the device area using the coordinate information (x, y) of its center point, the length information w, and the height information h, the expanded detection frame also represents the area using the coordinate information (x', y') of its center point, the length information w', and the height information h'. For example... Figure 4 As shown, when the original detection box represents the position and range of the device area using the coordinate information (x1, y1) and (x2, y2) of its diagonal points, the expanded detection box is also represented using the coordinate information (x1', y1') and (x2', y2') of its diagonal points.

[0062] For example, object detection algorithms, image segmentation algorithms, etc., can be used to determine the original detection box corresponding to the device region based on the installation image. For instance, the SAM (Segment Anything Model) visual large-scale model can be used to segment the device region in the installation image, and based on the determined device region, the coordinate information of the minimum bounding rectangle corresponding to the device region can be determined as the original detection box. For example, the Grounded-SAM model can be used to perform object detection on the device region, and the model can output the coordinate information of the upper left and lower right corners of the original detection box.

[0063] This disclosure expands the original detection frame corresponding to the device area to ensure that the expanded image can include the device area and the surrounding set area. The set area includes environmental information near the device installation location, which can reduce the processing of irrelevant areas during device installation detection and reduce the interference of irrelevant factors on device installation detection, making device installation detection more efficient and accurate.

[0064] As an optional embodiment, the expansion processing of the original detection box includes: The set boundary of the original detection frame is extended by a set distance, and the set boundary is determined according to the relative positional relationship between the reference object and the target device.

[0065] For example, the distance setting can be adjusted according to the actual situation. Distance in the image coordinate system can be described in pixels, and different image resolutions correspond to different set distances. For instance, if the image resolution is 96 DPI, the corresponding set distance could be 80 pixels; if the image resolution is 300 DPI, the corresponding set distance could be 200 pixels.

[0066] For example, the boundary can be determined based on the relative positional relationship between a reference object and the target device, wherein the reference object is, for example, a wall, floor, ceiling or other device.

[0067] For example, if the target device is the indoor unit of an air conditioner and the reference object is the top of the room, that is, the reference object is above the indoor unit, then the boundary can be the upper boundary of the original detection frame.

[0068] For example, the target device is the outdoor unit of an air conditioner, and the reference object is the obstacle or wall facing the air outlet of the outdoor unit, i.e., the reference object is in front of the indoor unit. In this case, since the shooting angle may be uncertain, the original detection frame can be expanded in all four directions. Therefore, the set boundary can be set as the four boundaries of the original detection frame.

[0069] In other examples, the target boundary can be determined by displaying a shooting frame on the terminal and prompting the user to align the target boundary of the shooting frame with a certain edge of the target device.

[0070] In other examples, the orientation of the target device in the installation image can be determined in advance, and then the set boundary can be determined. For example, in the outdoor unit of an air conditioner, the relative position of the outdoor unit's air outlet and the original detection frame can be determined through image recognition, and the boundary corresponding to the air outlet and the original detection frame can be determined as the set boundary.

[0071] As an optional embodiment, the expansion processing of the original detection box includes: Determine the first size information corresponding to the original detection frame; Based on the first size information, determine the expansion distance corresponding to the boundary to be expanded of the original detection frame; The original detection box is expanded according to the expansion distance corresponding to the boundary to be expanded; The step of determining the expansion distance corresponding to the expansion boundary of the original detection frame based on the first size information includes: Based on the first size information, the extension distance corresponding to the set boundary of the original detection frame is determined, and the set boundary is determined according to the relative positional relationship between the reference object and the target device.

[0072] For example, the first size information includes the length and width information of the original detection box, wherein the first size information is used to characterize the size and proportion of the original detection box in the mounting image. Based on the first size information, the extension distance of different boundaries of the original detection box can be determined to identify the target region in the mounting image.

[0073] For example, the boundary to be expanded is the boundary among all the boundaries of the original detection frame that needs to be expanded. The boundary to be expanded may include a defined boundary and other boundaries. The defined boundary is determined based on the relative positional relationship between the reference object and the target device. Other boundaries are at least one boundary among all the boundaries of the original detection frame other than the defined boundary. The boundaries of the original detection frame may include a left boundary, a right boundary, a top boundary, and a bottom boundary.

[0074] The defined boundary can be the relative direction between the reference object and the target device. For example, if the target device is the indoor unit of an air conditioner and the reference object is the top of the room (i.e., the reference object is above the indoor unit), then the defined boundary can be the upper boundary of the original detection frame.

[0075] Setting the extension distance based on the original pixels of the image may present some problems. For example, in an installation image, if the device area corresponding to the target device is too large, determining the target area based on the set distance may result in insufficient environmental information included in the target area. Similarly, if the device area corresponding to the target device in an installation image is too small, it may result in excessive installation environmental information included in the target area.

[0076] For example, when the original detection frame marks the position and extent of the device area using the coordinates (x, y) of its center point, its length w, and its height h, the first dimension information of the original detection frame can be determined to include the length w and height h. When the original detection frame marks the position and extent of the device area using the coordinates (x1, y1) and (x2, y2) of its diagonal points, the length information of the original detection frame can be determined by x2-x1, and the width information of the original detection frame can be determined by y2-y1.

[0077] For example, here we can extend only the defined boundary of the original detection frame, where the defined boundary is determined based on the relative positional relationship between the reference object and the target device. That is, in order to determine whether the distance between the target device and the reference object conforms to the defined installation specifications, the boundary of the original detection frame corresponding to the target device can be extended only in the direction corresponding to the reference object. And the target area is determined based on the area after extending the defined boundary by the extended distance.

[0078] For example, the extension distance corresponding to the set boundary is the distance extended on the set boundary of the original detection box. The extension distance can be set according to the size corresponding to the first size information. For example, the extension distance corresponding to the set boundary can be set to any one of the length of the original detection box, half of the length, the width, and half of the width.

[0079] For example, in the installation image of the target device taken by the installation engineer, if the image is taken facing the target device, the length and width ratio of the original detection frame is approximately the same as the actual length or width ratio of the target device.

[0080] It should be noted that even if the target device is tilted relative to the boundary of the installation image, it can still be used to inspect the device installation. The photographs taken are generally not frontal views. Figure 3 As shown, if the target device is tilted relative to the boundary of the installation image, and the set boundary is the lower boundary of the original detection box, then the S2 region corresponding to the original detection box already includes some environmental information around the target device. After expanding the set boundary of the original detection box based on the expansion distance corresponding to the set boundary, the expanded region is the S1 region, which is equivalent to adding environmental information in the expansion direction corresponding to the set boundary.

[0081] Here, the extension distance corresponding to the set boundary of the original detection frame determined based on the first size information can be correlated with the size of the original detection frame to make the size of the determined target area more appropriate, that is, to make the amount of installation environment information contained in the target area more appropriate. This avoids excessive interference from irrelevant factors in the equipment installation and testing, and also ensures that the target area contains sufficient installation environment information for the equipment installation and testing.

[0082] As an optional embodiment, determining the expansion distance corresponding to the expansion boundary of the original detection frame based on the first size information further includes: Based on the first size information and the set scaling factor, the expansion distances corresponding to the other boundaries of the original detection frame are determined respectively. The other boundaries are at least one boundary among all the boundaries of the original detection frame other than the set boundary.

[0083] For example, the extension of other boundaries can be used to ensure that the target area can include various aspects of the installation environment surrounding the device. For instance, when the boundary is set as the upper boundary of the original detection frame, the other boundaries can be at least one of the following: the left, right, and lower boundaries of the original detection frame. For instance, when the boundary is set as the upper and left boundaries of the original detection frame, the other directions can be at least one of the following: the right and lower boundaries of the original detection frame.

[0084] For example, the extension distances corresponding to the other boundaries are determined based on the first dimension information, and are used to determine the extension distances of the other boundaries of the original detection frame. The set scaling factor is a preset value that can be set according to actual conditions to control the extent of extension of the other boundaries of the original detection frame. For example, the scaling factor can be set to 0.1.

[0085] For example, if the other boundaries of the original detection box are the boundaries in the length direction, i.e., the left and right boundaries, the expansion distance can be determined based on the length information of the original detection box; if the other boundaries of the original detection box are the boundaries in the width direction, i.e., the top and bottom boundaries, the expansion distance can be determined based on the width information of the original detection box.

[0086] For example, the left and right boundaries of the original detection box can be extended by any of the following distances: the length value corresponding to the length information, half of the length value, the product of the length value and a set ratio, etc. At the same time, the upper and lower boundaries of the original detection box can be extended by any of the following distances: the width value corresponding to the width information, half of the width value, the product of the width value and a set ratio, etc.

[0087] For example, simply expanding the set boundary may result in less environmental information in the target area. By expanding multiple boundaries of the original detection box, more installation environment information can be provided for the detection of the device installation, and the amount of information in the background area in the original installation image can be reduced. This allows for flexible adaptation to different device installation environments and ensures that the target area containing the necessary information can be accurately determined under various circumstances.

[0088] For example, a reference object for the target device exists in the direction corresponding to the defined boundary of the original detection frame, while no reference object exists in the directions corresponding to other boundaries. Therefore, the defined boundary can be extended by a larger distance to ensure that the extended area corresponding to the defined boundary has more information for device installation detection; by setting a defined scaling factor, the distance can be extended by a smaller distance in the directions corresponding to other boundaries, which can be used to avoid more factors that are irrelevant to device installation detection.

[0089] For example, when the set boundary is the boundary in the length extension direction of the original detection frame, i.e., the left boundary and / or the right boundary, the corresponding extension distance on the set boundary of the original detection frame can be the length or half of the length; when the set boundary is the boundary in the width extension direction of the original detection frame, i.e., the upper boundary and / or the lower boundary, the corresponding extension distance on the set boundary of the original detection frame can be the width or half of the width.

[0090] For example, other boundaries may include boundaries in the length extension direction or boundaries in the width extension direction. For boundaries in the length extension direction, i.e., the left and / or right boundaries, the extension distance corresponding to the boundary can be set to the product of the length and a set scaling factor; for boundaries in the width extension direction, i.e., the upper and / or lower boundaries, the extension distance corresponding to the boundary can be set to the product of the width and a set scaling factor.

[0091] For example, by expanding the original detection box according to the expansion distance corresponding to each boundary, the coordinate points or position information corresponding to the expanded detection box can be obtained. Based on the coordinate points or position information corresponding to the expanded detection box, the target area for equipment installation inspection can be determined.

[0092] In addition, when using models or algorithms to detect whether a device conforms to the set installation specifications, compared to simply expanding the set boundary, expanding the original detection box in multiple directions to obtain the clipping area can provide richer reference information, enhancing the robustness of the model or algorithm and the reliability of the detection results.

[0093] As an optional embodiment, the step of expanding the original detection box according to the expansion distance corresponding to the boundary to be expanded further includes: Based on the second size information of the installation image and the original detection box, determine the maximum expansion distance corresponding to the boundary to be expanded; Based on the numerical relationship between the expansion distance corresponding to the boundary to be expanded and the maximum expansion distance, the target expansion distance corresponding to the boundary to be expanded is determined; The original detection box is expanded based on the target expansion distance corresponding to the boundary to be expanded.

[0094] For example, the second size information is the size of the mounting image, including width and height. The maximum expansion distance is the maximum distance that can be expanded for each boundary of the original detection box to be expanded. The maximum expansion distance is determined by the position of the original detection box in the mounting image and the size of the mounting image, and can be used to ensure that the expanded area does not exceed the range of the mounting image. The target expansion distance refers to the final expansion distance determined for each boundary of the original detection box based on the numerical relationship between the expansion distance corresponding to the boundary to be expanded and the maximum expansion distance.

[0095] For example, based on the position of the original detection box in the mounted image and the second size information of the mounted image, the maximum distance that each boundary of the original detection box can extend can be determined. For instance, if the original detection box is 100 pixels away from the left edge of the image, then the maximum extension distance corresponding to the left boundary is 100 pixels.

[0096] For example, for each boundary of the original detection box to be expanded, the numerical relationship between the expansion distance corresponding to the boundary to be expanded and the maximum expansion distance can be compared to determine the target expansion distance for each boundary. For instance, when the expansion distance corresponding to the boundary to be expanded is greater than the maximum expansion distance, the maximum expansion distance can be determined as the target expansion distance; when the expansion distance corresponding to the boundary to be expanded is less than or equal to the maximum expansion distance, the expansion distance corresponding to the boundary to be expanded can be determined as the target expansion distance. Then, based on the target expansion distance corresponding to each boundary of the original detection box, the original detection box is expanded to determine the target region in the installation image.

[0097] In the above operations, if the size information of the installation image is not considered, the target area determined after expanding the detection box may exceed the boundary of the installation image. The excess part can also be filled with black or white pixels. However, this filling operation may be redundant, and the pixels in the area exceeding the installation image are redundant information, which may reduce the efficiency of device installation detection.

[0098] Here, by taking into account the size information of the installation image, it is ensured that the target area after expanding the detection box will not exceed the range of the installation image, thus avoiding invalid filling operations and possible incorrect detection based on filled pixels, and further improving the efficiency and accuracy of device installation detection.

[0099] As an optional embodiment, determining the target region based on the expanded detection box includes: Based on the expanded detection box, the mask image corresponding to the installation image is determined; The target area is determined based on the mask image and the installation image.

[0100] For example, a mask image can be used to highlight specific regions in an image, and the pixel values ​​in the mask image can indicate whether a pixel belongs to the target region. For instance, pixels corresponding to the target region in the mask image are 1, and pixels outside the target region are 0.

[0101] For example, multiplying the pixel values ​​of the mask image and the mounting image yields the target region in the mounting image. Outside the target region of the multiplied image, black pixels still exist.

[0102] As an optional embodiment, determining the target region based on the expanded detection box includes: Based on the expanded detection box, the installation image is cropped to determine the target region.

[0103] For example, by cropping the installation image, a target area for equipment installation inspection can be determined, so that the equipment installation inspection focuses on the equipment and the designated area around it, reducing interference from background information and improving the accuracy and efficiency of the inspection.

[0104] In one example, the target region determined by the mask image and the mounting image contains more redundant information compared to the target region determined after cropping the image. That is, after the target region is determined by the mask image and the mounting image, when performing device mounting detection by image detection algorithms or image detection models, the model still calculates black pixels outside the target region, which may increase unnecessary calculations.

[0105] As an optional embodiment, cropping the installation image to determine the target area includes: The installation image is cropped to obtain the middle region image; The intermediate region image is processed to grayscale to determine the target region.

[0106] For example, grayscale processing is converting a color image to a grayscale image. Grayscale processing can reduce the complexity of an image and highlight edge and shape information within it.

[0107] For example, after cropping the installation image, the target area can be pre-processed with grayscale to avoid interference from factors such as image color and lighting. This allows the model to fully recognize the relative distance and / or relative angle information between the target device and surrounding reference objects, thereby improving the reliability of the detection results.

[0108] As an optional embodiment, the step of performing image recognition on the installation image to determine the original detection box corresponding to the device area includes: The installation image is subjected to image recognition using a target detection model to determine the original detection box corresponding to the device area.

[0109] For example, an object detection model can be used to identify target devices in an installation image and determine the location and extent of the target device in the installation image. The location and extent of the target device are represented by the original detection bounding box. The object detection model can be R-CNN, Fast R-CNN, Faster R-CNN, YOLO, SSD, Grounded-SAM, etc.

[0110] For example, the acquired installation image is input into a pre-trained object detection model. The object detection model automatically identifies the location of the target device in the installation image by analyzing the features in the image and generates an original detection box to mark the device region.

[0111] This disclosure enables the accurate identification of the device's location in an image through an object detection model, resulting in more accurate original detection boxes. It reduces the impact of human factors on the detection results, and the object detection model can be trained to adapt to different types of devices and various installation environments, exhibiting strong versatility and flexibility.

[0112] As an optional embodiment, the target device conforms to a set installation specification, including: The distance between the target device and the relative surfaces of the reference object is greater than a set distance; The angle between the target device and the relative surfaces of the reference object is less than a set angle.

[0113] For example, the target device is the device to be installed and tested, and the reference object is a reference object related to the installation location of the target device, which can be a wall, floor, ceiling, other equipment, etc., used to determine the relative positional relationship with the target device.

[0114] For example, the set distance refers to the minimum safe distance or working distance specified according to the design requirements, installation requirements, or relevant standards of the target equipment. If the distance between the target equipment and the reference object is greater than the distance threshold, it is considered to comply with the set installation specifications. If the distance between the target equipment and the reference object is less than the distance threshold, it is considered to violate the set installation specifications.

[0115] For example, the set angle refers to the minimum tilt angle specified according to the design requirements, installation requirements, or relevant standards of the target equipment. If the angle between the target equipment and the relative surfaces of the reference object is less than the set angle, it is considered to comply with the set installation specifications. If the angle between the target equipment and the relative surfaces of the reference object is greater than the set angle, it is considered to violate the set installation specifications.

[0116] As an optional embodiment, obtaining the installation image corresponding to the target device includes: in response to obtaining multiple images, performing image recognition on the multiple images using an object classification model to determine the installation image from the multiple images; and / or, The step of detecting whether the target device conforms to the set installation specifications based on the target area includes: determining whether the target device conforms to the set installation specifications based on the target area using an installation classification model.

[0117] For example, an object classification model can divide an image into two categories: an image containing the target device and an image not containing the target device.

[0118] For example, an object classification model can employ the lightweight MobileNet network and be trained using transfer learning to enable it to identify whether an image contains a target device. Transfer learning allows a model developed for task A to be reused in the development of a model for task B.

[0119] For example, the installation classification model can categorize the detection results of whether the target device conforms to the set installation specifications into two categories: that is, the target device conforms to the set installation specifications, or the target device does not conform to the set installation specifications.

[0120] For example, the installation classification model can use a ResNet network, and the ResNet network can be trained through transfer learning to enable it to identify whether the target device meets the set installation specifications.

[0121] For example, both the object classification model and the installation classification model can use a binary classification model, which is a supervised learning method in machine learning used to divide data into two mutually exclusive categories.

[0122] The training samples for the object classification model can include images containing the target device and images not containing the target device. Images containing the target device are considered positive samples, and images not containing the target device are considered negative samples.

[0123] For example, the training samples for the installation classification model can be installation images of the target device that have been manually labeled.

[0124] As an example, installation images of the target device that conform to the set installation specifications are considered positive samples, while those that do not are considered negative samples. If the set installation specification stipulates that the distance between the target device and the reference object is greater than a set distance, then installation images with a greater distance are considered positive samples, and those with a closer distance are considered negative samples. If the set installation specification stipulates that the angle between the opposing surfaces of the target device and the reference object is less than a set angle, then installation images with a smaller angle are considered positive samples, and those with a larger angle are considered negative samples.

[0125] As an example, such as Figure 5 The diagram illustrates a schematic of a MobileNet network structure. It shows how to remove the top layer of a pre-trained MobileNet network, create a custom three-layer fully connected network, and set the final activation function to a binary classification cross-entropy loss function, resulting in the following structure: Figure 5 A schematic diagram of the Euclidean network structure. The binary classification cross-entropy loss function is shown in equation (1):

[0126] in, It is used to characterize whether the installation image includes the actual label of the target device, and is either 0 or 1; is the predicted probability, which is between 0 and 1; N is the number of images.

[0127] As an example, such as Figure 6 The diagram illustrates a ResNet network structure. Layers 18, 34, 50, 101, and 152 represent the configurations of ResNet networks at different depths. The feature maps are pooled to a 1×1 size, then connected to a 1000-dimensional fully connected layer, and finally classified using an activation function. The number of residual blocks in conv2_x, conv3_x, conv4_x, and conv5_x varies depending on the ResNet network depth.

[0128] For example, a pre-trained ResNet152 model can be used, with the top layer removed and three fully connected layers stacked on top. The final activation function also uses the above formula (1), resulting in the following: Figure 6 The network structure is shown below. Among them, It is a real label used to characterize whether the target device conforms to the set installation specifications, and is either 0 or 1; is the predicted probability, which is between 0 and 1; N is the number of images.

[0129] As an optional embodiment, the target device is the indoor unit of an air conditioning unit, and the reference object is the ceiling or wall of the room; or, the target device is the outdoor unit of an air conditioning unit, and the reference object is the obstruction or wall facing the air outlet of the outdoor unit.

[0130] For example, two scenarios are presented here for testing whether a target device conforms to the established installation specifications. For air conditioning equipment installation testing, there are two scenarios: indoor unit installation testing and outdoor unit installation testing.

[0131] For indoor unit installation testing, the target equipment is the indoor unit of the air conditioner, and the reference object is the ceiling or wall of the room. For outdoor unit installation testing, the target equipment is the outdoor unit of the air conditioner, and the reference object is the obstruction or wall facing the outdoor unit's air outlet. The above-described equipment installation testing methods can be used to test the indoor and / or outdoor unit installations of the air conditioner.

[0132] As an optional example, see Figure 7-8 Taking an indoor air conditioner unit as an example, a schematic diagram of a testing method for equipment installation is shown.

[0133] For example, a MobileNet image classification model for air conditioner indoor units can be pre-trained. The trained model can identify photos of air conditioner indoor units from images of outdoor units, vacuum gauges, indoor units, and air conditioner specifications and models input by installation engineers, thereby reducing the complexity of the recognition model for indoor unit installation inspection.

[0134] For example, photos of air conditioner indoor units that are installed correctly or incorrectly can be collected beforehand based on expert labeling, and an image classification model can be trained on the model. Based on the cropped photos of the correct and incorrect air conditioner installations, a ResNet image classification model is trained to detect whether the air conditioner installation is correct or not, and output image labels and confidence scores.

[0135] For example, the existing target detection model SAM model can be used to identify the position coordinates of the indoor unit of the air conditioner in the image, and the output is the coordinate information of the upper left and lower right corners of the detection box in the image coordinate system.

[0136] like Figure 7 As shown, after the installation engineer uploads the installation image of the air conditioner indoor unit, the MobileNet image classification model can be used to detect whether the installation image contains the air conditioner indoor unit.

[0137] like Figure 8 As shown, the SAM (Search Engine Detection Model) is used to identify the original bounding boxes corresponding to the indoor unit of the air conditioner. Then, based on the identified original bounding boxes, the installation image corresponding to the indoor unit is cropped and processed into grayscale. The grayscale-processed target area is then detected using a ResNet model, and the model outputs a confidence score. If the confidence score is less than 0.5, the installation is considered abnormal; otherwise, it is considered normal. Photos with low confidence scores undergo manual review. These manually reviewed photos can be re-labeled and used for training the ResNet model, continuously improving the model's accuracy.

[0138] The process of cropping the installation image after identifying the indoor unit of the air conditioner involves the following steps: First, based on the Grounded-SAM model, the corner coordinates of the original detection frame corresponding to the indoor unit of the air conditioner are identified. tuples, where Represents the coordinates of the top left corner point. The coordinates represent the lower right corner of the air conditioner.

[0139] Taking the top left corner of the installation image as the origin of the coordinate system as an example, for For an outward expansion of a certain distance, assuming If it is the expanded new coordinate system, then the specific expansion method is as follows:

[0140]

[0141]

[0142]

[0143]

[0144]

[0145] in, Indicates the width of the image. Indicates the height of the image, 1 / This indicates setting a scaling factor, for example, setting it to 0.1.

[0146] Finally, after cropping the image based on the above formula, the image is converted to grayscale to remove the color information of the original image and avoid interference from RGB to the model training.

[0147] The MobileNet model is trained using all images of both air conditioners and non-air conditioners. Images of indoor air conditioners are used as positive samples, while images of outdoor units, pressure gauges, and other non-indoor unit images are excluded.

[0148] The ResNet model can be trained using manually labeled images of air conditioner indoor units that are installed correctly or incorrectly. Images of indoor units that are too close to the ceiling are considered positive samples, while images of indoor units that are too far from the ceiling are considered negative samples.

[0149] To address the complexity and diversity of image content, a phased model recognition scheme was adopted. Two different image classification models were trained based on convolutional neural networks to perform image recognition of air conditioner indoor units and indoor unit installation detection, respectively.

[0150] This disclosure identifies the target device based on the target detection model and performs image cropping to avoid interference from irrelevant factors such as curtains on the model's judgment. At the same time, grayscale image processing is performed to avoid interference from factors such as image color and lighting, allowing the model to fully learn the distance information between the air conditioner indoor unit and the room ceiling.

[0151] Reference Figure 9 , Figure 9 This is a block diagram illustrating a device-installed detection apparatus 900 according to an exemplary embodiment. Figure 9 As shown, the detection device 900 installed in the equipment includes an acquisition module 901, a determination module 902, and a detection module 903.

[0152] The acquisition module 901 is configured to acquire an installation image corresponding to the target device, wherein the installation image includes a device area and a background area corresponding to the target device. The determination module 902 is configured to determine a target region in the installation image, the target region including the device region and a set region surrounding the device region, the set region being smaller than the background region; The detection module 903 is configured to detect whether the target device conforms to the set installation specifications based on the target area.

[0153] As an optional embodiment, the determining module 902 is configured to: Image recognition is performed on the installation image to determine the original detection box corresponding to the device area. The original detection box is used to determine the position of the device area in the installation image. The original detection frame is expanded. The target region is determined based on the expanded detection bounding box.

[0154] As an optional embodiment, the determining module 902 is configured to: Determine the first size information corresponding to the original detection frame; Based on the first size information, determine the expansion distance corresponding to the boundary to be expanded of the original detection frame; The original detection box is expanded according to the expansion distance corresponding to the boundary to be expanded; The determining module 902 is further configured to: Based on the first size information, the extension distance corresponding to the set boundary of the original detection frame is determined, and the set boundary is determined according to the relative positional relationship between the reference object and the target device.

[0155] As an optional embodiment, the determining module 902 is further configured to: Based on the first size information and the set scaling factor, the expansion distances corresponding to the other boundaries of the original detection frame are determined respectively. The other boundaries are at least one boundary among all the boundaries of the original detection frame other than the set boundary.

[0156] As an optional embodiment, the determining module 902 is further configured to: Based on the second size information of the installation image and the original detection box, determine the maximum expansion distance corresponding to the boundary to be expanded; Based on the numerical relationship between the expansion distance corresponding to the boundary to be expanded and the maximum expansion distance, the target expansion distance corresponding to the boundary to be expanded is determined; The original detection box is expanded based on the target expansion distance corresponding to the boundary to be expanded.

[0157] As an optional embodiment, the determining module 902 is configured to: Based on the expanded detection box, the installation image is cropped to determine the target region.

[0158] As an optional embodiment, the determining module 902 is configured to: The installation image is subjected to image recognition using a target detection model to determine the original detection box corresponding to the device area.

[0159] As an optional embodiment, the acquisition module 901 is configured to: in response to acquiring multiple images, perform image recognition on the multiple images using an object classification model to determine the installation image from the multiple images; and / or, the detection module 903 is configured to: determine whether the target device conforms to the set installation specifications based on the target area using an installation classification model.

[0160] As an optional embodiment, the target device conforms to a set installation specification, including: The distance between the target device and the relative surfaces of the reference object is greater than a set distance; The angle between the target device and the relative surfaces of the reference object is less than a set angle.

[0161] As an optional embodiment, the target device is the indoor unit of an air conditioning unit, and the reference object is the ceiling or wall of the room; or, the target device is the outdoor unit of an air conditioning unit, and the reference object is the obstruction or wall facing the air outlet of the outdoor unit.

[0162] Regarding the detection device 900 installed in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments concerning the detection method for the installation of the device, and will not be elaborated upon here.

[0163] This disclosure also provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the steps of the device installation detection method provided in this disclosure.

[0164] This disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the device installation detection method described in the first aspect of the embodiments of this disclosure.

[0165] This disclosure also provides an electronic device, including: Storage device for storing computer programs; An execution device is used to execute the computer program to implement the device installation detection method described in this disclosure.

[0166] Reference Figure 10 , Figure 10 This is a block diagram illustrating an electronic device 1000 according to an exemplary embodiment. For example, the electronic device 1000 may be a computer, a tablet device, a server, etc.

[0167] like Figure 10 As shown, the electronic device 1000 may include one or more of the following components: processing component 1002, memory 1004, power supply component 1006, multimedia component 1008, audio component 1010, input / output interface 1012, sensor component 1014, and communication component 1016.

[0168] Processing component 1002 typically controls the overall operation of electronic device 1000, such as operations associated with display, telephone calls, data communication, camera operation, and recording operations. Processing component 1002 may include one or more processors 1020 to execute instructions to perform all or part of the steps of the methods described above. Furthermore, processing component 1002 may include one or more modules to facilitate interaction between processing component 1002 and other components. For example, processing component 1002 may include a multimedia module to facilitate interaction between multimedia component 1008 and processing component 1002.

[0169] Memory 1004 is configured to store various types of data to support the operation of electronic device 1000. Examples of such data include instructions for any application or method operating on electronic device 1000, contact data, phonebook data, messages, images, videos, etc. Memory 1004 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0170] Power supply component 1006 provides power to various components of electronic device 1000. Power supply component 1006 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 1000.

[0171] Multimedia component 1008 includes a screen that provides an output interface between the electronic device 1000 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 1008 includes a front-facing camera and / or a rear-facing camera. When the electronic device 1000 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0172] Audio component 1010 is configured to output and / or input audio signals. For example, audio component 1010 includes a microphone (MIC) configured to receive external audio signals when electronic device 1000 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 1004 or transmitted via communication component 1016. In some embodiments, audio component 1010 also includes a speaker for outputting audio signals.

[0173] Input / output interface 1012 provides an interface between processing component 1002 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, start buttons, and lock buttons.

[0174] Sensor assembly 1014 includes one or more sensors for providing state assessments of various aspects of electronic device 1000. For example, sensor assembly 1014 may detect the on / off state of electronic device 1000, the relative positioning of components such as the display and keypad of electronic device 1000, changes in position of electronic device 1000 or a component of electronic device 1000, the presence or absence of user contact with electronic device 1000, the orientation or acceleration / deceleration of electronic device 1000, and temperature changes of electronic device 1000. Sensor assembly 1014 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 1014 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 1014 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.

[0175] Communication component 1016 is configured to facilitate wired or wireless communication between electronic device 1000 and other devices. Electronic device 1000 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, communication component 1016 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 1016 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0176] In an exemplary embodiment, the electronic device 1000 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the detection method for the installation of the aforementioned device.

[0177] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 1004 including instructions, which can be executed by a processor 1020 of an electronic device 1000 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0178] In another exemplary embodiment, a computer program product is also provided, the computer program product comprising a computer program executable by a programmable device, the computer program having a code portion for performing the device installation detection method described above when executed by the programmable device.

[0179] Figure 11 This is a block diagram illustrating an apparatus 1100 for device installation detection according to an exemplary embodiment. For example, apparatus 1100 may be provided as a server. (Refer to...) Figure 11 The device 1100 includes a processing component 1122, which further includes one or more processors, and memory resources represented by memory 1132 for storing instructions, such as application programs, that can be executed by the processing component 1122. The application programs stored in memory 1132 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 1122 is configured to execute instructions to perform the aforementioned detection method installed on the device.

[0180] Device 1100 may also include a power supply component 1126 configured to perform power management of device 1100, a wired or wireless network interface 1150 configured to connect device 1100 to a network, and an input / output interface 1158. Device 1100 can operate on an operating system, such as Windows Server, stored in memory 1132. TM Mac OS X TM Unix TM Linux TM FreeBSD TM Or similar.

[0181] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of this disclosure. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0182] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for detecting equipment installation, characterized in that, include: Obtain an installation image corresponding to the target device, wherein the installation image includes a device area and a background area corresponding to the target device; Determine a target region in the installation image, the target region including the device region and a defined region surrounding the device region, the defined region being smaller than the background region; Based on the target area, detect whether the target device conforms to the set installation specifications.

2. The method according to claim 1, characterized in that, Determining the target region in the installation image includes: Image recognition is performed on the installation image to determine the original detection box corresponding to the device area. The original detection box is used to determine the position of the device area in the installation image. The original detection frame is expanded. The target region is determined based on the expanded detection bounding box.

3. The method according to claim 2, characterized in that, The process of expanding the original detection box includes: Determine the first size information corresponding to the original detection frame; Based on the first size information, determine the expansion distance corresponding to the boundary to be expanded of the original detection frame; The original detection box is expanded according to the expansion distance corresponding to the boundary to be expanded; The step of determining the expansion distance corresponding to the expansion boundary of the original detection frame based on the first size information includes: Based on the first size information, the extension distance corresponding to the set boundary of the original detection frame is determined, and the set boundary is determined according to the relative positional relationship between the reference object and the target device.

4. The method according to claim 3, characterized in that, The step of determining the expansion distance corresponding to the expansion boundary of the original detection frame based on the first size information further includes: Based on the first size information and the set scaling factor, the expansion distances corresponding to the other boundaries of the original detection frame are determined respectively. The other boundaries are at least one boundary among all the boundaries of the original detection frame other than the set boundary.

5. The method according to claim 3, characterized in that, Expanding the original detection box according to the expansion distance corresponding to the boundary to be expanded, further includes: Based on the second size information of the installation image and the original detection box, determine the maximum expansion distance corresponding to the boundary to be expanded; Based on the numerical relationship between the expansion distance corresponding to the boundary to be expanded and the maximum expansion distance, the target expansion distance corresponding to the boundary to be expanded is determined; The original detection box is expanded based on the target expansion distance corresponding to the boundary to be expanded.

6. The method according to claim 2, characterized in that, The determination of the target region based on the expanded detection bounding box includes: Based on the expanded detection box, the installation image is cropped to determine the target region.

7. The method according to claim 2, characterized in that, The step of performing image recognition on the installation image to determine the original detection box corresponding to the device area includes: The installation image is subjected to image recognition using a target detection model to determine the original detection box corresponding to the device area.

8. The method according to any one of claims 1-7, characterized in that, The step of obtaining the installation image corresponding to the target device includes: In response to acquiring multiple images, image recognition is performed on the multiple images using an object classification model to determine the installation image from the multiple images; And / or, The step of detecting whether the target device conforms to the set installation specifications based on the target area includes: By installing a classification model based on the target area, it is determined whether the target device conforms to the set installation specifications.

9. The method according to any one of claims 1-7, characterized in that, The target device conforms to the set installation specifications, including: The distance between the target device and the relative surfaces of the reference object is greater than a set distance; The angle between the target device and the relative surfaces of the reference object is less than a set angle.

10. The method according to claim 9, characterized in that, The target device is the indoor unit of an air conditioning unit, and the reference object is the ceiling or wall of the room; or... The target device is the outdoor unit of an air conditioning unit, and the reference object is the obstruction or wall facing the air outlet of the outdoor unit.

11. A detection device for equipment installation, characterized in that, The detection method for implementing the equipment installation according to any one of claims 1-11.

12. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the program instructions are executed by the processor, they implement the detection method for device installation as described in any one of claims 1-10.

13. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the detection method for device installation as described in any one of claims 1-10.

14. An electronic device, characterized in that, include: Storage device for storing computer programs; An execution device is used to execute the computer program to implement the detection method for device installation according to any one of claims 1-10.

Images