Method and device for reducing the undetected rate of elevator safety hazard detection

Abstract

The application discloses a method and device for reducing the missed detection rate of elevator safety hidden danger detection. The method comprises the following steps: in response to an elevator internal shielding detection event being triggered, acquiring at least two first change boundaries between first video frames collected by a monitoring device arranged in the elevator; determining the BC shape index of each of the at least two first change boundaries, respectively, and determining the boundary change characteristics between each of two adjacent first change boundaries in the at least two first change boundaries; if the BC shape index meets a preset spatial domain condition and / or the boundary change characteristics meet a preset time domain condition, it is determined that there is an object being shielded in the elevator. The scheme can accurately detect whether there is an object being shielded in the elevator.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to a method and apparatus for reducing the missed detection rate of elevator safety hazards. Background Technology

[0002] In elevator monitoring scenarios, electric scooters frequently enter elevators. However, these scooters pose several risks. First, they may collide with elevator doors and safety barriers, causing elevator malfunctions. Second, if the scooters are parked in hallways or charged at home after using the elevator, they could pose a fire risk. Furthermore, scooters parked in hallways obstruct safe evacuation routes, creating significant safety hazards for residents. Therefore, electric scooters are generally prohibited from entering elevators. However, to allow their scooters into elevators, users often cover them (e.g., with raincoats, barriers, or awnings).

[0003] In existing technologies, the presence of obstructions such as electric bicycles can be detected by monitoring changes in the brightness of images captured by surveillance cameras installed inside the elevator. However, this method has low accuracy and a high rate of false alarms and missed detections. Time-of-Flight (TOF) sensors can also be used to detect obstructions in elevators, but this method increases hardware costs and has limitations on the distance to the obstructing object, also leading to false alarms and missed detections. Summary of the Invention

[0004] This invention provides a method and apparatus for reducing the missed detection rate of elevator safety hazards, which can accurately detect whether there are objects that are not obstructing the view inside the elevator.

[0005] According to one aspect of the present invention, a method for reducing the missed detection rate of elevator safety hazards is provided, comprising:

[0006] In response to the triggering of an occlusion detection event inside the elevator, at least two first change boundaries between the first video frames collected by the monitoring equipment configured inside the elevator are acquired.

[0007] The BC shape index of each of the at least two first change boundaries is determined, and the boundary change characteristics between each two adjacent first change boundaries are determined.

[0008] If the BC shape index satisfies the preset spatial conditions, and / or the boundary change characteristics satisfy the preset temporal conditions, then it is determined that there is an obstruction inside the elevator.

[0009] According to another aspect of the present invention, a device for reducing the missed detection rate of elevator safety hazards is provided, comprising:

[0010] The first change boundary acquisition module is used to acquire at least two first change boundaries between the first video frames collected by the monitoring equipment configured in the elevator in response to the triggering of the occlusion detection event in the elevator.

[0011] The boundary change feature determination module is used to determine the BC shape index of each of the at least two first change boundaries, and to determine the boundary change features between each pair of adjacent first change boundaries.

[0012] An obstruction detection module is used to determine that an obstruction exists in the elevator if the BC shape index meets a preset spatial condition and / or the boundary change feature meets a preset temporal condition.

[0013] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0014] At least one processor; and

[0015] A memory communicatively connected to the at least one processor; wherein,

[0016] The memory stores a computer program that can be executed by the at least one processor, which is then executed by the at least one processor to enable the at least one processor to perform the method for reducing the missed detection rate of elevator safety hazards according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the method for reducing the missed detection rate of elevator safety hazards as described in any embodiment of the present invention.

[0018] The occlusion detection scheme of this invention, in response to an occlusion detection event triggered inside the elevator, acquires at least two first change boundaries between the first video frames collected by the monitoring equipment configured inside the elevator; determines the BC shape index of each of the at least two first change boundaries, and determines the boundary change characteristics between every two adjacent first change boundaries; if the BC shape index satisfies a preset spatial condition, and / or the boundary change characteristics satisfy a preset temporal condition, then it is determined that an occluded object exists inside the elevator. Through the technical solution provided by this invention, it is possible to accurately detect whether there is an object occluding the elevator that is not at the camera end.

[0019] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This invention provides a flowchart of a method for reducing the missed detection rate of elevator safety hazards in Embodiment 1.

[0022] Figure 2 This is a flowchart of a method for reducing the missed detection rate of elevator safety hazards provided in Embodiment 2 of the present invention;

[0023] Figure 3a This is a schematic diagram illustrating the changing boundary vector corresponding to the removal of obstruction in an embodiment of the present invention.

[0024] Figure 3b This is a schematic diagram illustrating the changing boundary vector when an obstruction is unobstructed in a reduced-size manner, as provided in an embodiment of the present invention.

[0025] Figure 4 This is a schematic diagram of a device for reducing the missed detection rate of elevator safety hazards provided in Embodiment 3 of the present invention;

[0026] Figure 5 A schematic diagram of the electronic device used to implement the method for reducing the missed detection rate of elevator safety hazards in this embodiment of the invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises 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 such processes, methods, products, or apparatus.

[0029] Example 1

[0030] Figure 1 This invention provides a flowchart of a method for reducing the missed detection rate of elevator safety hazards, as described in Embodiment 1. This embodiment is applicable to detecting whether there are objects obstructing the elevator interior. The method can be executed by a device for reducing the missed detection rate of elevator safety hazards. This device can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes:

[0031] S110. In response to the triggering of an occlusion detection event inside the elevator, acquire at least two first change boundaries between the first video frames collected by the monitoring equipment configured inside the elevator.

[0032] In this embodiment of the invention, when an elevator occlusion detection command is received, it can be determined that an elevator occlusion detection event has been triggered. In response to the triggering of the elevator occlusion detection event, at least two first change boundaries between the first video frames collected by the monitoring equipment configured inside the elevator are obtained. Specifically, the monitoring equipment configured inside the elevator collects first video frames in real time and forms a video stream based on a series of collected first video frames, determining the first change boundaries between two consecutive first video frames in the video stream. The first change boundary is the edge of a region where there is a significant difference between two consecutive first video frames. In this embodiment of the invention, the first change boundary can be determined based on two adjacent first video frames, or it can be determined based on a pre-set video frame interval for determining the change boundary. For example, when the video frame interval is 1, the first change boundary between the 1st and 3rd first video frames, the first change boundary between the 2nd and 3rd first video frames, ..., the first change boundary between the (n-2)th and nth first video frames are determined respectively. It should be noted that the embodiments of the present invention do not limit the image interval between the two first video frames for determining the first change boundary. For example, the first change boundary between two consecutive first video frames can be determined by methods such as pixel subtraction, histogram subtraction, edge detection algorithms, optical flow methods, and deep learning models such as convolutional neural networks, and this embodiment does not limit this.

[0033] Optionally, in response to an elevator occlusion detection event being triggered, the process includes: performing human detection on the first video frame to determine a human-shaped region in the first video frame; determining a dynamically changing region in the first video frame after removing the human-shaped region; and determining that the elevator occlusion detection event has been triggered if the area of ​​the dynamically changing region is within a preset area range. For example, human detection is performed on the first video frame using human detection technology to determine a human-shaped region in the first video frame, and the human-shaped region is removed from the first video frame to determine the dynamically changing region in the first video frame. The area of ​​the dynamically changing region is calculated, and it is determined whether the area of ​​the dynamically changing region is within a preset area range. If so, it indicates that the moving object following the person into the elevator may be an electric scooter, and in this case, an elevator occlusion detection event is triggered.

[0034] S120. Determine the BC shape index of each of the at least two first change boundaries, and determine the boundary change characteristics between each pair of adjacent first change boundaries.

[0035] In this embodiment of the invention, for each of at least two first changing boundaries, a BC shape index is determined, wherein the BC shape index is used to characterize the spatial regularity of the first changing boundary. Optionally, determining the BC shape index for each of the at least two first changing boundaries includes: for each of the at least two first changing boundaries, determining a first preset number of first sampling points on the first changing boundary; determining the centroid of the first changing boundary, and determining the radius from the centroid to each of the first sampling points; and calculating the BC shape index of the corresponding first changing boundary based on the first preset number of radius lengths.

[0036] In this embodiment of the invention, uniform sampling is performed on each first changing boundary, and n (i.e., a first preset number) first sampling points are determined on each first changing boundary. The first preset number can be a pre-set fixed value, or it can be determined according to the length of the first changing boundary and a preset ratio. This embodiment does not limit the number of first sampling points. The centroid of each first changing boundary is determined, and the radius length r between the centroid and each first sampling point on the first changing boundary is determined, so that a first preset number of radius lengths can be determined on each first changing boundary. The BC shape index of the first changing boundary is calculated based on the first preset number of radius lengths. For example, the BC shape index of the first changing boundary can be calculated according to the following formula: Among them, SI j r represents the BC shape index of the j-th first change boundary among at least two first change boundaries. i Let n represent the radius from the centroid of the j-th first change boundary to the i-th first sampling point on the j-th first change boundary, and n represent the number of first sampling points on the j-th first change boundary.

[0037] In this embodiment of the invention, boundary change features are determined between each pair of adjacent first change boundaries in at least two first change boundaries. These boundary change features characterize changes in shape, structure, etc., between adjacent first change boundaries. For example, corresponding feature points can be selected on each pair of adjacent first change boundaries. These feature points can be representative points such as points with large curvature or endpoints. This embodiment does not limit the type or selection method of the feature points. Based on the corresponding feature points on the two adjacent first change boundaries, a feature point change vector is constructed, and this feature point change vector is used as the boundary change feature between the two adjacent first change boundaries.

[0038] Optionally, determining the boundary change feature between each pair of adjacent first change boundaries in the at least two first change boundaries includes: for each pair of adjacent first change boundaries in the at least two first change boundaries, determining the sampling point vector composed of each pair of adjacent first sampling points on each of the two adjacent first change boundaries, and calculating the sampling point vector difference between each pair of adjacent sampling point vectors; calculating the rate of change of the sampling point vector difference between the two adjacent first change boundaries based on the corresponding two sampling point vector differences between the two adjacent first change boundaries; and using the rate of change of the sampling point vector difference as the boundary change feature between the two adjacent first change boundaries.

[0039] For example, any two adjacent first change boundaries among at least two first change boundaries are denoted as change boundary l0 and change boundary l1, and the n first sampling points determined on change boundary l0 are denoted as X1, X2, ..., X... n The n first sampling points determined on the change boundary l1 will be denoted as X1', X2', ..., X... n The sampling point vector is constructed based on each pair of adjacent first sampling points on the change boundary l0. And calculate the difference between the sampling point vectors of every two adjacent sampling point vectors on the change boundary l0, respectively. Similarly, a sampling point vector is constructed based on every two adjacent first sampling points on the change boundary l1. And calculate the difference between the sampling point vectors of every two adjacent sampling point vectors on the change boundary l1, respectively. Calculate the rate of change of the sampling point vector difference between changing boundary l0 and changing boundary l1 based on the vector difference between two corresponding sampling points on changing boundary l0 and changing boundary l1. (Sampling point vector difference on changing boundary l0) Vector difference with sampling points on the change boundary l1 Correspondingly, the vector difference of sampling points on the change boundary l0 Vector difference with sampling points on the change boundary l1 Corresponding to, ..., the vector difference of sampling points on the change boundary l0 Vector difference with sampling points on the change boundary l1 Correspondingly. Therefore, the rate of change of the sample point vector difference between the change boundary l0 and the change boundary l1 includes:

[0040] The rate of change of the vector difference between each sampling point between the changing boundary l0 and the changing boundary l1 is used as the boundary change feature between the changing boundary l0 and the changing boundary l1.

[0041] S130. If the BC shape index satisfies the preset spatial domain conditions, and / or the boundary change characteristics satisfy the preset temporal domain conditions, then it is determined that there is an obstruction in the elevator.

[0042] In this embodiment of the invention, it is determined whether the BC shape index of each first change boundary is within a preset index range, and the occupancy rate of the BC shape indices within the preset index range is calculated, that is, the ratio of the number of BC shape indices within the preset index range to the total number of all BC shape indices. When the occupancy rate of the BC shape indices within the preset index range is greater than a preset occupancy threshold, it can be determined that the BC shape index meets the preset spatial conditions. Optionally, the average BC shape index of all the BC shape indices of the first change boundaries is calculated, and it is determined whether the BC shape index is within the preset index range. If so, it is determined that the BC shape index meets the preset spatial conditions. Since if there is an object obstructing the elevator, users usually use flexible obstructions such as raincoats, barriers, awnings, and clothing to obstruct the object, when it is determined that the BC shape index meets the preset spatial conditions, it can be determined that there is an obstructed object in the elevator.

[0043] In this embodiment of the invention, it is determined whether the boundary change characteristics between any two adjacent first change boundaries satisfy a preset time-domain condition. If so, it can be determined that there is an obstruction within the elevator. Since the first change boundary is usually a closed edge line based on n first sampling points on each first change boundary, the vector difference rate of the n sampling points between any two adjacent first change boundaries can be determined in the above manner. For each pair of adjacent first change boundaries in at least two first change boundaries, it is determined whether the vector difference rate of the sampling points between each pair of adjacent first change boundaries is greater than a preset rate of change threshold, and the proportion of the vector change rates of the n sampling points that are greater than the preset rate of change threshold is counted. If the proportion is greater than a preset proportion threshold, the two adjacent first change boundaries can be counted as a valid change boundary group. The number of valid change boundary groups is determined from all pairs of adjacent first change boundaries determined from at least two first change boundaries. If the number of valid change boundary groups is greater than a preset threshold, it can be determined that the boundary change characteristics satisfy the preset time-domain condition.

[0044] Optionally, for each pair of adjacent first change boundaries in at least two first change boundaries, calculate the mean of the rate of change of the sampled point vectors of n sampled points between the two adjacent first change boundaries. Determine the mean rate of change of the sampling point vector Whether it exceeds a preset vector change rate threshold. If the number of sample point vector change rate averages exceeding the preset vector change rate threshold is greater than the preset threshold, then the boundary change feature can be determined to meet the preset time domain condition. Optionally, the average of the sample point vector change rate averages between all two adjacent first change boundaries can also be calculated. If the average of the sample point vector change rate averages between all two adjacent first change boundaries is greater than the preset threshold, then the boundary change feature can be determined to meet the preset time domain condition.

[0045] Optionally, for each pair of adjacent first change boundaries among at least two first change boundaries, the standard deviation of the rate of change of the sampling point vectors between the n sampling point vectors between the two adjacent first change boundaries is calculated, and it is determined whether the standard deviation of the rate of change of the sampling point vectors is greater than a preset threshold for the standard deviation of the vectors. If the number of sampling point vectors with standard deviations greater than the preset threshold is greater than the preset threshold, then it can be determined that the boundary change feature satisfies the preset time-domain condition. Optionally, the average of the standard deviations of the rate of change of the sampling point vectors between all adjacent first change boundaries can also be calculated. If the average of the standard deviations of the rate of change of the sampling point vectors between all adjacent first change boundaries is greater than the preset threshold, then it can be determined that the boundary change feature satisfies the preset time-domain condition.

[0046] Optionally, when the BC shape index of each first change boundary satisfies a preset spatial condition and the boundary change characteristics between any two adjacent first change boundaries satisfy a preset temporal condition, it can be determined that an obstruction exists within the elevator. The advantage of this setting is that it can further accurately determine whether a flexible obstruction exists within the elevator.

[0047] The method for reducing the false negative rate of elevator safety hazard detection according to embodiments of the present invention, in response to the triggering of an obstruction detection event inside the elevator, acquires at least two first change boundaries between the first video frames collected by the monitoring equipment configured inside the elevator; determines the BC shape index of each of the at least two first change boundaries, and determines the boundary change characteristics between each pair of adjacent first change boundaries; if the BC shape index satisfies a preset spatial condition, and / or the boundary change characteristics satisfy a preset temporal condition, then it is determined that there is an obstruction inside the elevator, and the elevator weighing module detects whether the obstruction is an electric bicycle. Through the technical solution provided by the embodiments of the present invention, it is possible to accurately detect whether there is an object obstructing the elevator that is not at the camera end.

[0048] In some embodiments, if an obstruction is determined to exist within the elevator, the elevator weighing module further detects whether the obstruction is an electric scooter. When a person and the obstruction enter the elevator, the elevator weighing module can detect the elevator weight increment (i.e., the sum of the person's and the obstruction's weights) before and after their entry. The number of people in the first video frame is determined, and based on preset human body weights and the number of people, the target human body weight of the person who entered the elevator simultaneously with the obstruction is calculated. The weight of the obstruction within the elevator can be estimated using the elevator weight increment and the target human body weight, where the weight of the obstruction is the difference between the elevator weight increment and the target human body weight. It is determined whether the weight of the obstruction is within a preset electric scooter weight range; if so, it can be determined that the obstruction is highly likely to be an electric scooter.

[0049] In some embodiments, after determining that an obstruction exists within the elevator, the method further includes: obtaining the elevator increment through an elevator weighing module; determining the number of human-shaped pixels in the first video frame, and calculating the weight of the human body inside the elevator based on the number of human-shaped pixels and a pre-set human-shaped pixel weight; calculating the weight of the obstruction based on the elevator increment and the human body weight, and determining whether the obstruction is an electric bicycle based on the weight of the obstruction. For example, human detection is performed on the first video frame to determine the human-shaped region contained in the first video frame, and the number of pixels contained in the human-shaped region is determined, which is then used as the number of human-shaped pixels. A pre-set human-shaped pixel weight is obtained, where the human-shaped pixel weight is used to characterize the weight of the human body corresponding to one pixel. The weight of the human body inside the elevator is calculated based on the number of human-shaped pixels and the human-shaped pixel weight, where the weight of the human body inside the elevator is the product of the number of human-shaped pixels and the human-shaped pixel weight. Understandably, because people entering the elevator vary in size, the number of human-shaped pixels they occupy in the first video frame differs. Therefore, by using the number of human-shaped pixels and a pre-set proportion of human-shaped pixels, the weight of the person inside the elevator can be accurately calculated. The difference between the elevator's weight increment and the human weight is calculated and used as the weight of the obstructed object. Then, based on the weight of the obstructed object, it is determined whether the obstructed object is an electric scooter.

[0050] Optionally, determining whether the obscured object is an electric scooter based on its weight includes: determining the number of obscured object pixels in the first video frame; calculating the pixel weight of the obscured object based on its weight and the number of pixels; and determining whether the obscured object is an electric scooter based on the pixel weight and a pre-set pixel weight for an electric scooter. This approach improves the accuracy of detecting whether an obscured object is an electric scooter. For example, the first video frame is analyzed to determine the obscured object region, where the obscured object region can be a dynamically changing region in the first video frame after removing human figures. The number of pixels contained in the obscured object region is taken as the number of obscured object pixels. The pixel weight of the obscured object is calculated based on its weight and the number of pixels, where the pixel weight is the ratio of the weight of the obscured object to the number of pixels. The system obtains a pre-set pixel weight ratio for the electric scooter and calculates the difference between the pixel weight ratio of the obscured object and the electric scooter. When the difference is within a preset range, the obscured object is identified as an electric scooter. However, when the difference is outside the preset range, it indicates that although the weight of the obscured object is similar to that of the electric scooter, the spatial volume of the obscured object does not match that of the electric scooter. Therefore, the obscured object is not an electric scooter.

[0051] Example 2

[0052] Figure 2 This is a flowchart of a method for reducing the missed detection rate of elevator safety hazards provided in Embodiment 2 of the present invention, as shown below. Figure 2 As shown, the method includes:

[0053] S210. In response to the triggering of an occlusion detection event inside the elevator, acquire at least two first change boundaries between the first video frames collected by the monitoring equipment configured inside the elevator.

[0054] S220. Determine the BC shape index of each of the at least two first change boundaries, and determine the boundary change characteristics between each pair of adjacent first change boundaries.

[0055] S230. If the BC shape index satisfies the preset spatial domain conditions, and / or the boundary change characteristics satisfy the preset temporal domain conditions, then it is determined that there is an obstruction in the elevator.

[0056] S240, prompt the personnel inside the elevator to remove the obstruction of the obstructed object, and after a preset time period, obtain at least two second change boundaries between the second video frames collected by the monitoring equipment.

[0057] In this embodiment of the invention, when it is determined that an obstruction exists within the elevator, in order to further verify whether the obstruction actually exists, an audible and visual alarm is used to prompt the people inside the elevator to remove the obstruction, that is, to prompt the people inside the elevator to remove the obstruction, thereby determining whether the obstruction is an electric bicycle through electric bicycle identification. After a preset time period (e.g., 10 seconds) after prompting the people inside the elevator to remove the obstruction, at least two second change boundaries between the second video frames collected by the monitoring equipment are acquired. The method for determining the second change boundaries is similar to the method for determining the first change boundaries described in the above embodiment, and will not be repeated here.

[0058] S250. Determine a second preset number of second sampling points on each of the second change boundaries.

[0059] In this embodiment of the invention, uniform sampling is performed on each second change boundary, and m (i.e., the second preset number) second sampling points are determined on each second change boundary. The second preset number can be a pre-set fixed value, or it can be determined according to the length of the second change boundary and a preset ratio. In this embodiment, the number of second sampling points is not limited.

[0060] S260. For each pair of adjacent second change boundaries among the at least two change boundaries, determine the boundary change vector composed of two second sampling points corresponding to the two adjacent second change boundaries.

[0061] For example, any two adjacent second change boundaries among at least two second change boundaries are denoted as change boundary B0 and change boundary B1, and the m second sampling points determined on change boundary B0 are denoted as Y1, Y2, ..., Y... m The m second sampling points determined on the change boundary B1 will be denoted as Y1', Y2', ..., Y... m The second sampling point Y1 on the change boundary B0 corresponds to the second sampling point Y1' on the change boundary B1, the second sampling point Y2 on the change boundary B0 corresponds to the second sampling point Y2' on the change boundary B1, ..., the second sampling point Y1' on the change boundary B0 corresponds to the second sampling point Y1' on the change boundary B1. m The second sampling point Y on the change boundary B1 m 'Corresponding. Therefore, the boundary change vector consisting of the two second sampling points corresponding to the change boundary B0 and the change boundary B1 includes:'

[0062] S270. If the boundary change vector satisfies the preset vector change characteristic condition, then it is determined that the occlusion of the occluded object has been removed.

[0063] The preset vector change feature conditions can refer to the vector change features in the standard state corresponding to different methods of removing occlusion. The change boundary vector satisfying the preset vector change feature conditions means that the change boundary vector conforms to pre-set vector change features. Specifically, the similarity between the change boundary vector and the pre-set vector change features can be calculated. If the similarity is greater than a similarity threshold, then the change boundary vector is determined to conform to the pre-set vector change features, and the change boundary vector satisfies the preset vector change feature conditions.

[0064] Understandably, when an object is unobstructed, it typically involves one or a combination of the following two scenarios: the obstructing object is moved away from the top surface of the obstructed object, or the obstructing object is folded down from its original size. Therefore, the preset vector change feature conditions can include either an obstructing object removal type or an obstructing object shrinking type. For example, taking an umbrella as the obstructing object, if it is determined that the umbrella has moved away from the top surface of the obstructed object, or that the umbrella used to obstruct the obstructed object has changed from open to closed, then it is determined that the obstruction of the obstructed object has been removed. Figure 3a This is a schematic diagram illustrating the changing boundary vector corresponding to the removal of obstruction in an embodiment of the present invention. Figure 3b This is a schematic diagram illustrating the changing boundary vector when an obstruction is reduced in size and then removed, as provided in an embodiment of the present invention.

[0065] Optionally, the boundary change vector satisfies a preset vector change characteristic condition, including: calculating the standard deviation of the boundary change vector based on a determined second preset number of boundary change vectors; if the standard deviation of the boundary change vector is less than a preset standard deviation threshold, then determining that the boundary change vector satisfies the preset vector change characteristic condition; or, determining the average value of the sampling points on each of the second change boundaries based on a second preset number of second sampling points on each of the second change boundaries, and determining the centripetal vector composed of each second sampling point on each of the second change boundaries and the corresponding average value of the sampling points; for each of the at least two change boundaries, determining the average value vector of the sampling points based on the average value of the two sampling points on the two adjacent second change boundaries; determining the boundary vector sensitivity between the two adjacent second change boundaries based on the boundary change vector, the average value vector of the sampling points, and the centripetal vector; if the boundary vector sensitivity is less than a preset sensitivity threshold, then determining that the boundary change vector satisfies the preset vector change characteristic condition.

[0066] In this embodiment of the invention, a set of changing boundary vectors Let them be a1, a2, ..., a m Then the vector average of each boundary vector can be obtained by... The standard deviation of each boundary vector can be calculated using the following formula: For each pair of adjacent second change boundaries in at least two second change boundaries, determine whether the standard deviation of the boundary change vector between each pair of adjacent second change boundaries is less than a preset standard deviation threshold. If all the standard deviations of the boundary change vectors are less than the preset standard deviation threshold, or if the proportion of the standard deviations of the boundary change vectors less than the preset standard deviation threshold is greater than the preset threshold, then it can be determined that the standard deviation of the boundary change vectors satisfies the preset vector change characteristic condition, that is, it can be determined that the occlusion is... Figure 3a The removal method shown removes the occlusion. Optionally, a preset ratio A is obtained, such as A∈[1,3]. In each change boundary vector, change boundary vectors that deviate from the average value of the vector or the average value of the sampling points and are greater than or equal to Aδ are removed, and the feature points or sampling points corresponding to the change boundary vector are also removed.

[0067] In this embodiment of the invention, the average value of the second sampling points on each second change boundary is determined, such as the average value of each second sampling point Y1, Y2, ..., Y on the change boundary B0. m The average value of the sampling points is denoted as . The second sampling points Y1', Y2', ..., Y on the change boundary B1 are... m The average value of the sample points is denoted as '. Determine the centripetal vector formed by each second sampling point and its corresponding average value on each second change boundary. For example, the centripetal vectors on change boundary B0 can be denoted as... The centripetal vectors on the changing boundary B1 can be denoted as: Where i = 1, 2, ..., m. For each pair of adjacent second change boundaries in at least two change boundaries, a sample point average vector is determined based on the average of two sample points on the two adjacent second change boundaries. For example, based on the average of sample points on change boundary B0... Compared with the average value of the sampling points of the boundary of change B1, Determine the average vector of sampling points Let this be denoted as b. Based on the boundary change vector, the average value vector of the sampling points, and the centripetal vector, the boundary vector sensitivity between two adjacent second change boundaries is determined. For example, the determined boundary vector sensitivity between change boundary B0 and change boundary B1 can be expressed as: Let the centripetal vector be denoted as c, then the boundary vector sensitivity can be calculated using the following formula: If the boundary vector sensitivity is less than the preset sensitivity threshold, then the boundary change vector is determined to satisfy the preset vector change characteristic condition, which means that the occlusion is determined to be... Figure 3b The shrinking method shown removes the obstruction.

[0068] In this embodiment of the invention, if it is determined that the obstruction of the object has been removed, the image captured by the monitoring device can be obtained, and the image can be detected by electric vehicle detection technology to determine whether the obstructed object is an electric vehicle. If so, an alarm can be issued.

[0069] The technical solution provided by the embodiments of the present invention can not only accurately detect whether there are objects obstructing the elevator from the camera end, but also prompt the people entering the elevator to remove the obstruction when it is determined that there are objects obstructing the elevator. It can also accurately determine whether the obstructed object has been truly removed, thereby helping to further confirm whether the obstructed object is an electric bicycle through image recognition technology.

[0070] Example 3

[0071] Figure 4 This is a schematic diagram of a device for reducing the missed detection rate of elevator safety hazards provided in Embodiment 3 of the present invention. Figure 4 As shown, the device includes:

[0072] The first change boundary acquisition module 410 is used to acquire at least two first change boundaries between the first video frames collected by the monitoring equipment configured in the elevator in response to the triggering of the occlusion detection event in the elevator.

[0073] The boundary change feature determination module 420 is used to determine the BC shape index of each of the at least two first change boundaries, and to determine the boundary change features between each pair of adjacent first change boundaries.

[0074] The obstruction detection module 430 is used to determine that there is an obstruction in the elevator if the BC shape index meets a preset spatial condition and / or the boundary change feature meets a preset temporal condition.

[0075] Optionally, the boundary change feature determination module is used for:

[0076] For each of the at least two first change boundaries, a first preset number of first sampling points are determined on the first change boundary;

[0077] Determine the centroid of the first change boundary, and determine the radius from the centroid to each of the first sampling points;

[0078] The BC shape index of the first changing boundary is calculated based on the first preset number of the radius lengths.

[0079] Optionally, the boundary change feature determination module is used for:

[0080] For each of the at least two first change boundaries, determine the sampling point vector composed of each pair of adjacent first sampling points on each of the two adjacent first change boundaries, and calculate the sampling point vector difference between each pair of adjacent sampling point vectors;

[0081] The rate of change of the sampling point vector difference between the two adjacent first change boundaries is calculated based on the vector difference between the two corresponding sampling points between the two adjacent first change boundaries.

[0082] The rate of change of the vector difference between the sampling points is used as the boundary change feature between the two adjacent first change boundaries.

[0083] Optional, also includes:

[0084] The elevator increment acquisition module is used to acquire the elevator increment through the elevator weighing module after determining that there is an obstruction inside the elevator.

[0085] The human body weight calculation module is used to determine the number of human-shaped pixels in the first video frame, and to calculate the weight of the human body in the elevator based on the number of human-shaped pixels and a preset proportion of human-shaped pixels.

[0086] The electric vehicle obstruction detection module is used to calculate the weight of the obstructed object based on the elevator increment and the human body weight, and to determine whether the obstructed object is an electric vehicle based on the weight of the obstructed object.

[0087] Optional, the electric vehicle obstruction detection module is used for:

[0088] Determine the number of pixels of the occluded object in the first video frame;

[0089] Calculate the pixel weight of the obscured object based on the weight of the obscured object and the number of pixels of the obscured object;

[0090] Based on the pixel weight of the obscured object and the preset pixel weight of the electric vehicle, it is determined whether the obscured object is an electric vehicle.

[0091] Optionally, in response to an occlusion detection event being triggered inside the elevator, the following may be included:

[0092] Perform human detection on the first video frame to determine the human-shaped regions in the first video frame;

[0093] Determine the dynamically changing region in the first video frame after removing the human-shaped region;

[0094] If the area of ​​the dynamically changing region is within a preset area range, then the elevator occlusion detection event is determined to be triggered.

[0095] Optional, also includes:

[0096] The second change boundary acquisition module is used to prompt the people in the elevator to remove the obstruction of the obstruction after determining that there is an obstruction in the elevator, and to acquire at least two second change boundaries between the second video frames collected by the monitoring device after a preset time period.

[0097] The second sampling point determination module is used to determine a second preset number of second sampling points on each of the second change boundaries;

[0098] The boundary change vector determination module is used to determine, for each of the at least two adjacent second change boundaries, a boundary change vector composed of two second sampling points corresponding to the two adjacent second change boundaries;

[0099] The occlusion removal determination module is used to determine that the occluded object has been unoccluded if the boundary change vector satisfies a preset vector change characteristic condition.

[0100] Optionally, the occlusion release determination module is used for:

[0101] Calculate the standard deviation of the boundary change vectors based on the determined second preset number of boundary change vectors;

[0102] If the standard deviation of the boundary change vector is less than a preset standard deviation threshold, then the boundary change vector is determined to satisfy a preset vector change characteristic condition; or,

[0103] Based on the second preset number of second sampling points on each second change boundary, determine the average value of the sampling points on the corresponding second change boundary, and determine the centripetal vector composed of each second sampling point on each second change boundary and the corresponding average value of the sampling points.

[0104] For each pair of adjacent second change boundaries among the at least two change boundaries, a sampling point average value vector is determined based on the average value of two sampling points on the two adjacent second change boundaries;

[0105] The boundary vector sensitivity between two adjacent second change boundaries is determined based on the boundary change vector, the average value vector of the sampling points, and the centripetal vector.

[0106] If the boundary vector sensitivity is less than a preset sensitivity threshold, then the boundary change vector is determined to satisfy a preset vector change characteristic condition.

[0107] The device for reducing the missed detection rate of elevator safety hazards provided in the embodiments of the present invention can execute the method for reducing the missed detection rate of elevator safety hazards provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method.

[0108] Example 4

[0109] Figure 5 A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0110] like Figure 5 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0111] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0112] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, central processing unit (CPU), graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as methods for reducing the false negative rate of elevator safety hazard detection.

[0113] In some embodiments, the method for reducing the false negative rate of elevator safety hazard detection can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for reducing the false negative rate of elevator safety hazard detection described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the method for reducing the false negative rate of elevator safety hazard detection by any other suitable means (e.g., by means of firmware).

[0114] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0115] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0116] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0117] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0118] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0119] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0120] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0121] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for reducing the missed detection rate of elevator safety hazards, characterized in that, include: In response to the triggering of an occlusion detection event inside the elevator, at least two first change boundaries between the first video frames collected by the monitoring equipment configured inside the elevator are acquired. The BC shape index of each of the at least two first change boundaries is determined, and the boundary change characteristics between each two adjacent first change boundaries are determined. If the BC shape index satisfies the preset spatial conditions, and / or the boundary change characteristics satisfy the preset temporal conditions, then it is determined that there is an obstruction inside the elevator.

2. The method according to claim 1, characterized in that, Determining the BC shape index of each of the at least two first change boundaries includes: For each of the at least two first change boundaries, a first preset number of first sampling points are determined on the first change boundary; Determine the centroid of the first change boundary, and determine the radius from the centroid to each of the first sampling points; The BC shape index of the first changing boundary is calculated based on the first preset number of the radius lengths.

3. The method according to claim 2, characterized in that, Determining the boundary change characteristics between every two adjacent first change boundaries in the at least two first change boundaries includes: For each pair of adjacent first change boundaries among the at least two first change boundaries, a sampling point vector composed of each pair of adjacent first sampling points on each of the two adjacent first change boundaries is determined, and the sampling point vector difference between each pair of adjacent sampling point vectors is calculated. The rate of change of the sampling point vector difference between the two adjacent first change boundaries is calculated based on the vector difference between the two corresponding sampling points between the two adjacent first change boundaries. The rate of change of the vector difference between the sampling points is used as the boundary change feature between the two adjacent first change boundaries.

4. The method according to claim 1, characterized in that, After determining that there is an obstruction inside the elevator, the process also includes: Elevator increment is obtained through elevator weighing module; The number of human-shaped pixels in the first video frame is determined, and the weight of the human body inside the elevator is calculated based on the number of human-shaped pixels and the pre-set proportion of human-shaped pixels. The weight of the obstructed object is calculated based on the elevator increment and the human body weight, and it is determined whether the obstructed object is an electric vehicle based on the weight of the obstructed object.

5. The method according to claim 4, characterized in that, Determining whether the obstructed object is an electric bicycle based on its weight includes: Determine the number of pixels of the occluded object in the first video frame; Calculate the pixel weight of the obscured object based on the weight of the obscured object and the number of pixels of the obscured object; Based on the pixel weight of the obscured object and the preset pixel weight of the electric vehicle, it is determined whether the obscured object is an electric vehicle.

6. The method according to claim 1, characterized in that, In response to an occlusion detection event being triggered inside the elevator, including: Perform human detection on the first video frame to determine the human-shaped regions in the first video frame; Determine the dynamically changing region in the first video frame after removing the human-shaped region; If the area of ​​the dynamically changing region is within a preset area range, then the elevator occlusion detection event is determined to be triggered.

7. The method according to claim 1, characterized in that, After determining that there is an obstruction inside the elevator, the process also includes: The system prompts personnel inside the elevator to remove the obstruction from the object, and after a preset time period, it acquires at least two second change boundaries between the second video frames collected by the monitoring device. A second preset number of second sampling points are determined on each of the second change boundaries; For each pair of adjacent second change boundaries among the at least two change boundaries, a boundary change vector composed of two second sampling points corresponding to the two adjacent second change boundaries is determined; If the boundary change vector satisfies the preset vector change characteristic condition, then it is determined that the occlusion of the occluded object has been removed.

8. The method according to claim 7, characterized in that, The boundary change vector satisfies preset vector change feature conditions, including: Calculate the standard deviation of the boundary change vectors based on the determined second preset number of boundary change vectors; If the standard deviation of the boundary change vector is less than a preset standard deviation threshold, then the boundary change vector is determined to satisfy a preset vector change characteristic condition; or, Based on the second preset number of second sampling points on each second change boundary, determine the average value of the sampling points on the corresponding second change boundary, and determine the centripetal vector composed of each second sampling point on each second change boundary and the corresponding average value of the sampling points. For each pair of adjacent second change boundaries among the at least two change boundaries, a sampling point average value vector is determined based on the average value of two sampling points on the two adjacent second change boundaries; The boundary vector sensitivity between two adjacent second change boundaries is determined based on the boundary change vector, the average value vector of the sampling points, and the centripetal vector. If the boundary vector sensitivity is less than a preset sensitivity threshold, then the boundary change vector is determined to satisfy a preset vector change characteristic condition.

9. A device for reducing the missed detection rate of elevator safety hazards, characterized in that, include: The first change boundary acquisition module is used to acquire at least two first change boundaries between the first video frames collected by the monitoring equipment configured in the elevator in response to the triggering of the occlusion detection event in the elevator. The boundary change feature determination module is used to determine the BC shape index of each of the at least two first change boundaries, and to determine the boundary change features between each pair of adjacent first change boundaries. An obstruction detection module is used to determine that an obstruction exists in the elevator if the BC shape index meets a preset spatial condition and / or the boundary change feature meets a preset temporal condition.

10. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method for reducing the missed detection rate of elevator safety hazards as described in any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the method for reducing the missed detection rate of elevator safety hazards as described in any one of claims 1-8.

Images