A method, apparatus and storage medium for detecting abandoned items in elevators

Abstract

This invention discloses a method, apparatus, and storage medium for detecting leftover objects in elevators. The method includes: acquiring raw image data that meets background conditions inside the elevator car; drawing a detection area located at the bottom of the car and a light reference area located on the side wall of the car in the raw image data; calculating a first brightness value of the detection area and a second brightness value of the light reference area in the raw image data; querying one or more background image groups; selecting background image data that matches the brightness of the raw image data from the background image groups based on the first and second brightness values ​​of the raw image data, as target image data; and comparing the detection area of ​​the target image data with the detection area of ​​the raw image data to identify whether there are leftover objects in the car. This embodiment effectively improves the efficiency of detecting leftover objects in elevators.

Description

Technical Field

[0001] This invention relates to the technical field of elevators, and more particularly to a method, device, and storage medium for detecting items left behind in elevators. Background Technology

[0002] As residential buildings, shopping malls and other buildings become taller, elevators have become one of the most commonly used vertical transportation tools. During the process of riding the elevator, people sometimes leave their belongings in the car due to carelessness or other reasons. If these items are not cleaned up in time, it will affect people's elevator experience.

[0003] Currently, image data is collected inside the elevator car, and a classification network in deep learning is used to detect the types of debris left in the image data.

[0004] However, people may leave a wide variety of items in the elevator car, which requires a high degree of generalization of the classification network and a high degree of reliance on high-performance processing, increasing hardware costs and energy consumption. In addition, the lighting in the elevator car changes frequently, and when the carpet is changed, the floor is mopped, or items are covered (such as being wrapped or placed in a corner), false detection or missed detection is likely to occur, resulting in low efficiency in detecting leftover items in the elevator. Summary of the Invention

[0005] In view of this, the present invention provides a method, apparatus and storage medium for detecting abandoned objects in elevators, so as to improve the efficiency of detecting abandoned objects in elevators.

[0006] A first aspect of the present invention provides a method for detecting abandoned objects in an elevator, comprising:

[0007] Raw image data that meets the background conditions are collected inside the elevator car; the background conditions are that there are no passengers in the car and the car door is closed.

[0008] The detection area located at the bottom of the car and the light reference area located on the side wall of the car are plotted in the original image data.

[0009] Calculate a first brightness value for the detection area and a second brightness value for the light reference area from the original image data;

[0010] Query one or more background image groups; the background image data of the multiple frames in the background image group are all original image data that meet the background conditions and were collected in the elevator car in history.

[0011] Based on the first and second brightness values ​​of the original image data, background image data that matches the brightness of the original image data is selected from the background image group and used as target image data.

[0012] The detection area of ​​the target image data is compared with the detection area of ​​the original image data to identify whether there are any objects left in the car.

[0013] A second aspect of the present invention provides an apparatus for detecting abandoned objects in an elevator, comprising:

[0014] The image acquisition module is used to acquire raw image data that meets the background conditions inside the elevator car; the background conditions are that there are no passengers in the car and the car door is closed.

[0015] The region recognition module is used to draw the detection region located at the bottom of the car and the light reference region located on the side wall of the car in the original image data;

[0016] A brightness statistics module is used to calculate a first brightness value of the detection area and a second brightness value of the light reference area in the original image data.

[0017] The background image group query module is used to query one or more background image groups; the background image group contains multiple frames of background image data, which are all original image data that meet the background conditions and were historically collected inside the elevator car.

[0018] The target image data filtering module is used to filter background image data that matches the brightness of the original image data in the background image group based on the first brightness value and the second brightness value of the original image data, and use them as target image data.

[0019] The detection area comparison module is used to compare the detection area of ​​the target image data with the detection area of ​​the original image data to identify whether there are any objects left in the car.

[0020] A third aspect of the present invention provides an electronic device, the electronic device comprising:

[0021] At least one processor; and

[0022] A memory communicatively connected to the at least one processor; wherein,

[0023] The memory stores a computer program that can be executed by the at least one processor to enable the at least one processor to perform the method for detecting abandoned objects in an elevator as described in the first aspect above.

[0024] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for detecting abandoned objects in an elevator as described in the first aspect above.

[0025] A fifth aspect of the present invention provides a computer program product comprising a computer program that, when executed by a processor, implements the method for detecting abandoned objects in an elevator as described in the first aspect above.

[0026] In this embodiment, raw image data meeting background conditions is acquired inside the elevator car; the background conditions are: no passengers inside the car and the car door is closed; a detection area located at the bottom of the car and a light reference area located on the side wall of the car are drawn in the raw image data; a first brightness value of the detection area and a second brightness value of the light reference area are calculated in the raw image data; one or more background image groups are queried; the multiple frames of background image data in the background image group are all raw image data that meet the background conditions and were historically acquired inside the elevator car; based on the first and second brightness values ​​of the raw image data, background image data that matches the brightness of the raw image data is selected from the background image group as target image data; the detection area of ​​the target image data is compared with the detection area of ​​the raw image data to identify whether there are any objects left inside the car. This embodiment finds background image data with similar recent brightness distribution based on the linear relationship between the bottom and side walls of the elevator car in terms of brightness. It then detects objects in the current image data. The background image data is constantly changing, which can effectively adapt to situations such as carpet replacement, floor cleaning, and object obstruction in the elevator car. The brightness intelligently adapts to changes in the elevator car, which can improve the accuracy of detecting objects in the elevator car and reduce false detections or missed detections. The operation of detecting objects mainly involves comparing images and does not rely on a classification network, which can reduce the dependence on high-performance processing, reduce hardware costs and energy consumption, thereby effectively improving the efficiency of detecting objects in elevators.

[0027] 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

[0028] 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.

[0029] Figure 1 This is a flowchart of a method for detecting abandoned objects in an elevator, provided in Embodiment 1 of the present invention.

[0030] Figure 2 This is a system architecture diagram for detecting leftover objects in a car, provided in Embodiment 1 of the present invention.

[0031] Figure 3 This is an example diagram of original image data that does not meet the background conditions, provided in Embodiment 1 of the present invention.

[0032] Figure 4 This is an example diagram of original image data that meets the background conditions provided in Embodiment 1 of the present invention.

[0033] Figure 5 This is an example diagram of a fitting detection region provided in Embodiment 1 of the present invention.

[0034] Figure 6 This is an example diagram of statistical brightness of the detection area and the light reference area provided in Embodiment 1 of the present invention.

[0035] Figure 7 This is a flowchart of a method for detecting abandoned objects in an elevator, provided in Embodiment 2 of the present invention.

[0036] Figure 8 This is an example diagram of a brightness change function provided in Embodiment 2 of the present invention.

[0037] Figure 9 This is a schematic diagram of a device for detecting leftover objects in an elevator, provided in Embodiment 3 of the present invention.

[0038] Figure 10 This is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. Detailed Implementation

[0039] 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.

[0040] 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 used interchangeably where appropriate so that the embodiments of the invention described herein can cover implementations in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover 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.

[0041] Example 1

[0042] See Figure 1 The diagram illustrates a flowchart of a method for detecting abandoned objects in an elevator according to Embodiment 1 of the present invention. This method can be executed by a device for detecting abandoned objects in an elevator. 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:

[0043] Step 101: Collect raw image data that meets the background conditions inside the elevator car.

[0044] Different types of buildings have different transportation needs for people, pets, and goods. Therefore, different types of elevators can be deployed in buildings according to different transportation needs, such as passenger elevators, freight elevators, sightseeing elevators, etc.

[0045] Generally speaking, an elevator is a complex system. Different types of elevators contain different components, and different maintenance strategies can be configured for different components.

[0046] In one example, the components of a certain type of elevator configuration include: elevator control system, call buttons distributed on each floor, car (including car doors), motor for pulling the car (also known as traction machine), control cabinet, speed governor, door operator, car frame, car door, counterweight guide rail, car guide rail, guide rail support, traveling cable, counterweight device, compensating chain (cable), landing door, guide device for compensating chain (cable), buffer, etc.

[0047] In some types of elevators, components such as traction machines, control cabinets, speed governors, and traveling cables can be omitted.

[0048] These components can be divided into different sets according to their functions, thus forming various subsystems that support the operation of the elevator. The elevator control system is connected to the multiple systems of the elevator via wired means such as serial port or serial clock line (SCL). The elevator control system monitors each system and controls the operation of each subsystem, so that the car moves in the hoistway and reaches each floor of the building.

[0049] In one example, the controller includes a door system, a frequency converter system, a call system, and a traction system. The door system controls the elevator doors, which include car doors and hall doors on each floor. The car doors are of the same type as the hall doors and open and close simultaneously. The frequency converter system controls the frequency converter. The call system controls the logic for internal calls (calling the elevator from inside the car) and external calls (calling the elevator from the hall). The traction system controls the vertical movement of the car in the hoistway (vertically upward or vertically downward).

[0050] like Figure 2 As shown, the elevator manager or owner can choose whether to install an edge terminal on the elevator based on factors such as the elevator's load status. If no edge terminal is installed, the controller maintains the original control logic and does not affect the normal operation of the elevator. If an edge terminal is installed, a suitable computing device can be selected as the elevator's edge terminal according to the needs. The edge terminal is combined with the original elevator control system to form a new elevator control system and redefine the elevator's operating logic.

[0051] Generally, edge terminals are computing devices with strong computing capabilities, such as computers, servers, or embedded devices. In addition, depending on the different intelligent services, edge terminals can be equipped with graphics processing units (GPUs) or embedded neural network processors (NPUs).

[0052] An edge terminal refers to a new business platform built at the network edge near the elevator, providing storage, computing, and network resources. It offloads some critical business applications to the access network edge to reduce bandwidth and latency losses caused by network transmission and multi-level forwarding. Located between the user and the cloud (server), the edge terminal is closer to the user (data source) than the traditional cloud, featuring miniaturization, distribution, and user-friendliness. Massive amounts of data (such as audio data) no longer need to be uploaded to the cloud for processing; data processing can be performed at the network edge, reducing request response time, reducing network bandwidth, and ensuring data security and privacy.

[0053] In addition, the edge terminal can realize algorithm functions and model reasoning, communicate with the original elevator control system, and provide the original elevator control system with artificial intelligence (AI) and complex computing capabilities; the edge terminal can also communicate with the cloud to realize algorithm functions and model updates, and relay the function calls of the original elevator control system.

[0054] In practical applications, a camera can be installed at the top of the elevator car. The camera's field of view covers the bottom of the car and the car door. The camera collects raw image data of the car interior, forming a video stream, which is then transmitted to an edge terminal. The edge terminal transmits the video stream data to a server via a mobile network or other means. The operation of detecting leftover objects is then performed on the edge terminal or server. At this point, machine learning or deep learning techniques can be used to filter out raw image data that meets the background conditions from all the raw image data.

[0055] The background conditions are that there are no passengers in the elevator car and the car door is closed. In this case, if there are objects in the car, they may be items left behind by passengers.

[0056] For example, a target detection network can be set up for the car, and sample image data can be collected inside the car. The sample image data can be labeled (such as the status of people and the car doors (including open and closed)). The labeled sample image data can be used to conduct supervised training of the target detection network.

[0057] The structure of the target detection network is not limited to manually designed neural networks, such as YOLO (You Only Look Once), R-CNN (Region Convolutional Neural Networks) series, FPN (Feature Pyramid Networks), etc. It can also be a neural network optimized by model quantization methods, a neural network that searches for the characteristics of passengers and doors in the elevator car using NAS (Neural Architecture Search) methods, etc. This embodiment does not impose any restrictions on this.

[0058] Furthermore, the object detection network can be a pre-trained neural network. Based on this, the object detection network can be fine-tuned using a small amount of labeled sample image data. This reduces the cost of labeling samples and training the object detection network while ensuring the accuracy of object detection.

[0059] In this example, the raw image data can be input into the object detection network to detect whether there are people (i.e., passengers) in the elevator car and the state of the car doors.

[0060] like Figure 3 As shown, if the original image data detects the presence of a passenger in the elevator car and the car door is in the "open" state, then it can be determined that the original image data does not meet the background conditions.

[0061] like Figure 4 As shown, if there is no passenger in the elevator car and the car door is closed, then the original image data can be determined to meet the background conditions.

[0062] Step 102: Draw the detection area at the bottom of the car and the light reference area on the side wall of the car in the original image data.

[0063] In this embodiment, a detection area M located at the bottom of the car and a light reference area N located on the side wall of the car can be drawn in the original image data. When a passenger falls an object, it usually falls to the bottom of the car. Therefore, the passenger's object can usually be detected in the detection area M.

[0064] The light reference area N is a non-transparent area, belonging to the background area of ​​the elevator car, and is used to reflect the ambient light intensity inside the car. Therefore, its height is usually higher than a certain threshold (such as 2 meters) so that passengers will not block the light reference area N. The area of ​​the light reference area N is smaller than the area of ​​the detection area M. There is at least one light reference area N. When there are multiple light reference areas N, sampling multiple light reference areas N evenly on the side wall of the car can improve the stability of the brightness in the light reference area N.

[0065] In practical implementation, a detection area can be fitted to the bottom of the car in the original image data. For ease of processing, the detection area M is kept rectangular or approximately rectangular. Due to the camera's installation angle, the bottom of the car in the original image data exhibits some distortion, such as... Figure 5 As shown, if the detection region M is not rectangular, the original image data can be rotated or otherwise manipulated to make the detection region M approximately rectangular.

[0066] A first threshold is obtained by taking a first proportion of the area of ​​the detection region, and a second threshold is obtained by taking a second proportion of the area of ​​the detection region.

[0067] The values ​​of the first ratio and the second ratio are both in the range of (0,1), and the first ratio is greater than the second ratio.

[0068] In the original image data, a light reference region is determined located on the side wall of the car, with an area between a first threshold and a second threshold.

[0069] For example, suppose the first ratio is 1 / 4 and the second ratio is 1 / 12, then S M / 4>S N >S M / 12, where S M S represents the area of ​​the detection region. N This represents the area of ​​the light reference region.

[0070] like Figure 6 As shown, for a specific car, under the condition that the camera's installation position, FOV (Field of View), focal length, and other factors are fixed, the situation inside the car is relatively fixed in the original image data. Therefore, the bottom and side walls of the car are also relatively fixed. Thus, the detection area M and the light reference area (including N1, N2, and N3) can be pre-drawn offline. When detecting debris in real time, the detection area and the light reference area are loaded into the original image data.

[0071] Step 103: Calculate the first brightness value of the detection area and the second brightness value of the light reference area in the original image data.

[0072] In the original image data, the brightness value of the detection area as a whole is calculated and recorded as the first brightness value. Generally, the average brightness value of each pixel in the detection area is calculated and used as the first brightness value Lm[x], where x is the sequence number of the original image data.

[0073] In the original image data, the brightness value of the light reference area as a whole is calculated and recorded as the second brightness value. Generally, the average brightness value of each pixel in the light reference area is calculated and used as the second brightness value Ln[x], where x is the sequence number of the original image data.

[0074] When there are multiple light reference regions in the original image data, the average value of the second brightness value of the multiple light reference regions can be taken as the final second brightness value.

[0075] For example, such as Figure 6 As shown, the first brightness value of the detection area M is 115.4, the second brightness value of the light reference area N1 is 207.0, the second brightness value of the light reference area N2 is 90.1, and the second brightness value of the light reference area N1 is 41.6.

[0076] Inside the car, the ambient light is uniform. Therefore, Lm[x] and Ln[x] are both positively correlated with the changes in light inside the car, and there is a certain linear relationship between Lm[x] and Ln[x].

[0077] Step 104: Query one or more background image groups.

[0078] In this embodiment, one or more background image sets can be maintained for the car, and the original image data can be used to continuously update one or more background image sets to adapt to changes in the environment inside the car, such as pasting advertisements or replacing carpets.

[0079] Within the same background image group, there are multiple background image groups with a linear relationship. The background image data of the multiple frames in the background image group are all original image data that meet the background conditions and were collected in the elevator car in the past.

[0080] Step 105: Based on the first and second brightness values ​​of the original image data, select background image data from the background image group that matches the brightness of the original image data, and use them as target image data.

[0081] Since the background image data is the original image data from a recent period of time, the background image data contains the detection area and its first brightness value, and the light reference area and its second brightness value.

[0082] In this embodiment, the first and second brightness values ​​of the original image data and the first and second brightness values ​​of the background image data in the background image group can be compared with each other to filter the background image data that matches the original image data in terms of brightness as the target image data. That is, the brightness distribution of the target image data is the same as or similar to the brightness distribution of the original image data.

[0083] In one embodiment of the present invention, step 105 may include the following steps:

[0084] Step 1051: Configure priority for the background image group based on the first brightness value and the second brightness value of the original image data.

[0085] If a background image group exists, it can be assumed that the background image group has the highest priority.

[0086] If there are multiple (i.e. at least two) background image groups, the similarity between the first and second brightness values ​​of the original image data and the first and second brightness values ​​of the background image groups can be evaluated as a whole. Based on the similarity, the background image groups can be prioritized, and target image data that matches the original image data in brightness can be filtered according to the priority. The target image data can be used as a reference to detect residual objects in the original image data, thereby improving the efficiency of residual object detection.

[0087] In the initial state (such as the first detection of leftovers after car maintenance), query the brightness change function configured for the background image group; where the brightness change function represents the linear relationship between the first brightness value and the second brightness value in multiple frames of background image data in the same background image group.

[0088] The second brightness value from the original image data is substituted into the brightness change function to fit the target desired brightness value.

[0089] The difference between the first brightness value in the target image data and the target expected brightness value is calculated as the target brightness error, where the target expected brightness value is the expected brightness value of the detection area in the target image data under the constraint of the brightness value of the light reference area in the target image data.

[0090] For example, the brightness change function can be expressed as: Lm=a×Ln+b, where Lm is the first brightness value, Ln is the second brightness value, a is the slope, and b is the offset.

[0091] In this example, let the first brightness value of the original image data be Lm[x] and the second brightness value be Ln[x]. Substitute the second brightness value Ln[x] into the brightness change function to calculate the target expected brightness value Lm[x]′=a×Ln[x]+b. At this time, the target brightness error c=|Lm[x]-Lm[x]′|.

[0092] The background image group is assigned a priority based on the target brightness error; the priority is negatively correlated with the target brightness error, that is, the larger the target brightness error, the lower the priority, and vice versa.

[0093] For example, there are three background image groups T1, T2 and T3. The target brightness error of background image group T1 is c1, the target brightness error of background image group T2 is c2 and the target brightness error of background image group T3 is c3. If c1 < c2 < c3, then in terms of priority, T1 > T2 > T3.

[0094] Under normal circumstances, the raw image data captured by the camera is relatively stable in a short period of time and will not change abruptly. Therefore, in non-initial state, the priority of the background image group when detecting residues in the previous frame of raw image data (i.e., the priority of the background image group when detecting residues for the last time) can be queried and used as the priority of the background image group when detecting residues in the current frame of raw image data. In this way, the calculation of target brightness error and other operations are eliminated, which can reduce the amount of computation and save resources.

[0095] Step 1052: Read the background image group according to priority and use it as the target image group.

[0096] Step 1053: In the target image group, calculate the difference between the second brightness value of the original image data and the second brightness value of the background image data as the sidewall brightness error.

[0097] Step 1054: In the target image group, determine the background image data with the smallest sidewall brightness error that matches the original image data in brightness, and use it as the target image data.

[0098] In this embodiment, background image groups can be read sequentially according to priority, and are referred to as target image groups.

[0099] In the current target image group, the difference between the second brightness value of the original image data and the second brightness value of each background image data is calculated and denoted as the brightness error.

[0100] Let the second brightness value of the original image data be Ln[x], and the second brightness value of each background image data in the target image group be Ln[y]. Then the brightness error ΔLn=|Ln[x]-Ln[y]|.

[0101] In the target image group, the brightness errors of each sidewall are compared, and the background image data with the smallest sidewall brightness error is selected. That is, the background image data whose second brightness value is closest to the original image data is selected. This background image data matches the original image data in brightness and is used as the target image data.

[0102] Step 106: Compare the detection area of ​​the target image data with the detection area of ​​the original image data to identify whether there are any objects left in the car.

[0103] In this embodiment, the detection area of ​​the target image data can be compared with the detection area of ​​the original image data, and the presence of any remaining objects in the car can be identified based on the distribution information of the similarities and differences between the detection areas of the target image data and the detection areas of the original image data.

[0104] like Figure 2 As shown, when an object is detected in the elevator car, a notification message can be generated and pushed to the client and the monitoring personnel. The monitoring personnel can retrieve the video stream data inside the elevator car from the server for manual review. If an object is found during the manual review, subsequent processing such as cleaning up the object, registering the lost item, and tracking the owner can be carried out.

[0105] In the specific implementation, the original image data and the target image data are converted to grayscale respectively. The detection area in the original image data is converted into the first grayscale area, and the detection area in the target image data is converted into the second grayscale area.

[0106] To improve the efficiency of real-time detection of leftover objects in elevator cars, the target image data can be processed into grayscale offline. When detecting leftover objects in elevator cars in real time, the second grayscale area can be read directly.

[0107] The first similarity R between the first gray-level region and the second gray-level region is calculated using methods such as squared difference, normalized cross-correlation, and normalized correlation coefficient.

[0108] For example, the first similarity R in a certain way takes the range of [-1, 1]. The larger the value, the higher the matching degree between the first gray area and the second gray area.

[0109] If the first similarity R is greater than or equal to the preset first threshold Y0 (i.e., R≥Y0), it is determined that there are no leftover objects in the car, and the priority of the background image group corresponding to the target image data is adjusted to the highest level.

[0110] When there are at least three background image groups, the background image groups other than the background image group corresponding to the target image data have a low impact on the detection of the residue. The priority of other background image groups can be adjusted in any way. For example, the priority of the background image group with the highest priority can be set to the original priority of the background image group corresponding to the target image data, etc. This embodiment does not limit this.

[0111] If the first similarity R is less than the preset first threshold Y0 (i.e., R < Y0), then switch to the target image data in the next priority target image group; return to perform grayscale processing on the original image data and the target image data respectively.

[0112] If the first similarity R of all target image data is less than the preset first threshold Y0 (i.e., R < Y0), then the detection region in the original image data is divided into multiple first sub-regions according to the specified segmentation method (such as a nine-square grid). Each first sub-region is numbered.

[0113] For the target image data in the highest priority background image group, the detection region in the target image data is divided into multiple second sub-regions according to a specified segmentation method (such as a nine-square grid). Each second sub-region is numbered.

[0114] For the same number, the second similarity Ri between the first and second sub-regions is calculated using methods such as squared difference, normalized cross-correlation, and normalized correlation coefficient, where i is the position number.

[0115] In multiple consecutive frames of original image data, if the second similarity Ri corresponding to the number is less than the preset second threshold Y1 (i.e., Ri < Y1), then the number is marked as a change position. There may be a residual, that is, the state of the change position remains unchanged for a certain period of time (e.g., 3 seconds).

[0116] If the number of changing positions is less than the preset third threshold, it indicates that a small area at the bottom of the car has changed in a short period of time, which is consistent with the characteristics of leftovers, and it is determined that there are leftovers in the car.

[0117] If the number of changing positions is greater than or equal to the preset third threshold, it indicates that a large area of ​​the bottom of the car has changed in a short period of time, which meets the characteristics of changing the background. Then, it is determined that the background of the bottom of the car will be changed, such as changing the carpet at the bottom of the car, pasting advertisements at the bottom of the car, etc.

[0118] The third threshold is related to the number of positions; that is, the third threshold tends to be close to the number of positions but is less than the number of positions. For example, when the number of positions is 9, the third threshold is 7.

[0119] In this embodiment, raw image data meeting background conditions is acquired inside the elevator car; the background conditions are: no passengers inside the car and the car door is closed; a detection area located at the bottom of the car and a light reference area located on the side wall of the car are drawn in the raw image data; a first brightness value of the detection area and a second brightness value of the light reference area are calculated in the raw image data; one or more background image groups are queried; the multiple frames of background image data in the background image group are all raw image data that meet the background conditions and were historically acquired inside the elevator car; based on the first and second brightness values ​​of the raw image data, background image data that matches the brightness of the raw image data is selected from the background image group as target image data; the detection area of ​​the target image data is compared with the detection area of ​​the raw image data to identify whether there are any objects left inside the car. This embodiment finds background image data with similar recent brightness distribution based on the linear relationship between the bottom and side walls of the elevator car in terms of brightness. It then detects objects in the current image data. The background image data is constantly changing, which can effectively adapt to situations such as carpet replacement, floor cleaning, and object obstruction in the elevator car. The brightness intelligently adapts to changes in the elevator car, which can improve the accuracy of detecting objects in the elevator car and reduce false detections or missed detections. The operation of detecting objects mainly involves comparing images and does not rely on a classification network, which can reduce the dependence on high-performance processing, reduce hardware costs and energy consumption, thereby effectively improving the efficiency of detecting objects in elevators.

[0120] Example 2

[0121] See Figure 7 The diagram illustrates a flowchart of a method for detecting abandoned objects in an elevator according to Embodiment 2 of the present invention. This embodiment adds an operation of updating background image data based on the previous embodiments. Figure 7 As shown, the method includes:

[0122] Step 701: Collect raw image data that meets the background conditions inside the elevator car.

[0123] The background conditions are that there are no passengers in the elevator car and the elevator car door is closed.

[0124] Step 702: Draw the detection area at the bottom of the car and the light reference area on the side wall of the car in the original image data.

[0125] Step 703: Calculate the first brightness value of the detection area and the second brightness value of the light reference area in the original image data.

[0126] Step 704: Mark the original image data as background image data at preset time intervals.

[0127] In this embodiment, the original image data can be collected from the video stream data generated by the camera at certain time intervals (such as 30 minutes) and updated to background image data. That is, at each preset time interval (such as 30 minutes), the original image data is marked as background image data, the time of generating the original image data is recorded, and it is stored in the form of a cache queue, database, etc.

[0128] Step 705: If the amount of background image data reaches a preset upper limit, then a linear brightness change function is fitted based on the first brightness value and the second brightness value in the background image data.

[0129] In this embodiment, the number of current background image data can be counted. If the number of background image data reaches a preset upper limit (e.g., 30), the background image data can be re-divided into background image groups. Furthermore, a linear brightness change function can be fitted based on the first brightness value and the second brightness value in the background image data of the same background image group.

[0130] In the specific implementation, a brightness point is constructed using the first brightness value in the background image data as the y-coordinate and the second brightness value in the background image data as the x-coordinate.

[0131] The second brightness value in the background image data is substituted into the linear brightness change function to calculate the expected brightness value of the sample. The expected brightness value of the sample is the expected brightness value of the detection area in the background image data under the constraint of the brightness value of the light reference area in the background image data.

[0132] Based on the expected brightness value of the sample, a preset standard brightness error is added and subtracted to obtain the error range. That is, the expected brightness value of the sample is the range of reasonable fluctuation of the brightness value of the detection area in the background image data.

[0133] Since the environment inside the car is relatively simple, an upper limit (e.g., 3) can be set for the number of background image groups. Under the condition that the first brightness value in the background image data is within the error range, the brightness points are divided into one or more clusters. That is, considering that there is an upper limit to the number of clusters, the first brightness value in the background image data is within the error range as a constraint and the goal of minimizing the number of background image groups is to solve the clustering problem for the brightness points, and obtain one or more clusters, each of which has multiple brightness points.

[0134] If the parameters (such as slope, offset, etc.) in the brightness variation function are fitted using brightness points from the same cluster, then the background image data to which the brightness points belong will be divided into the same background image group within the same cluster.

[0135] For example, the brightness change function can be expressed as: Lm=a×Ln+b, where Lm is the first brightness value, Ln is the second brightness value, a is the slope to be fitted, and b is the offset to be fitted.

[0136] In this example, let the first brightness value of the original image data be Lm[x], the second brightness value be Ln[x], and the labeled brightness error be c. The labeled brightness error is the maximum error introduced when fitting the brightness change function, which is usually an empirical value. Then, the first brightness value in the background image data is within the error range and can be expressed as: a×Ln[x]+bc<Lm[x]<a×Ln[x]+b+c.

[0137] like Figure 8 As shown, considering situations such as replacing the carpet at the bottom of the car and turning off the car lights when there are no passengers, this example uses three background image sets to fit three brightness change functions, that is, to fit three sets of parameters to the brightness change functions.

[0138] Of course, if one or two background image groups can cover all the brightness points, the remaining background image groups can be set to null, that is, no brightness points are included in the background image group.

[0139] In this coordinate system, the x-axis represents the second brightness value Ln, and the y-axis represents the first brightness value Lm. Different brightness points may have the same second brightness value Ln but different first brightness values ​​Lm. For example, in the background image cluster1, a certain brightness point has Ln of 2 and Lm of approximately 8, while in the background image cluster1, a certain brightness point has Ln of 2 and Lm of approximately 20.

[0140] The brightness variation function fitted using the background image group cluster2 reflects a general linear relationship, where both the light reference area N and the detection area M increase with the increase of ambient light brightness. The brightness variation function fitted using the background image group cluster1 can be understood as the bottom of the car being covered with a darker carpet, where the light reference area N increases with the increase of ambient light brightness, while the detection area M does not change significantly with the increase of ambient light brightness. The brightness variation function fitted using the background image group cluster2 is a reservation, supplement, and correction for general scenes.

[0141] Step 706: Remove the earliest portion of the background image data.

[0142] In this embodiment, the earliest generated portion of the background image data can be removed from locations such as cache queues and databases.

[0143] The number of background image data removed is related to the upper limit of the number of background image data. For example, if the upper limit of the number of background image data is 30, then the 10 earliest generated frames of background image data will be removed.

[0144] Example 3

[0145] See Figure 9 The diagram shows a structural schematic of a device for detecting abandoned objects in an elevator, according to Embodiment 3 of the present invention. Figure 9 As shown, the device includes:

[0146] Image acquisition module 901 is used to acquire raw image data that meets background conditions inside the elevator car; the background conditions are that there are no passengers in the car and the car door is closed.

[0147] The region recognition module 902 is used to draw a detection region located at the bottom of the car and a light reference region located on the side wall of the car in the original image data;

[0148] The brightness statistics module 903 is used to calculate a first brightness value of the detection area and a second brightness value of the light reference area in the original image data.

[0149] Background image group query module 904 is used to query one or more background image groups; the background image group contains multiple frames of background image data, which are all original image data that meet the background conditions and were historically collected inside the elevator car.

[0150] The target image data filtering module 905 is used to filter background image data that matches the brightness of the original image data in the background image group based on the first brightness value and the second brightness value of the original image data, and use them as target image data.

[0151] The detection area comparison module 906 is used to compare the detection area of ​​the target image data with the detection area of ​​the original image data to identify whether there are any objects left in the car.

[0152] In one embodiment of the present invention, the region identification module 902 includes:

[0153] A detection region fitting module is used to fit a detection region to the bottom of the car in the original image data;

[0154] The first threshold generation module is used to take a first proportion of the area of ​​the detection region to obtain a first threshold.

[0155] The second threshold generation module is used to take a second ratio on the area of ​​the detection region to obtain a second threshold; wherein the first ratio is greater than the second ratio.

[0156] The light reference area determination module is used to determine, in the original image data, a light reference area located on the side wall of the car, and whose area is between the first threshold and the second threshold.

[0157] In one embodiment of the present invention, the target image data filtering module 905 includes:

[0158] The priority configuration module is used to configure a priority for the background image group based on the first brightness value and the second brightness value of the original image data;

[0159] The target image group reading module is used to read the background image group according to the priority, and use it as the target image group;

[0160] The sidewall brightness error calculation module is used to calculate the difference between the second brightness value of the original image data and the second brightness value of the background image data in the target image group, as the sidewall brightness error;

[0161] The sidewall brightness error filtering module is used to determine, in the target image group, the background image data with the smallest sidewall brightness error that matches the original image data in brightness, and use it as the target image data.

[0162] In one embodiment of the present invention, the priority configuration module includes:

[0163] The brightness variation function query module is used to query the brightness variation function configured for the background image group; the brightness variation function represents the linear relationship between a first brightness value and a second brightness value in multiple frames of the background image data;

[0164] The target desired brightness value fitting module is used to substitute the second brightness value in the original image data into the brightness change function to calculate the target desired brightness value;

[0165] The target brightness error calculation module is used to calculate the difference between the first brightness value in the original image data and the target expected brightness value, as the target brightness error;

[0166] The target brightness error configuration module is used to configure a priority for the background image group according to the target brightness error; the priority is negatively correlated with the target brightness error.

[0167] In one embodiment of the present invention, the priority configuration module further includes:

[0168] The priority query module is used to query the priority of the background image group when detecting residues in the original image data of the previous frame, and use it as the priority of the background image group when detecting residues in the original image data of the current frame.

[0169] In one embodiment of the present invention, the detection area comparison module 906 includes:

[0170] The first grayscale region conversion module is used to convert the detection region in the original image data into a first grayscale region;

[0171] The second grayscale region conversion module is used to convert the detection region in the target image data into a second grayscale region;

[0172] The first similarity calculation module is used to calculate the first similarity between the first gray area and the second gray area;

[0173] The first object determination module is used to determine that there is no object left in the car if the first similarity is greater than or equal to a preset first threshold, and to adjust the priority of the background image group corresponding to the target image data to the highest level.

[0174] The target image group switching module is used to switch to the target image data in the background image group of the next priority if the first similarity is less than a preset first threshold; and then return to execute the second grayscale region conversion module.

[0175] The first sub-region division module is used to divide the detection region in the original image data into multiple first sub-regions if the first similarity of all the target image data is less than a preset first threshold; each first sub-region has a number.

[0176] The second sub-region division module is used to divide the detection area in the target image data in the background image group with the highest priority into multiple second sub-regions; each second sub-region is numbered.

[0177] The second similarity calculation module is used to calculate the second similarity between the first sub-region and the second sub-region for the same number;

[0178] The change bit annotation module is used to annotate the change bit of the number if the second similarity corresponding to the number is less than a preset second threshold in multiple consecutive frames of the original image data.

[0179] The second abandoned object determination module is used to determine that there is abandoned object in the car if the number of the changed positions is less than a preset third threshold.

[0180] The background switching determination module is used to determine the background switching at the bottom of the car if the number of changing positions is greater than or equal to a preset third threshold.

[0181] In one embodiment of the present invention, it further includes:

[0182] The background image data marking module is used to mark the original image data as background image data at preset time intervals.

[0183] The brightness change function fitting module is used to fit a linear brightness change function based on the first brightness value and the second brightness value in the background image data if the amount of background image data reaches a preset upper limit value.

[0184] The background image data removal module is used to remove the earliest portion of the background image data.

[0185] In one embodiment of the present invention, the brightness change function fitting module includes:

[0186] A brightness point construction module is used to construct brightness points using a first brightness value in the background image data as the y-coordinate and a second brightness value in the background image data as the x-coordinate.

[0187] The sample expected brightness value fitting module is used to substitute the second brightness value in the background image data into a linear brightness change function to fit the sample expected brightness value.

[0188] The error range generation module is used to add a preset standard brightness error and subtract a preset standard brightness error from the expected brightness value of the sample to obtain the error range.

[0189] A clustering module is used to divide the brightness points into one or more clusters, provided that the first brightness value in the background image data is within the error range.

[0190] A parameter fitting module is used to fit the parameters in the brightness change function using the brightness points in the same cluster;

[0191] The background image group division module is used to divide the background image data to which the brightness points belong into the same background image group within the cluster.

[0192] The device for detecting abandoned objects in elevators provided in this embodiment of the invention can execute the method for detecting abandoned objects in elevators provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method for detecting abandoned objects in elevators.

[0193] Example 4

[0194] See Figure 10 This diagram illustrates a structural schematic of an electronic device according to an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, blade servers, mainframe computers, and other suitable computers. 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.

[0195] like Figure 10 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.

[0196] 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.

[0197] 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, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as methods for detecting abandoned objects in an elevator.

[0198] In some embodiments, the method for detecting abandoned objects in an elevator may 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 may 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 detecting abandoned objects in an elevator described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the method for detecting abandoned objects in an elevator by any other suitable means (e.g., by means of firmware).

[0199] 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.

[0200] 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.

[0201] 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.

[0202] 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).

[0203] 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.

[0204] 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.

[0205] Example 5

[0206] This invention also provides a computer program product comprising a computer program that, when executed by a processor, implements the method for detecting abandoned objects in an elevator as provided in any embodiment of this invention.

[0207] In implementing the computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0208] 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.

[0209] 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 detecting abandoned items in an elevator, characterized in that, include: Collect raw image data that meets the background conditions inside the elevator car; The background conditions are that there are no passengers in the elevator car and the elevator car door is closed; The detection area located at the bottom of the car and the light reference area located on the side wall of the car are plotted in the original image data. Calculate a first brightness value for the detection area and a second brightness value for the light reference area from the original image data; Query one or more background image groups; the background image data of the multiple frames in the background image group are all original image data that meet the background conditions and were collected in the elevator car in history. Based on the first and second brightness values ​​of the original image data, background image data that matches the brightness of the original image data is selected from the background image group and used as target image data. The detection area of ​​the target image data is compared with the detection area of ​​the original image data to identify whether there are any objects left in the car.

2. The method according to claim 1, characterized in that, The step of drawing the detection area located at the bottom of the car and the light reference area located on the side wall of the car in the original image data includes: Fit a detection region to the bottom of the car in the original image data; A first threshold is obtained by taking a first proportion of the area of ​​the detection region; A second threshold is obtained by taking a second proportion of the area of ​​the detection region; wherein the first proportion is greater than the second proportion. In the original image data, a light reference region is determined that is located on the side wall of the car and has an area between the first threshold and the second threshold.

3. The method according to claim 1, characterized in that, The step of selecting background image data from the background image group that matches the brightness of the original image data based on the first and second brightness values ​​of the original image data, and using this as target image data, includes: The background image group is assigned a priority based on the first brightness value and the second brightness value of the original image data; The background image group is read according to the stated priority and used as the target image group; In the target image group, the difference between the second brightness value of the original image data and the second brightness value of the background image data is calculated as the sidewall brightness error; In the target image group, the background image data with the smallest sidewall brightness error is determined to match the original image data in brightness and is used as the target image data.

4. The method according to claim 3, characterized in that, The step of configuring a priority for the background image group based on the first brightness value and the second brightness value of the original image data includes: The brightness variation function configured for the background image group is queried; the brightness variation function represents the linear relationship between a first brightness value and a second brightness value in multiple frames of the background image data; The second brightness value in the original image data is substituted into the brightness change function to calculate the target desired brightness value; The difference between the first brightness value in the original image data and the target desired brightness value is calculated as the target brightness error; The background image group is assigned a priority according to the target brightness error; the priority is negatively correlated with the target brightness error.

5. The method according to claim 4, characterized in that, The step of configuring a priority for the background image group based on the first brightness value and the second brightness value of the original image data further includes: The priority of the background image group when detecting remnants in the original image data of the previous frame is used as the priority of the background image group when detecting remnants in the original image data of the current frame.

6. The method according to claim 3, characterized in that, The step of comparing the detection area of ​​the target image data with the detection area of ​​the original image data to identify whether there are any objects left in the elevator car includes: The detection area in the original image data is converted into a first grayscale area; The detection region in the target image data is converted into a second grayscale region; Calculate the first similarity between the first grayscale region and the second grayscale region; If the first similarity is greater than or equal to a preset first threshold, it is determined that there are no leftovers in the car, and the priority of the background image group corresponding to the target image data is adjusted to the highest level. If the first similarity is less than a preset first threshold, then switch to the target image data in the next priority target image group; return to the step of converting the detection area in the target image data into a second grayscale area; If the first similarity of all the target image data is less than a preset first threshold, then the detection region in the original image data is divided into multiple first sub-regions; each first sub-region is numbered. For the target image data in the background image group with the highest priority, the detection area in the target image data is divided into multiple second sub-regions; each second sub-region is numbered. For the same number, calculate the second similarity between the first sub-region and the second sub-region; In multiple consecutive frames of the original image data, if the second similarity corresponding to the number is less than a preset second threshold, then the number is marked with a change bit. If the number of the changed positions is less than a preset third threshold, it is determined that there are leftovers in the car; If the number of changing positions is greater than or equal to a preset third threshold, then the background at the bottom of the car is switched.

7. The method according to any one of claims 1-6, characterized in that, Also includes: At preset time intervals, the original image data is marked as background image data; If the amount of background image data reaches a preset upper limit, a linear brightness change function is fitted based on the first brightness value and the second brightness value in the background image data. Remove the earliest portion of the background image data.

8. The method according to claim 7, characterized in that, The step of fitting a linear brightness change function based on the first brightness value and the second brightness value in the background image data includes: A brightness point is constructed using the first brightness value in the background image data as the y-coordinate and the second brightness value in the background image data as the x-coordinate. The second brightness value in the background image data is substituted into a linear brightness change function to fit the expected brightness value of the sample; Based on the expected brightness value of the sample, add a preset standard brightness error and subtract the preset standard brightness error respectively to obtain the error range; If the first brightness value in the background image data is within the error range, the brightness points are divided into one or more clusters. The parameters in the brightness variation function are fitted using the brightness points within the same cluster; Within the cluster, the background image data to which the brightness points belong are grouped into the same background image group.

9. 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 executable by the at least one processor, which enables the at least one processor to perform the method for detecting abandoned objects in an elevator as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method for detecting abandoned objects in an elevator as described in any one of claims 1-8.

Images