Anti-pinch folding and unfolding furniture, anti-pinch control method and device and related system

Abstract

The invention discloses anti-pinch folding and unfolding furniture, an anti-pinch control method, an anti-pinch control device and a related system.The anti-pinch folding and unfolding furniture comprises a fixing part, a folding and unfolding part, a processor, a folding and unfolding driving mechanism and an anti-pinch detection device, and the processor is used for receiving and responding to a folding control instruction and sending an image collection instruction to the anti-pinch detection device; the anti-pinch detection device is used for receiving and responding to the image acquisition instruction, acquiring a first image of the folding and unfolding space and sending the first image to the processor; the processor is used for receiving the first image and detecting whether a space occupying object exists in the folding and unfolding space or not according to the first image; if it is detected that the space occupying object exists, the space occupying state detection operation continues after waiting; if it is detected that no space occupying object exists, a folding action instruction is sent to a folding and unfolding driving mechanism; and the folding and unfolding driving mechanism is used for receiving the folding action instruction and executing folding operation. In this way, the safety and attractiveness of the anti-pinch folding and unfolding furniture are improved.

Description

Technical Field

[0001] This application relates to the field of smart furniture technology, and in particular to an anti-pinch folding furniture, an anti-pinch control method, a device, and a related system. Background Technology

[0002] As living standards continue to improve, people's pursuit of quality of life is also increasing. In the field of smart homes, people not only want furniture to have increasingly diversified functions, but also put forward high standards for furniture safety. For example, in retractable sofas or chairs, the movable units can be moved by electric drive mechanisms to allow the sofa or chair to switch between sitting and reclining positions.

[0003] However, in the reclining extended position, the movable unit extends away from the main body of the sofa or chair, exposing the space between the movable unit and the main body. Foreign objects may enter this space from the outside, preventing the sofa or chair from returning to its original position. Furthermore, if a pet or child enters this space, they may be pinched or injured during folding. Current foldable furniture uses protective panels on the movable unit to prevent foreign objects from entering the folding space. However, this design not only increases the freedom of movement of the foldable furniture but also affects its aesthetics.

[0004] Therefore, how to improve the safety and aesthetics of anti-pinch folding furniture is an urgent issue to be addressed. Summary of the Invention

[0005] This application provides an anti-pinch folding furniture, an anti-pinch control method, a device, and a related system, which improves the safety and aesthetics of foldable furniture.

[0006] In a first aspect, embodiments of this application provide an anti-pinch folding furniture, comprising a furniture fixing part, a furniture unfolding part, a processor, an unfolding drive mechanism, and an anti-pinch detection device disposed on a first end face of the furniture fixing part. The first end face is the end face of the furniture unfolding part facing the unfolding space, the unfolding space being the space enclosed by the first end face and the second end face when the anti-pinch folding furniture is in the unfolded state, and the second end face being the inner end face of the furniture fixing part. The processor is configured to receive and respond to a retraction control command for the furniture retraction section; and send a first image acquisition command to the anti-pinch detection device. The anti-pinch detection device is used to receive and respond to the first image acquisition command; acquire a first image of the retractable space; and send the first image to the processor. The processor is configured to receive the first image and detect, based on the first image, whether a first space occupant object exists in the retractable space; If the first space occupant is detected, the image acquisition and space occupancy status detection operations are performed on the expanded and contracted space after a preset time period; and / or, a prompt message is output to remind the user to move the first space occupant. If the first space occupant is not detected, a retraction action command is sent to the retraction drive mechanism; The retraction drive mechanism is used to receive the retraction action command and start executing the retraction operation.

[0007] Secondly, embodiments of this application provide an anti-pinch control method applied to retractable furniture. The anti-pinch retractable furniture includes a furniture fixing part, a furniture retractable part, a processor, a retractable drive mechanism, and an anti-pinch detection device disposed on a first end face of the furniture retractable part. The first end face is the end face of the furniture fixing part facing the retractable space. The retractable space refers to the space enclosed by the first end face and the second end face of the anti-pinch retractable furniture in the unfolded state. The second end face refers to the inner end face of the furniture retractable part. The method includes: The anti-pinch detection device is controlled to receive and respond to a first image acquisition command; acquire a first image of the retraction space; and send the first image to the processor. The processor is controlled to receive the first image and detect whether there is a space occupant in the retractable space based on the first image; if the space occupant is detected, the image acquisition and space occupant status detection operations are performed on the retractable space after a preset first time period; and / or, a prompt message is output to remind the user to move the space occupant; if the space occupant is not detected, a retracting action command is sent to the retractable drive mechanism. The retraction drive mechanism receives the retraction action command and executes the retraction operation.

[0008] Thirdly, this application provides an anti-pinch control device applied to a processor of the anti-pinch folding furniture. The anti-pinch folding furniture includes a furniture fixing part, a furniture folding part, the processor, a folding drive mechanism, and an anti-pinch detection device disposed on a first end face of the furniture folding part. The first end face is the end face of the furniture fixing part facing the folding space. The folding space refers to the space enclosed by the first end face and the second end face of the anti-pinch folding furniture in the unfolded state. The second end face refers to the inner end face of the furniture folding part. The anti-pinch control device includes: The acquisition unit is configured to control the anti-pinch detection device to receive and respond to a first image acquisition command; acquire a first image of the retraction space; and send the first image to the processor. The control unit is configured to control the processor to receive the first image, and detect whether there is a space occupant in the retractable space based on the first image; if the space occupant is detected, the control unit waits for a preset first time period and then continues to perform image acquisition and space occupancy status detection operations on the retractable space; and / or outputs a prompt message to remind the user to move the space occupant; if the space occupant is not detected, the control unit sends a retracting action command to the retractable drive mechanism. An execution unit is used to control the retraction drive mechanism to receive the retraction action command and execute the retraction operation.

[0009] Fourthly, embodiments of this application provide an electronic device including a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the programs include instructions for performing steps in any method of the second aspect of embodiments of this application.

[0010] Fifthly, embodiments of this application provide an anti-pinch control system, wherein the anti-pinch control system can perform some or all of the steps described in any method of the second aspect of embodiments of this application.

[0011] By implementing the embodiments of this application, the following beneficial effects are achieved: This application describes an anti-pinch folding furniture, an anti-pinch control method, a device, and a related system. The anti-pinch folding furniture includes a furniture fixing part, a furniture unfolding part, a processor, an unfolding drive mechanism, and an anti-pinch detection device disposed on a first end face of the furniture fixing part. The first end face is the end face of the furniture unfolding part facing the unfolding space, which refers to the space enclosed by the first and second end faces when the anti-pinch folding furniture is in its unfolded state. The second end face is the inner end face of the furniture fixing part. The processor is used to receive and respond to a folding control command for the unfolding part and to send a first image acquisition command to the anti-pinch detection device. The anti-pinch detection device is used for... The system receives and responds to a first image acquisition command, acquires a first image of the retractable space, and sends the first image to a processor. The processor receives the first image and detects whether a first space-occupying object exists in the retractable space based on the first image. If a first space-occupying object is detected, the system waits for a preset time period and then continues to perform image acquisition and space-occupying status detection operations on the retractable space, and / or outputs a prompt message to remind the user to move the first space-occupying object. If no first space-occupying object is detected, a retraction action command is sent to the retraction drive mechanism. The retraction drive mechanism receives the retraction action command and begins to execute the retraction operation. Thus, on the one hand, by detecting the retractable space through the detection module in the anti-pinch detection device and outputting prompt messages or action commands from the retraction drive mechanism based on the detection results, injuries to pets or children caused by the retraction of furniture can be avoided, improving the safety of anti-pinch retractable furniture. On the other hand, by installing the anti-pinch detection device on the anti-pinch retractable furniture, redundant structures are avoided to ensure the safety of the anti-pinch furniture, thereby improving the aesthetics of the anti-pinch retractable furniture. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a system architecture diagram of an anti-pinch furniture unfolding system provided in an embodiment of this application; Figure 2 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application; Figure 3 This is a flowchart illustrating a processor execution method for anti-pinch retractable furniture provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of an anti-pinch detection device provided in an embodiment of this application; Figure 5This is a flowchart illustrating an anti-pinch control method for anti-pinch retractable furniture provided in an embodiment of this application; Figure 6 This is a schematic diagram of the functional interaction relationship of an anti-pinch detection method provided in an embodiment of this application; Figure 7 This is a structural schematic diagram of an anti-pinch folding furniture provided in an embodiment of this application; Figure 8 This is a block diagram of the functional modules of an anti-pinch control device provided in an embodiment of this application. Detailed Implementation

[0014] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0015] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. 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 includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0016] It should be understood that the term "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this document indicates that the preceding and following related objects are in an "or" relationship. In the embodiments of this application, "multiple" refers to two or more.

[0017] In the embodiments of this application, "at least one item" or its similar expression refers to any combination of these items, including any combination of a single item or a plurality of items. "One or more" means one or more, while "multiple" means two or more. For example, "at least one item" of a, b, or c can represent the following seven cases: a, b, c; a and b; a and c; b and c; a, b, and c. Each of a, b, and c can be an element or a set containing one or more elements.

[0018] In this application, the term "connection" refers to various connection methods, such as direct connection or indirect connection, to achieve communication between devices. This application does not impose any limitations on this.

[0019] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0020] The following is an explanation of the relevant terms used in this application: Edge processing algorithms: Edge processing algorithms are a technology in the field of digital image processing. Essentially, they identify and extract the background of objects in an image or the "edges" between different areas of an object by analyzing the gray values, colors, or texture changes of pixels in the image. That is, the set of pixels where pixel attributes change drastically.

[0021] Existing folding furniture poses a risk of pinching or injuring pets or other objects within its folding space when it is being folded up. Currently, some folding furniture uses force-sensing anti-pinch solutions. Specifically, a processor detects that the folding section is contacting an object that is being obstructed or subjected to force based on force sensing data. If this is detected, the folding section stops folding and unfolds to remove pressure on the object. However, because pinching an object could cause it to be startled or deformed, current anti-pinch folding furniture still has shortcomings in terms of intelligence and safety.

[0022] To address the aforementioned problems, this application provides an anti-pinch folding furniture, an anti-pinch control method, a device, and a related system. The anti-pinch folding furniture includes a furniture fixing part, a furniture unfolding part, a processor, an unfolding drive mechanism, and an anti-pinch detection device disposed on a first end face of the furniture fixing part. The first end face is the end face of the furniture unfolding part facing the unfolding space. The unfolding space refers to the space enclosed by the first and second end faces when the anti-pinch folding furniture is in its unfolded state. The second end face refers to the inner end face of the furniture fixing part. The processor receives and responds to a folding control command for the unfolding part and sends a first image acquisition command to the anti-pinch detection device. The anti-pinch detection... The device receives and responds to a first image acquisition command, acquires a first image of the retractable space, and sends the first image to a processor. The processor receives the first image and detects whether a first space-occupying object exists in the retractable space based on the first image. If a first space-occupying object is detected, it waits for a preset time period and then continues to perform image acquisition and space-occupying status detection operations on the retractable space, and / or outputs a prompt message to remind the user to move the first space-occupying object. If no first space-occupying object is detected, it sends a retraction action command to the retractable drive mechanism. The retractable drive mechanism receives the retraction action command and begins to execute the retraction operation. In this way, anti-pinch retractable furniture can accurately detect and identify the space-occupying status of a space-occupying object in the retractable space without contacting or squeezing it. The anti-pinch retractable furniture can control the retractable part to stop the retraction action before contacting or squeezing the space-occupying object. This improves the safety of the anti-pinch retractable furniture. Furthermore, since the anti-pinch detection mechanism detects the space occupied by the furniture during the unfolding process, there is no need to design redundant structures to prevent the furniture from being pinched, thus improving the aesthetics of the anti-pinch furniture.

[0023] The following is combined with Figure 1 The system architecture of an anti-pinch furniture retractable device according to an embodiment of this application will be described. Figure 1 This is a system architecture diagram of an anti-pinch folding furniture according to an embodiment of this application. The system architecture 100 of the anti-pinch folding furniture includes an anti-pinch detection module 110, a processor 120, a folding drive mechanism 130, and an alarm module 140.

[0024] The anti-pinch detection module 110 is the front-end sensing unit of the anti-pinch furniture, used to perceive and monitor the state within the furniture's unfolding space in real time to avoid potential pinching risks during the unfolding process. The anti-pinch detection module 110 includes a camera module 111, which can be a regular visible light camera, a night vision camera, or an infrared camera. Under normal lighting conditions, a regular visible light camera is sufficient for image acquisition; however, when the furniture is in a poorly lit indoor environment or at night, a night vision or infrared camera can provide a clearer image, avoiding detection errors caused by dim lighting. After receiving an image acquisition command, the camera module 111 can acquire image information within the unfolding space in real time and transmit the image data to the processor 120 for further processing. In one possible embodiment, the anti-pinch detection module 110 is fixedly connected to the furniture's unfolding part via an installation structure, which can be a magnetic structure or a fastener structure. When a magnetic structure is used, the user can directly attach the anti-pinch detection module 110 to the target position of the furniture, which is convenient for disassembly and maintenance. When a fastener structure is used, the anti-pinch detection module 110 can be stably installed at the target position of the furniture movement module to ensure that the camera module 111 can move synchronously with the extension and retraction part, thereby continuously monitoring the potential pinch area.

[0025] The processor 120 is the processing and control unit for the anti-pinch furniture retraction system. It receives image information collected by the anti-pinch detection module 110, performs image processing and target recognition operations, and generates corresponding control commands based on the analysis results. The processor 120 includes a receiving module 121, a processing module 122, and a transmitting module 123. The receiving module 121 receives image data collected by the anti-pinch detection module 110 and performs buffering and preliminary analysis. Since the camera module 111 may output a video stream or multiple consecutive frames of images, the receiving module 121 needs to have data stream parsing and frame synchronization capabilities to ensure the extraction of key frames and maintain temporal consistency. The processing module 122 processes the received images, specifically including image denoising, feature extraction, target recognition, and risk assessment. First, the processing module 122 preprocesses the image based on a preset denoising algorithm to remove noise components introduced by insufficient lighting, jitter, or electromagnetic interference. Next, the processing module 122 performs feature extraction to extract the texture, contour, and motion features of the target object to determine whether there are any objects within the retraction space that could cause pinching. The sending module 123 converts the processing results into control commands and transmits them to the retraction drive mechanism 130 and the alarm module 140. The sending module 123 must have low latency and high reliability communication capabilities to ensure that an anti-pinch response can be triggered immediately when a dangerous situation is detected. In one possible embodiment, the sending module 123 supports wireless communication (such as Bluetooth, Wi-Fi) and wired communication (such as CAN bus) for flexible configuration in different furniture application scenarios.

[0026] The retraction / extension drive mechanism 130 is used to execute corresponding retraction / extension actions according to the control commands issued by the processor 120. Specifically, when the processor 120 detects that there are no space-occupying objects in the retraction / extension space, the sending module 123 will issue a retraction operation command, and the retraction / extension drive mechanism 130 will then execute the retraction action; if a potential pinching risk is detected, the processor 120 will send a stop command or an unfolding command to the retraction / extension drive mechanism 130 to stop the current retraction operation or unfold the furniture component in reverse, thereby creating conditions for the removal of space-occupying objects. In one possible embodiment, the retraction / extension drive mechanism 130 includes execution units such as a motor, lead screw, and reduction gear, and supports multi-speed control so that the retraction / extension action can be executed at a lower speed in medium-risk situations, thereby reducing the pinching risk. The alarm module 140 is an intuitive feedback unit for the system to interact with external users. When a space-occupying object is detected, the alarm module 140 will execute corresponding warning operations according to different risk levels. For example, in low-risk situations, the alarm module 140 can emit a warning signal through a buzzer and LED lights to remind the user to pay attention to and remove the obstacle in time; in high-risk situations, the alarm module 140 will emit a continuous high-intensity audible and visual alarm to ensure that the user can take immediate action.

[0027] As can be seen, the aforementioned system architecture for anti-pinch folding furniture enables image acquisition, data processing, command issuance, and user interaction. On one hand, the anti-pinch detection module 110 ensures real-time detection of the folding space; on the other hand, the processor 120, through image processing and intelligent recognition, achieves rapid judgment and classification response to potential pinching risks. Combined with the coordinated execution of the folding drive mechanism 130 and the alarm module 140, the safety and reliability of the folding furniture are effectively improved in actual use. The overall architecture has high scalability and flexibility, enhancing the aesthetics of folding furniture (sofas, beds, tables, and chairs) without altering its structure.

[0028] The following is combined with Figure 2 The electronic devices in the embodiments of this application will be described. Figure 2 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application, such as... Figure 2 As shown, the electronic device 200 includes one or more digital signal processors 210, a memory 220, a communication interface 230, and one or more programs 221. The digital signal processor 210 is communicatively connected to the memory 220 and the communication interface 230 via an internal communication bus.

[0029] The digital signal processor 210 can be a central processing unit (CPU), a general-purpose processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, cells, and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc. The communication unit can be a communication interface, a transceiver, a transceiver circuit, etc., and the storage unit can be a memory.

[0030] The memory 220 can be volatile memory or non-volatile memory, or it can include both. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0031] The one or more programs 221 are stored in the memory 220 and configured to be executed by the digital signal processor 210. The one or more programs 221 include instructions for performing any step in the processor execution method embodiment of the anti-pinch retractable furniture described below.

[0032] It is understood that the electronic device 200 may include more or fewer structural elements than those shown in the above block diagram, such as a power module, physical buttons, a Wi-Fi module, a speaker, a Bluetooth module, sensors, a display module, etc., without limitation. It is understood that the electronic device may incorporate elements such as... Figure 1 The system architecture of an anti-pinch furniture unfolding system is described above.

[0033] After understanding the software and hardware architecture of this application, the following will be combined with... Figure 3 The execution logic of the processor in an anti-pinch folding furniture embodiment of this application will be described. Figure 3This is a flowchart illustrating a processor execution method for anti-pinch folding furniture according to an embodiment of this application. The processor execution method is applied to a processor for anti-pinch folding furniture. The anti-pinch folding furniture further includes a furniture fixing part, a furniture folding part, a folding drive mechanism, and an anti-pinch detection device disposed on a first end face of the furniture fixing part. The first end face is the end face of the furniture folding part facing the folding space. The folding space refers to the space enclosed by the first end face and the second end face when the anti-pinch folding furniture is in its unfolded state. The second end face refers to the inner end face of the furniture fixing part. The processor is configured to receive and respond to a retraction control command for the furniture retraction section; and send a first image acquisition command to the anti-pinch detection device. The anti-pinch detection device is used to receive and respond to the first image acquisition command; acquire a first image of the retractable space; and send the first image to the processor. The processor is configured to receive the first image and detect, based on the first image, whether a first space occupant object exists in the retractable space; If the first space occupant is detected, the image acquisition and space occupancy status detection operations are performed on the expanded and contracted space after a preset time period; and / or, a prompt message is output to remind the user to move the first space occupant. If the first space occupant is not detected, a retraction action command is sent to the retraction drive mechanism; The retraction drive mechanism is used to receive the retraction action command and start executing the retraction operation; The processor execution method for the anti-pinch retractable furniture specifically includes the following steps: Step S310: Receive and respond to the retraction control command for the furniture retraction section.

[0034] The retraction control command is a motion control request issued by the user to the anti-pinch retraction furniture. It can be input via physical buttons, a remote control, a mobile application, or a voice control system; there are no specific limitations. The retraction control command is transmitted to the processor through the anti-pinch retraction furniture's control interface and is parsed and scheduled in the processor's task queue. Since the retraction section involves the driving of motion mechanisms, the retraction control command includes not only a start signal but also call identifiers for the target motion mode, execution speed parameters, and safety monitoring parameters to ensure the diversity and safety of control during the retraction process.

[0035] Specifically, upon receiving the retraction control command, the processor first performs a conditional judgment. This judgment combines the current state information of the furniture (such as the position of the retraction section, the working status of the drive mechanism, and whether there are any space-occupying objects detected by the anti-pinch detection device) to determine whether it is permissible to immediately enter the retraction operation process. For example, if the anti-pinch detection device detects a living object in the retraction space, the processor will delay executing the command and enter a safety waiting process to avoid the risk of pinching. Furthermore, during the command response process, the processor sends a first image acquisition command to the anti-pinch detection device to obtain real-time visual information of the retraction space, ensuring that the retraction operation is not performed despite the presence of space-occupying objects. Verification using multimodal data (including images and dynamic feature information) enhances safety. Simultaneously, the processor calls the motion control module of the retraction drive mechanism, loading execution parameters that match the retraction control command. For example, under normal conditions, the drive mechanism performs the retraction operation at a standard speed; however, when a potentially low-risk object (such as a small, stationary object) is detected, it will execute at a reduced initial retraction speed until the anti-pinch detection device confirms that the risk has been eliminated.

[0036] It should be noted that the processor maintains continuous communication with the anti-pinch detection device throughout the entire execution process. If a new object is detected entering the folding space during the folding process, the current folding task will be immediately interrupted, and different response strategies will be triggered according to the risk level, such as shutdown, reverse unfolding, or audible and visual alarms. This ensures the normal operation of the furniture's folding function while minimizing the risk of injury to users or pets.

[0037] Step S320: Send a first image acquisition command to the anti-pinch detection device.

[0038] The first image acquisition command, generated by the processor, triggers the detection module of the anti-pinch detection device to initiate image acquisition within a specified time period. This command may include a start signal and settings for acquisition parameters, such as exposure time, frame rate, resolution, and gain control parameters, to adapt to imaging needs under different ambient lighting conditions. In practical applications, unfolded furniture is often located in indoor environments with insufficient or uneven lighting. Directly acquiring the original image may result in a dark image, excessive noise, and blurred edges, thus affecting subsequent object detection. Therefore, the parameters of the first image acquisition command are adaptively adjusted according to the current environmental conditions, such as increasing the exposure time or activating night vision supplementary lighting under low light conditions.

[0039] Specifically, upon receiving the first image acquisition command, the anti-pinch detection device responds and initiates the image acquisition process. The detection module of the anti-pinch detection device is typically a camera, which can be a standard camera, an infrared camera, or a night vision camera, depending on the configuration, to ensure effective image information is acquired under different lighting conditions. If a standard camera is used, after image acquisition, a preset low-light image enhancement and noise reduction algorithm is used to compensate for dark details and reduce random noise. If a night vision or infrared camera is used, a grayscale image or infrared thermal image can be directly output, ensuring the identification of space-occupying objects even in completely dark environments. After completing image acquisition, the anti-pinch detection device transmits the first image data to the processor in real time via the communication module, allowing the processor to process the first image.

[0040] For easier understanding, please refer to Figure 4 , Figure 4 This is a schematic diagram of the structure of an anti-pinch detection device provided in an embodiment of this application, as shown below. Figure 4 As shown, the anti-pinch detection device 400 includes a detection module 410 and an installation structure 420. The detection module 410 acquires image information of the spatial area where the anti-pinch furniture may pose a pinching risk during its unfolding (lifting) process. The installation structure 420 fixes or installs the detection module 410 to a preset position (fixed part) of the anti-pinch furniture to ensure that the detection module 410 can collect image information related to the pinching risk in real time during the furniture's movement. Furthermore, the anti-pinch detection device 400 also includes a communication module for establishing a communication connection between the anti-pinch detection device 400 and the anti-pinch furniture. The communication module can use wired or wireless communication. Preferably, the communication module is a wireless communication unit, such as Wi-Fi, Bluetooth, ZigBee, or other short-range wireless communication methods, to enable data interaction between the detection module 410 and the control system within the furniture. Through the communication module, the image information acquired by the detection module 410 can be transmitted in real time to the processor of the target furniture, allowing the processor to analyze and process the image to determine whether there is an obstructing object in the target area and generate a corresponding control strategy accordingly. Preferably, the detection module 410 is a camera, which can be a regular camera, a night vision camera, or an infrared camera, and is not limited thereto. The mounting structure 420 is used to fix the detection module 410 to the target installation position of the anti-pinch folding furniture. The mounting structure 420 is a magnetic structure. When the user needs to install the anti-pinch detection device 400, it can be directly fixed to the furniture surface by magnetic attraction without additional tools, and the installation operation is simple.

[0041] Step S330: Receive the first image.

[0042] The first image refers to image data acquired by the anti-pinch detection device in response to the first image acquisition command and transmitted to the processor via the communication module. This image data can be a single static image, a continuous sequence of multiple images, or a video stream transmitted in real time. Multiple image sequences and video streams can more comprehensively reflect the dynamic changes in the retraction space, ensuring the reliability of subsequent object detection and risk assessment.

[0043] When receiving the first image, the processor determines different receiving and buffering mechanisms based on the data type. For a single-frame image, it is directly written to the buffer unit, with a timestamp and acquisition parameters appended. For a multi-frame image sequence, it is stored in chronological order to form a complete frame sequence data structure, supporting optical flow analysis and motion vector calculation. For a video stream, image data is received in segments using a streaming media receiving protocol, and a circular queue is established in the buffer to ensure real-time performance and continuity. For multi-frame image sequences and video streams, the processor automatically extracts time-series features during reception, appends acquisition time information to each frame, and maintains frame continuity in the buffer. This mechanism facilitates the extraction of dynamic information through optical flow estimation algorithms or deep learning spatiotemporal feature models, thereby enabling the differentiation between biological and non-biological objects.

[0044] It should be noted that the processor is not limited to passively receiving image data, but also includes anomaly handling and data management strategies during the reception process. When transmission delay, network jitter, or lens obstruction is detected, the processor will perform frame interpolation, frame loss marking, or activate redundancy detection devices according to preset fault tolerance mechanisms to ensure the integrity and continuity of the image data link.

[0045] Step S340: Detect whether a first space occupant exists in the retractable space based on the first image; if the first space occupant is detected, wait for a preset time period and then continue to perform image acquisition and space occupancy status detection operations on the retractable space; and / or, output a prompt message to remind the user to move the first space occupant; if the first space occupant is not detected, send a retracting action command to the retractable drive mechanism.

[0046] The first space occupant refers to a target object within the unfolding space that may affect the unfolding operation, including biological objects (such as infants, children, and pets) and non-biological objects (such as clothing, toys, and tools). After receiving the first image, the processor performs preprocessing, feature extraction, and classification based on preset image processing and target detection algorithms to determine whether there are occupant objects in the unfolding space. Specifically, the preprocessing stage includes operations such as noise reduction, brightness enhancement, and edge enhancement to improve image quality; the feature extraction stage combines black-and-white and color image information and uses a fusion algorithm to extract contour features, texture features, and dynamic features; the classification stage, based on a trained target detection model, identifies whether there is a target region belonging to the occupant object category.

[0047] When a preemptive object is detected in the folding / retrieval space, the processor triggers a safety waiting mechanism. This mechanism is designed based on the real-world scenario where a user or pet might actively leave the folding / retrieval space within a short period. For example, if a pet is detected briefly entering the folding / retrieval area, the processor waits for a preset time period (e.g., 1-3 seconds) before re-performing image acquisition and status detection. This effectively avoids frequent interruptions to the folding / retrieval operation due to momentary interference. If a preemptive object is still detected after the waiting period, the processor outputs a prompt message. This message can be sent to the user via an audible and visual alarm module, a display terminal, or a mobile terminal, reminding them to remove or move the object. Furthermore, when no preemptive object is detected in the folding / retrieval space, the processor sends a folding action command to the folding / retrieval drive mechanism. This command includes not only a folding start signal but also parameters such as folding speed, drive current threshold, and safety linkage indicators, ensuring that the folding process is conducted under safety monitoring. Upon receiving the folding action command, the drive mechanism will drive the folding / retrieval unit into folding operation according to a preset execution mode until the target position is reached.

[0048] In one possible embodiment, detecting whether a first space occupant exists in the retractable space based on the first image specifically includes the following steps: 341. The first image is denoised based on a preset denoising algorithm to obtain a first denoised image; 342. Extract the black-and-white image information and color image information from the first denoised image; 343. Determine the pixel area in the black and white image information based on a preset first brightness threshold to obtain the first area; 344. Determine the pixel area in the color image information based on a preset second brightness threshold to obtain a second area; 345. Determine the first weight corresponding to the black and white image information and the second weight corresponding to the color image information based on the first area and the second area; 346. Determine the target color fusion image based on a preset hierarchical fusion algorithm, the first weight, the second weight, the black and white image information, and the color image information; 347. The target color fusion image is enhanced based on a preset edge processing algorithm to obtain a first enhanced image; 348. Perform target detection on the first enhanced image based on a preset target detection algorithm to obtain the detection result; 349. Determine whether the first space occupant exists in the retraction space based on the detection results.

[0049] The preset noise reduction algorithm can be a combination of multi-frame noise reduction and single-frame noise reduction. For multi-frame noise reduction, the processor can acquire continuous image frames within a preset time window and eliminate random noise through spatiotemporal filtering. Multi-frame noise reduction can distinguish between static noise and dynamic targets. Static noise is eliminated through inter-frame averaging, while dynamic targets are preserved between frames, thus avoiding loss of detail. Single-frame noise reduction can be a combination of Gaussian filtering and bilateral filtering. Gaussian filtering is used to smooth low-frequency noise, while bilateral filtering can preserve edge and texture features, preventing image blurring. Since the anti-pinch detection device is located on the first end face (close to the retractable space), there may be insufficient lighting or interference from shadows, and the original image is prone to graininess or noise accumulation. If not processed, the accuracy of subsequent feature extraction and target recognition will be greatly reduced. Through multi-level noise reduction, more than 90% of random noise can be eliminated while ensuring clear details, thus obtaining a reliable first-level denoised image.

[0050] In this system, black and white image information is primarily used to reflect global brightness features and edge contours, while color image information is used to identify the color and texture features of objects. The processor separates and stores the brightness and color channels of the image separately. This separation ensures that edge detection can still be performed using black and white image information in low-light environments, while combining color information allows for further differentiation of objects such as clothing, hair, and skin, ensuring robustness of the detection. A first brightness threshold is used to distinguish foreground targets from background areas in the black and white image. For example, if the brightness threshold is set to 20 cd / m², areas below this threshold are considered background, and areas above this threshold are considered potential placeholders. The processor calculates the pixel area of ​​the target area to obtain a first area parameter, which directly reflects the spatial size of potential placeholders in the image, and can initially eliminate some small noise or irrelevant interfering objects. A second brightness threshold is used for color image segmentation, usually set in conjunction with the distribution characteristics of the color histogram. For example, for dark areas, the weights depend more on the black and white image; for bright areas, they depend more on the color image. The calculated second area is used to represent the range of possible target areas in the color channel. Combining the first and second areas avoids misjudgments caused by a single threshold. For example, in dim environments where color images fail, compensation can be made using the area of ​​the black and white image. In bright light, color information can effectively restore the true boundaries of the target. The processor automatically calculates the relative proportion of the two areas based on the brightness distribution. When the proportion of black and white area is high, the black and white image is given a higher weight (e.g., 0.8-0.9), and when the proportion of color area is high, the color image is given a higher weight (e.g., 0.6-0.7). The advantage of adaptive weighting is that it ensures that the fused image retains both brightness details and color textures, achieving a fusion effect with strong adaptability to lighting conditions.

[0051] The hierarchical fusion algorithm involves fusing low-frequency and high-frequency layers. The low-frequency layer primarily relies on global brightness information provided by the black-and-white image, while the high-frequency layer overlays texture and edge features from the color image. The resulting target color fusion image combines brightness accuracy with detail integrity, making it suitable for subsequent feature recognition. The hierarchical fusion algorithm can utilize deep learning models (such as the EEMEFN edge enhancement network) to improve the fusion effect, especially in low light or complex backgrounds, preserving key features such as facial contours and hair details. Edge processing algorithms can include Canny edge detection, Unsharp Mask sharpening, or deep edge extraction methods based on convolutional neural networks. The enhanced first image highlights object boundaries and texture features, making it easier for subsequent object detection models to identify spatially occupying objects. Object detection algorithms can be traditional feature classification methods, such as HOG+SVM, or deep learning algorithms, such as YOLO and Faster R-CNN. The processor inputs the enhanced image into the object detection model to obtain possible target bounding boxes and their category labels. The detection results include not only the presence or absence of the target but also its category (human, animal, or inanimate object) and location coordinates. If the detection result shows that there is a target bounding box, and the target category is a potential risk object (such as an infant, pet, or clothing), then it is determined that there is a first space occupant in the storage space; if the detection result is empty or only a background area exists, then the space is determined to be in an idle state.

[0052] In one possible embodiment, the step of performing target detection on the first enhanced image based on a preset target detection algorithm to obtain a detection result specifically includes the following steps: 3481. A preset feature extraction algorithm is used to extract features from the first enhanced image to obtain multiple first features; 3482. Extract the features of the eyes, the features of the texture, and the features of the body contour from the plurality of first features respectively to obtain the features of the eyes, the features of the texture, and the features of the body contour; 3483. Determine the pupil category among the aforementioned eye features; 3484. Determine the texture category in the texture features; 3485. Determine the morphological category in the body contour features; 3486. Determine the detection result based on the target detection algorithm, the pupil category, the texture category, and the morphology category.

[0053] The preset feature extraction algorithm can be a traditional image processing algorithm (such as SIFT, HOG, LBP, etc.) or a deep learning-based convolutional feature extraction method (such as the intermediate layer feature vectors of CNN). By performing multi-scale convolution, gradient calculation, and edge analysis on the first enhanced image, several first features reflecting the image content are obtained. These features include local texture, edge contours, brightness contrast, and spatial distribution. Specifically, eye features are used to identify the presence of humans or animals, and their parameters include pupil shape, bright spot reflection, and iris contour. Texture features are used to identify the material properties of the object's surface, such as skin, hair, and clothing. Body contour features are used to identify the overall geometric shape of the object, such as human, animal, or irregular shapes. Pupil categories can include round pupils and non-round pupils. Round pupils typically correspond to humans, while non-round pupils (such as vertical ovals or irregular shapes) typically correspond to pets such as cats and dogs. Through geometric feature analysis of the eye region, the processor can determine the attributes of the occupier object based on the pupil category.

[0054] To ensure accuracy, the system also incorporates secondary verification based on the reflective characteristics of the eyeball (i.e., the brightness parameter of the light spot). A stable circular outline with low light spot brightness is identified as a human pupil; a non-circular outline or abnormal reflection is identified as a pet pupil. Texture categories include skin texture, hair texture, and clothing texture. Skin texture typically appears as a smooth structure with fine pores, hair texture consists of numerous high-frequency fine lines, and clothing texture appears as regularly repeating lines or fibrous structures. Local image regions are statistically analyzed and modeled using a gray-level co-occurrence matrix (GLCM) or a texture convolution kernel from a deep convolutional neural network to determine their corresponding texture categories. Morphological categories primarily include human and animal forms. Human forms have relatively straight edges (such as the head and limb outlines), while animal forms typically exhibit irregular outlines or have hair extending outwards.

[0055] Specifically, the processor uses edge detection algorithms combined with geometric fitting methods to identify the overall outline of an object, and then determines its category based on the shape and proportion of the outline. For example, when the aspect ratio of the outline is within a specific range (such as the torso proportion of a child), it is classified as a human form; when the outline edges show a hair-like jitter, it is classified as an animal form. The object detection algorithm integrates the classification results of the three types of features to make a final judgment on the detected object. If eye features are present and the pupil category is round, combined with skin or clothing texture and human-shaped outline, it is classified as a human; if the pupil category is non-circular, combined with hair texture and animal-shaped outline, it is classified as a pet; if no eye features are detected and the texture category is clothing texture but without obvious morphological features, it is classified as a non-biological object. This fusion judgment process can be implemented through rule-based decision trees or through a trained multimodal classification model (such as ensemble random forests or neural networks) to improve the accuracy of object recognition in complex scenes.

[0056] In one possible embodiment, the pupil category includes circular pupils and non-circular pupils; the texture category includes skin texture, hair texture, and clothing texture; the morphology category includes human morphology and animal morphology. Determining the detection result based on the target detection algorithm, the pupil category, the texture category, and the morphology category specifically includes the following steps: 34861. Determine the ocular reflection features corresponding to the pupil category based on the plurality of first features; 34862. The target detection algorithm is used to determine the brightness parameters of the eyeball spot based on the eyeball reflection characteristics; 34863. If the brightness parameter of the eyeball spot is greater than the preset third brightness threshold, and the pupil category is the non-circular pupil, and the morphology category is the animal morphology, and the texture category is the fur texture, then the detection result is determined to be a pet; 34864. If the brightness parameter of the eyeball spot is less than or equal to the third brightness threshold, and the pupil category is the circular pupil, and the shape category is the human shape, and the texture category is the skin texture, then the detection result is determined to be an infant; 34865. If the brightness parameter of the eyeball spot is less than or equal to the third brightness threshold, and the pupil category is the circular pupil, and the shape category is the human shape, and the texture category is the clothing texture, then the detection result is determined to be a child. 34866. If the texture category is the clothing texture, and the pupil category does not exist, and the morphology category does not exist, then the detection result is determined to be non-biological.

[0057] Among them, the ocular reflectivity feature is a key indicator extracted by detecting the brightness, shape, and distribution of light spots in the eye region. Due to the curved surface of the eyeball, it produces significant specular reflection under different lighting conditions. The processor extracts the number, area, and intensity of reflective spots by calculating grayscale gradients and analyzing brightness peaks of pixels in the eye region, and uses these as parameters for ocular reflectivity features. The extraction of ocular reflectivity features can effectively distinguish between living and non-living objects. For example, while non-living surfaces such as glass and metal have strong reflectivity, their shape is stable and does not change with the pupil outline, whereas ocular reflectivity is closely related to the pupil position, thus serving as an effective feature for identifying living organisms. The light spot brightness parameter is a quantitative indicator obtained by normalizing the brightness values ​​of the reflective area, used to measure the intensity of the ocular light spot. Specifically, the average grayscale value of the light spot area is calculated, and contrast correction is performed by combining it with the surrounding background brightness to avoid interference from ambient light. If the brightness parameter of the light spot exceeds a certain threshold, which is set to 128 (255 / 2≈128), it indicates that the eye reflection is strong, which usually corresponds to the high reflectivity of animal eyes in low light environments; if the brightness of the light spot is weak, it is more in line with the reflection characteristics of human eyes.

[0058] Specifically, the logic for identifying a pet based on detection results utilizes prior information, namely, typical characteristics of animal eyes: non-circular pupils (such as the vertical pupils of felines), high reflectivity, fur texture, and irregular animal body contours. The purpose of using prior information is to improve the interpretability of the judgment. Specifically, this judgment is mainly applied to common household pets such as cats and dogs. When they enter the storage space, it can promptly identify them and trigger anti-pinch strategies to prevent injury. The logic for identifying an infant based on detection results combines typical visual characteristics of infants: round pupils, low-reflectivity eyes, human-shaped contours, and skin texture. Because infants are small and have a large exposed skin area, the matching degree between skin texture and contour shape can be used to accurately identify the object as an infant. The logic for identifying a child based on detection results is applicable to children wearing clothing. When the pupil detection shows a round shape and the body shape conforms to a human shape, but the texture detection result shows clothing texture, further analysis of body proportions (such as head-to-body ratio and height range) can determine that it is a child, not an adult. The logic for determining whether a detection result is non-biological is used to exclude interference from non-biological objects. For example, clothing, toys, or plush items may visually present a similar texture to clothing, but because they lack pupils and overall morphological features, they will not be identified as living things. This determination mechanism can significantly reduce the false alarm rate and avoid frequent interruptions to the unfolding and folding operations due to the detection of harmless objects, thereby ensuring the normal user experience of the furniture.

[0059] In one possible embodiment, after the retraction operation is initiated, the processor of the anti-pinch retraction furniture executes a method that further includes the following steps: A1. Obtain a second image of the retractable space using an anti-pinch detection device; A2. Detect whether there is a second space occupant in the expansion / contraction space based on the second image; A3. If no second space placeholder is detected, continue to perform the collapse operation until the current collapse task is completed; A4. If the presence of the second space occupant is detected, the retraction drive mechanism is controlled to stop performing the retraction operation, and / or the retraction drive mechanism is controlled to perform the unfolding operation so that the second space occupant can move to a space outside the retraction space.

[0060] The second image refers to the dynamic visual information acquired by the anti-pinch detection device during the retraction process of the retractable component. The second image can be a single-frame image sequence or a video stream; no limitation is made here. When acquiring the second image, the anti-pinch detection device automatically performs ambient brightness compensation and focus adjustment to ensure image clarity and brightness stability, thus providing reliable input for subsequent target detection. The logic for determining the second spatial occupant is similar to that of the first spatial occupant, but with greater emphasis on real-time performance and dynamism. Since the retractable space continuously shrinks during the retraction process, there is a significant risk of pinching if a new biological or non-biological object enters the space. Therefore, the processor performs rapid detection based on the second image, which can employ lightweight target detection algorithms (such as models based on YOLO-Lite, MobileNet, or adaptive threshold segmentation) to minimize detection latency, thereby achieving near real-time spatial occupant recognition.

[0061] Specifically, the detection steps include: image preprocessing (noise reduction, brightness enhancement), feature extraction (texture, edge, pupil features), classification and recognition (biological / non-biological), and outputting the detection confidence score. If the detection confidence score is higher than a preset threshold (e.g., 90%), it is directly determined that a second space occupier exists; if the confidence score is in the ambiguous range (e.g., 60%–90%), multiple frames of images are acquired and the recognition accuracy is improved through a temporal fusion algorithm. If the detection results during the retraction operation are consistently negative for any occupier, the processor controls the drive mechanism to retract the object at a predetermined speed and stroke, and then enters a standby state. Upon receiving a detection result indicating the presence of an occupier, the processor issues an emergency stop command to the retraction drive mechanism, causing it to stop within the shortest braking distance to avoid clamping the object. The unfolding operation is suitable for situations where the object is still in a dangerous area and has not moved on its own. In this case, the processor issues an unfolding command to restore the retraction space to a larger state, providing sufficient space for the object to move. For example, when the object of the test is an infant or pet, the unfolding operation can ensure that they are safely removed from the danger zone, thereby minimizing the risk of pinching injury.

[0062] In one possible embodiment, the second space occupant includes non-biological and biological objects. The steps of controlling the retraction / expansion drive mechanism to stop performing the retraction operation and / or controlling the retraction / expansion drive mechanism to perform the expansion operation specifically include: 41. If the second space occupant is the non-biological object, then control the retraction drive mechanism to stop performing the retraction operation and output information about the non-biological object to prompt the user to perform the corresponding action. 42. If the second space occupant is the biological object, then the dynamic characteristics of the biological object are determined based on the second image; 43. Determine the target location information and target motion information of the biological object based on the dynamic characteristics; 44. Determine the risk level corresponding to the biological object based on the target motion information and the target location information; 45. Determine the response strategy corresponding to the risk level based on the preset mapping relationship between the risk level and the preset response strategy; 46. ​​Based on the response strategy, control the retraction / extension drive mechanism to stop performing the retraction operation, and / or control the retraction / extension drive mechanism to perform the extension operation.

[0063] Non-biological objects typically include clothing, toys, books, daily necessities, or tools. These objects do not move on their own and therefore do not require unfolding. To avoid damage or obstruction, the processor will prioritize stopping the unfolding mechanism upon identifying such objects, ensuring that the unfolding components do not mechanically interfere with the objects. Prompt messages can be sent via audible and visual alarms, furniture control panels, or mobile apps to remind users to remove non-biological objects. For example, when clothing is detected remaining in the unfolding path, the system can issue a voice prompt, "Please remove the obstacle," thereby enhancing the user experience.

[0064] Biological objects primarily include infants, children, and pets. Dynamic features refer to the object's displacement trajectory, posture changes, and acceleration parameters exhibited in consecutive image frames. The processor extracts the object's motion speed, direction, and behavior patterns by performing optical flow analysis and keypoint tracking on a multi-frame sequence of the second image. The extraction of dynamic features not only determines whether the biological object is stationary but also predicts its potential movement trends. For example, when an infant is crawling, its dynamic features exhibit low-speed, random crawling; pets such as cats and dogs exhibit high-speed trajectories with sudden changes in direction. Target location information includes the biological object's position coordinates relative to the boundary of the enclosure space, and target motion information includes its direction of movement, speed, and acceleration. Specifically, the combination of target location information and motion information can be used to predict in real time whether an object is about to enter a danger zone. For example, when a child is detected at the edge of the enclosure space and moving towards the center, the system will prioritize raising the risk level to ensure more proactive protective actions are taken. The risk levels are divided into three categories: low, medium, and high. Low risk: the object is far from the folding / unfolding space or is stationary. Medium risk: the object is near the edge of the folding / unfolding space but moving slowly, or is within the space but shows a clear tendency to move away. High risk: the object is located in the core area of ​​the folding / unfolding space and continues to move towards the folding / unfolding components, posing a direct danger of being trapped. Each risk level corresponds to a specific response strategy: In low risk, the processor sends a message to the folding / unfolding drive mechanism to continue the folding operation while maintaining monitoring; in medium risk, the processor sends a message to the folding / unfolding drive mechanism to immediately stop the folding operation and issue a warning message to alert the user; in high risk, the processor sends a message to the folding / unfolding drive mechanism to not only immediately stop the folding operation but also execute the unfolding operation to ensure the biological object quickly escapes the danger zone.

[0065] In one possible embodiment, determining the dynamic characteristics of the biological object based on the second image specifically includes the following steps: 421. Extract multiple consecutive images from the second image to obtain a continuous image sequence; 422. Extract the image corresponding to the biological object from the continuous image sequence to obtain the target image sequence; 423. Based on a preset optical flow estimation algorithm, perform vector calculations on each image in the target image sequence to obtain a motion vector parameter sequence; 424. Determine the motion trajectory of the biological object based on the motion vector parameter sequence to obtain the first motion curve of the biological object; 425. Calculate the dynamic texture features of each target image in the target image sequence to obtain a texture feature set; 426. Input the texture feature set into a preset spatiotemporal feature model to obtain multiple spatiotemporal dynamic parameters; 427. Determine the first average value, first maximum value, and first minimum value of the plurality of spatiotemporal dynamic parameters; 428. Determine the average value of the first motion curve to obtain the second average value; 429. Determine the difference between the first average value and the second average value to obtain the first difference value; The adjustment value corresponding to the first difference is determined based on the preset mapping relationship between the first difference and the adjustment value, and the target adjustment value is obtained. 4210. Fine-tune the first motion curve according to the target adjustment value to obtain the target motion curve; 4211. Determine the dynamic characteristics of the biological object based on the target motion curve.

[0066] The estimation algorithm performs vector calculations on each second image in the target image sequence to obtain a sequence of motion vector parameters. The optical flow estimation algorithm can use sparse optical flow (Lucas-Kanade) to calculate keypoint displacements or dense optical flow (Farneback) to calculate global pixel motion. The motion vector parameters include horizontal velocity components, vertical velocity components, and displacement direction angles, used to describe the motion state of the biological object on the image plane. The processor performs optical flow calculations on each frame of the target image, obtaining a set of vectors, which are then arranged according to time sequence to form a sequence of motion vector parameters for trajectory fitting. The motion trajectory is represented by a two-dimensional coordinate sequence, where the horizontal and vertical coordinates correspond to the relative position of the target in the expansion and contraction space, and t is the timestamp. The first motion curve is smoothed using a Kalman filter or Bayesian filter algorithm to reduce trajectory jitter caused by noise points in a single frame. Dynamic texture features include local binary mode, gray-level co-occurrence matrix features, and frequency domain energy distribution, used to describe the changes in surface texture over time. Among them, baby skin has a smooth texture and little variation under light; pet fur shows high-frequency local texture jumps; and clothing surfaces may show regular texture lines.

[0067] Specifically, the processor calculates texture feature vectors for each frame of the target image and combines them on the time axis to form a texture feature set, providing input for spatiotemporal feature modeling. The spatiotemporal dynamic parameters include multi-dimensional indicators such as velocity change rate, direction shift frequency, and texture fluctuation amplitude, used to comprehensively reflect the dynamic behavior pattern of the object. If the object is an infant, the spatiotemporal dynamic parameters exhibit low-speed, low-frequency, and highly random fluctuations; if the object is a pet, they exhibit high-speed, sudden direction switching, with significant differences in parameter value distribution. The first average value represents the stability of the overall motion trend, the first maximum value reflects instantaneous extreme actions (such as rapid jumps), and the first minimum value represents a near-static state. If the difference between the maximum and minimum values ​​is too large, it indicates that the object exhibits sudden movement behavior, requiring increased weighting in risk assessment. The second average value is the overall average velocity or displacement trend parameter of the motion trajectory curve. If the second average value is close to zero, it indicates that the object is mostly stationary or moving slowly; if its value is large, it indicates that the object is continuously moving rapidly. Then, the second average value is compared with the first average value to further evaluate the stability of the trajectory. If the difference is too large, it indicates that the object's movement exhibits strong non-linear behavior (such as sudden collision), and the risk level should be increased. If the difference is close to zero, it indicates that the trajectory and dynamic parameters are highly consistent, and the movement pattern is relatively stable. The mapping relationship can be established using large-scale experimental data. For example, the movement difference threshold is 0.2 for infants, 0.35 for children, and 0.5 for pets. When the threshold is exceeded, curve correction is required. The target adjustment value is a correction factor used to fine-tune the smoothness and trend parameters of the movement curve, preventing misjudgment caused by a single extreme movement. The adjustment value can be obtained through table lookup, linear interpolation, or neural network prediction to flexibly adapt to different objects. Fine-tuning methods include weighted smoothing, curve fitting correction, and outlier removal to ensure that the trajectory more realistically reflects the object's movement trend. The final target movement curve will serve as the dynamic feature output of the biological object, including average speed, directional stability, trajectory smoothness, and sudden movement indicators.

[0068] In one possible embodiment, determining the risk level of the biological object based on the target motion information and the target location information specifically includes the following steps: 441. Obtain the first position information of the easily pinched position of the anti-pinch folding furniture; 442. Determine the target distance based on the target location information and the first location information; 443. Determine the movement speed of the biological object based on the target motion information; 444. If the movement speed is less than or equal to a preset first speed threshold, and the target distance is less than or equal to a preset first distance threshold, then the risk level is determined to be low risk. 445. If the movement speed is between the first speed threshold and the preset second speed threshold, and the target distance is between the first distance threshold and the preset second distance threshold, then the risk level is determined to be medium risk. 446. If the movement speed is greater than or equal to the second speed threshold and the target distance is greater than or equal to the second distance threshold, then the risk level is determined to be high risk.

[0069] The "easily pinched location" refers to the clamping area formed between the retractable and fixed parts of the furniture during the retraction process. Geometrically, this area corresponds to the edge of the retraction space or the "joint movement" point. The primary location information is typically represented in three-dimensional coordinates, with the reference point being the inner end face of the fixed part. By pre-setting this location as a high-risk sensitive area, objects near this area can be prioritized during detection. The target location information is the biological object's position coordinates obtained through the aforementioned dynamic feature extraction steps, representing the object's relative position within the retraction space. The target distance is calculated by measuring the Euclidean distance between the object's coordinates and the easily pinched location coordinates. Movement speed reflects both whether the object is rapidly approaching the retractable part and its movement trend (approaching / moving away from the danger zone). If the velocity direction vector is consistent with the direction of the clamping area, the risk level needs to be increased.

[0070] Specifically, the processor uses a sliding window method to smooth the speed calculation, eliminating the impact of occasional frame jitter. For example, it initializes the average speed over 0.5 seconds as the current speed to avoid incorrect risk assessments due to single-frame errors. The first speed threshold is a preset lower speed limit, which can be set within the range of 0.1m / s–0.3m / s, suitable for situations where an infant is crawling slowly or a pet is stationary. The first distance threshold is the minimum safe range near the clamping point (e.g., 10cm). When the object's speed is low and it has entered this range, it is still considered low-risk because it will not immediately pose a clamping hazard. The second speed threshold is set within the range of 0.5m / s–0.8m / s, similar to the speed of a child walking or a pet jogging. The second distance threshold is set to 20–30cm, indicating that the object is close to the clamping point but still within an interventionable buffer space. When the object is in this risk zone, the processor will interrupt the retraction operation and trigger an audible and visual alarm to alert the user. For example, if a child approaches the edge of furniture at a moderate speed, the processor will immediately stop retraction and issue a warning. A high-risk status indicates that the object is approaching the clamping point at a high speed, posing an immediate risk of being clamped. Emergency braking and deployment must be triggered immediately. The risk assessment employs a dual-threshold mechanism combining speed and distance to ensure that even if the object has not yet entered the clamping point, potential danger due to high-speed movement can be addressed in advance.

[0071] In one possible embodiment, the anti-pinch furniture further includes an early warning module. The response strategy includes a low-risk response strategy, a medium-risk response strategy, and a high-risk response strategy. The step of generating and controlling the retraction drive mechanism to stop performing the retraction operation, and / or controlling the retraction drive mechanism to perform the unfolding operation, based on the response strategy, specifically includes the following steps: 461. If the response strategy is the low-risk response strategy, then control the retraction drive mechanism to perform the retraction operation, and control the early warning module to perform the audible and visual early warning operation; 462. If the response strategy is the medium-risk response strategy, then control the retraction drive mechanism to perform the retraction operation at a preset first retraction speed; 463. If the response strategy is the high-risk response strategy, then control the retraction drive mechanism to stop performing the retraction operation; after the retraction drive mechanism stops performing the retraction operation, control the retraction drive mechanism to perform the unfolding operation at a preset first unfolding speed.

[0072] The warning module may include a buzzer, LED indicator, miniature projection light spot device, or an integrated voice broadcast unit. The audio-visual warning operation includes emitting a continuous low-intensity sound (e.g., a buzzer less than 50dB) and flashing indicator light at a frequency of 1–2Hz to attract the user's attention without causing noise disturbance. Specifically, when a low-risk object is detected, the processor will simultaneously activate the audio-visual warning, alerting the user that "a low-speed object is in the unfolding area; please be careful." In this way, the normal function of the furniture is maintained, while also allowing the user to intervene promptly when necessary to remove any potential obstacles.

[0073] The medium-risk response strategy primarily addresses scenarios where a child slowly approaches the clamping point or a pet moves at a moderate speed. While there is some danger in these situations, the object's speed and position are not yet high enough to trigger an immediate stop. To balance safety and efficiency, the processor reduces the retraction speed. The first retraction speed is a speed-limiting mode relative to the standard speed; for example, if the standard speed is 10 cm / s, the first retraction speed can be set to 3–5 cm / s. By reducing the speed of the drive mechanism, the buffer time for the user or object to react can be significantly extended. Specifically, when the processor determines a medium-risk scenario, it issues a speed-limiting command, and the drive mechanism enters a slow-speed operation state while maintaining real-time monitoring of the retraction space. If the risk level increases during deceleration (e.g., a child suddenly accelerates into the clamping point), the processor immediately interrupts the retraction operation and escalates to a high-risk response; if the risk level decreases, normal speed can be resumed. The high-risk response strategy is used for scenarios where a biological object rapidly enters or is about to enter the clamping area. For example, a pet suddenly rushes into the retraction space, or a child approaches the edge of furniture at near-running speed. In these situations, the processor must immediately take maximum protective measures. The command to stop the retraction operation is an emergency stop command. The processor ensures that the retraction components completely stop operating in the shortest possible time by cutting off the drive current, activating the braking unit, and locking the current position. After the emergency stop, the processor will issue an unfolding command, causing the retraction components to move in the opposite direction at a preset first unfolding speed, actively expanding the retraction space and providing an escape path for the living object. For example, when a pet's tail or paw is detected near a clamping point, the unfolding operation can quickly release space and prevent injury. The first unfolding speed is typically 5–8 cm / s, which can quickly create space without causing mechanical impact due to excessive speed.

[0074] For easier understanding, please refer to Figure 5 , Figure 5 This is a flowchart illustrating an anti-pinch control method for anti-pinch folding furniture provided in an embodiment of this application. The anti-pinch control method is applied to a processor of the anti-pinch folding furniture. The anti-pinch folding furniture includes a furniture fixing part, a furniture folding part, a processor, a folding drive mechanism, and an anti-pinch detection device disposed on the first end face of the furniture folding part. The first end face is the end face of the furniture fixing part facing the folding space. The folding space refers to the space enclosed by the first end face and the second end face when the anti-pinch folding furniture is in its unfolded state. The second end face refers to the inner end face of the furniture folding part. The anti-pinch control method specifically includes the following steps: S510, control the anti-pinch detection device to receive and respond to the first image acquisition command; acquire the first image of the retraction space; and send the first image to the processor; S520, control the processor to receive the first image, and detect whether there is a space occupant in the retractable space based on the first image; if the space occupant is detected, wait for a preset first time period and then continue to perform image acquisition and space occupancy status detection operations on the retractable space; and / or, output a prompt message to remind the user to move the space occupant; if the space occupant is not detected, send a retracting action command to the retractable drive mechanism; S530, control the retraction drive mechanism to receive the retraction action command and execute the retraction operation.

[0075] Specifically, space-occupying objects include both biological and non-biological objects. During the detection process, the processor can not only identify the presence of space-occupying objects but also classify their types. When the detection result indicates that the space-occupying object is a biological object (such as an infant or pet), the processor will prioritize outputting a prompt message and sending a pause operation to the retraction drive mechanism to avoid potential pinching risks. When the space-occupying object is a non-biological object (such as a toy or miscellaneous items), the processor outputs an audible and visual prompt to remind the user to manually remove the obstacle. Thus, by combining multi-level detection with a graded response strategy, the anti-pinch control method proposed in this application can significantly improve the safety of retractable furniture during use.

[0076] For easier understanding, please refer to Figure 6 , Figure 6 This is a functional interaction diagram of an anti-pinch detection method provided in an embodiment of this application. As can be seen, the anti-pinch detection method involves functional interaction between the anti-pinch detection device, the processor, the retraction / extension drive mechanism, and the alarm module. Regarding user interaction, the user sends a retraction / extension command to the anti-pinch retraction furniture, and the anti-pinch detection device in the furniture responds to the image acquisition command issued by the processor. Simultaneously, to reduce the energy consumption of the anti-pinch furniture, the retraction / extension furniture can also provide low-power management, i.e., entering a sleep / wake-up state. When the user does not need to retract or extend the furniture, it enters a sleep state to reduce power consumption.

[0077] Specifically, regarding the anti-pinch detection device and processor processing, the anti-pinch detection device monitors the furniture's unfolding space in real time by capturing images through a camera and sends the captured image information to the processor via a communication module. The processor preprocesses the images captured by the camera, including image noise reduction and feature extraction. Then, based on the preprocessed images, the processor detects occupants in the unfolding space and determines the motion commands for the unfolding drive mechanism based on these occupants. That is, the processor generates relevant commands to control the unfolding drive mechanism based on the occupant's position, and then performs other operations based on the occupant's status. For example, when an object is detected in the image, the processor further determines whether the object is an occupant and whether it is a living organism (such as an infant, child, or pet) or a non-living object (such as a toy, clothing, or miscellaneous items). In terms of unfolding discrimination, the response strategy of the unfolding drive mechanism is determined based on the occupant's information, and motion commands for the unfolding drive mechanism are generated. Regarding the movement of the retraction / extension drive mechanism, it receives motion commands from the processor and then controls the mechanism to execute corresponding retraction / extension commands, including commands to retract, stop, or extend. Simultaneously, for early warning and alerting, the alarm module triggers corresponding audible and visual prompts and plays relevant alarm information. Furthermore, the alarm module also implements interactive feedback functionality for the anti-pinch detection method. After receiving prompts from the processor, the alarm module can execute audible and visual alarm operations based on different risk levels. For example, in low-risk situations, the alarm module alerts the user through intermittent buzzer beeping and flashing indicator lights.

[0078] visible, Figure 6 The functional interaction diagram of an anti-pinch detection method clearly illustrates the interaction logic between the anti-pinch detection device, processor, retraction / extension drive mechanism, and alarm module. By combining image acquisition, data processing, command issuance, and prompt feedback, it can not only detect potential clamping objects in time before the retraction / extension action is executed, but also quickly take corresponding measures after the risk is detected, thereby effectively ensuring the safety performance of users when using anti-pinch retraction furniture.

[0079] For easier understanding, please refer to Figure 7 , Figure 7This is a structural schematic diagram of an anti-pinch folding furniture provided in an embodiment of this application. As can be seen, the anti-pinch folding furniture 700 mainly includes a foldable electric leg rest (folding section), a camera (anti-pinch detection device), a vulnerable-to-pinch area (folding space), and a reclining chair body (fixing section). Specifically, the electric leg rest of the anti-pinch folding furniture 700 unfolds and folds via a folding drive mechanism (motor drive), and can flexibly adjust the angle and height according to user needs, thereby meeting the comfortable sitting and reclining needs of users of different body types. During its unfolding process, due to the relative displacement between the leg rest and the main body of the furniture, a potentially narrow space area is inevitably formed, which is the vulnerable-to-pinch area. If children, pets, or other objects occupy this area during the unfolding process, a pinching risk is highly likely. Therefore, an anti-pinch detection mechanism is set in the vulnerable-to-pinch area to ensure the safety of the furniture during automatic unfolding. In this structure, the camera, as part of the anti-pinch detection device, is installed on the lower side of the foldable electric leg rest, and its position corresponds precisely to the vulnerable-to-pinch area. The camera can move and adjust its angle synchronously with the movement of the leg rest, enabling real-time acquisition of image information about the vulnerable area during different stages of the leg rest's unfolding and retraction. The camera can be configured as a standard visible light camera, a night vision camera, or an infrared camera to adapt to detection tasks under different lighting conditions. The acquired image information is transmitted to the processor for subsequent recognition and analysis to determine whether an object is present in the vulnerable area. The vulnerable area is located between the folding edge of the electric leg rest and the main body of the furniture, and is the part of the leg rest's movement trajectory most prone to trapping external objects. In normal use scenarios, if a user's leg, a child's finger, or a pet enters this area while the leg rest is unfolding, there is a high risk of injury. By placing a camera in this area to acquire image information and combining it with image processing algorithms for real-time detection, potential trapping objects can be detected in a timely manner, effectively preventing accidents. Furthermore, the image data acquired by the camera is processed by the image processing module in the processor for noise reduction, feature extraction, and target detection. For example, based on preset brightness thresholds and texture features, the system can identify the presence of skin texture, hair texture, or clothing texture within the vulnerable area to distinguish between biological and non-biological objects. When a biological object is detected, the anti-pinch furniture will immediately control the folding drive mechanism to stop the current folding action and trigger the alarm module to issue an audible and visual alert, reminding the user to handle the situation promptly. When the detected object is a non-biological object (such as a toy or miscellaneous items), the system can prompt the user to manually remove the obstacle via audible and visual alerts to ensure a smooth folding process. Furthermore, to improve the stability of anti-pinch detection, the camera can also combine multi-frame image analysis or video stream detection to continuously monitor the vulnerable area. By extracting the motion vector features of the object through optical flow estimation algorithms, it can further determine whether the target object is in a dynamic state and classify the risk level accordingly.

[0080] As can be seen, by executing the processor execution method for anti-pinch folding furniture provided in this application embodiment, the anti-pinch folding furniture can accurately detect and identify the space-occupying object in the folding space without contacting or squeezing it. The anti-pinch folding furniture can control the folding part to stop the folding action before contacting or squeezing the space-occupying object. This improves the safety of the anti-pinch folding furniture. Furthermore, since the anti-pinch detection mechanism detects the space-occupying object in the folding space, there is no need to design redundant structures to prevent the folding furniture from being pinched, thus improving the aesthetics of the anti-pinch furniture.

[0081] The above primarily describes the solutions of the embodiments of this application from the perspective of the method execution process. It is understood that, in order to achieve the above functions, the electronic device includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments provided herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0082] This application embodiment can divide the electronic device into functional units according to the above method example. For example, each function can be divided into a separate functional unit, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the unit division in this application embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0083] When dividing each function into modules according to its corresponding function. Figure 8 This application provides a functional block diagram of an anti-pinch control device 800, which includes: The acquisition unit 810 is configured to control the anti-pinch detection device to receive and respond to the first image acquisition command; acquire a first image of the retraction space; and send the first image to the processor. The control unit 820 is configured to control the processor to receive the first image, and detect whether there is a space occupant in the retractable space based on the first image; if the space occupant is detected, the processor waits for a preset first time period and then continues to perform image acquisition and space occupancy status detection operations on the retractable space; and / or outputs a prompt message to remind the user to move the space occupant; if the space occupant is not detected, the control unit 820 sends a retracting action command to the retractable drive mechanism. The execution unit 830 is used to control the retraction drive mechanism to receive the retraction action command and execute the retraction operation.

[0084] In one possible embodiment, the control unit 820, in detecting whether a first space occupant exists in the retractable space based on the first image, is specifically configured to: The first image is denoised based on a preset denoising algorithm to obtain a first denoised image; Extract black-and-white and color image information from the first denoised image; The pixel area in the black and white image information is determined based on a preset first brightness threshold to obtain the first area; The pixel area in the color image information is determined based on a preset second brightness threshold to obtain the second area; The first weight corresponding to the black and white image information and the second weight corresponding to the color image information are determined based on the first area and the second area. The target color fusion image is determined based on a preset hierarchical fusion algorithm, the first weight, the second weight, the black and white image information, and the color image information; The target color fusion image is enhanced based on a preset edge processing algorithm to obtain a first enhanced image; The first enhanced image is subjected to target detection based on a preset target detection algorithm to obtain the detection result; Based on the detection results, it is determined whether the first space occupant exists in the retraction space.

[0085] In one possible embodiment, the control unit 820, in the process of performing target detection on the first enhanced image based on a preset target detection algorithm to obtain a detection result, is specifically configured to: A preset feature extraction algorithm is used to extract features from the first enhanced image to obtain multiple first features; The features of the eyes, the features of the texture, and the features of the body contour are extracted from the plurality of first features respectively to obtain the eye features, texture features, and body contour features; Determine the pupil category among the aforementioned eye features; Determine the texture category in the texture features; Determine the morphological category in the body contour features; The detection result is determined based on the target detection algorithm, the pupil category, the texture category, and the morphology category.

[0086] In one possible embodiment, the pupil category includes circular pupils and non-circular pupils; the texture category includes skin texture, hair texture, and clothing texture; the morphology category includes human morphology and animal morphology. The control unit 820, in determining the detection result based on the target detection algorithm, the pupil category, the texture category, and the morphology category, is specifically used for: Based on the plurality of first features, the ocular reflective features corresponding to the pupil category are determined; The target detection algorithm is used to determine the brightness parameters of the eyeball spot based on the eyeball reflection features; If the brightness parameter of the eyeball spot is greater than the preset third brightness threshold, and the pupil category is the non-circular pupil, and the morphology category is the animal morphology, and the texture category is the fur texture, then the detection result is determined to be a pet; If the brightness parameter of the eyeball spot is less than or equal to the third brightness threshold, and the pupil category is the round pupil, and the shape category is the human shape, and the texture category is the skin texture, then the detection result is determined to be an infant; If the brightness parameter of the eyeball spot is less than or equal to the third brightness threshold, and the pupil category is the round pupil, and the shape category is the human shape, and the texture category is the clothing texture, then the detection result is determined to be a child; If the texture category is the clothing texture, and the pupil category does not exist, and the morphology category does not exist, then the detection result is determined to be non-biological.

[0087] In one possible embodiment, after the retraction operation is initiated, the control unit 820 is further configured to: A second image of the retraction / expansion space is obtained through the anti-pinch detection device; Based on the second image, detect whether there is a second space occupant in the expansion / contraction space; If no second space placeholder is detected, the collapse operation continues until the current collapse task is completed; If the presence of the second space occupant is detected, the retraction drive mechanism is controlled to stop performing the retraction operation, and / or the retraction drive mechanism is controlled to perform an unfolding operation so that the second space occupant can move to a space outside the retraction space.

[0088] In one possible embodiment, the control unit 820, which occupies the second space including both non-biological and biological objects, is specifically configured to: stop the retraction operation of the retraction drive mechanism and / or extend the retraction drive mechanism to perform the extension operation. If the second space occupant is the non-biological object, then the retraction drive mechanism is controlled to stop performing the retraction operation and output information about the non-biological object to prompt the user to perform the corresponding action. If the second space occupant is the biological object, then the dynamic characteristics of the biological object are determined based on the second image; The target location information and target motion information of the biological object are determined based on the dynamic characteristics. The risk level of the biological object is determined based on the target motion information and the target location information. The response strategy corresponding to the risk level is determined based on the mapping relationship between the preset risk level and the preset response strategy. The response strategy generates a control to stop the retraction operation of the retraction drive mechanism and / or controls the retraction drive mechanism to perform an unfolding operation.

[0089] In one possible embodiment, the control unit 820, in determining the dynamic characteristics of the biological object based on the second image, is specifically configured to: Multiple consecutive images are extracted from the second image to obtain a continuous image sequence; The target image sequence is obtained by extracting the image corresponding to the biological object from the continuous image sequence; Based on a preset optical flow estimation algorithm, vector calculations are performed on each image in the target image sequence to obtain a motion vector parameter sequence; The motion trajectory of the biological object is determined based on the motion vector parameter sequence, and the first motion curve of the biological object is obtained. Calculate the dynamic texture features of each target image in the target image sequence to obtain a texture feature set; The texture feature set is input into a preset spatiotemporal feature model to obtain multiple spatiotemporal dynamic parameters; Determine the first average value, first maximum value, and first minimum value of the plurality of spatiotemporal dynamic parameters; Determine the average value of the first motion curve to obtain the second average value; The difference between the first average value and the second average value is determined to obtain the first difference; The adjustment value corresponding to the first difference is determined based on the preset mapping relationship between the first difference and the adjustment value, and the target adjustment value is obtained. The first motion curve is fine-tuned according to the target adjustment value to obtain the target motion curve; The dynamic characteristics of the biological object are determined based on the target motion curve.

[0090] In one possible embodiment, the control unit 820, in determining the risk level corresponding to the biological object based on the target motion information and the target location information, is specifically configured to: Obtain first position information of the easily pinched position of the anti-pinch folding furniture; The target distance is determined based on the target location information and the first location information; The movement speed of the biological object is determined based on the target motion information; If the movement speed is less than or equal to a preset first speed threshold, and the target distance is less than or equal to a preset first distance threshold, then the risk level is determined to be low risk. If the movement speed is between the first speed threshold and the preset second speed threshold, and the target distance is between the first distance threshold and the preset second distance threshold, then the risk level is determined to be medium risk. If the movement speed is greater than or equal to the second speed threshold and the target distance is greater than or equal to the second distance threshold, then the risk level is determined to be high risk.

[0091] In one possible embodiment, the anti-pinch furniture further includes a warning module, and the response strategy includes a low-risk response strategy, a medium-risk response strategy, and a high-risk response strategy. Specifically, the control unit 820, in generating and controlling the retraction drive mechanism to stop performing the retraction operation and / or controlling the retraction drive mechanism to perform the unfolding operation based on the response strategy, is used for: If the response strategy is the low-risk response strategy, then the retraction drive mechanism is controlled to perform the retraction operation, and the early warning module is controlled to perform the audible and visual early warning operation. If the response strategy is the medium-risk response strategy, then the retraction drive mechanism is controlled to perform the retraction operation at a preset first retraction speed; If the response strategy is the high-risk response strategy, then the retraction drive mechanism is controlled to stop performing the retraction operation; after the retraction drive mechanism stops performing the retraction operation, the retraction drive mechanism is controlled to perform the unfolding operation at a preset first unfolding speed.

[0092] It should be noted that the specific functional implementation of the anti-pinch control device 800 is described above. Figure 3 The description of a processor execution method for anti-pinch furniture unfolding and retracting, for example, the acquisition unit 810 is used to implement the relevant content of execution S330, which will not be repeated here. The various units or modules in the anti-pinch control device 800 can be individually or entirely merged into one or more other units or modules, or some of the units(s) can be further divided into multiple functionally smaller units or modules, which can achieve the same operation without affecting the technical effect of the embodiments of the present invention. The above-mentioned units or modules are based on logical function division. In practical applications, the function of one unit (or module) is implemented by multiple units (or modules), or the function of multiple units (or modules) is implemented by one unit (or module).

[0093] As can be seen, the anti-pinch control device provided in this application embodiment is applied to anti-pinch folding furniture. The anti-pinch folding furniture includes a furniture fixing part, a furniture folding part, a processor, a folding drive mechanism, and an anti-pinch detection device disposed on the first end face of the furniture fixing part. The first end face is the end face of the furniture folding part facing the folding space. The folding space refers to the space enclosed by the first end face and the second end face when the anti-pinch folding furniture is in the unfolded state. The second end face refers to the inner end face of the furniture fixing part. The processor is used to receive and respond to folding control commands for the folding part and send a first image acquisition command to the anti-pinch detection device. The anti-pinch detection device is used to... The system receives and responds to a first image acquisition command, acquires a first image of the retractable space, and sends the first image to a processor. The processor receives the first image and detects whether a first space-occupying object exists in the retractable space based on the first image. If a first space-occupying object is detected, it waits for a preset time period and then continues to perform image acquisition and space-occupying status detection operations on the retractable space, and / or outputs a prompt message to remind the user to move the first space-occupying object. If no first space-occupying object is detected, it sends a retraction action command to the retractable drive mechanism. The retractable drive mechanism receives the retraction action command and begins to execute the retraction operation. In this way, the anti-pinch retractable furniture can accurately detect and identify the space-occupying status of a space-occupying object in the retractable space without contacting or squeezing it. The anti-pinch retractable furniture can control the retractable part to stop the retraction action before contacting or squeezing the space-occupying object. This improves the safety of the anti-pinch retractable furniture. Furthermore, since the anti-pinch detection mechanism detects the space occupied by the furniture during the unfolding process, there is no need to design redundant structures to prevent the furniture from being pinched, thus improving the aesthetics of the anti-pinch furniture.

[0094] This application also provides an anti-pinch control system, wherein the anti-pinch control system can perform some or all of the steps of any method described in the embodiments of this application.

[0095] It should be noted that, for the sake of simplicity, the above embodiments are all described as a series of actions. Those skilled in the art should understand that this application is not limited to the described order of actions, as some steps in the embodiments of this application can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions, steps, modules, or units involved are not necessarily essential to the embodiments of this application.

[0096] In the above embodiments, the descriptions of each embodiment in this application have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0097] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

[0098] The steps of the methods or algorithms described in the embodiments of this application can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in RAM, flash memory, ROM, EPROM, electrically erasable programmable read-only memory (EEPROM), registers, hard disk, portable hard disk, read-only optical disk (CD-ROM), or any other form of storage medium well known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Furthermore, the ASIC can reside in a terminal device or management device. Alternatively, the processor and storage medium can exist as discrete components in the terminal device or management device.

[0099] Those skilled in the art will recognize that, in one or more of the examples above, the functions described in the embodiments of this application can be implemented, in whole or in part, by software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. This computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., digital video discs (DVDs)), or semiconductor media (e.g., solid-state disks (SSDs)).

[0100] The modules / units included in the various devices and products described in the above embodiments can be software modules / units, hardware modules / units, or a combination of both. For example, for devices and products applied to or integrated into a chip, all modules / units can be implemented using hardware methods such as circuits, or at least some modules / units can be implemented using software programs that run on a processor integrated within the chip, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits. For devices and products applied to or integrated into a chip module, all modules / units can be implemented using hardware methods such as circuits. Different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or different components of the chip module, or at least some modules / units can be implemented using hardware methods such as circuits. The implementation is achieved through a software program that runs on the processor integrated within the chip module. The remaining modules / units (if any) can be implemented using hardware methods such as circuits. For various devices and products applied to or integrated into terminal equipment, each of their modules / units can be implemented using hardware methods such as circuits. Different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or different components within the terminal equipment. Alternatively, at least some modules / units can be implemented through a software program that runs on the processor integrated within the terminal equipment, while the remaining modules / units (if any) can be implemented using hardware methods such as circuits.

[0101] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the embodiments of this application. It should be understood that the above descriptions are merely specific embodiments of the embodiments of this application and are not intended to limit the protection scope of the embodiments of this application. Any modifications, equivalent substitutions, improvements, etc., made on the basis of the technical solutions of the embodiments of this application should be included within the protection scope of the embodiments of this application.

Claims

1. A type of anti-pinch folding furniture, characterized in that, The anti-pinch folding furniture includes a furniture fixing part, a furniture folding part, a processor, a folding drive mechanism, and an anti-pinch detection device disposed on the first end face of the furniture fixing part. The first end face is the end face of the furniture folding part facing the folding space. The folding space refers to the space enclosed by the first end face and the second end face when the anti-pinch folding furniture is in the unfolded state. The second end face refers to the inner end face of the furniture fixing part. The processor is configured to receive and respond to a retraction control command for the furniture retraction section; and send a first image acquisition command to the anti-pinch detection device. The anti-pinch detection device is used to receive and respond to the first image acquisition command; acquire a first image of the retraction space; and send the first image to the processor. The processor is configured to receive the first image and detect, based on the first image, whether a first space occupant object exists in the retractable space; If the first space occupant is detected, the image acquisition and space occupancy status detection operations are performed on the expanded and contracted space after a preset time period; and / or, a prompt message is output to remind the user to move the first space occupant. If the first space occupant is not detected, a retraction action command is sent to the retraction drive mechanism; The retraction drive mechanism is used to receive the retraction action command and start performing the retraction operation.

2. The anti-pinch furniture as described in claim 1, characterized in that, In the process of detecting whether a first space occupant object exists in the expanded space based on the first image, the processor is specifically used for: The first image is denoised based on a preset denoising algorithm to obtain a first denoised image; Extract the black-and-white image information and color image information from the first denoised image; The pixel area in the black and white image information is determined based on a preset first brightness threshold to obtain the first area; The pixel area in the color image information is determined based on a preset second brightness threshold to obtain the second area; The first weight corresponding to the black and white image information and the second weight corresponding to the color image information are determined based on the first area and the second area. The target color fusion image is determined based on a preset hierarchical fusion algorithm, the first weight, the second weight, the black and white image information, and the color image information; The target color fusion image is enhanced based on a preset edge processing algorithm to obtain a first enhanced image; The first enhanced image is subjected to target detection based on a preset target detection algorithm to obtain the detection result; Based on the detection results, it is determined whether the first space occupant exists in the retraction space.

3. The anti-pinch furniture as described in claim 2, characterized in that, In the process of performing target detection on the first enhanced image based on a preset target detection algorithm to obtain the detection result, the processor is specifically used for: A preset feature extraction algorithm is used to extract features from the first enhanced image to obtain multiple first features; The features of the eyes, the features of the texture, and the features of the body contour are extracted from the plurality of first features respectively to obtain the eye features, texture features, and body contour features; Determine the pupil category among the aforementioned eye features; Determine the texture category in the texture features; Determine the morphological category in the body contour features; The detection result is determined based on the target detection algorithm, the pupil category, the texture category, and the morphology category.

4. The anti-pinch furniture as described in claim 3, characterized in that, The pupil category includes round pupils and non-round pupils; the texture category includes skin texture, hair texture, and clothing texture; the morphology category includes human morphology and animal morphology. Determining the detection result based on the target detection algorithm, the pupil category, the texture category, and the morphology category includes: Based on the plurality of first features, the ocular reflective features corresponding to the pupil category are determined; The target detection algorithm is used to determine the brightness parameters of the eyeball spot based on the eyeball reflection features; If the brightness parameter of the eyeball spot is greater than the preset third brightness threshold, and the pupil category is the non-circular pupil, and the morphology category is the animal morphology, and the texture category is the fur texture, then the detection result is determined to be a pet; If the brightness parameter of the eyeball spot is less than or equal to the third brightness threshold, and the pupil category is the round pupil, and the shape category is the human shape, and the texture category is the skin texture, then the detection result is determined to be an infant; If the brightness parameter of the eyeball spot is less than or equal to the third brightness threshold, and the pupil category is the round pupil, and the shape category is the human shape, and the texture category is the clothing texture, then the detection result is determined to be a child; If the texture category is the clothing texture, and the pupil category does not exist, and the morphology category does not exist, then the detection result is determined to be non-biological.

5. The anti-pinch furniture as described in any one of claims 1-4, characterized in that, After the collapse operation is initiated, the processor is further configured to: A second image of the retraction / expansion space is obtained through the anti-pinch detection device; Based on the second image, detect whether there is a second space occupant in the expansion / contraction space; If no second space placeholder is detected, the collapse operation continues until the current collapse task is completed; If the presence of the second space occupant is detected, the retraction drive mechanism is controlled to stop performing the retraction operation, and / or the retraction drive mechanism is controlled to perform an unfolding operation so that the second space occupant can move to a space outside the retraction space.

6. The anti-pinch furniture as described in claim 5, characterized in that, The second space occupant includes non-biological objects and biological objects. During the process of controlling the retraction / expansion drive mechanism to stop performing the retraction operation and / or controlling the retraction / expansion drive mechanism to perform the expansion operation, the processor is specifically used for: If the second space occupant is the non-biological object, then the retraction drive mechanism is controlled to stop performing the retraction operation and output information about the non-biological object to prompt the user to perform the corresponding action. If the second space occupant is the biological object, then the dynamic characteristics of the biological object are determined based on the second image; The target location information and target motion information of the biological object are determined based on the dynamic characteristics. The risk level of the biological object is determined based on the target motion information and the target location information. The response strategy corresponding to the risk level is determined based on the mapping relationship between the preset risk level and the preset response strategy. The response strategy generates a control to stop the retraction operation of the retraction drive mechanism and / or controls the retraction drive mechanism to perform an unfolding operation.

7. The anti-pinch furniture as described in claim 6, characterized in that, Determining the dynamic characteristics of the biological object based on the second image includes: Multiple consecutive images are extracted from the second image to obtain a continuous image sequence; The target image sequence is obtained by extracting the image corresponding to the biological object from the continuous image sequence; Based on a preset optical flow estimation algorithm, vector calculations are performed on each image in the target image sequence to obtain a motion vector parameter sequence; The motion trajectory of the biological object is determined based on the motion vector parameter sequence, and the first motion curve of the biological object is obtained. Calculate the dynamic texture features of each target image in the target image sequence to obtain a texture feature set; The texture feature set is input into a preset spatiotemporal feature model to obtain multiple spatiotemporal dynamic parameters; Determine the first average value, first maximum value, and first minimum value of the plurality of spatiotemporal dynamic parameters; Determine the average value of the first motion curve to obtain the second average value; The difference between the first average value and the second average value is determined to obtain the first difference; The adjustment value corresponding to the first difference is determined based on the preset mapping relationship between the first difference and the adjustment value, and the target adjustment value is obtained. The first motion curve is fine-tuned according to the target adjustment value to obtain the target motion curve; The dynamic characteristics of the biological object are determined based on the target motion curve.

8. The anti-pinch furniture as described in claim 6, characterized in that, Determining the risk level of the biological object based on the target motion information and the target location information includes: Obtain first position information of the easily pinched position of the anti-pinch folding furniture; The target distance is determined based on the target location information and the first location information; The movement speed of the biological object is determined based on the target motion information; If the movement speed is less than or equal to a preset first speed threshold, and the target distance is less than or equal to a preset first distance threshold, then the risk level is determined to be low risk. If the movement speed is between the first speed threshold and the preset second speed threshold, and the target distance is between the first distance threshold and the preset second distance threshold, then the risk level is determined to be medium risk. If the movement speed is greater than or equal to the second speed threshold and the target distance is greater than or equal to the second distance threshold, then the risk level is determined to be high risk.

9. The anti-pinch furniture as described in any one of claims 6-8, characterized in that, The anti-pinch furniture also includes an early warning module. The response strategy includes a low-risk response strategy, a medium-risk response strategy, and a high-risk response strategy. The step of generating and controlling the retraction drive mechanism to stop performing the retraction operation, and / or controlling the retraction drive mechanism to perform the unfolding operation, based on the response strategy, includes: If the response strategy is the low-risk response strategy, then the retraction drive mechanism is controlled to perform the retraction operation, and the early warning module is controlled to perform the audible and visual early warning operation. If the response strategy is the medium-risk response strategy, then the retraction drive mechanism is controlled to perform the retraction operation at a preset first retraction speed; If the response strategy is the high-risk response strategy, then the retraction drive mechanism is controlled to stop performing the retraction operation; after the retraction drive mechanism stops performing the retraction operation, the retraction drive mechanism is controlled to perform the unfolding operation at a preset first unfolding speed.

10. A method for preventing pinching during the unfolding of furniture, characterized in that, A processor is applied to the anti-pinch folding furniture, the anti-pinch folding furniture including a furniture fixing part, a furniture folding part, the processor, a folding drive mechanism, and an anti-pinch detection device disposed on a first end face of the furniture folding part. The first end face is the end face of the furniture fixing part facing the folding space. The folding space refers to the space enclosed by the first end face and the second end face of the anti-pinch folding furniture in the unfolded state. The second end face refers to the inner end face of the furniture folding part. The method includes: The anti-pinch detection device is controlled to receive and respond to a first image acquisition command; acquire a first image of the retraction space; and send the first image to the processor. The processor is controlled to receive the first image and detect whether there is a space occupant in the retractable space based on the first image; if the space occupant is detected, the image acquisition and space occupant status detection operations are performed on the retractable space after a preset first time period; and / or, a prompt message is output to remind the user to move the space occupant; if the space occupant is not detected, a retracting action command is sent to the retractable drive mechanism. The retraction drive mechanism receives the retraction action command and executes the retraction operation.

Images