Intelligent forklift, intelligent forklift control method and medium

Abstract

The embodiment of the present application provides an intelligent forklift, an intelligent forklift control method and a medium, and relates to the field of time-of-flight camera detection technology. In the intelligent forklift, a hydraulic cylinder (3) is fixed to a vehicle body (1) and drives a door frame (2) to move up and down along the vehicle body (1); a transport component (5) includes a fixed fork leg and a lifting fork leg, the lifting fork leg is fixed to the bottom of the door frame (2), and the fixed fork leg is fixed to the bottom of the vehicle body (1); a time-of-flight camera assembly (4) includes an upper hanging plate (41), a special-shaped bracket (44) and a time-of-flight camera (46), the upper hanging plate (41) is fixed to the door frame (2), the special-shaped bracket (44) slides up and down along the upper hanging plate (41), and is fixed to the upper hanging plate (41) when it slides down to a specified position, and the time-of-flight camera (46) is fixed to the special-shaped bracket (44); a delayed telescopic mechanism of the time-of-flight camera assembly enables the intelligent forklift to detect a dangerous area below the fork leg.

Description

Technical Field

[0001] The present application relates to the field of TOF camera detection technology, and in particular to an intelligent forklift, an intelligent forklift control method, and a medium. Background Art

[0002] When a smart forklift is lifting cargo, if a person or other vehicle gets close to the space under the fork legs, it can pose a significant safety hazard. Therefore, in practice, sensors are typically installed at the bottom of the fork legs to detect environmental information during the lifting process, preventing accidents.

[0003] In general, the sensor installed at the bottom of the fork leg usually adopts a laser sensor, an ultrasonic sensor or a TOF (Time of Flight) camera (i.e., a depth camera). However, in the current method of using sensors to detect environmental information, the vacuum area under the fork leg during the lifting process of the intelligent forklift fork leg cannot be detected by the sensor, so there are cases of false positives or missed reports. In addition, since the information of the sensor relative to the environment changes continuously during the lifting or lowering process of the intelligent forklift fork leg, there is a problem of mixing static and dynamic boundaries. Therefore, the method of using sensors to detect environmental information in the prior art is also very likely to cause image disproportion, resulting in a decrease in feature recognition rate. Summary of the Invention

[0004] The present invention provides an intelligent forklift, an intelligent forklift control method and a medium. Through the delayed telescopic mechanism of the TOF camera assembly 4, the problem in the prior art that a vacuum danger zone exists under the fork legs during the lifting process of the intelligent forklift and cannot be detected is solved.

[0005] In a first aspect, an embodiment of the present application provides an intelligent forklift, which includes a body 1, a mast 2, a hydraulic cylinder 3, a TOF camera assembly 4, a handling component 5, and a main control board, wherein:

[0006] The hydraulic cylinder 3 is fixed to the vehicle body 1 and drives the gantry 2 to move up and down along the vehicle body 1;

[0007] The transport component 5 includes a fixed fork leg and a lifting fork leg, the lifting fork leg is fixed to the bottom of the mast 2, and the fixed fork leg is fixed to the bottom of the vehicle body 1;

[0008] A first marking point is provided on each fixed fork leg, and a second marking point is provided on each lifting fork leg;

[0009] The TOF camera assembly 4 includes an upper hanging plate 41, a special-shaped bracket 44, and a TOF camera. The upper hanging plate 41 is fixed to the gantry 2. The special-shaped bracket 44 slides up and down along the upper hanging plate 41 and is fixed to the upper hanging plate 41 when it slides down to a specified position. The TOF camera is fixed to the special-shaped bracket 44.

[0010] The above-mentioned main control board is used to control the above-mentioned TOF camera component 4. After controlling the above-mentioned TOF camera component 4 to collect the target image, the above-mentioned fixed fork leg in the above-mentioned target image is identified according to the above-mentioned first marking point, and the above-mentioned lifting fork leg in the above-mentioned target image is identified according to the above-mentioned second marking point.

[0011] The above-mentioned smart forklift, through the structural design of the TOF component, that is, the special-shaped bracket 44 in the TOF camera slides up and down along the upper hanging plate 41, and is fixed on the upper hanging plate 41 when it slides down to the specified position, so that the TOF camera moves a certain distance after the mast 2 and the fork legs move during the lifting process of the smart forklift, and then moves with the mast 2 and the fork legs, realizing the delayed extension and retraction function of the TOF camera component 4, so that the smart forklift can detect the dangerous area under the fork legs.

[0012] As an optional embodiment, the TOF camera assembly 4 further includes a linear guide rail 42 and a slider 43, wherein:

[0013] The linear guide rail 42 is fixed on the upper hanging plate 41;

[0014] The sliding block 43 is fixed on the special-shaped bracket 44 and moves up and down in the linear guide rail 42 , driving the special-shaped bracket 44 to slide up and down along the upper hanging plate 41 .

[0015] As an optional embodiment, balls are installed inside the slider 43 , and the balls roll in the raceway, driving the slider 43 to move in the linear guide rail 42 .

[0016] As an optional embodiment, the TOF camera assembly 4 further includes a ball screw 45, wherein:

[0017] The bottom of the upper hanging plate 41 is provided with a flange and a positioning hole, and the positioning hole is located on the flange;

[0018] The above-mentioned ball screw 45 is fixed to the side of the special-shaped bracket 44 close to the above-mentioned upper hanging plate 41. The steel ball on the side of the above-mentioned ball screw 45 close to the above-mentioned positioning hole expands and contracts under the action of a spring and external force. When the above-mentioned special-shaped bracket 44 slides down along the above-mentioned upper hanging plate 41 to a position where the above-mentioned ball screw 45 is aligned with the above-mentioned positioning hole, the above-mentioned ball screw 45 is fixed in the above-mentioned positioning hole, so that the above-mentioned special-shaped bracket 44 is fixed on the above-mentioned upper hanging plate 41.

[0019] As an optional embodiment, a limiting edge is provided on one side of the special-shaped bracket 44 close to the upper hanging plate 41. When the special-shaped bracket 44 slides down to the limiting edge and contacts the vehicle body 1, the ball screw 45 disengages from the positioning hole, thereby releasing the fixation of the special-shaped bracket 44 from the upper hanging plate 41.

[0020] In a second aspect, an embodiment of the present application provides an intelligent forklift control method, which is applied to the above-mentioned intelligent forklift, and the method includes:

[0021] Get the target image captured by the TOF camera;

[0022] Determining a first boundary based on a first marking point in the target image, and determining a second boundary based on a second marking point in the target image, wherein the first boundary and the second boundary are parallel;

[0023] Correcting the target image based on the first boundary and the second boundary;

[0024] The corrected target image is detected using a detection algorithm, and the intelligent forklift is controlled when a human figure is detected.

[0025] The above method determines the first boundary (i.e., the dynamic boundary relative to the TOF camera) based on the first marking point in the target image captured by the TOF camera, determines the second boundary (i.e., the static boundary relative to the TOF camera) based on the second marking point, and corrects the target image based on the first boundary and the second boundary. That is, the target image is corrected by comprehensively considering the static and dynamic features in the reference forklift, which can make the generated image more accurate and reduce the degree of image imbalance caused by the mixing problem of dynamic and static boundaries, thereby increasing the accuracy of subsequent human recognition.

[0026] As an optional implementation, determining the first boundary according to the first marking point in the target image, and determining the second boundary according to the second marking point in the target image, includes:

[0027] Determining an image recognition base point based on the first marking point and the second marking point in the target image;

[0028] Based on the image recognition base point, coordinate transformation is performed on the target image to obtain a transformed image in a pixel coordinate system corresponding to the target image;

[0029] Determining position information of a first marking point and position information of a second marking point in the converted image;

[0030] A first boundary is determined based on the position information of the first marking point in the converted image, and a second boundary is determined based on the position information of the second marking point in the converted image.

[0031] As an optional implementation, correcting the target image based on the first boundary and the second boundary includes:

[0032] Determining a hybrid correction coefficient based on a first distance between the first boundary and the second boundary; wherein the hybrid correction coefficient is proportional to the first distance;

[0033] According to the mixed correction coefficient, the position coordinates of each pixel in the target image are corrected.

[0034] The above method continuously updates the hybrid correction coefficient based on the distance between the first boundary and the second boundary as the lifting height changes, and corrects the image according to the updated hybrid correction coefficient, effectively improving the image restoration degree.

[0035] As an optional implementation, a detection algorithm is used to perform human figure recognition on the corrected target image, and the intelligent forklift is controlled when a human figure is recognized, including:

[0036] determining an area between the first boundary and the second boundary in the corrected target image as a key area;

[0037] Using a detection algorithm to determine whether there is a circular area in the above-mentioned key area;

[0038] When the circular area exists in the key area, it is determined that a human figure is recognized and the intelligent forklift is controlled.

[0039] By dividing the key detection areas and conducting detection on the key areas, the accuracy of human recognition is guaranteed, the overall computational complexity is reduced, and the algorithm operation speed and detection efficiency are improved.

[0040] As an optional implementation, determining that the intelligent forklift is controlled when a human figure is recognized includes:

[0041] When it is determined that a human figure is recognized, a second distance between the circular area and the intelligent forklift is calculated;

[0042] When it is determined that the second distance is less than a preset threshold, the intelligent forklift is controlled to stop working and an alarm is triggered;

[0043] When it is determined that the second distance is not less than the preset threshold, a message reminder is triggered.

[0044] The above method, when determining that a human figure is recognized, determines the corresponding intelligent forklift control method based on the distance between the human figure and the intelligent forklift. When the distance is less than a set threshold, the intelligent forklift is controlled to stop working and an alarm is triggered, significantly improving the safety and stability of the use of the intelligent forklift.

[0045] In a third aspect, an embodiment of the present application provides an intelligent forklift control device, the device comprising:

[0046] An acquisition unit, used to acquire a target image captured by a TOF camera;

[0047] a determining unit, configured to determine a first boundary according to a first marking point in the target image, and to determine a second boundary according to a second marking point in the target image, wherein the first boundary and the second boundary are parallel;

[0048] a correction unit, configured to correct the target image based on the first boundary and the second boundary;

[0049] The control unit is used to detect the corrected target image using a detection algorithm and control the above-mentioned intelligent forklift when a human figure is detected.

[0050] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having computer program instructions stored thereon, which implement any step of the above-mentioned intelligent forklift control method when executed by a processor.

[0051] In a fifth aspect, an embodiment of the present application further provides a computer program product, comprising a computer program, which is stored in a computer-readable storage medium; when the intelligent forklift reads the computer program from the computer-readable storage medium, the computer program is executed, causing the intelligent forklift to perform any step in the above-mentioned intelligent forklift control method. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0053] Figure 1 A schematic structural diagram of an intelligent forklift provided in an embodiment of the present application;

[0054] Figure 2 A schematic structural diagram of a transport component provided in an embodiment of the present application;

[0055] Figure 3 A schematic structural diagram of a TOF camera assembly provided in an embodiment of the present application;

[0056] Figure 4 A schematic diagram of the structure of a main control board provided in an embodiment of the present application;

[0057] Figure 5 A flow chart of an intelligent forklift control method provided in an embodiment of the present application;

[0058] Figure 6 A schematic diagram of a coordinate transformation relationship provided in an embodiment of the present application;

[0059] Figure 7 A schematic diagram of the structure of a key area and a non-key area provided in an embodiment of the present application;

[0060] Figure 8 A schematic diagram of a feature point and reference boundary provided in an embodiment of the present application;

[0061] Figure 9 A schematic structural diagram of an intelligent forklift control device provided in an embodiment of the present application.

[0062] icon:

[0063] 1: Car body; 2: Mast; 3: Hydraulic cylinder;

[0064] 4: TOF camera assembly; 41: upper hanging plate; 42: linear guide rail; 43: slider;

[0065] 44: Special-shaped bracket; 45: Ball screw; 46: TOF camera;

[0066] 5: transport component; 51-52: second marking point; 53-54: first marking point;

[0067] 6: Cargo location. DETAILED DESCRIPTION

[0068] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application will be further described below with reference to the accompanying drawings and examples. However, the example embodiments can be implemented in various forms and should not be understood as being limited to the embodiments set forth herein; on the contrary, these embodiments are provided to make the present application more comprehensive and complete, and to fully convey the concepts of the example embodiments to those skilled in the art. The same figure marks in the figures represent the same or similar structures, and their repeated descriptions will be omitted. The words expressing position and direction described in this application are all explained using the accompanying drawings as examples, but changes can be made as needed, and all changes are included in the scope of protection of this application. The drawings in this application are only used to illustrate relative position relationships and do not represent true proportions.

[0069] In this application, unless otherwise specified or limited, the terms "installed," "connected," "connect," "fixed," etc. should be understood in a broad sense. For example, they can refer to fixed connection, detachable connection, or integration; mechanical connection or electrical connection; direct connection or indirect connection through an intermediate medium; internal communication between two elements or interaction between two elements, unless otherwise specified. Those skilled in the art will understand the specific meanings of the above terms in this application based on specific circumstances.

[0070] It should be noted that when an element is referred to as being "fixed to" or "disposed on" another element, it may be directly connected to the other element or there may be an intermediate element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be an intermediate element.

[0071] When a smart forklift is lifting cargo, if a person or other vehicle gets close to the space under the fork legs, it can pose a significant safety hazard. Therefore, in practice, sensors are typically installed at the bottom of the fork legs to detect environmental information during the lifting process, preventing accidents.

[0072] Typically, sensors installed at the bottom of the fork legs are laser sensors, ultrasonic sensors, or time-of-flight cameras. Laser sensors, due to their line or surface scanning methods, are not easily able to capture three-dimensional space and are therefore prone to missed detections. Furthermore, current methods of using sensors to detect environmental information prevent the vacuum area beneath the fork legs during the lifting process of smart forklifts, leading to false or missed detections.

[0073] Time-of-flight (TOF) technology uses light to measure distance. This involves emitting continuous light pulses toward the object being measured. A sensor then receives the reflected signals and calculates the distance to the target by calculating the light pulse's round-trip time of flight. A 3D camera based on TOF technology is a new, compact stereoscopic imaging device. This type of camera can simultaneously capture both intensity and depth information about the object being measured. Due to its low manufacturing cost, simple data processing, fast response, and insensitivity to light, it is widely used in gaming and entertainment, virtual reality, motion recognition and tracking, autonomous robot navigation, and industrial automated assembly.

[0074] A TOF camera, broadly speaking, measures the time of flight of a light pulse between a sensor and an object's surface. The product of this measured time and the speed of light is twice the required distance. A TOF camera operates by actively transmitting a modulated light signal to the object being measured. After reflection, the light signal is received by a photodetector. The phase difference between the transmitted and received signals is calculated based on the charge accumulated on the detector, yielding the distance between the object and the camera.

[0075] In order to solve the problem that the vacuum area under the fork leg of the above-mentioned intelligent forklift cannot be detected by the sensor during the lifting process of the fork leg, thereby causing false alarms or missed alarms, an embodiment of the present application provides an intelligent forklift with a TOF camera. The structure of the intelligent forklift is as follows: Figure 1 shown.

[0076] like Figure 1 As shown, the intelligent forklift provided in the embodiment of the present application mainly includes: a body 1, a door frame 2, a hydraulic cylinder 3, a TOF camera assembly 4, a handling component 5 and a main control board, wherein the main control board is not in the Figure 1 Mark in.

[0077] The above-mentioned gantry 2 is welded by multiple steel bars and square steels, with pulleys installed on both sides. It can move up and down in the vehicle body guide rails and is used to fix the fork legs (the fork legs refer to the lifting fork legs in the transport component 5) and other components;

[0078] The hydraulic cylinder 3 is divided into two parts, left and right, which are fixed to the vehicle body 1. Specifically, the lower end of the hydraulic cylinder 3 is fixed to the bottom of the vehicle body 1, and a clamp is locked on the vehicle body 1 in the middle of the cylinder. The upper end is fixed to the jacking plate. The hydraulic cylinder is driven by hydraulic pressure to realize the lifting function of the jacking plate.

[0079] The jacking plate is installed on the gantry 2 and is connected to the push rods of the hydraulic cylinder 3 on both sides, which is used to transmit the telescopic movement of the hydraulic cylinder 3 to the gantry 2, driving the gantry 2 to move up and down along the vehicle body 1;

[0080] The above-mentioned transport component 5 includes a fixed fork leg and a lifting fork leg. Usually, the number of the fixed fork leg and the fork leg is two. The lifting fork leg is fixed to the bottom of the door frame 2 by welding. As the door frame 2 moves up and down, it is used to insert into the bottom hole of the storage bracket to lift the goods (the position of the goods is as shown in the figure). Figure 2 6); The fixed fork leg is fixed to the bottom of the vehicle body 1 by welding, does not move during the lifting process of the smart forklift, and is in direct contact with the ground, which is used to increase the anti-overturning performance of the smart forklift;

[0081] Each fixed fork leg is provided with a first marking point, and each lifting fork leg is provided with a second marking point, such as Figure 2As shown, first marking points 53 and 54 are provided on the fixed fork legs, and second marking points 51 and 52 are provided on the lifting fork legs. The first marking points are generally located at the same position on the multiple fixed fork legs, i.e., each first marking point is located at the same distance from the bottom of the corresponding fixed fork leg as the distances from the other first marking points to the bottom of the corresponding fixed fork legs. Similarly, the second marking points are generally located at the same position on the multiple lifting fork legs, i.e., each second marking point is located at the same distance from the bottom of the corresponding lifting fork leg as the distances from the other second marking points to the bottom of the corresponding lifting fork legs. The arrangement of the positions of the first and second marking points ensures that the first boundary connecting the first marking points and the second boundary connecting the second marking points are parallel.

[0082] It should be noted that the first and second marking points can be different colors from the prongs on which they are located, or can be special structures located on the corresponding prongs, such as protrusions. The shapes of the first and second marking points can be circular, square, or other shapes, and the shapes of the first and second marking points can be the same or different.

[0083] The main control board is used to control the TOF camera assembly 4, and after controlling the TOF camera assembly 4 to capture the target image, the position of the fixed fork leg in the target image is identified according to the first marking point, and the position of the lifting fork leg in the target image is identified according to the second marking point;

[0084] The TOF camera assembly 4 is installed at the bottom of the lifting fork leg to detect the environmental information at the front end of the fork leg; Figure 3 As shown, the TOF camera assembly specifically includes an upper hanging plate 41, a special-shaped bracket 44 and a TOF camera 46. The upper hanging plate 41 is a T-shaped sheet metal part, which is fixed to the door frame 2 by screws. The special-shaped bracket 44 is a special-shaped sheet metal part, which can slide up and down along the upper hanging plate 41 and is fixed on the upper hanging plate 41 when it slides down to the specified position. The TOF camera is fixed on the special-shaped bracket 44 and is used to detect the environmental information in front of the fork foot of the intelligent forklift.

[0085] In some embodiments, the TOF camera assembly 4 further includes a linear guide rail 42 and a slider 43, wherein: the linear guide rail 42 is fixed to the upper hanging plate through a countersunk hole, and is used to cooperate with the slider 43 to form a sliding pair, driving the special-shaped bracket 44 to slide up and down;

[0086] The slider 43 is fixed to the special-shaped bracket 44 with screws and moves up and down within the linear guide 42, driving the special-shaped bracket 44 to slide up and down along the upper hanging plate 41. Specifically, the slider 43 is equipped with a ball bearing that rolls within a raceway. The slider 43 is mounted on the linear guide 42 through a corresponding notch, and the rolling of the ball bearing allows the slider 43 to move freely on the linear guide 42.

[0087] In some embodiments, a flange and a positioning hole are provided at the bottom of the upper hanging plate 41. The positioning hole is located on the flange and is used to cooperate with the special-shaped bracket 44 to achieve a self-locking function; wherein, the self-locking function is that when the special-shaped bracket 44 slides down to reach the specified position, the positions of the upper hanging plate 41 and the special-shaped bracket 44 are relatively fixed and no longer have relative displacement. After reaching the specified position, the special-shaped bracket 44 will move with the upper hanging plate 41.

[0088] In some embodiments, the TOF camera assembly 4 further includes a ball screw 45 , wherein:

[0089] The ball screw 45 is fixed to the side of the special-shaped bracket 44 close to the upper hanging plate. The steel ball on the side of the ball screw 45 close to the positioning hole expands and contracts under the action of the spring and external force. When the special-shaped bracket 44 slides down along the upper hanging plate 41 to the position where the ball screw 45 is aligned with the positioning hole, the ball screw 45 is fixed in the positioning hole, so that the special-shaped bracket 44 is fixed on the upper hanging plate 41, thereby realizing the above-mentioned self-locking function.

[0090] In some embodiments, a limiting edge is provided on one side of the shaped bracket 44 near the upper mounting plate 41 to limit the lower limit of the shaped bracket 44. When the shaped bracket 44 slides down to a specified height, the limiting edge contacts and supports the vehicle body 1. At this time, the ball screw 45 disengages from the positioning hole, releasing the fixed relationship between the shaped bracket 44 and the upper mounting plate 41 (i.e., the self-locking function is released).

[0091] The present application also provides a schematic diagram of the structure of a main control board, such as Figure 4 As shown, the main control board may include at least one processor and at least one memory. The memory stores program code, which, when executed by the processor, causes the processor to perform the steps of the intelligent forklift control method described in the following various exemplary embodiments of this specification.

[0092] Refer to the following Figure 4 The main control board 400 according to this embodiment of the present application will be described. Figure 4 The main control board 400 shown is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0093] like Figure 4 As shown, the main control board 400 is presented as a general-purpose device. Components of the main control board 400 may include, but are not limited to: the at least one processor 401 described above, the at least one memory 402 described above, and a bus 403 connecting different system components (including the memory 402 and the processor 401). The memory stores program code. When the program code is executed by the processor, the processor performs the following steps:

[0094] Get the target image captured by the TOF camera;

[0095] Determining a first boundary according to a first marking point in the target image, and determining a second boundary according to a second marking point in the target image, wherein the first boundary and the second boundary are parallel;

[0096] Correcting the target image based on the first boundary and the second boundary;

[0097] The detection algorithm is used to detect the corrected target image and control the intelligent forklift when a human figure is detected.

[0098] Bus 403 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, and a processor or local bus using any of a variety of bus architectures.

[0099] The memory 402 may include a readable medium in the form of a volatile memory, such as a random access memory (RAM) 4021 and / or a cache memory 4022 , and may further include a read-only memory (ROM) 4023 .

[0100] The memory 402 may also include a program / utility 4025 having a set (at least one) of program modules 4024, such program modules 4024 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0101] The main control board 400 can also communicate with one or more external devices 404 (e.g., a keyboard, pointing device, etc.), one or more devices that enable a user to interact with the main control board 400, and / or any device that enables the main control board 400 to communicate with one or more other devices (e.g., a router, a modem, etc.). This communication can occur via an input / output (I / O) interface 405. Furthermore, the main control board 400 can communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network such as the Internet) via a network adapter 406. As shown, the network adapter 406 communicates with other modules of the main control board 400 via bus 403. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the main control board 400, including but not limited to microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0102] In the embodiment of the present application, the smart forklift adopts the structural design of the TOF component, that is, the special-shaped bracket in the TOF camera slides up and down along the upper hanging plate, and is fixed to the upper hanging plate when it slides down to the specified position. During the lifting process of the smart forklift, the TOF camera moves a certain distance after the mast and the lifting fork legs move, and then moves with the mast and the lifting fork legs, thereby realizing the delayed extension and retraction function of the TOF camera component, so that the smart forklift can detect the dangerous area under the fork legs.

[0103] In the existing technology, since the information of the sensor relative to the environment constantly changes during the lifting or lowering of the fork legs of the intelligent forklift, there is a problem of static and dynamic boundary mixing. Therefore, the method of using sensors to detect environmental information in the existing technology is also very likely to cause image disproportion, resulting in a decrease in feature recognition rate.

[0104] In order to solve the problem in the prior art that image proportions are imbalanced and feature recognition rate is reduced due to the mixing of static and dynamic boundaries, this application proposes an intelligent forklift control method based on the structure of the above-mentioned intelligent forklift. By presetting calibration points (i.e., first marking points and second marking points) on the fork leg shape, the static boundary (i.e., second boundary) and dynamic boundary (i.e., first boundary) are accurately identified, and the image recognized by the TOF camera is continuously corrected as the lifting height of the intelligent forklift changes. It can adapt to the scene where static boundaries and dynamic boundaries are mixed during the lifting process, effectively improve the image restoration degree, reduce the degree of image proportion imbalance, and thus improve the feature recognition rate.

[0105] like Figure 5 As shown, the present application provides an intelligent forklift control method, which is applied to the above-mentioned intelligent forklift, and specifically includes the following steps:

[0106] Step 501: Acquire a target image captured by a TOF camera;

[0107] During the operation of the intelligent forklift, when the intelligent forklift moves the body to the starting position corresponding to the goods to be transported and is ready to lift the fork legs upward, the main control board controls the TOF camera to turn on. The TOF camera first performs a self-check when it is turned on, and then performs image acquisition after completing the self-check.

[0108] Specifically, the TOF camera emits a modulated light signal and receives the light signal reflected back from the object being measured. Distance measurement is achieved by calculating the time difference f(t) between the emission time and the reception time.

[0109] In a specific implementation, the TOF camera collects planar image information and illumination information, wherein the planar image information is the pixel point collected by the TOF camera, and the illumination information is the time difference between the emission time and the reception time. After collecting the planar image information and illumination information, the depth value corresponding to the pixel point is calculated using the time difference corresponding to the pixel point in the illumination information. According to the above method, the depth value corresponding to each pixel point in the planar image information is calculated. After obtaining the depth value of each pixel point, the camera coordinate system is established with the specified position (the specified position can use the TOF camera) as the coordinate origin, such as Figure 6 As shown, the camera coordinate system is a three-dimensional coordinate system Oc-XcYcZc, and according to the depth value of each pixel point, the position coordinates corresponding to each pixel point in the camera coordinate system are obtained to obtain the target image.

[0110] Step 502: determining a first boundary based on a first marking point in the target image, and determining a second boundary based on a second marking point in the target image, wherein the first boundary and the second boundary are parallel;

[0111] Specifically, first, an image recognition base point is determined according to the first marking point and the second marking point in the target image. The image recognition base point is usually set to the position where the TOF camera emits the light signal.

[0112] Then, based on the image recognition base point, the target image is transformed to obtain a transformed image in the pixel coordinate system corresponding to the target image; that is, with the image recognition base point as the coordinate origin, the target image in the camera coordinate system is transformed into a transformed image in the pixel coordinate system, specifically, the coordinates of each collected pixel point are transformed from the camera coordinate system to the pixel coordinate system.

[0113] like Figure 6 As shown, the camera coordinate system is Oc-XcYcZc and the pixel coordinate system is O I -uv, the physical image coordinate system is O R -xy, the world coordinate system is Ow-XwYwZw. Assuming the camera coordinate system as the world coordinate system and the pixel coordinate system as the physical image coordinate system, the conversion relationship between the pixel coordinate system and the camera coordinate system is:

[0114]

[0115] Where: f is the focal length of the camera; dx and dy are the pixel sizes of the sensor in the x and y directions respectively; (c x ,c y ) is the position of the imaging optical center.

[0116] That is, for any pixel point captured by TOF, the functional relationship between its position coordinate P(X, Y, Z) in the camera coordinate system and its position coordinate p(x, y) in the pixel coordinate system (physical image coordinate system) is: {P(X), P(Y), P(Z)} = {A*p(x), A*p(y), A*f(t)c}, where A is the proportional coefficient, f(t) is the time difference between the sending time and the receiving time of the TOF camera wide signal, and c is the speed of light.

[0117] The proportionality coefficient A is calculated based on the depth value of any marking point in the image and the actual distance between the marking point and the light signal transmitting end of the TOF camera.

[0118] After obtaining a conversion image in a pixel coordinate system corresponding to the target image, position information of a first marking point and position information of a second marking point in the conversion image are determined.

[0119] Finally, a first boundary is determined based on the position information of the first marking point in the converted image, and the first boundary is determined based on the position information of the second marking point in the converted image, and the second boundary is determined based on the position information of the second marking point in the converted image, and the second boundary is determined based on the position information of the second marking point in the converted image.

[0120] Step 503: Correct the target image based on the first boundary and the second boundary;

[0121] Specifically, a first distance between the first boundary and the second boundary is first calculated, a hybrid correction coefficient is determined based on the first distance, and then the position coordinates of each pixel point in the target image are corrected according to the hybrid correction coefficient; wherein the hybrid correction coefficient is proportional to the first distance.

[0122] As can be seen from the structure of the above-mentioned intelligent forklift, when setting the first marking point and the second marking point, the first boundary connecting the first marking point and the second boundary connecting the second marking point are parallel to each other, so the first distance between the first boundary and the second boundary can be directly calculated.

[0123] Moreover, when the TOF camera assembly moves together with the lifting fork, since the first marking point is located on the fixed fork, and the fixed fork does not move during the lifting process of the intelligent forklift, that is, the position of the first marking point relative to the TOF camera is in a moving state, the first boundary is a dynamic boundary; the second marking point is located on the lifting fork, and the lifting fork moves together with the TOF camera during the lifting process of the intelligent forklift, so the second boundary is a static boundary. In this application, the target image is corrected based on the dynamic boundary and the static boundary, which can make the target image more accurate.

[0124] It should be noted that, when the lifting fork just starts to move and the TOF camera assembly is delayed in extending and retracting, the above-mentioned first boundary is a static boundary and the above-mentioned second boundary is a dynamic boundary. In the embodiment of the present application, the specific process of the above-mentioned intelligent forklift control method is mainly introduced by taking the TOF camera assembly and the lifting fork moving together as an example. When the above-mentioned TOF camera assembly is delayed in extending and retracting, the specific process of the intelligent forklift control method is similar.

[0125] Step 504 : Detect the corrected target image using a detection algorithm, and control the intelligent forklift when a human figure is detected.

[0126] Specifically, we first divide the regions into Figure 7 As shown, the first boundary (i.e. Figure 7 The dynamic boundary line) and the second boundary (i.e. Figure 7 The area between the static boundary line in Figure 7 The key identification area is divided into non-key (identification) areas except the key area, wherein the key area corresponds to the dangerous area of ​​the intelligent forklift. The left part of the figure is a side view of the forklift, wherein AB is a schematic diagram of the above-mentioned first boundary (i.e., the dynamic boundary line), CD is a schematic diagram of the above-mentioned second boundary (i.e., the static boundary line), E represents the position of the TOF camera assembly in the side view, and the range inside the GEF is the field of view of the TOF camera under one condition. In some embodiments, after determining the first boundary and the second boundary, the first boundary and the second boundary in the pixel coordinate system are converted into the first boundary and the second boundary in the camera coordinate system, and the corrected target image in the camera coordinate system is divided into regions according to the first boundary and the second boundary in the camera coordinate system.

[0127] Then, a detection algorithm is used to determine whether there is a circular area in the key area; when there is a circular area in the key area, it is determined that a human figure is recognized and the intelligent forklift is controlled; when it is determined that no human figure is recognized, the intelligent forklift is kept in normal working state.

[0128] By dividing key areas and conducting detection in these areas, the accuracy of human recognition is guaranteed, the overall computational complexity is reduced, and the algorithm operation speed and detection efficiency are improved.

[0129] After the aforementioned division into key and non-key areas, the intrusion recognition algorithm can be further weighted based on the recognition accuracy of each area. This means maintaining the recognition accuracy of key areas while reducing it in non-key areas. This ensures the computational accuracy of key detection areas while reducing the overall computational effort, improving the algorithm's speed and detection efficiency.

[0130] The above method determines the first boundary (i.e., the dynamic boundary relative to the TOF camera) based on the first marking point in the target image captured by the TOF camera, determines the second boundary (i.e., the static boundary relative to the TOF camera) based on the second marking point, and corrects the target image based on the first boundary and the second boundary. That is, the target image is corrected by comprehensively considering the static and dynamic features in the reference forklift, which can make the generated image more accurate and reduce the degree of image imbalance caused by the mixing problem of dynamic and static boundaries, thereby increasing the accuracy of subsequent human recognition.

[0131] The following combination Figure 8 According to a specific example, the process of determining the first boundary according to the position information of the first marking point in the converted image, determining the second boundary according to the position information of the second marking point in the converted image, and correcting the target image based on the first boundary and the second boundary is described in detail.

[0132] It should be noted that, in this application, the above process is described only by taking the case where the first boundary and the second boundary are parallel to the horizontal axis as an example. In other cases, the method of determining the first boundary and the second boundary is similar.

[0133] Figure 8 In the figure, feature points p1 and p2 are the two second marking points on the lifting fork legs, and their coordinates in the pixel coordinate system are p1(x1, y1) and p2(x2, y2) respectively. Feature points p3 and p4 are the two first marking points on the fixed fork legs, and their coordinates in the pixel coordinate system are p3(x3, y3) and p4(x4, y4) respectively, where y1=y2 and y3=y4.

[0134] The height function f1(h) = A*·y1 of the first boundary corresponding to the lifting fork is determined based on the coordinates p1(x1, y1) of the feature point p1 and the coordinates p2(x2, y2) of the feature point p2. The first boundary rises and falls together with the TOF camera, so it does not change with the change of the lifting height, that is, it is a static boundary.

[0135] For example, assuming that the coordinates of the feature point p1 are p1 (30, 50) and the coordinates of the feature point p2 are p2 (60, 50), the height function of the first boundary is obtained as f1 (h) = 50A.

[0136] Based on the coordinates of feature point p3 (x3, y3) and feature point p4 (x4, y4), the height function of the second boundary corresponding to the lifting fork is determined as f2(h) = A*·f(t)·y3. The second boundary does not rise or fall with the TOF camera, so it changes with the lifting height, that is, it is a dynamic boundary.

[0137] For example, assuming that the coordinates of the feature point p3 are p3 (30, 100) and the coordinates of the feature point p4 are p4 (60, 100), the height function of the second boundary is obtained as f2(h)=100A·f(t).

[0138] After determining the first boundary and the second boundary, the height function f1(h) of the first boundary y1 in the image and the height function f2(h) of the second boundary y2 in the image can be determined, thereby obtaining the height difference change function Δf(h) = f1(h) - f2(h) of the first boundary and the second boundary. On this basis, the boundary mixing correction coefficient α is determined. The mixing correction coefficient α is a variable related to the height difference change function Δf(h) and is proportional to the height difference change function Δf(h).

[0139] Then, based on the mixed correction coefficient α, the position coordinates of each pixel in the target image are corrected. The correction expression is: {P(X), P(Y), P(Z)} = {Aα*p(x), Aα*p(y), Aα*f(t)c}.

[0140] In step 504, the process of determining whether to control the intelligent forklift when a human figure is detected specifically includes:

[0141] When it is determined that a human figure is detected, a second distance between the circular area and the intelligent forklift is calculated;

[0142] The above-mentioned second distance is the distance between the preset position in the circular area and the preset position in the smart forklift. Specifically, it can be the distance between the center point of the circular area and the TOF camera in the smart forklift, or it can be the distance between the leftmost point in the circular area and a fork foot in the smart forklift. Among them, the preset position in the circular area and the preset position in the smart forklift are not limited in this application, but it should be noted that different preset positions in different circular areas and preset positions in the smart forklift correspond to different preset thresholds.

[0143] When it is determined that the second distance is less than a preset threshold, the intelligent forklift is controlled to stop working and an alarm is triggered;

[0144] If it is determined that the second distance is less than the preset threshold, it can be considered that the human figure is in the danger zone under the fork leg. At this time, the intelligent forklift is controlled to stop working and an alarm sound is issued to prompt the person to leave the forklift danger zone.

[0145] When it is determined that the second distance is not less than the preset threshold, a message reminder is triggered;

[0146] If it is determined that the second distance is not less than the preset threshold, it can be considered that the human figure is not in the dangerous area under the fork leg. At this time, a message reminder is displayed or a voice message reminder is triggered to remind people or vehicles to stay away from the forklift area.

[0147] In some embodiments, the coordinate range of the candidate area corresponding to the area under the fork foot where there may be danger can also be determined in advance. When it is determined that the proportion of the detected circular area located in the candidate area in the circular area is greater than a preset value, the second distance between the circular area and the intelligent forklift is calculated to reduce the amount of calculation and improve the recognition rate.

[0148] Specifically, the value range of P(Z) (i.e., the corresponding value range of the candidate area under the Z coordinate axis) can be determined based on the actual size of the fork and the actual position of the area under the fork that may be dangerous, and then the value ranges of P(X) and P(Y) (i.e., the corresponding value ranges of the candidate area under the X coordinate axis and the Y coordinate value) can be determined to reduce the recognition range.

[0149] The above method identifies human image information based on the human image features (circular features) and specific distance range in the usage scenario, reduces the missed reporting rate, improves the recognition rate and usage performance, and significantly improves the safety and stability of the use of smart forklifts.

[0150] Based on the same disclosed concept, the embodiment of the present application also provides an intelligent forklift control device. Since the device is the device in the method in the embodiment of the present application, and the principle of solving the problem by the device is similar to that of the method, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0151] Figure 9 For a schematic diagram of an intelligent forklift control device provided in an embodiment of the present application, please refer to Figure 9 , an embodiment of the present application provides an intelligent forklift control device, the device comprising:

[0152] An acquisition unit 901 is used to acquire a target image captured by a TOF camera;

[0153] a determining unit 902 configured to determine a first boundary based on a first marking point in the target image, and to determine a second boundary based on a second marking point in the target image, wherein the first boundary and the second boundary are parallel;

[0154] a correction unit 903, configured to correct the target image based on the first boundary and the second boundary;

[0155] The control unit 904 is configured to detect the corrected target image using a detection algorithm, and control the intelligent forklift when a human figure is detected.

[0156] Optionally, the determining unit 902 is configured to determine a first boundary according to a first marking point in the target image, and to determine a second boundary according to a second marking point in the target image, including:

[0157] Determining an image recognition base point based on the first marking point and the second marking point in the target image;

[0158] Based on the image recognition base point, coordinate transformation is performed on the target image to obtain a transformed image in a pixel coordinate system corresponding to the target image;

[0159] Determining position information of a first marking point and position information of a second marking point in the converted image;

[0160] A first boundary is determined based on the position information of the first marking point in the converted image, and a second boundary is determined based on the position information of the second marking point in the converted image.

[0161] Optionally, the correction unit 903 is configured to correct the target image based on the first boundary and the second boundary, including:

[0162] Determining a hybrid correction coefficient based on a first distance between the first boundary and the second boundary; wherein the hybrid correction coefficient is proportional to the first distance;

[0163] According to the mixed correction coefficient, the position coordinates of each pixel in the target image are corrected.

[0164] Optionally, the correction unit 903 is configured to perform human figure recognition on the corrected target image using a detection algorithm, and control the intelligent forklift when a human figure is recognized, including:

[0165] determining an area between the first boundary and the second boundary in the corrected target image as a key area;

[0166] Using a detection algorithm to determine whether there is a circular area in the above-mentioned key area;

[0167] When the circular area exists in the key area, it is determined that a human figure is recognized and the intelligent forklift is controlled.

[0168] Optionally, the control unit 904 is configured to control the intelligent forklift when a human figure is recognized, including:

[0169] When it is determined that a human figure is recognized, a second distance between the circular area and the intelligent forklift is calculated;

[0170] When it is determined that the second distance is less than a preset threshold, the intelligent forklift is controlled to stop working and an alarm is triggered;

[0171] When it is determined that the second distance is not less than the preset threshold, a message reminder is triggered.

[0172] In some possible implementations, various aspects of the intelligent forklift control method provided in the present application may also be implemented in the form of a program product, which includes program code. When the program product is run on a computer device, the program code is used to enable the computer device to execute the steps of the intelligent forklift control method according to various exemplary embodiments of the present application described above in this specification.

[0173] The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0174] The program product for monitoring of the embodiment of the present application can be a portable compact disc read-only memory (CD-ROM) and include program code, and can be run on a device. However, the program product of the present application is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0175] A readable signal medium may include a data signal transmitted in baseband or as part of a carrier wave, which carries readable program code. Such a transmitted data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium that can transmit, propagate, or transfer a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0176] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0177] The program code for performing the operations of the present application can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user device, partially on the user device, as a separate software package, partially on the user device and partially on a remote device, or entirely on the remote device or server. In cases involving a remote device, the remote device can be connected to the user device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external device (for example, using an Internet service provider to connect through the Internet).

[0178] It should be noted that although several units or subunits of the device are mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, depending on the embodiment of the application, the features and functions of two or more units described above can be embodied in a single unit. Conversely, the features and functions of a single unit described above can be further divided and embodied by multiple units.

[0179] Furthermore, although the operations of the method of the present application are described in a particular order in the accompanying drawings, this does not require or imply that the operations must be performed in this particular order, or that all illustrated operations must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0180] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0181] The present application is described with reference to the flowcharts and block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowcharts and block diagrams, as well as the combination of processes and boxes in the flowcharts and block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts. Figure 1 A process or multiple processes and boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0182] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and boxes Figure 1 The function specified in one or more boxes.

[0183] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 A process or multiple processes and boxes Figure 1 A step that specifies a function in one or more boxes.

[0184] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0185] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

Claims

1. An intelligent forklift control method, characterized in that: Applied to an intelligent forklift, the intelligent forklift comprises a vehicle body (1), a door frame (2), a hydraulic cylinder (3), a TOF camera assembly (4), a transport component (5) and a main control board, wherein: The hydraulic cylinder (3) is fixed to the vehicle body (1) and drives the gantry (2) to move up and down along the vehicle body (1); the transport component comprises a fixed fork leg and a lifting fork leg, the lifting fork leg is fixed to the bottom of the gantry (2), and the fixed fork leg is fixed to the bottom of the vehicle body (1); A first marking point is provided on each fixed fork leg, and a second marking point is provided on each lifting fork leg; The TOF camera assembly (4) comprises an upper hanging plate (41), a special-shaped bracket (44) and a TOF camera (46); the upper hanging plate (41) is fixed to the door frame (2); the special-shaped bracket (44) slides up and down along the upper hanging plate (41) and is fixed to the upper hanging plate (41) when sliding down to a specified position; and the TOF camera (46) is fixed to the special-shaped bracket (44); The main control board is used to control the TOF camera assembly (4), and after controlling the TOF camera assembly (4) to capture a target image, identifies the fixed fork leg in the target image according to the first marking point, and identifies the lifting fork leg in the target image according to the second marking point; The method comprises: Get the target image captured by the TOF camera; determining a first boundary according to a first marking point in the target image, and determining a second boundary according to a second marking point in the target image, wherein the first boundary and the second boundary are parallel; Correcting the target image based on the first boundary and the second boundary; The corrected target image is detected using a detection algorithm, and the intelligent forklift is controlled when a human figure is detected.

2. The method according to claim 1, characterized in that The TOF camera assembly (4) further comprises a linear guide rail (42) and a slider (43), wherein: The linear guide rail (42) is fixed on the upper hanging plate (41); The slider (43) is fixed on the special-shaped bracket (44) and moves up and down in the linear guide rail (42), driving the special-shaped bracket (44) to slide up and down along the upper hanging plate (41).

3. The method according to claim 2, characterized in that Balls are installed inside the slider (43), and the balls roll in the raceway, driving the slider (43) to move in the linear guide rail (42).

4. The method according to claim 1, wherein The TOF camera assembly (4) further includes a ball screw (45), wherein: The bottom of the upper hanging plate (41) is provided with a flange and a positioning hole, and the positioning hole is located on the flange; The ball screw (45) is fixed to a side of the special-shaped bracket (44) close to the upper hanging plate (41), and the steel ball on the side of the ball screw (45) close to the positioning hole is expanded and contracted under the action of a spring and an external force. When the special-shaped bracket (44) slides down along the upper hanging plate (41) to a position where the ball screw (45) is aligned with the positioning hole, the ball screw (45) is fixed in the positioning hole, so that the special-shaped bracket (44) is fixed on the upper hanging plate (41).

5. The method according to claim 4, characterized in that A limited edge is provided on one side of the special-shaped bracket (44) close to the upper hanging plate (41); when the special-shaped bracket (44) slides down to the limited edge and contacts the vehicle body (1), the ball screw (45) disengages from the positioning hole, thereby releasing the fixation between the special-shaped bracket (44) and the upper hanging plate (41).

6. The method according to claim 1, characterized in that Determining a first boundary according to a first marking point in the target image, and determining a second boundary according to a second marking point in the target image, comprising: Determining an image recognition base point according to the first marking point and the second marking point in the target image; Based on the image recognition base point, coordinate transformation is performed on the target image to obtain a transformed image in a pixel coordinate system corresponding to the target image; Determining position information of a first marking point and position information of a second marking point in the converted image; A first boundary is determined according to position information of a first marking point in the converted image, and a second boundary is determined according to position information of a second marking point in the converted image.

7. The method according to claim 6, characterized in that Correcting the target image based on the first boundary and the second boundary includes: determining a hybrid correction coefficient based on a first distance between the first boundary and the second boundary, wherein the hybrid correction coefficient is proportional to the first distance; The position coordinates of each pixel in the target image are corrected according to the mixed correction coefficient.

8. The method according to claim 1, characterized in that Performing human figure recognition on the corrected target image using a detection algorithm, and controlling the intelligent forklift when a human figure is recognized, including: determining an area between the first boundary and the second boundary in the corrected target image as a key area; Determine whether there is a circular area in the key area using a detection algorithm; When the circular area exists in the key area, it is determined that a human figure is recognized, and the intelligent forklift is controlled.

9. The method according to claim 8, characterized in that Controlling the intelligent forklift when a human figure is identified includes: When it is determined that a human figure is recognized, calculating a second distance between the circular area and the intelligent forklift; When it is determined that the second distance is less than a preset threshold, the intelligent forklift is controlled to stop working and an alarm is triggered; When it is determined that the second distance is not less than a preset threshold, a message reminder is triggered.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

Images