Visual perception-based surgical instrument classification method

Abstract

The invention relates to the field of intelligent sorting, in particular to a surgical instrument classification method based on visual perception, and the method comprises the steps: obtaining a target path of a carrying task needing to be executed by a carrying mechanical arm, controlling the carrying mechanical arm to carry out sorting verification based on the target path for various target sorting instruments, and carrying out the morphological analysis; the method comprises the following steps: determining a stress interference section and stress interference characteristics according to a single-class target sorting instrument, further determining a carrying stress class of the single-class target sorting instrument, and subsequently monitoring and controlling a carrying mechanical arm on the basis of the carrying stress class aiming at the carrying stress class of the target sorting instrument. The operation efficiency of the carrying mechanical arm is guaranteed, the monitoring precision is guaranteed, then the surgical instruments are placed and arranged, the service life of the reused diagnosis and treatment instruments is prolonged, and use safety and intact functions of the reused diagnosis and treatment instruments are guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent sorting, in particular to a surgical instrument classification method based on visual perception. BACKGROUND

[0002] In the process of recycling, classifying and cleaning surgical instruments, the types of instruments are diverse, so classification is needed before centralized mechanical cleaning, disinfection or manual cleaning, disinfection, and then the surgical instruments are placed and arranged to prolong the service life of reusable medical instruments and ensure the safe use and good function of reusable medical instruments. Therefore, the sorting technology of surgical instruments is valued by people.

[0003] For example, Chinese patent publication No. CN114535105A discloses an intelligent centralized sorting system for surgical instruments based on image recognition, which includes a conveying assembly, an identification assembly, a sorting assembly, a first storage box and a second storage box. The conveying assembly includes a support, a conveying line and a plurality of hooks. The identification assembly includes a detection frame, a plurality of cameras and a comparator. The sorting assembly includes a mover, a lifter and a jaw. The cleaned instruments are hung on the hooks, the conveying line drives the hooks and the instruments to move, the cameras obtain pictures of the cleaned instruments from different angles, the comparator is connected with the plurality of cameras and is used for identifying the detection images obtained by the cameras with standard images, the jaw can grab the instruments after detection, the mover drives the lifter to place the instruments, and the cleaning status of the instruments can be automatically detected to improve work efficiency.

[0004] However, the existing technology still has the following problems, The traditional sorting task execution mode for instruments seriously depends on preprogrammed fixed paths. However, such pre-set paths lack adaptive adjustment capability when facing different types of target sorting instruments. For example, surgical instruments are mostly irregular shaped parts, and the mass difference of different parts can be large. When the conveying path is complex, the influence of inertia is superimposed, which causes the conveying robot arm to easily produce swing, stress mutation and other abnormalities, which may cause the goods to fall off, affect the deviation monitoring of the conveying robot arm, and reduce the positioning accuracy. SUMMARY

[0005] Therefore, the present application provides a surgical instrument classification method based on visual perception to overcome the problem that in the existing technology, surgical instruments are mostly irregular shaped parts, and the mass difference of different parts can be large. When the conveying path is complex, the influence of inertia is superimposed, which causes the conveying robot arm to easily produce swing, stress mutation and other abnormalities, which may cause the goods to fall off, affect the deviation monitoring of the conveying robot arm, and reduce the positioning accuracy.

[0006] To achieve the above purpose, the present application provides a surgical instrument classification method based on visual perception, which comprises: Obtain the target path for the handling robot arm to perform the handling task; The sorting verification of various target sorting devices is carried out based on target path control of the handling robot arm. This includes handling a single type of target sorting device based on the target path, identifying several force interference segments in the target path, and obtaining the force interference characteristics of the single type of target sorting device in the force interference segments. Morphological analysis is performed on various target sorting devices, including determining the morphological center, constructing a rotation reference line, and determining the maximum contour difference between the two sides of the rotation reference line during rotation. Based on the force interference characteristics and maximum contour difference of single-target sorting equipment, the handling force characterization value is calculated to determine the handling force category of single-target sorting equipment. In response to the target sorting equipment that needs to be moved, the force category of the target sorting equipment is determined, and the handling robot arm is monitored and controlled based on the force category; The handling parameters of the handling robot are adjusted based on the force interference characteristics, a verification space surrounding the force interference segment is constructed based on the force interference segment, and the abnormality of the handling robot is detected based on the relative positional relationship between the verification space and the real-time spatial coordinates of the end of the handling robot. The force interference characteristics include the swing amplitude and stress fluctuation value, and the range of the verification space is determined based on the transport force characterization value corresponding to the force interference segment.

[0007] Furthermore, the process of identifying several force interference segments in the target path includes, Determine the end point and inflection point of the trajectory in the target path; Determine the trajectory segment of a predetermined length before reaching the end point of the trajectory; Determine the trajectory segment of predetermined length before and after the trajectory inflection point; Each of the aforementioned trajectory segments is identified as a force interference segment.

[0008] Furthermore, the process of obtaining the force interference characteristics of a single-type target sorting device in the force interference segment includes, The force amplitude at the end of the handling robotic arm is collected in real time when the target sorting device is moved to the corresponding force interference section, and the variance of the force amplitude is determined as the stress fluctuation value. The real-time coordinates of the end of the handling robot arm are collected when the target sorting device is moved to the corresponding force interference section, and the swing amplitude of the end of the robot arm is determined based on the real-time coordinates.

[0009] Furthermore, the process of determining the maximum contour difference between the two sides during the rotation of the rotation reference line includes, Acquire image data corresponding to the target sorting device, extract the outline of the target sorting device, and determine the corresponding shape center; A virtual rotation reference line is constructed with the shape center, and the rotation reference line is rotated with the shape center. Determine the ratio of the area difference of several contours on both sides when the rotation reference line is rotated to different positions; The maximum contour area difference ratio is defined as the maximum contour difference.

[0010] Furthermore, based on the force interference characteristics and maximum contour difference of a single-type target sorting device, the process of calculating the handling force characterization value includes: Based on the force interference characteristics and maximum contour differences of single-type target sorting equipment, the process of calculating the handling force characterization value includes: Determine the force interference characteristics corresponding to each force interference segment, and calculate the average swing amplitude and the average stress fluctuation value; The ratio of the average swing amplitude to the swing amplitude threshold is calculated as the first force interference factor; The ratio of the average stress fluctuation value to the stress fluctuation threshold is calculated as the second force interference factor; The ratio of the maximum contour difference to the predetermined difference threshold is calculated as the third force interference factor; The weighted summation of the first force interference factor, the second force interference factor, and the third force interference factor yields the transport force characterization value.

[0011] Furthermore, the process of determining the handling force category of a single-target sorting device includes, If the handling force characterization value corresponding to the single-type target sorting device is greater than or equal to the preset handling force threshold, it is determined to be a discrete handling force category. If the handling force characterization value corresponding to the single-type target sorting device is less than the preset handling force threshold, it is determined to be a non-discrete handling force category.

[0012] Furthermore, the handling force category of the target sorting device is determined, and the handling robotic arm is monitored and controlled based on the handling force category. If the handling force category is discrete handling force category, then monitoring and control of the handling robot arm will be performed.

[0013] Furthermore, adjusting the handling parameters of the handling robot arm based on the force interference characteristics includes: The handling speed of the robotic arm is reduced, and the reduction amount is positively correlated with the handling force characterization value.

[0014] Furthermore, the process of constructing a verification space surrounding the force interference segment includes, A cylindrical space is generated based on the force interference segment, and the cylindrical space is determined as the verification space. The center of any cross section of the cylindrical space perpendicular to the force interference segment is a circle, and the center of each circle is on the force interference segment. The range of the verification space is positively correlated with the transport force characterization value.

[0015] Furthermore, the process of detecting whether there are anomalies in the handling robot arm based on the relative positional relationship between the verification space and the real-time spatial coordinates of the end effector of the handling robot arm includes, Determine the set of coordinates corresponding to the verification space. If the real-time spatial coordinates of the end effector of the handling robot are not within the set of coordinates, it is determined that the handling robot has an anomaly.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: by obtaining the target path of the handling robot arm to perform the handling task, the handling robot arm is controlled based on the target path to perform sorting verification for various target sorting devices, the force interference segment and force interference characteristics are determined, and the handling force category of a single type of target sorting device is further determined. Subsequently, based on the handling force category of the target sorting device, the handling robot arm is monitored and controlled. The present invention adaptively adjusts the handling parameters and detection strategies to address the impact of inertia and internal shaking of the target sorting device on the handling process and monitoring accuracy when performing the handling task, thereby ensuring the operating efficiency of the handling robot arm, ensuring the monitoring accuracy of the handling robot arm, and improving the stability and reliability of the handling robot arm's working process.

[0017] In particular, this invention uses a target path-controlled handling robot arm for handling verification and monitoring to determine the force interference segment. In reality, the force interference segment is determined based on the trajectory end point and trajectory inflection point. Due to the differences in the shape of the target sorting equipment, different parts may have different masses, resulting in significant differences due to inertia. This situation does not occur uniformly along the entire target path, but is most severe at the end of acceleration / deceleration (corresponding to the trajectory end point) and the moment of change of direction (corresponding to the trajectory inflection point). Based on this, the force interference characteristics of the force interference segment are obtained, reflecting the impact of the differences in the shape of the equipment to be handled and the inertia superposition process on the handling robot arm during the handling process. This provides data support for subsequent classification of handling force categories, thereby ensuring the operating efficiency of the handling robot arm, ensuring the monitoring accuracy of the handling robot arm, and improving the stability and reliability of the handling robot arm's working process.

[0018] In particular, this invention determines the force interference characteristics of the force interference segment and quantitatively collects two key features: swing amplitude and stress fluctuation value. The swing amplitude directly reflects the spatial drift of the end effector of the robotic arm and is the most intuitive manifestation of swaying. The stress fluctuation value (calculated by the variance of the force amplitude) reveals the degree of drastic change in the dynamic load borne by the internal structure of the robotic arm. The influence of the superposition relationship of the object's swaying on the handling robotic arm is quantified from two dimensions, thereby classifying the handling force categories. Subsequently, the handling robotic arm is monitored and controlled based on the handling force categories, thereby ensuring the operating efficiency of the handling robotic arm, ensuring the monitoring accuracy of the handling robotic arm, and improving the stability and reliability of the handling robotic arm's working process.

[0019] In particular, for discrete handling force categories, the significant differences in the shape of the target sorting equipment and the severe dynamic interference to the handling robot arm under the superimposed inertial influence cause disturbances at the end of the handling robot arm, affecting its accuracy and making it prone to errors in anomaly monitoring. Based on this, an adaptive verification space is constructed with a certain allowable error, allowing a certain amount of disturbance in the handling robot arm. This is used to determine whether there are anomalies in the handling robot arm, reducing false triggers and improving the reliability of anomaly detection. At the same time, the handling parameters of the handling robot arm are adaptively adjusted so that the handling robot arm can adapt to the corresponding type of target sorting equipment, reducing the impact on the handling robot arm during the handling process, thereby ensuring the operating efficiency of the handling robot arm, ensuring the monitoring accuracy of the handling robot arm, and improving the stability and reliability of the handling robot arm's working process. Attached Figure Description

[0020] Figure 1 This is a schematic diagram illustrating the steps of a visual perception-based surgical instrument classification method according to an embodiment of the invention. Figure 2 A logic block diagram for determining the handling force category of a single-target sorting device according to an embodiment of the invention; Figure 3 This is a logic block diagram illustrating the monitoring and control of a handling robot arm based on the type of handling force, as described in an embodiment of the invention. Figure 4 This is a logic block diagram for detecting whether a handling robot arm has an abnormality, according to an embodiment of the invention. Detailed Implementation

[0021] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0022] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0023] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0024] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the term "connection" should be interpreted broadly. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0025] Please see Figure 1 The diagram illustrates the steps of a visual perception-based surgical instrument classification method according to an embodiment of the present invention. The visual perception-based surgical instrument classification method according to an embodiment of the present invention includes: Step S1: Obtain the target path for the handling robot arm to perform the handling task; Step S2: Perform sorting verification for various target sorting devices by controlling the handling robot arm based on the target path, including handling a single type of target sorting device based on the target path, determining several force interference segments in the target path, and obtaining the force interference characteristics of the single type of target sorting device in the force interference segments. Step S3 involves performing morphological analysis on various target sorting devices, including determining the morphological center, constructing a rotation reference line, and determining the maximum contour difference between the two sides of the rotation reference line during rotation. Step S4: Based on the force interference characteristics and maximum contour difference of the single-type target sorting device, calculate the handling force characterization value and determine the handling force category of the single-type target sorting device. Step S5: In response to the target sorting device that needs to be moved, determine the force category of the target sorting device and monitor and control the handling robot arm based on the force category. Step S6: Adjust the handling parameters of the handling robot arm based on the force interference characteristics, construct a verification space surrounding the force interference segment based on the force interference segment, and detect whether there is any abnormality in the handling robot arm based on the relative positional relationship between the verification space and the real-time spatial coordinates of the end of the handling robot arm. The force interference characteristics include the swing amplitude and stress fluctuation value, and the range of the verification space is determined based on the transport force characterization value corresponding to the force interference segment.

[0026] Specifically, the specific structure of the handling robotic arm is not limited. It can be a robotic arm with a gripping structure at the end. The degree of freedom and operation mode of the robotic arm can be selected by those skilled in the art according to the site environment, which will not be elaborated here.

[0027] Specifically, the target path for the required handling task is predetermined. The handling task includes moving the target sorting device to the storage location. The movement path of the robotic arm end effector during the handling process can be detected to obtain the target path, which will not be elaborated further.

[0028] Specifically, there is no limitation on the classification method of target sorting equipment. In actual production, it is usually necessary to grasp multiple types of target sorting equipment. Based on this, sorting equipment with the same target are grouped into one category for easy centralized analysis, which will not be elaborated further.

[0029] Specifically, the process of determining several force interference segments in the target path includes, Determine the end point and inflection point of the trajectory in the target path; Determine the trajectory segment of a predetermined length before reaching the end point of the trajectory; Determine the trajectory segment of predetermined length before and after the trajectory inflection point; Each of the aforementioned trajectory segments is identified as a force interference segment.

[0030] In practice, the predetermined length is determined in advance. The purpose is to select the trajectory end point and the trajectory inflection point near the trajectory, so as to facilitate the observation of the situation when the end of the handling robot moves to the corresponding trajectory segment. In practice, the predetermined length is set to 0.05 times the total length of the target path.

[0031] This invention utilizes a target path-controlled handling robot arm for handling verification and monitoring, identifying force interference segments. In practice, these segments are determined based on the trajectory end point and trajectory inflection point. Due to the differences in the shape of the target sorting equipment, different parts may have varying masses, resulting in significant differences due to inertia. This situation does not occur uniformly along the entire target path but is most pronounced at the endpoints of acceleration / deceleration (corresponding to the trajectory end point) and at the instants of direction change (corresponding to the trajectory inflection point). Based on this, the force interference characteristics of the force interference segments are obtained, reflecting the impact of the differences in the shape of the equipment to be handled and the inertial superposition process on the handling robot arm during handling. This provides data support for subsequent classification of handling force categories, thereby ensuring the operating efficiency and monitoring accuracy of the handling robot arm, and improving the stability and reliability of the handling robot arm's operation.

[0032] Specifically, the process of obtaining the force interference characteristics of a single-type target sorting device in the force interference segment includes, The force amplitude at the end of the handling robotic arm is collected in real time when the target sorting device is moved to the corresponding force interference section, and the variance of the force amplitude is determined as the stress fluctuation value. The real-time coordinates of the end of the handling robot arm are collected when the target sorting device is moved to the corresponding force interference section, and the swing amplitude of the end of the robot arm is determined based on the real-time coordinates.

[0033] Specifically, there is no limitation on the method of monitoring the force amplitude at the end of the robotic arm. In practice, the end of the robotic arm is usually an end effector, such as a gripper or welding torch. Therefore, a six-dimensional force sensor can be set at the movable joint near the end of the robotic arm to monitor the force on the end of the robotic arm in multiple directions. When calculating the force amplitude, the resultant force in multiple directions can be calculated and used as the force amplitude. This will not be elaborated further.

[0034] Specifically, the process of determining the maximum contour difference between the two sides during the rotation of the rotation reference line includes, Acquire image data corresponding to the target sorting device, extract the outline of the target sorting device, and determine the corresponding shape center; A virtual rotation reference line is constructed with the shape center, and the rotation reference line is rotated with the shape center. Determine the ratio of the area difference of several contours on both sides when the rotation reference line is rotated to different positions; The maximum contour area difference ratio is defined as the maximum contour difference.

[0035] Specifically, the image data can be two-dimensional images. It is only necessary to consider the overall outline of the surgical instruments. There is no limitation on the method of determining the shape center. The corresponding geometric center can be identified after the outline of the target sorting instrument is identified. This will not be elaborated further.

[0036] It is understandable that the rotation reference line is a virtual straight line that passes through the center of the shape and then rotates around the center of the shape. When it rotates to different positions, there are contours on both sides of the rotation reference line, so the difference ratio of the contour area on both sides can be calculated.

[0037] Specifically, there is no limitation on the method for determining the swing amplitude of the robotic arm's end effector. The swing behavior generated by the robotic arm can be captured based on the real-time coordinates of the end effector, and the swing amplitude can then be determined. This will not be elaborated further.

[0038] Specifically, the process of calculating the handling force characterization value based on the force interference characteristics and maximum contour difference of a single-type target sorting device includes: Based on the force interference characteristics and maximum contour differences of single-type target sorting equipment, the process of calculating the handling force characterization value includes: Determine the force interference characteristics corresponding to each force interference segment, and calculate the average swing amplitude and the average stress fluctuation value; The ratio of the average swing amplitude to the swing amplitude threshold is calculated as the first force interference factor; The ratio of the average stress fluctuation value to the stress fluctuation threshold is calculated as the second force interference factor; The ratio of the maximum contour difference to the predetermined difference threshold is calculated as the third force interference factor; The weighted summation of the first force interference factor, the second force interference factor, and the third force interference factor yields the transport force characterization value.

[0039] The swing amplitude threshold and stress fluctuation threshold are predetermined, wherein, Pre-determine several operation records of the handling robot under abnormal working conditions, including the force interference characteristics of several force interference segments during operation; The mean swing amplitude and the mean stress fluctuation value are calculated. The swing amplitude threshold is set as the product of the mean swing amplitude and the error coefficient, and the stress fluctuation threshold is set as the product of the mean stress fluctuation value and the error coefficient. The error coefficient is selected in the range [0.85, 0.95], preferably 0.9.

[0040] The predetermined difference threshold is set in advance to reflect situations where there is a large difference in the area of ​​the two contours, so as to reflect the situation where the equipment to be transported may be seriously deflected to one side due to inertia during the actual handling process. In practice, the predetermined difference threshold is selected in the range [0.5, 0.8], preferably 0.6.

[0041] When performing weighted summation, to comprehensively consider all factors, the weights for the first force interference factor, the second force interference factor, and the third force interference factor are 0.3, 0.3, and 0.4, respectively.

[0042] Specifically, please refer to Figure 2 As shown, Figure 2 The present invention provides a logic block diagram for determining the handling force category of a single-target sorting device according to an embodiment of the invention. The process for determining the handling force category of a single-target sorting device includes: If the handling force characterization value corresponding to the single-type target sorting device is greater than or equal to the preset handling force threshold, it is determined to be a discrete handling force category. If the handling force characterization value corresponding to the single-type target sorting device is less than the preset handling force threshold, it is determined to be a non-discrete handling force category.

[0043] The handling force threshold is selected within the range [1.15, 1.3].

[0044] This invention identifies the force interference characteristics of the force interference segment and quantitatively collects two key features: swing amplitude and stress fluctuation value. The swing amplitude directly reflects the spatial drift of the robotic arm's end effector and is the most intuitive manifestation of swaying. The stress fluctuation value (calculated by the variance of the force amplitude) reveals the degree of drastic change in the dynamic load borne by the internal structure of the robotic arm. The influence of the superposition relationship of object swaying on the handling robotic arm is quantified from two dimensions, thereby classifying the handling force categories. Subsequently, the handling robotic arm is monitored and controlled based on the handling force categories, thereby ensuring the operating efficiency of the handling robotic arm, ensuring the monitoring accuracy of the handling robotic arm, and improving the stability and reliability of the handling robotic arm's working process.

[0045] Specifically, please refer to Figure 3 As shown, Figure 3 This is a logic block diagram illustrating the monitoring and control of a robotic arm based on the type of handling force according to an embodiment of the invention. The process involves determining the handling force type of the target sorting device and monitoring and controlling the robotic arm based on that force type. If the handling force category is discrete handling force category, then monitoring and control of the handling robot arm will be performed.

[0046] Specifically, adjusting the handling parameters of the handling robot arm based on the force interference characteristics includes: The handling speed of the robotic arm is reduced, and the reduction amount is positively correlated with the handling force characterization value.

[0047] In implementation, optional, The ratio of the difference between the characteristic value of the handling force and the threshold value of the handling force is used as the speed adjustment coefficient; The product of the initial speed and the speed adjustment coefficient is determined as the reduction in transport speed.

[0048] Understandably, the transport speed is the speed at which the robotic arm moves after grasping the target sorting device, and the difference ratio is the ratio of the difference between two values ​​to the mean of the two values.

[0049] Specifically, the process of constructing a verification space surrounding the force interference segment includes: A cylindrical space is generated based on the force interference segment, and the cylindrical space is determined as the verification space. The center of any cross section of the cylindrical space perpendicular to the force interference segment is a circle, and the center of each circle is on the force interference segment. The range of the verification space is positively correlated with the transport force characterization value.

[0050] It is understandable that a circle can be created at the beginning of the force interference segment, and the circle can be moved along the force interference segment. During the movement, the circle is always perpendicular to the force interference segment, and thus the corresponding verification space can be obtained. Of course, other methods can also be used, as long as they can meet the corresponding requirements.

[0051] In implementation, optional, The range of the verification space can be defined by adjusting the diameter of the cylindrical space section. The larger the diameter, the larger the range. In practice, the ratio of the handling force characterization value to the handling force threshold can be calculated as the range adjustment coefficient. The product of the reference diameter and the range adjustment factor is calculated to obtain the adjusted diameter, and then the range of the verification space is determined. In practice, the volume of the verification space can be regarded as the range, which will not be elaborated further.

[0052] The reference diameter is determined based on the swing amplitude threshold and is set to 1.1 times the swing amplitude threshold.

[0053] Specifically, please refer to Figure 4 As shown, Figure 4 This is a logic block diagram illustrating the detection of abnormalities in a robotic arm according to an embodiment of the invention. The process of detecting abnormalities in the robotic arm based on the relative positional relationship between the verification space and the real-time spatial coordinates of the robotic arm's end effector includes: Determine the set of coordinates corresponding to the verification space. If the real-time spatial coordinates of the end effector of the handling robot are not within the set of coordinates, it is determined that the handling robot has an anomaly.

[0054] For discrete handling force categories, the characteristics of target sorting equipment vary greatly in shape, and the dynamic interference to the handling robot arm under the superimposed inertial influence is quite serious, causing disturbance at the end of the handling robot arm, affecting the accuracy of the handling robot arm, and easily leading to errors in the monitoring of abnormalities of the handling robot arm. Based on this, an adaptive verification space is constructed, and a certain allowable error is set to allow a certain amount of disturbance in the handling robot arm. This is used to determine whether there is an abnormality in the handling robot arm, reducing false triggers and improving the reliability of abnormality detection. At the same time, the handling parameters of the handling robot arm are adaptively adjusted so that the handling robot arm can adapt to the corresponding type of target sorting equipment, reducing the impact on the handling robot arm during the handling process, thereby ensuring the operating efficiency of the handling robot arm, ensuring the monitoring accuracy of the handling robot arm, and improving the stability and reliability of the handling robot arm's working process.

[0055] The technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0056] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A visual perception-based surgical instrument classification method, characterized in that, acquiring a target path required for a carrying manipulator to perform a carrying task; controlling the carrying manipulator based on the target path to verify each type of target sorting instrument, including carrying single-type target sorting instruments based on the target path, determining a plurality of force interference sections in the target path, and acquiring force interference characteristics of the single-type target sorting instruments in the force interference sections; performing morphology analysis on each type of target sorting instrument, including determining a morphology center, constructing a rotating reference line, and determining the maximum profile difference on both sides of the rotating reference line during rotation; calculating a carrying force representation value based on the force interference characteristics and the maximum profile difference of the single-type target sorting instruments to determine the carrying force category of the single-type target sorting instruments; in response to a target sorting instrument needing to be carried, determining the carrying force category of the target sorting instrument, and monitoring and controlling the carrying manipulator based on the carrying force category; adjusting the carrying parameters of the carrying manipulator based on the force interference characteristics, constructing a verification space surrounding the force interference sections based on the force interference sections, and detecting whether the carrying manipulator is abnormal based on the relative position relationship between the verification space and the real-time spatial coordinates of the end of the carrying manipulator; wherein the force interference characteristics include a swing amplitude and a stress fluctuation value, and the range of the verification space is determined based on the carrying force representation value corresponding to the force interference section.

2. The visual perception based surgical instrument classification method of claim 1, wherein, The process of determining a plurality of force interference sections in the target path includes, determining the end point and inflection point of the trajectory in the target path; determining a trajectory section of a predetermined length before reaching the end point of the trajectory; determining a trajectory section of a predetermined length before and after the inflection point of the trajectory; determining each trajectory section as a force interference section.

3. The visual perception based surgical instrument classification method of claim 2, wherein, The process of acquiring the force interference characteristics of the single-type target sorting instruments in the force interference sections includes, real-time acquisition of the force amplitude of the end of the carrying manipulator when the target sorting instrument is carried to the corresponding force interference section, and determination of the variance of the force amplitude as the stress fluctuation value; real-time acquisition of the real-time coordinates of the end of the carrying manipulator when the target sorting instrument is carried to the corresponding force interference section, and determination of the swing amplitude of the end of the carrying manipulator based on the real-time coordinates.

4. The visual perception based surgical instrument classification method of claim 1, wherein, The process of determining the maximum profile difference on both sides of the rotating reference line during rotation includes, acquiring image data corresponding to the target sorting instrument, extracting the profile of the target sorting instrument to determine the corresponding morphology center; constructing a virtual rotating reference line with the morphology center, and rotating the rotating reference line with the morphology center; determining a plurality of profile area difference ratios on both sides of the rotating reference line when rotated to different positions; determining the maximum profile area difference ratio as the maximum profile difference.

5. The visual perception based surgical instrument classification method of claim 4, wherein, The process of calculating the carrying force representation value based on the force interference characteristics and the maximum profile difference of the single-type target sorting instruments includes, determining the force interference characteristics corresponding to each force interference section, calculating the mean value of the swing amplitude and the mean value of the stress fluctuation value; calculating the ratio of the mean value of the swing amplitude to the swing amplitude threshold value as the first force interference factor; calculating the ratio of the mean value of the stress fluctuation value to the stress fluctuation threshold value as the second force interference factor; calculating the ratio of the maximum profile difference to the predetermined difference threshold value as the third force interference factor; The first stress interference factor, the second stress interference factor, and the third stress interference factor are weighted and summed to obtain a carrying stress characteristic value.

6. The visual perception based surgical instrument classification method of claim 5, wherein, The process of determining the carrying stress category of the single-class target sorting device includes, If the carrying stress characteristic value corresponding to the single-class target sorting device is greater than or equal to a preset carrying stress threshold, it is determined as a discrete carrying stress category. If the carrying stress characteristic value corresponding to the single-class target sorting device is less than the preset carrying stress threshold, it is determined as a non-discrete carrying stress category.

7. The visual perception based surgical instrument classification method of claim 1, wherein, The carrying stress category of the target sorting device is determined, and the carrying mechanical arm is monitored and controlled based on the carrying stress category, wherein If the carrying stress category is a discrete carrying stress category, the carrying mechanical arm is monitored and controlled.

8. The visual perception based surgical instrument classification method of claim 1, wherein, The carrying parameters of the carrying mechanical arm are adjusted based on the stress interference characteristics, including The carrying speed of the carrying mechanical arm is reduced, and the reduction amount is positively correlated with the carrying stress characteristic value.

9. The visual perception based surgical instrument classification method of claim 1, wherein, The process of constructing a verification space surrounding the stress interference segment based on the stress interference segment includes A columnar space is generated based on the stress interference segment, the columnar space is determined as the verification space, the columnar space is the center of any cross section perpendicular to the stress interference segment, and each center is on the stress interference segment. The range of the verification space is positively correlated with the carrying stress characteristic value.

10. The visual perception based surgical instrument classification method of claim 1, wherein, The process of detecting whether the carrying mechanical arm is abnormal based on the relative position relationship between the verification space and the real-time spatial coordinates of the end of the carrying mechanical arm includes A coordinate set corresponding to the verification space is determined, and if the real-time spatial coordinates of the end of the carrying mechanical arm are not in the coordinate set, it is determined that the carrying mechanical arm is abnormal.

Images