Crane intelligent examination system based on vision

Abstract

The invention discloses an intelligent crane examination system based on vision, relates to the technical field of image processing, and solves the technical problems that a traditional crane examination system mainly depends on manual evaluation, is low in efficiency and is easily influenced by human subjective factors, so that the evaluation result is not fair and accurate enough. Identity information of examinees is verified; collecting a mark image in the process that the examinee transports the moving goods from the starting position to the preset warehousing position through the crane; on the basis of the mark image and the assessment indexes of the crane, obtaining various assessment indexes of the examinee; scoring each examination index of the examinee to obtain an examination result of the examinee; according to the method, the examination results of the examinees are obtained by automatically scoring the various examination indexes of the examinees, so that the evaluation efficiency is improved, subjectivity and unfairness possibly caused by artificial scoring are avoided, and the objectivity and accuracy of the evaluation results are ensured.

Description

Technical Field

[0001] The invention belongs to the field of image processing, and in particular is a vision-based intelligent crane testing system. Background Art

[0002] With the rapid development of modern industrial technology, cranes, as an important material handling equipment, have been widely used in industrial production, logistics warehousing, port terminals and other fields. However, the operation of cranes requires high professional skills and a rigorous attitude. Once improperly operated, it is very easy to cause safety accidents. Therefore, the training and assessment of crane operators is particularly important.

[0003] The traditional crane test system mainly relies on manual evaluation, which is not only inefficient but also easily affected by human subjective factors, resulting in unfair and inaccurate evaluation results. Therefore, the present invention provides a vision-based intelligent crane test system. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a vision-based intelligent crane testing system to solve the technical problem that the traditional crane testing system mainly relies on manual judgment, has low efficiency and is easily affected by human subjective factors, resulting in unfair and inaccurate judgment results.

[0005] To achieve the above-mentioned object, the first aspect of the present invention provides a vision-based intelligent crane examination system, comprising an information collection module, and an identity verification module and a scoring module connected thereto; Identity verification module: used to verify the candidate's identity information; Information collection module: used to collect images of markers during the test and obtain marked images; based on the marked images and assessment indicators, obtain the results of various assessment indicators of the examinee; the assessment indicators include moving speed, swing amplitude and storage results; the storage results include storage failure and storage success; It should be noted that: the markers are moving goods, sliding components, lifting ropes and hooks; Scoring module: used to score the various assessment indicators of candidates and obtain the assessment results of candidates based on the scores of various assessment indicators.

[0006] Preferably, the verification of the examinee's identity information includes: Obtain the candidate's identity information and capture the candidate's facial image through the camera in the examination area to verify whether the candidate's identity information, the captured facial image and the registration information are consistent; if yes, the candidate's identity authentication is successful; if not, the candidate's identity authentication is failed.

[0007] The present invention reads identity information through the ID card recognition system, ensuring that the ID card information submitted by the examinee can be accurately matched with the registration information in the database; the face image acquisition and comparison technology further enhances the accuracy of authentication, because even if the identity information is forged, the facial features are difficult to copy, which greatly reduces the possibility of taking the exam on behalf of others. The multiple verification mechanism provides higher security for the exam. Any mismatch will trigger an alarm and prevent the examinee from entering the examination room, thereby effectively preventing cheating. Identity authentication is achieved through technical means, which reduces human intervention and reduces security issues caused by human errors or negligence.

[0008] Preferably, the collecting of images of the crane during the test includes: The camera collects images of the marker during its movement in real time, and divides the marker images according to movement characteristics according to the movement direction of the crane; wherein the movement characteristics include vertical rising distance, vertical descending distance and horizontal moving distance; the camera is arranged on the front and side of the test area.

[0009] Preferably, dividing the marked images according to the movement characteristics according to the movement direction of the crane includes: The first mark position preset for moving cargo on the crane is obtained as a vertical movement reference point, and the second mark position preset for the sliding component on the crane is obtained as a horizontal movement reference point; Establishing a coordinate system of the marked image, extracting the coordinates of the vertical moving reference point and the horizontal moving reference point in the marked image; wherein the coordinate system of the marked image takes the central pixel point of the marked image as the origin of the coordinate axis, and a number of pixels as a coordinate unit; According to the time sequence, the distance between the vertical movement reference point and the horizontal movement reference point in two adjacent marked images is calculated, and the difference processing is performed to obtain the vertical movement distance value of the vertical movement reference point; when the vertical movement distance value between the two adjacent marked images is not 0, the marked image is marked as a vertical rising image or a vertical falling image; otherwise, the marked image is not marked; wherein, when the vertical movement distance value between the two adjacent marked images is less than 0, the marked image is marked as a vertical rising image; when the vertical movement distance value between the two adjacent marked images is greater than or equal to 0, the marked image is marked as a vertical falling image; and, when the coordinates of the horizontally moving reference point change in the horizontal direction, marking the marked image as a horizontal marked image; The marked images of the same moving feature are integrated to obtain a set of moving feature images, and the images in the set are arranged in chronological order; wherein the moving feature image set includes a vertically rising image set, a vertically falling image set and a horizontally marked image set; when a marked image is marked with more than one moving feature, the moving feature marked for the first or last time is retained.

[0010] The present invention can ensure a fixed reference point when analyzing a marked image by setting a vertical movement reference point and a horizontal movement reference point, thereby improving the accuracy of the analysis; and establish a coordinate system of the image, analyze the position changes of the vertical movement reference point and the horizontal movement reference point, and then identify the characteristics of the marked image, such as vertical rise, vertical descent or horizontal movement, etc., to lay the foundation for the subsequent analysis of various assessment indicators of the examinee through the various movement characteristics of the marked image.

[0011] Preferably, the method for obtaining the indicator result of the moving speed includes the following steps: Obtaining the initial image and the final image in each moving feature image, and extracting the coordinates of the vertical moving reference point and the horizontal moving reference point in the initial image and the final image respectively; wherein the initial image is the image of the moving feature when it starts to move; and the final image is the image of the moving feature when it ends to move; Extract the coordinates of the vertical moving reference point and the horizontal moving reference point of the initial image in the vertical rising distance, calculate the distance between the vertical moving reference point and the horizontal moving reference point in the initial image, and mark it as the initial distance; and extract the coordinates of the vertical moving reference point and the horizontal moving reference point of the termination image in the vertical rising distance, calculate the distance between the vertical moving reference point and the horizontal moving reference point in the termination image, and mark it as the termination distance; calculate the absolute value of the difference between the initial distance and the termination distance, and calculate the ratio of the absolute value of the difference to the acquisition time difference between the initial image and the termination image to obtain the moving speed of the moving goods rising vertically; Extract the coordinates of the vertical moving reference point and the horizontal moving reference point of the initial image in the vertical descent distance, calculate the distance between the vertical moving reference point and the horizontal moving reference point in the initial image, and mark it as the initial distance; and extract the coordinates of the vertical moving reference point and the horizontal moving reference point of the termination image in the vertical descent distance, calculate the distance between the vertical moving reference point and the horizontal moving reference point in the termination image, and mark it as the termination distance; calculate the absolute value of the difference between the initial distance and the termination distance, and calculate the ratio of the absolute value of the difference to the acquisition time difference between the initial image and the termination image to obtain the moving speed of the moving cargo in vertical descent; Calculate the distance between the horizontal movement reference point of the initial image of the horizontal movement and the horizontal movement reference point in the final image, and divide the calculated distance by the acquisition time difference between the initial image of the horizontal movement and the final image to obtain the moving speed of the moving cargo in the horizontal movement; The moving speed of the moving goods in each direction is compared with the preset speed threshold range, the moving speed that is not within the preset speed threshold range is marked as abnormal speed, and the number of abnormal speeds is counted.

[0012] The present invention calculates the initial distance and the terminal distance of the moving feature based on the coordinates of the vertical moving reference point and the horizontal moving reference point in the initial image and the terminal image, accurately reflects the movement of the moving goods in the vertical and horizontal directions, and divides the distance difference by the acquisition time difference to obtain the moving speed of the moving goods in each direction, and monitors the rising and falling speeds of the moving goods in the vertical direction and the horizontal direction in real time, thereby evaluating the examinee's ability to control the speed when operating a crane to move goods.

[0013] Preferably, the step of obtaining the initial image and the final image of each moving feature comprises: Obtaining the initial image and the final image in each moving feature image, including: Determine whether there are multiple moving features of the marked image. If yes, keep the moving features of the first mark and discard the moving features of the subsequent marks; if no, do not perform any operation; Mark the first marked image in the vertical ascent image set as the initial image of the vertical ascent, and mark the last image as the termination image of the vertical ascent; Mark the last marked image in the horizontal marked image set as the initial image of vertical descent, and mark the last image in the vertical descent image set as the termination image of vertical descent; Mark the ending image of vertical rise as the initial image of horizontal movement; mark the initial image of vertical fall as the ending image of horizontal movement; or, Determine whether there are multiple moving features of the marked image. If yes, keep the moving features marked last and discard the moving features marked later. If no, do not perform any operation. Mark the first marked image in the vertically ascending image set as the initial image of vertically ascending, and mark the first image in the horizontally marked image set as the terminal image of vertically ascending; The first marked image in the vertical descent image set is marked as the initial image of vertical descent, and the last image is marked as the termination image of vertical ascent; The ending image of vertical rise is marked as the initial image of horizontal movement; the initial image of vertical fall is marked as the ending image of horizontal movement.

[0014] When a marked image is marked with more than one moving feature, the moving feature marked for the first time or the last time shall prevail, thereby ensuring the consistency of processing and avoiding confusion caused by multiple markings; by clearly defining the initial and terminal images of each moving feature, the continuity between different moving features is achieved, which helps to analyze the complete path of the moving goods and lays the foundation for analyzing the assessment indicators.

[0015] Preferably, the method for obtaining the index result of the swing amplitude comprises the following steps: The angle between the crane's rope and the relative vertical line in the marked image is calculated to obtain the swing amplitude of the moving cargo when the crane is running, and the swing amplitude of the moving cargo is compared with the preset swing threshold range. The swing amplitude that is not within the preset swing threshold range is marked as abnormal swing, and the number of abnormal swings is counted; wherein the relative vertical line is a straight line in the marked image that passes through the horizontal moving reference point and is perpendicular to the X-axis.

[0016] The present invention can monitor the swing amplitude of the moving cargo in real time by calculating the angle between the crane's rope and the relative vertical line, reflecting the examinee's operating skill level; the swing amplitude of the moving cargo directly reflects the examinee's stability and accuracy when operating the crane, and is an important indicator for evaluating his or her operating skills. By monitoring the swing amplitude, the examinee's subtle differences in the operation process can be clearly understood, thereby more accurately evaluating his or her skill level.

[0017] Preferably, the method for obtaining the storage result includes the following steps: The edge information of the moving goods in the marked image and the edge information of the preset warehousing position are extracted respectively. When the edge information of the moving goods exceeds the edge information of the preset warehousing position, the moving status of the moving goods is marked as warehousing failure; when the edge information of the moving goods is within the edge information of the preset warehousing position, it is marked as warehousing success.

[0018] During the process of moving goods into storage, the present invention realizes accurate extraction and comparison of edge information of the moving goods and edge information of the preset storage position, thereby ensuring the accuracy of the evaluation result. Once the edge of the moving goods exceeds or presses on the edge of the preset storage position, it is judged as a storage failure, thereby promptly reflecting the deficiencies of the examinee in the storage operation skills.

[0019] Preferably, the step of scoring each assessment indicator of the examinee to obtain the examinee's assessment result includes: According to the scoring rules for each assessment indicator formulated by experts, the candidates' various assessment indicators are scored. The total score of each assessment indicator is the candidate's assessment score, which is compared with the passing score to obtain the final assessment result; among them, the candidate's assessment score will generate an electronic transcript; the electronic transcript includes the score of each assessment indicator and the total score; the assessment result is passed or failed.

[0020] The present invention ensures the uniformity and objectivity of the scoring criteria through the scoring rules formulated by experts. Each assessment indicator has a clear scoring standard, which reduces the unfairness caused by subjective judgment and makes the assessment results more fair and accurate; by scoring the various assessment indicators of the candidates, the skill level and comprehensive quality of the candidates can be fully and carefully understood. Each assessment indicator represents the ability of the candidate in different aspects, and the comprehensive score can more comprehensively reflect the actual level of the candidate; and an electronic transcript is generated to facilitate the management and query of the scores. The electronic transcript lists the scores and total scores of each assessment indicator in detail, so that the candidates can clearly understand their scores, which is helpful for the candidates to conduct self-assessment and reflection.

[0021] Preferably, the second aspect of the present invention provides a vision-based crane intelligence testing method, comprising the following steps: Step 1: Verify the candidate’s identity information; Step 2: Collect the marked images of the examinee in the process of transporting the mobile goods from the starting position to the preset storage position by the crane; based on the marked images of the crane and the assessment indicators, obtain the examinee's various assessment indicators; Step 3: Score each assessment indicator of the candidate and obtain the candidate's assessment result.

[0022] Compared with the prior art, the present invention has the following beneficial effects: The present invention firstly verifies the identity information of the examinee and verifies the successful entry into the examination stage, thereby ensuring the seriousness and security of the examination, effectively preventing cheating such as taking the examination on behalf of another examinee, and maintaining the fairness and justice of the examination; secondly, the camera collects the marked image of the crane during the examination in real time, providing real and intuitive data support for the subsequent evaluation of assessment indicators; then, based on the marked image and assessment indicators of the crane, various assessment indicators of the examinee, such as moving speed, swing amplitude and warehousing results, are obtained, thereby realizing the automation and intelligence of the assessment, which not only improves the assessment efficiency, but also reduces the interference of human factors, and ensures the objectivity and fairness of the assessment results; finally, various assessment indicators of the examinee are scored, and the assessment results of the examinee are obtained, which provides clear feedback to the examinee, helps the examinee understand his or her strengths and weaknesses in operating the crane, and points out the direction for subsequent skill improvement; the present invention automatically captures key information in the crane operation process, such as moving speed, swing amplitude and warehousing results, through visual technology, thereby realizing the automated evaluation of the examination, and greatly improving the evaluation efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 creative work.

[0024] Figure 1 It is a schematic diagram of the system framework structure of the present invention; Figure 2 It is a schematic diagram of the horizontal moving distance and the vertical moving distance of the present invention; Figure 3 It is a schematic diagram of the swing amplitude of the crane of the present invention during operation; Figure 4 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0025] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] See also Figure 1 , the first aspect of the present invention provides a vision-based crane intelligent examination system, including an information collection module, and an identity authentication module and a scoring module connected thereto; The identity verification module verifies the candidate’s identity information; Specifically, after entering the examination area, the candidate places his / her ID card on the ID card recognition area, the examination supervision platform reads the identity information, and the camera in the examination area collects the candidate's facial image. If the read identity information and the collected facial image are consistent with the registration information, the candidate's identity authentication is successful and he / she enters the examination; otherwise, the candidate's identity authentication fails and he / she cannot enter the examination.

[0027] Among them, identity information includes the candidate's name, age, address, ID card photo, etc.; registration information includes registration identity information and registration face picture; and the candidate's ID card photo, registration photo and the candidate's face image collected on site must be consistent.

[0028] It should be noted that the examination supervision platform is an integrated information system used to manage and monitor the examination process to ensure the fairness and security of the examination. It uses a variety of technical means (such as identity verification, video monitoring, data recording, etc.) to realize functions such as confirmation of candidate identities, supervision of the examination process, and post-examination.

[0029] The information collection module collects images of the markers during the test to obtain a marker image; Specifically, a camera is set on the front and side of the test area respectively to collect images of the crane during its movement in real time to obtain marked images; wherein the cameras are installed at the same height, and the field of view covers the entire crane, and the size of the collected images is the same.

[0030] It should be noted that the cameras are set on the front and sides in order to fully collect changes in the crane's operation. For example, when the crane is transporting moving goods, it will swing forward, backward, left or right. If the front of the test area is the front of the crane, the collected images can observe the left and right swing; if the side of the test area is the side of the crane, the collected images can observe the front and back swing.

[0031] The marked images are divided according to the moving direction of the crane according to the moving features; wherein the moving features include vertical rising distance, vertical descending distance and horizontal moving distance.

[0032] Specifically, a vertical movement reference point and a horizontal movement reference point of the moving direction of the moving cargo are set; wherein the vertical movement reference point is set at the center of the moving cargo; the horizontal movement reference point is set on the sliding assembly on the crane beam, and a certain position on the sliding assembly is used as the horizontal movement reference point; wherein the sliding assembly is a device for pulling the moving cargo to move; Establishing a coordinate system of the marked image, extracting the coordinates of the vertical moving reference point and the horizontal moving reference point in the marked image; wherein the coordinate system of the marked image takes the central pixel point of the marked image as the origin of the coordinate axis, and a number of pixels as a coordinate unit; According to the time sequence, the distance between the vertical movement reference point and the horizontal movement reference point in two adjacent marked images is calculated, and the difference processing is performed to obtain the vertical movement distance of the vertical movement reference point; when the vertical movement distance between two adjacent marked images is not 0, the vertical movement reference point moves upward or downward in the vertical direction, and the marked image is a vertical rising image or a vertical falling image; wherein, when the vertical movement distance between two adjacent marked images is less than 0, the marked image has a vertical rising feature; when the vertical movement distance between two adjacent marked images is greater than or equal to 0, the marked image has a vertical falling feature; And, when the coordinates of the horizontal movement reference point change in the horizontal direction, the marked image has a horizontal movement feature; Images with the same moving features are integrated to obtain moving feature image sets, and the images in the sets are arranged in chronological order; wherein the moving feature image set includes a vertically rising image set or a vertically falling image set and a horizontally marked image set.

[0033] Among them, if a marked image is marked with more than one moving feature, the moving feature marked for the first time or the last time shall prevail; and the presence of more than one moving feature in the marked image will only occur at the junction of the end of the rise and the beginning of the translation, and at the junction of the end of the translation and the beginning of the descent.

[0034] See also Figure 2 For example, the center pixel of each marked image is used as the origin of the coordinate axis (0,0), 1 coordinate unit = 10 pixels, X axis: horizontal direction, rightward is positive, leftward is negative, Y axis: vertical direction, upward is positive, downward is negative; Assume that the test is an assessment of a bridge crane. Now, several labeled images of a candidate operating a crane are collected. The labeled images are arranged in time. If two adjacent labeled images are marked as labeled image A and labeled image B (assuming that the collection time of labeled image A is earlier than that of labeled image B), the horizontal movement coordinate points (the center point coordinates of the sliding component on the crossbeam of the bridge crane) of labeled images A and B are (XAS, YAS) and (XBS, YBS) respectively, and the vertical movement coordinate points (the center coordinates of the moving goods) are (XAC, YAC) and (XBC, YBC) respectively. Calculate the vertical movement distance of the vertically moving reference point in the A and B marker images = ; When this vertical moving distance is less than 0, it indicates that the crane from the A-marked image to the B-marked image is performing an ascending operation, and the A-marked and B-marked images are marked as vertical ascending images; When the vertical moving distance is greater than or equal to 0, it indicates that the crane from the A-marked image to the B-marked image is performing a descending operation, and the A-marked and B-marked images are marked as vertical descending images.

[0035] Assuming that for two adjacent images, if the horizontal reference point of the image at the previous moment is on the left side of the horizontal reference point of the image at the next moment, the crane moves to the right; otherwise, the crane moves to the left.

[0036] The information collection module also obtains various assessment indicators of the examinee based on the marked image and assessment indicators of the crane; the assessment indicators include moving speed, swing amplitude and moving storage results; the storage results include storage failure and storage success; Specifically, an initial image and a terminal image of each moving feature are obtained; If a marked image is marked with more than one moving feature, the moving feature marked first shall prevail; The first marked image in the vertical ascent image set is the initial image of the vertical ascent, and the last image is the termination image of the vertical ascent; The last marked image in the horizontal marked image set is the initial image of the vertical descent, and the last image in the vertical descent image set is the termination image of the vertical descent; The ending image of vertical rise is the initial image of horizontal movement; the initial image of vertical fall is the ending image of horizontal movement.

[0037] If a marked image is marked with more than one moving feature, the moving feature marked last shall prevail; The first marked image in the vertical ascending image set is the initial image of the vertical ascending, and the first image in the horizontal marked image set is the terminal image of the vertical ascending; The first marked image in the vertical descent image set is the initial image of the vertical descent, and the last image is the termination image of the vertical ascent; The ending image of vertical rise is the initial image of horizontal movement; the initial image of vertical fall is the ending image of horizontal movement.

[0038] It should be noted that the operating sequence of a crane for transporting mobile goods is generally to first rise to a certain height, then move horizontally to just above the designated placement position of the mobile goods, and then slowly descend to place the mobile goods at the designated position.

[0039] The various assessment indicators for candidates are obtained as follows: Establishing the coordinate system of the marked image, respectively extracting the coordinates of the vertical moving reference point and the horizontal moving reference point in the initial image and the final image; Calculate the distance between the vertical moving reference point and the horizontal moving reference point in the initial image of vertical ascent / descent, marked as the initial distance, and the distance between the vertical moving reference point and the horizontal moving reference point in the final image, marked as the final distance, divide the absolute value of the difference between the initial distance and the final distance by the acquisition time difference between the initial image and the final image of vertical ascent / descent, and obtain the moving speed of the moving cargo in vertical ascent / descent; It should be noted that the initial / terminal distance is the length of the extension / contraction of the telescopic steel rope when the crane is performing the ascent / descent operation, that is, the distance that the mobile cargo moves in the vertical direction, and then divided by the acquisition time difference between the initial image and the terminal image of the vertical ascent / descent, that is, the time taken from the initial movement to the end of the movement, to calculate the moving speed of the vertical ascent / descent of the mobile cargo.

[0040] The distance between the horizontal movement reference point of the initial image of the horizontal movement and the horizontal movement reference point in the termination image is calculated and divided by the acquisition time difference between the initial image of the horizontal movement and the termination image to obtain the moving speed of the moving goods in the horizontal movement.

[0041] It should be noted that the horizontal movement reference point only moves in the horizontal direction, so the vertical coordinate value is fixed. The absolute value of the difference between the horizontal coordinate at the initial movement and the horizontal coordinate at the end of the movement is the distance the moving goods move in the horizontal direction. Divide it by the time taken from the initial movement to the end of the movement to calculate the horizontal movement speed of the moving goods.

[0042] The moving speed of the moving goods in each direction is compared with the speed threshold range, the moving speed that is not within the speed threshold range is marked as an abnormal speed, and the number of abnormal speeds is counted.

[0043] The swing amplitude of the crane is analyzed as follows: The angle between the crane's rope and the relative vertical line in the marked image is calculated to obtain the swing amplitude of the moving cargo when the crane is running, and the swing amplitude of the moving cargo is compared with the preset swing threshold range. The swing amplitude that is not within the preset swing threshold range is marked as abnormal swing, and the number of abnormal swings is counted; wherein the relative vertical line is a straight line in the marked image that passes through the horizontal moving reference point and is perpendicular to the X-axis.

[0044] Please note that Figure 3 , draw a straight line on the X-axis through the coordinates of the center point of the sliding component, and the crane's rope and the straight line form a right triangle. The Pythagorean theorem is used to calculate the angle between the rope and the straight line, that is, the amplitude of the swing of the moving cargo, marked as β.

[0045] And, the results of analyzing the incoming goods are as follows: The edge information of the moving goods in the marked image and the edge information of the preset warehousing position are extracted respectively. When the edge information of the moving goods exceeds the edge information of the preset warehousing position, the moving status of the moving goods is marked as warehousing failure; when the edge information of the moving goods is within the edge information of the preset warehousing position, it is marked as warehousing success.

[0046] The scoring module scores each assessment indicator of the examinee and obtains the examinee's assessment result.

[0047] According to the scoring rules for each assessment indicator formulated by experts, the candidates' various assessment indicators are scored. The total score of each assessment indicator is the candidate's assessment score, which is compared with the passing score to obtain the final assessment result; among them, the candidate's assessment score will generate an electronic transcript; the electronic transcript includes the score of each assessment indicator and the total score; the assessment result is passed or failed.

[0048] For example: Assume that the expert sets the full score of the crane assessment to be 100 points, and the passing score is 60 points. If there is an abnormality in the assessment index, 5 points will be deducted, and if the moving of goods into the warehouse fails, 10 points will be deducted. If the data of each assessment index of a candidate operating a crane are 5 times of abnormal speed, 5 times of abnormal swing, and successful moving of goods into the warehouse, then the candidate's assessment score = 100-10 5=50 points<60 points, then the candidate's result is unqualified.

[0049] See also Figure 4 The second aspect of the present invention provides a vision-based crane intelligent testing method, comprising the following steps: Step 1: Verify the candidate’s identity information; Step 2: Collect the marked images of the examinee in the process of transporting the mobile goods from the starting position to the preset storage position by the crane; based on the marked images of the crane and the assessment indicators, obtain the examinee's various assessment indicators; Step 3: Score each assessment indicator of the candidate and obtain the candidate's assessment result.

[0050] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0051] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A vision-based crane intelligent testing system, characterized in that: It includes an information collection module, and an identity authentication module and a scoring module connected thereto; Identity verification module: used to verify the candidate's identity information; Information collection module: used to collect images of markers during the test in real time through cameras to obtain marker images, and divide the crane images according to movement characteristics according to the movement direction of the crane; wherein the movement characteristics include vertical rising distance, vertical descending distance and horizontal moving distance; the cameras are set at the front and side of the test area; Based on the marked images and assessment indicators, the test takers' assessment indicator results are obtained; the marked objects include moving goods, sliding components, lifting ropes and hooks; the assessment indicators include moving speed, swing amplitude and storage results; the storage results include storage failure and storage success; Scoring module: used to score the various assessment indicators of candidates and obtain the assessment results of candidates based on the scores of various assessment indicators.

2. A vision-based crane intelligent testing system according to claim 1, characterized in that: The identity information of the examinee to be verified includes: Obtain the candidate's identity information and capture the candidate's facial image through the camera in the examination area to verify whether the candidate's identity information, the captured facial image and the registration information are consistent; if yes, the candidate's identity authentication is successful; if not, the candidate's identity authentication is failed.

3. The vision-based crane intelligent testing system according to claim 1 is characterized in that: The step of dividing the marked image according to the moving direction of the crane according to the moving features includes: A first mark position preset on the crane moving cargo is obtained as a vertical movement reference point, and a second mark position preset on the crane sliding component is obtained as a horizontal movement reference point; Establishing a coordinate system of the marked image, extracting the coordinates of the vertical moving reference point and the horizontal moving reference point in the marked image; wherein the coordinate system of the marked image takes the central pixel point of the marked image as the origin of the coordinate axis, and a number of pixels as a coordinate unit; According to the time sequence, the distance between the vertical movement reference point and the horizontal movement reference point in two adjacent marked images is calculated, and the difference processing is performed to obtain the vertical movement distance value of the vertical movement reference point; when the vertical movement distance value between the two adjacent marked images is not 0, the marked image is marked as a vertical rising image or a vertical falling image; otherwise, the marked image is not marked; wherein, when the vertical movement distance value between the two adjacent marked images is less than 0, the marked image is marked as a vertical rising image; when the vertical movement distance value between the two adjacent marked images is greater than or equal to 0, the marked image is marked as a vertical falling image; and, when the coordinates of the horizontally moving reference point change in the horizontal direction, marking the marked image as a horizontal marked image; The marked images of the same moving feature are integrated to obtain a set of moving feature images, and the images in the set are arranged in chronological order; wherein the moving feature image set includes a vertically rising image set, a vertically falling image set and a horizontally marked image set; when a marked image is marked with more than one moving feature, the moving feature marked for the first or last time is retained.

4. A vision-based crane intelligent testing system according to claim 3, characterized in that: The method for obtaining the index result of the moving speed includes the following steps: Obtaining the initial image and the final image in each moving feature image, and extracting the coordinates of the vertical moving reference point and the horizontal moving reference point in the initial image and the final image respectively; wherein the initial image is the image of the moving feature when it starts to move; and the final image is the image of the moving feature when it ends to move; Extract the coordinates of the vertical moving reference point and the horizontal moving reference point of the initial image in the vertical rising distance, calculate the distance between the vertical moving reference point and the horizontal moving reference point in the initial image, and mark it as the initial distance; and extract the coordinates of the vertical moving reference point and the horizontal moving reference point of the termination image in the vertical rising distance, calculate the distance between the vertical moving reference point and the horizontal moving reference point in the termination image, and mark it as the termination distance; calculate the absolute value of the difference between the initial distance and the termination distance, and calculate the ratio of the absolute value of the difference to the acquisition time difference between the initial image and the termination image to obtain the moving speed of the moving goods rising vertically; Extract the coordinates of the vertical moving reference point and the horizontal moving reference point of the initial image in the vertical descent distance, calculate the distance between the vertical moving reference point and the horizontal moving reference point in the initial image, and mark it as the initial distance; and extract the coordinates of the vertical moving reference point and the horizontal moving reference point of the termination image in the vertical descent distance, calculate the distance between the vertical moving reference point and the horizontal moving reference point in the termination image, and mark it as the termination distance; calculate the absolute value of the difference between the initial distance and the termination distance, and calculate the ratio of the absolute value of the difference to the acquisition time difference between the initial image and the termination image to obtain the moving speed of the moving cargo in vertical descent; Calculate the distance between the horizontal movement reference point of the initial image of the horizontal movement and the horizontal movement reference point in the final image, and divide the calculated distance by the acquisition time difference between the initial image of the horizontal movement and the final image to obtain the moving speed of the moving cargo in the horizontal movement; The moving speed of the moving goods in each direction is compared with the preset speed threshold range, the moving speed that is not within the preset speed threshold range is marked as abnormal speed, and the number of abnormal speeds is counted.

5. The vision-based crane intelligent testing system according to claim 4 is characterized in that: The step of obtaining the initial image and the final image in each moving feature image comprises: Determine whether there are multiple moving features of the marked image. If yes, keep the moving features of the first mark and discard the moving features of the subsequent marks; if no, do not perform any operation; Mark the first marked image in the vertical ascent image set as the initial image of the vertical ascent, and mark the last image as the termination image of the vertical ascent; Mark the last marked image in the horizontal marked image set as the initial image of vertical descent, and mark the last image in the vertical descent image set as the termination image of vertical descent; Mark the ending image of vertical rise as the initial image of horizontal movement; mark the initial image of vertical fall as the ending image of horizontal movement; or, Determine whether there are multiple moving features of the marked image. If yes, keep the moving features marked last and discard the moving features marked later. If no, do not perform any operation. Mark the first marked image in the vertically ascending image set as the initial image of vertically ascending, and mark the first image in the horizontally marked image set as the terminal image of vertically ascending; The first marked image in the vertical descent image set is marked as the initial image of vertical descent, and the last image is marked as the termination image of vertical ascent; The ending image of vertical rise is marked as the initial image of horizontal movement; the initial image of vertical fall is marked as the ending image of horizontal movement.

6. The vision-based crane intelligent testing system according to claim 1 is characterized in that: The method for obtaining the index result of the swing amplitude includes the following steps: The angle between the crane's rope and the relative vertical line in the marked image is calculated to obtain the swing amplitude of the moving cargo when the crane is running, and the swing amplitude of the moving cargo is compared with the preset swing threshold range. The swing amplitude that is not within the preset swing threshold range is marked as abnormal swing, and the number of abnormal swings is counted; wherein the relative vertical line is a straight line in the marked image that passes through the horizontal moving reference point and is perpendicular to the X-axis.

7. The vision-based crane intelligent testing system according to claim 1 is characterized in that: The method for obtaining the storage result includes the following steps: The edge information of the moving goods in the marked image and the edge information of the preset warehousing position are extracted respectively. When the edge information of the moving goods exceeds the edge information of the preset warehousing position, the moving status of the moving goods is marked as warehousing failure; when the edge information of the moving goods is within the edge information of the preset warehousing position, it is marked as warehousing success.

8. The vision-based crane intelligent testing system according to claim 1 is characterized in that: The above-mentioned scoring of each assessment index of the examinee to obtain the examinee's assessment result includes: According to the scoring rules for each assessment indicator formulated by experts, the candidates' various assessment indicators are scored. The total score of each assessment indicator is the candidate's assessment score, which is compared with the passing score to obtain the final assessment result; among them, the candidate's assessment score will generate an electronic transcript; the electronic transcript includes the score of each assessment indicator and the total score; the assessment result is passed or failed.

Images