Task scheduling method and system based on zero-work system

Abstract

The invention relates to a task scheduling method and system based on a zero-work system, and relates to the technical field of data processing, and the method comprises the steps: receiving a task issuing instruction, and calling a task level and a task type based on the task issuing instruction; determining a capability level based on the task type, and querying whether preset capability information of the order receiving user is smaller than the capability level or not; if yes, filtering the task for the order receiving user; if not, a task access permission is opened for the order receiving user; receiving a receiving instruction of an order receiving user based on the open task receiving permission, generating a task package based on the receiving instruction, and taking off the task; and acquiring identity information based on the task package, and after the identity information is consistent with preset order receiving user information, matching a supervision method and performing supervision. The method has the effect of improving the task completion rate.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a task scheduling method and system based on a gig system. Background Art

[0002] The gig market is an intelligent platform that provides comprehensive services such as task outsourcing, labor recruitment, and commission settlement to a large number of employers.

[0003] Currently, when companies need to outsource, they mainly rely on publishing the required tasks, allowing others to accept orders at will. After accepting the orders, they complete the tasks according to the company's requirements, thereby completing the outsourced task requirements of the company within the specified time.

[0004] Regarding the above-mentioned related technologies, when an enterprise assigns tasks under such an operating model, anyone can accept the task, even those without experience. This method of task assignment has a low completion rate after accepting the task because people can accept the task at will, and there is room for improvement. Summary of the Invention

[0005] In order to improve the task completion rate, the present invention provides a task scheduling method and system based on a gig system, which improves the task completion rate by matching users of the same level as the task type.

[0006] In a first aspect, the present invention provides a task scheduling method and system based on a gig economy system, which adopts the following technical solutions: A task scheduling method based on a gig system, comprising: S1: Receive a task release instruction, and retrieve a task level and task type based on the task release instruction; S2: Determine the ability level based on the task type and the task level, and query whether the preset ability information of the order-receiving user is less than the ability level; if so, filter the task for the order-receiving user; if not, grant the order-receiving user permission to accept the task; S3: Based on the open task acceptance authority, receiving the acceptance instruction of the order-accepting user, generating a task package based on the acceptance instruction and removing the task; S4: Collect identity information based on the task package. After the identity information is consistent with the preset order-taking user information, match the supervision method and perform supervision.

[0007] By adopting the above technical solution, users of the same level as the task type are matched, and the order-taking permissions are opened to these users so that they can take orders. At the beginning of the task, the identity of the order-taking user is identified to determine that it is the person who is performing the task, thereby matching people with the same level as the task, thereby improving the completion rate of the task.

[0008] Optionally, further comprising: S40: matching out a supervision instruction based on the access instruction, and collecting a working image based on the supervision instruction; S41: based on a preset finished product feature, a processing image is framed out from the working image, a corresponding finished product image is matched out based on the processing image, and a processing feature is matched out based on the finished product image; S42: based on a comparison relationship between the processing feature and the finished product image, a detection area in the processing image is determined; S43: a feature to be processed in the detection area is collected; S44: a production score is determined based on a comparison between the feature to be processed and the processing feature; S45: if the production score is lower than a preset required score, a modification reminder is issued, and a modification method of the processing feature is determined based on the processing feature; if the production score is not lower than the preset required score, detection is continued, and a processing time is recorded.

[0009] By adopting the above technical solution, the supervision instruction is matched out to supervise the processing feature of the user when processing the product, so as to determine whether the product processed by the user meets the requirements, and if not, the modification is prompted, thereby improving the production accuracy.

[0010] Optionally, further comprising: S410: based on a preset arm feature, a working image is identified and marked from the working image, and an intersection on an extension line of adjacent positions is calculated; S411: in a preset unit time, intersections of the same part are sequentially connected to obtain a movement trend, and a moving direction is determined based on the movement trend; S412: when the moving direction is consistent with a preset approaching direction, a warning distance between the intersection and a preset warning line is collected; S413: when the warning distance is less than a preset reference distance, a warning is given; when the working image does not contain the finished product feature, a disappearance time is accumulated; S4140: if the disappearance time is not greater than a preset production time, detection is continued; S4141: if the disappearance time is greater than the preset production time, an alarm is issued, the product is prompted to be shown by the order receiving user, a show image is collected, and a detection time is recorded; S415: based on the show image and the working image before the accumulated disappearance time, an image change area is determined; S416: a cost feature quantity in the image change area is collected, and is defined as a new quantity, a new efficiency is calculated based on the new quantity and the detection time; S4170: If the newly added efficiency is greater than the preset production efficiency, an alarm is issued and recorded; S4171: If the newly added efficiency is not greater than the preset production efficiency, a prompt is given and the number of detections is recorded. If the number of detections is greater than the preset maximum number of detections or cheating production is detected, the task is stopped and the user is prompted to stop the task.

[0011] By adopting the above technical solution, whether cheating occurs is determined by detecting the movement trend of the intersection of the arm features in the working image. When the finished product features cannot be detected, the disappearance time is recorded. When the disappearance time is greater than the production time, the additional efficiency is determined. When the additional efficiency is greater than the preset production efficiency, an alarm is issued, thereby monitoring whether cheating occurs during user production.

[0012] Optionally, also include: S4100: Selecting arm features in the image based on the working image and defining the image as a detection image; S4101: Determine a frame selection area based on the detection image; S4102: When the frame selection area reaches a preset alert area, determining a covering position of the detection image in the working image; determining a gap image in the working image based on the covering position, determining an avoidance vector based on the gap image; and determining a preset rotation angle of the camera device based on the avoidance vector. S4103: performing an evasive movement according to the evasive vector, rotating according to the rotation angle, and updating the working image; S4104: If the frame selection area of ​​the updated working image is the same as the frame selection area of ​​the original image, supervision is performed using a preset auxiliary supervision method.

[0013] By adopting the above technical solution, when the area of ​​the arm feature detected in the working image is larger than the preset area, the required avoidance vector and rotation angle are determined, and the rotation is performed, so that there is no obstruction in the working image to prevent occlusion.

[0014] Optionally, also include: S41040: Segmenting the gap image using a preset segmentation line to obtain segmented regions; selecting a segmented region with a large area from the segmented regions and defining it as a rotation region; and determining a rotation angle according to the rotation region. S41041: determining a preset movable reflector based on the rotation angle, and determining a reflection angle of the movable reflector based on a preset reflection position and the processed image; S41042: Control the movable reflector to rotate based on the reflection angle, and update the working image according to the rotation angle.

[0015] By adopting the above technical solution, the gap area is segmented to obtain segmented areas, a large rotation area is selected from the segmented areas, and the rotation angle is determined according to the rotation area. The active reflector is rotated to update the working image, so that the processed image is detected by mirror reflection when it is covered and cannot be moved, thereby preventing the order-taking user from cheating.

[0016] Optionally, also include: S400: Acquiring a contrast image on a preset contrast area in the working image; S401: determining a contrast color difference based on the contrast image and a preset reference contrast feature; S403: When the contrast color difference is less than a preset maximum contrast color difference value, issuing a replacement instruction; S404: replacing the preset protective film based on the replacement instruction and updating the working image.

[0017] By adopting the above technical solution, the contrast color difference is determined by comparing the image of the preset contrast area with the reference feature. If the color difference is less than the maximum contrast color difference value, the protective film is replaced, thereby reducing the situation where the protective film is dusted due to long-term non-use, resulting in unclear working images being collected.

[0018] Optionally, also include: S405: outputting a cleaning instruction when replacing, controlling a preset light emitting device to illuminate the protective film in response to the cleaning instruction, and collecting the intensity of light passing through the protective film; S406: Selecting a cleaning area based on the light intensity and a preset light transmittance, and determining a cleaning intensity according to the light intensity; S407: Based on the cleaning area, controlling a preset cleaning device to align with the cleaning area, cleaning the protective film according to the cleaning intensity, and controlling the light-emitting device to emit light; S408: Stop cleaning when the light intensity is not less than the preset light transmission intensity.

[0019] By adopting the above technical solution, after the protective film is replaced, the protective film is passed through light to determine the light intensity, so as to frame the cleaning area of ​​the protective film and clean the cleaning area. After cleaning, light is passed through again. When the light intensity is not less than the preset light transmission intensity, light transmission is stopped, so that the protective film covered with dust can be cleaned after replacement and can be used again.

[0020] Optionally, also include: S50: selecting a coverage area from the working image frame based on a preset coverage feature, defining it as a coverage image, and outputting a removal instruction; S51: Based on the removal instruction, retrieve historical working images within a preset unit time, select a close image showing a magnification trend based on the historical working images, and match the covering material based on the close image; S52: Matching the blowing intensity based on the covering material; S53: Controlling a preset blowing device to blow up the covering material according to the blowing intensity, and updating the working image; S54: issuing an alarm when the working image does not change.

[0021] By adopting the above technical solution, when a covering feature is detected, the historical working image is retrieved by removing the instruction to determine the covering material, and the blowing intensity is matched to blow away the covering material. An alarm is issued when it cannot be blown away. Therefore, when the working image is covered due to environmental reasons, the covering material can be blown away by blowing to ensure supervision of the production process.

[0022] Optionally, also include: S440: Determine, based on the processing time, a completion time point of the finished product feature in the working image and an appearance time point of a preset blank feature in the working image. Calculating the user's stage efficiency based on the completion time point and the appearance time point; S441: When a processing feature is detected, the number of detections is recorded. When the number of detections reaches a preset reference number, To determine the initial time point and completion time point of each processing time, and calculate the stage efficiency of each processing time when processing different processing features to determine the most efficient stage; S4420: If the highest efficiency stages are different, extracting the working images of the respective highest efficiency stages and recording common features in the working images; S44200: If the common feature appears in the other working images, check its sorting order. If the sorting order is greater than that of the working images without the common feature, record it as a beneficial feature; S44201: Determine a simulation method based on the beneficial features and output a simulation instruction; S44202: Based on the simulation instruction, prompt the customer to simulate the production environment using the simulation method; S4421: If the stages of the highest production efficiency are the same, determining the task station according to the highest production efficiency; S44210: Prompt the order-receiving user to arrive at the task station based on the task station; and supervise the order-receiving user after the user arrives at the task station.

[0023] By adopting the above technical solution, the stage in which the user is most efficient is determined. If the stages are different, the user is prompted to simulate the matched beneficial features. If the stages are the same, the user is directed to the matched workstation for production. This improves the user's production efficiency through reasonable allocation or simulation, thereby improving the task completion rate.

[0024] In a second aspect, the present application provides a task scheduling system based on a gig economy system, which adopts the following technical solutions: A task scheduling system based on a gig system, comprising: The acquisition module is used to obtain task release instructions, acceptance instructions, identity information, features to be processed, warning distance, comparison images, light intensity, removal instructions, and simulation instructions; A memory for storing a production program for any of the above-mentioned new energy motor housings; The processor is configured to load, execute, and implement the program stored in the memory.

[0025] In summary, this application includes at least one of the following beneficial technical effects: 1. By matching the task level and type when the task is released, matching people with the same task level, thereby improving the completion rate of the task; 2. By determining the rotation area to determine the rotation angle, and rotating the movable mirror to update the working image, the processed image is detected through mirror reflection when it is covered and cannot be moved, preventing cheating by the order-taking user; 3. Improve user productivity by reasonably allocating or simulating the environment, thereby improving the task completion rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a structural diagram of an embodiment of the present invention; Figure 2 It is a structural schematic diagram of the camera of the present invention; Figure 3 The present invention is a flowchart of a task scheduling method based on a gig system.

[0027] The parts indicated by the numerical symbols in the above drawings are as follows: 1. movable reflector; 2. camera device; 3. bracket; 4. protective film; 5. replacement area. DETAILED DESCRIPTION

[0028] The present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0029] The present application embodiment discloses a task scheduling method and system based on a gig system, referring to Figure 1The application controls the camera 2 on the support 3 to rotate and drives the movable mirror 1 installed on the support 3 to rotate when the image does not change, so as to rotate when the movable mirror cannot be moved to avoid covering.

[0030] Referring to Figure 1 and Figure 2 When it is detected that the protective film 4 in front of the camera is covered with dust, the motor in the replacement area 5 is controlled to rotate to drive the protective film 4 to move, and the protective film 4 is moved to the back of the camera, and the protective film 4 for replacement in the back of the camera is moved to the front of the camera lens.

[0031] Referring to Figure 3 A task scheduling method based on a zero-work system, comprising the following steps: Step 1: receiving a task publishing instruction, and calling a task level and a task category based on the task publishing instruction.

[0032] The task level refers to the minimum level that can complete the task, and the task category refers to the classification of the task type. The task level and the task category can be manually input by the employing unit, which is not described herein.

[0033] The task publishing instruction refers to the instruction of the employing unit publishing the task. The task publishing instruction can be received by the server after the employing unit publishes the task. After receiving the task publishing instruction, the task level and the task category of the corresponding task are called.

[0034] Step 2: determining the ability level based on the task category and the task level, and querying whether the preset ability information of the order receiving user is less than the ability level. If it is less than, the task is filtered for the order receiving user, and if it is not less than, the task receiving permission is opened to the order receiving user.

[0035] The order receiving user refers to the user receiving the order. The user who needs to receive the order in the platform is the order receiving user.

[0036] The ability level refers to the level of the user's ability. The ability level can be determined by querying the ability information of the user to determine whether the ability level of the task category of the user meets the task level. If it meets, the task receiving permission is opened to the order receiving user, and if it does not meet, the task is filtered for the order receiving user.

[0037] The open task receiving permission refers to the permission of the order receiving user who meets the requirement to receive the task. When the ability level of the corresponding task category of the user reaches the task level, the task receiving permission is opened.

[0038] The ability information refers to the data recording the ability of each task category of the order receiving user. The ability information can be determined by querying the historical order, work content and work type ability of the user. The ability information can be used to match the ability level.

[0039] Determine the ability level based on the task category and the task level, which means that the corresponding ability level needs to be matched according to the category of the published task, for example, if the task is a manual production, the task level of the manual production of the order receiving user is matched.

[0040] Step 3: Based on the open task access permission, receive the access instruction of the order receiving user, generate a task package based on the access instruction, and take down the task.

[0041] The access instruction is an instruction issued by the user after receiving the task. The access instruction can be obtained by identifying whether the order receiving user has received the order, issuing the access instruction after receiving the order, issuing the order instruction after the user receives the order, taking down the task corresponding to the instruction, and issuing the task package to the user for completing the task. The task package refers to the content required by the task, and the task package is started when the task starts.

[0042] Step 4: Based on the task package, collect identity information, and after the identity information is consistent with the preset order receiving user information, match the supervision method and perform supervision.

[0043] The identity information refers to the identity information of the user who starts the task. The identity information can be collected by a camera controlled by the task package after starting the task package. The model of the camera is selected by the staff according to the actual situation, which is not described here.

[0044] The order receiving user information refers to the information of the order receiving user, which can be determined by the order receiving user in advance, which is not described here.

[0045] After starting the task package, the identity information of the user who starts the task is collected. If the collected identity information after starting is not consistent with the order receiving user information, the user is prevented from working. If they are consistent, the supervision method is matched for supervision.

[0046] The supervision method refers to a method for supervising the production of the order receiving user. The supervision method determines the corresponding supervision method according to the task category. The supervision method in this embodiment is the supervision method for manual production.

[0047] The supervision method comprises the following steps: Step 40: Match the supervision instruction based on the access instruction, and collect the working image based on the supervision instruction.

[0048] The supervision instruction refers to the instruction for supervision. The supervision instruction can be matched after receiving the access instruction. The working image refers to the image of the order receiving user working. The working image can be collected by using a camera. The model of the camera is selected by the staff according to the actual situation, which is not described here.

[0049] Step 41: A processed image is selected from the working image based on the preset finished product features, a corresponding finished product image is matched based on the processed image, and the processing features are matched based on the finished product image.

[0050] Finished product characteristics refer to the characteristics of the finished product. Finished product characteristics can be manually input by staff and will not be described in detail here.

[0051] By labeling a large number of images with finished product features, inputting the labeled images with finished product features into the YOLO large model, repeatedly stacking images using the Pytorch network architecture to extract finished product features, and calculating the error between the results and the data, it is put into use when the error is less than 1%.

[0052] The working image is input into the large model. When the finished product features are identified, the finished product features are marked and framed out from the working image. The framed image is the processed image.

[0053] The finished product image refers to the image after processing. The finished product image can be matched by querying the finished product record table. The finished product record table records the image of the finished product features corresponding to the task, and the finished product image is determined based on the prototype features that appear during processing.

[0054] Processing features refer to the features when users process products, which can be queried and determined from the feature data table. The feature data table refers to the processing features corresponding to different finished product images determined in advance through experiments.

[0055] Step 42: Determine a detection area in the processed image based on a comparison relationship between the processed features and the finished product image.

[0056] The inspection area is the area used to detect processing features. By inserting the finished product image into the processed image and selecting the different areas between the finished product image and the processed image, this area is the required inspection area.

[0057] Step 43: Collect the features to be processed in the inspection area.

[0058] The features to be processed refer to the features that need to be processed when the order-taking user processes the product. The features to be processed can be determined by identifying the features processed by the order-taking user in the detection area through image recognition technology.

[0059] Step 44: Determine the production score based on the comparison between the to-be-machined features and the machined features.

[0060] The production score is the score generated after the production of the feature. It is determined by comparing the feature to the feature to be processed. The completed feature to be processed is compared with the feature to be processed, and any incorrect positions or shapes are highlighted to reduce the production score. Otherwise, the score remains unchanged.

[0061] Step 45: If the production score is lower than the preset required score, a modification reminder is issued, and the modification method of the processing feature is determined based on the processing feature; if the production score is not lower than the preset required score, the inspection is continued and the processing time is recorded.

[0062] The requirement score refers to the score of the minimum production requirement for the production. The requirement score is manually entered by the staff and will not be detailed here.

[0063] A modification reminder is a reminder used to prompt the order-taking user to modify the product. When the production score is lower than the required score, the control audio will emit a prompt sound to prompt the order-taking user to modify the product. The model of the audio is selected by the staff based on actual conditions and will not be elaborated here.

[0064] The modification method refers to a method used to modify product features. The modification method can identify the processing methods of the order-taking user through the working image, determine the methods that reduce the score of the order-taking user, and match the correct processing methods to enable the order-taking user to carry out production.

[0065] Production time refers to the time it takes for the order-taking user to complete the production of a product. The production time can be determined by the time recorded by the camera, so I will not go into details here.

[0066] The cheating detection method includes the following steps: Step 410: Identify and mark the arm features from the working image based on the preset features, and calculate the intersection points on the extension lines of adjacent positions.

[0067] The arm feature refers to the feature of the arm. The arm feature is input in advance by the staff and will not be described in detail here. The intersection point refers to the intersection point between the arms. The arm extension line can be used as the extension line, and the point where the extension line intersects is the intersection point.

[0068] Step 411: In a preset unit time, the intersection points of the same part are sequentially connected to obtain a movement trend; and the moving direction is determined based on the movement trend.

[0069] Unit time refers to a pre-set time period. This setting is determined by the staff based on actual circumstances and is not detailed here. Movement trend refers to the direction of arm movement. This is determined by connecting the intersection points in the work image within the unit time period and determining the next movement direction based on the order of the connections. This movement direction is the movement trend. Movement direction refers to the direction of movement of the movement trend. The movement direction can be determined by extending the connected line segments.

[0070] Step 412: When the moving direction is consistent with the preset approach direction, collect the warning distance between the intersection point and the preset warning line.

[0071] The approach direction refers to the direction in which the intersection approaches the warning line. The approach direction is manually entered by the staff and is not detailed here. The warning line is a boundary line used to alert the order-taking user to possible fraud. The warning line is the line segment at the edge of the production table that the camera identifies during production.

[0072] When the moving direction is consistent with the preset approach direction, it means that the user may have a cheating tendency by moving in this direction. At this time, the distance between the intersection and the warning line is collected.

[0073] Step 413: When the warning distance is less than the preset reference distance, a warning is issued; when the working image does not contain the finished product feature, the disappearance time is accumulated.

[0074] When the warning distance is less than the preset reference distance, it indicates that the user who accepts the order may engage in cheating production. At this time, a warning is issued to prompt the user who accepts the order to increase the distance between his arm and the warning line.

[0075] When the finished product feature is not included in the working image, it means that the user does not obey the warning, and the disappearance time of the finished product feature is recorded.

[0076] The disappearance time refers to the time when the finished product feature disappears. The disappearance time is obtained by recording the time point when the finished product feature disappears and continuously recording the time. For example, the finished product feature disappears at 10:00, and all the time after the finished product feature disappears is the disappearance time.

[0077] Step 4140: If the disappearance time is not greater than the preset production time, continue testing.

[0078] Production time refers to the time it takes to complete cheating production. The production time is entered by the staff based on actual conditions and will not be elaborated here.

[0079] If the disappearance time is not greater than the preset production time, it means that the disappearance time is too short and the cheating production cannot be completed. At this time, the production process of the order-taking user is continued to be detected.

[0080] Step 4141: If the disappearance time is greater than the preset production time, an alarm is issued to prompt the order-taking user to display the product and collect the display image, and record the detection time.

[0081] If the disappearance time is greater than the preset production time, it means that the user who accepts the order during this period may engage in cheating production. At this time, an alarm is issued to prompt the user to display the product.

[0082] The display image refers to the image displayed by the user. The display image can be obtained through a camera and will not be described in detail here.

[0083] The detection time refers to the entire duration of the disappearance period, that is, the point in time when the production feature appears during the disappearance period. The entire disappearance period is recorded as the detection time. For example, if the disappearance period starts at 10:00 and time recording begins, and the user displays the product at 10:20, the detection time is 20 minutes.

[0084] Step 415: Determine the image change area based on the display image and the working image before the cumulative disappearance time.

[0085] The image change region refers to the image region that changes within the detection time. By comparing the working image and the display image before the disappearance time, the image change region can extract some finished product features that are not in the working image. The partial finished product features are the image change region.

[0086] Step 416: Collect the number of cost features in the image change area and define it as the new number, and calculate the new efficiency based on the new number and the detection time.

[0087] The number of cost features collected in the image change region represents the number of finished product features added during the detection time. This new number refers to the number of newly added finished product features. For example, if a user orders a flower, and the flower has 10 petals before the finished product feature disappears and 20 petals when it reappears, the new number is 20 - 10 = 10 petals.

[0088] The newly added efficiency refers to the production efficiency of the order-taking user within the testing time. The newly added efficiency can be determined by calculating the quotient of the newly added quantity and the testing time. For example, if 10 petals are completed in 20 minutes of testing time, the production efficiency is 10 / 20=0.5 petals / minute.

[0089] Step 4170: If the newly added efficiency is greater than the preset production efficiency, an alarm is issued and recorded.

[0090] Production efficiency refers to the production efficiency of the order-taking user. Production efficiency can be determined by querying the efficiency record table of the order-taking user. The efficiency record table refers to the production efficiency range of the tasks previously accepted by the order-taking user recorded in advance.

[0091] Step 4171: If the newly added efficiency is not greater than the preset production efficiency, a prompt is given and the number of detections is recorded. If the number of detections is greater than the preset maximum number of detections or cheating production is detected, the task is stopped and the user is prompted to stop the task.

[0092] If the newly added efficiency is not greater than the preset production efficiency, it means that the user has not engaged in cheating production. At this time, the number of users who accept orders is recorded. When the number is greater than the maximum detection number, the user is prompted to stop the task.

[0093] The number of inspections refers to the number of times the finished product features cannot be detected in the working image. The number of inspections is determined by the number of inspection times. Each time an inspection time occurs, the number of inspections is recorded. The maximum number of inspections refers to the number of inspections allowed. The maximum number of inspections is manually entered by the staff and is not detailed here.

[0094] The cover detection method includes the following steps: Step 4100: Based on the working image, the arm features in the image are selected and defined as a detection image.

[0095] The detection image refers to an image for detecting arm features. The detection image is determined by identifying the arm features in the working image and selecting the arm features according to the largest circumscribed quadrilateral.

[0096] Step 4101: Determine the selection area based on the detected image.

[0097] The frame selection area refers to the area of ​​the detection image. The frame selection area can be determined by calculating the length and width of the largest circumscribed quadrilateral. The length of the largest circumscribed quadrilateral in the detection image is 5cm and the width is 7cm. The frame selection area is 5x7=35cm 2 .

[0098] Step 4102: When the framed area reaches the preset alert area, determine the covering position of the detection image in the working image; determine the gap image in the working image based on the covering position, and determine the avoidance vector based on the gap image; determine the preset rotation angle of the camera device 2 based on the avoidance vector.

[0099] The caution area refers to the maximum area used for the caution frame selection. This area is manually entered by staff and is not detailed here. When the frame selection area reaches the caution area, it indicates that the user accepting the order may intend to cover the camera with their arm. The cover position refers to the location of the detection image. By detecting the position of the frame selection area in the working image, the position of the entire detection image in the working image is defined as the cover position. The gap image refers to the remaining image in the working image excluding the detection image. By identifying the cover image at the cover position in the working image, the cover image is removed from the working image. The remaining image is the gap image.

[0100] The avoidance vector refers to the direction and distance the camera moves when avoiding the detection image. By examining the detection image removed from the gap image and comparing the distance from the detection image to the edge of the gap image, the longer distance is defined as the avoidance direction. Image recognition technology is used to measure the maximum distance from the detection image to the edge of the working image. This maximum distance is then entered into the movement distance data table to determine the movement distance. The movement distance data table contains movement distances corresponding to different maximum distances determined in advance through experiments. The smaller the maximum distance, the longer the movement distance.

[0101] The rotation angle refers to the angle at which the camera device 2 rotates. The rotation angle can be determined by querying a rotation data table. The rotation data table refers to the rotation angles corresponding to different moving distances determined in advance through experiments. The longer the moving distance, the greater the rotation angle.

[0102] The camera device 2 refers to a camera used to collect working images. The model of the camera is selected by the staff according to actual conditions and will not be described in detail here.

[0103] Step 4103: Perform avoidance movement according to the avoidance vector, rotate according to the rotation angle, and update the working image.

[0104] The camera device 2 is controlled to move according to the avoidance vector and rotate according to the rotation angle while moving. When the rotation is completed, the working image is updated.

[0105] Step 4104: If the frame selection area of ​​the updated working image is the same as the frame selection area of ​​the original image, supervision is performed using a preset auxiliary supervision method.

[0106] The frame area of ​​the updated working image is the same as the frame area of ​​the original image, indicating that the camera device 2 cannot be moved. At this time, the preset auxiliary supervision method is used for supervision. The auxiliary supervision method can be found in steps 41040 to 41042.

[0107] The auxiliary supervision method includes the following steps: Step 41040: Segment the gap image using a preset segmentation line to obtain segmented regions; select a segmented region with a large area from the segmented regions and define it as a rotation region; and determine a rotation angle based on the rotation region.

[0108] The segmentation line refers to a line segment used to divide the gap area. The segmentation area is the area formed after the gap area is divided by the segmentation line. The segmentation area is formed by identifying the shortest vertical distance that can be divided in the gap area and connecting them to divide the two areas.

[0109] If the detection image is located in the center, the split line cannot be determined, and in this case, the gap region is the region that is not connected on the left and right sides. In this case, the rotation region is the gap region with a larger area. If the areas are the same, the left or right side is randomly selected for rotation.

[0110] The rotation region refers to a region used to determine the rotation direction. By comparing the areas of the split regions, the split region with a larger area is defined as the rotation region.

[0111] The rotation angle refers to the angle used to rotate the camera 2. The rotation angle has only two values, positive 90 degrees and negative 90 degrees. The rotation angle is determined based on the position of the rotation region. If the rotation region is on the left side, the rotation angle is negative 90 degrees. If the rotation region is on the right side, the rotation angle is positive 90 degrees.

[0112] Step 41041: Determine the preset movable mirror 1 based on the rotation angle, and determine the reflection angle of the movable mirror 1 based on the preset reflection position and the processing image.

[0113] The movable mirror 1 refers to a rotatable mirror used to reflect the working environment. The movable mirror 1 can be selected by the worker according to the actual situation, which is not described here. The movable mirror 1 has two, one on the left and one on the right. The movable mirror 1 is installed on the support 3. There is a motor below the movable mirror 1 for rotating the mirror. The motor and the support are selected by the worker according to the actual situation, which is not described here.

[0114] The reflection position refers to the position of the mirror. The reflection position is manually input by the worker and is not described here.

[0115] The processing image position refers to the position of the processing image in the working image. The processing image can be identified and framed by image recognition technology. The framed position in the working image is the processing image position.

[0116] The reflection angle refers to the angle used to move the mirror. The reflection angle can be obtained by inputting the reflection position of the mirror and the processing image into the reflection angle data table. The reflection angle data table refers to the reflection angle corresponding to different reflection positions and processing images determined by experiments in advance. Under the condition that the processing image position remains unchanged, the longer the reflection distance, the smaller the reflection angle. Under the condition that the reflection distance position remains unchanged, the farther the processing image position from the mirror, the smaller the reflection angle.

[0117] Step 41042: Control the rotation of the movable mirror 1 based on the reflection angle, and update the working image according to the rotation angle.

[0118] The rotation motor under the movable mirror 1 is controlled to rotate by the reflection angle to drive the movable mirror 1 to rotate. The model of the rotation motor is selected by the worker according to the actual situation, which is not described here.

[0119] After the rotation is completed, the camera device 2 is controlled to capture the image in the movable reflector 1 to update the working image.

[0120] The verification method comprises the following steps: Step 400: Acquire a contrast image on a preset contrast area in the working image.

[0121] The comparison area is the area used to detect dust accumulation. The comparison area is pre-set by the staff and is not described in detail here. The comparison image is an image of the comparison area. The image of the comparison area in the working image can be identified using image recognition technology. This image is the comparison image.

[0122] Step 401: Determine contrast color difference based on the contrast image and a preset reference contrast feature.

[0123] The protective film 4 refers to a tough soft film used to protect the camera lens. The protective film 4 is selected by the staff according to actual conditions and will not be described in detail here.

[0124] The benchmark contrast feature refers to a feature used to detect whether there is dust accumulation. The benchmark contrast feature is a small square pattern pre-set on the protective film 4. After the protective film 4 is photographed by the camera, a small square for comparison will appear in the lower right corner of the working image. The small square is covered with dust, and the state of the small square displayed by the shooting is the benchmark contrast feature.

[0125] RGB values ​​refer to the values ​​used for red, green, and blue. RGB values ​​can be determined through image recognition technology and will not be described in detail here.

[0126] Contrast color difference refers to the difference between the color difference in the contrast image and the color difference of the reference contrast feature. The contrast color difference can be calculated by identifying the RGB values ​​of the contrast image and the RGB values ​​of the reference contrast feature through image recognition technology, and the color difference between the two is calculated. This color difference is the contrast color difference.

[0127] Substitute into the formula: E=√((R1-R2) 2 + (G1-G2) 2 + (B1-B2) 2 ).

[0128] For example, the color of an image is represented by the components of red (R), blue (G), and green (B). The baseline contrast feature simulates the RGB value of a dusty photo. For example, if the RGB value of the baseline contrast feature is (174.238.238) and the RGB value of the comparison image is (187.255.255), then the color difference is E, E = √((187-174) 2 + (255-238) 2 + (255-238)2 )=√(132+172+172), further calculation is E=√(169+289+289)=√747≈27.

[0129] Step 403: When the contrast color difference is less than the preset maximum contrast color difference value, a replacement instruction is issued.

[0130] The maximum contrast color difference value refers to the maximum value of the contrast color difference value. The maximum contrast color difference value is manually input by the staff and will not be described in detail here.

[0131] When the calculated contrast color difference value is less than the preset maximum contrast color difference value, it indicates that the protective film 4 needs to be replaced, and the server issues a replacement instruction.

[0132] Step 404: Replace the preset protective film 4 based on the replacement instruction and update the working image.

[0133] The replacement device refers to a device used to replace the protective film 4. The replacement device is selected by the staff according to actual conditions and will not be described in detail here.

[0134] After receiving the replacement instruction, the replacement device is controlled to start, and the protective film 4 is driven to move by rotating the replacement device, thereby replacing the protective film 4.

[0135] The cleaning method comprises the following steps: Step 405 : Output a cleaning instruction when replacing, control a preset light emitting device to illuminate the protective film 4 in response to the cleaning instruction, and collect the intensity of light passing through the protective film 4 .

[0136] The cleaning instruction is an instruction for cleaning the protective film 4. The cleaning instruction is issued after the replacement instruction is completed. The lighting device refers to the light bulb used for lighting. The model of the light bulb is selected by the staff according to the actual situation and is not detailed here. The light intensity refers to the intensity of the light passing through the protective film 4. The light intensity can be determined by a luminance meter. The model of the luminance meter is selected by the staff according to the actual situation and is not detailed here.

[0137] Step 406: A cleaning area is selected based on the light intensity and the preset light transmittance, and the cleaning intensity is determined according to the light intensity.

[0138] The light transmission intensity refers to the light intensity emitted by the light emitting device when it passes through the protective film 4. The light transmission intensity is selected by the staff according to the actual situation and will not be described in detail here. The cleaning area is the area that needs to be cleaned. The cleaning area can be framed to select the area where the light intensity is lower than the light transmission intensity. This area is the light transmission area.

[0139] The cleaning intensity refers to the water pressure used to clean the protective film 4. The cleaning intensity can be determined by consulting a cleaning data table, which records the cleaning intensities corresponding to different light intensities determined in advance through experiments. The lower the light intensity, the greater the cleaning intensity.

[0140] Step 407: Based on the cleaning area, the preset cleaning device is controlled to align with the cleaning area, and the protective film 4 is cleaned according to the cleaning intensity, and the light-emitting device is controlled to emit light.

[0141] The cleaning device refers to a water gun and a water pump for cleaning the protective film 4. The models of the water gun and the water pump are selected by the staff according to the actual situation and are not described in detail here. The water gun is controlled to aim at the cleaning area, and the water pump is controlled to set the cleaning intensity to spray water to rinse the protective film 4, and the light-emitting device is controlled to emit light.

[0142] Step 408: Stop cleaning when the light intensity is not less than the preset light transmission intensity.

[0143] When the detected light intensity is not lower than the transmitted light intensity, the water gun is controlled to stop cleaning.

[0144] The air blowing removal method comprises the following steps: Step 50: The coverage area selected from the working image frame based on the preset coverage feature is defined as a coverage image, and a removal instruction is output.

[0145] Coverage features refer to features that are covered by the working image. Coverage features are preset and input by the staff and are not detailed here. A cover image refers to the area covered by the cover feature. The cover image can be used to select the cover feature from the working image using image recognition technology. A removal command is used to remove the cover feature and is issued after the cover image is recognized.

[0146] Step 51: Based on the removal instruction, retrieve historical work images within a preset unit time. Select a magnified close-up image based on the historical work images, and match the covering material based on the close-up image. The unit time refers to the pre-set time period for retrieval. The unit time is set by the staff based on actual circumstances and is not detailed here. Historical work images refer to work images within a pre-set unit time period. Historical work images can be determined by retrieving work images within that unit time period.

[0147] A close image is an image that grows from small to large. You can close an image by calling up historical working images, playing them back in time, and selecting a gradually enlarged image. The gradually enlarged image is the close image.

[0148] Covering material refers to the material covering the camera. The covering material can be matched by matching the covering record table by bringing the close image into the covering record table.

[0149] Step 52: Match the blowing intensity based on the covering material.

[0150] The blowing intensity refers to the intensity used to blow away the covering material. The blowing intensity can be obtained by querying the blowing data table. The blowing data table refers to the blowing intensities corresponding to different covering materials determined in advance through experiments.

[0151] Step 53: Control the preset blowing device to blow up the covering material according to the blowing intensity, and update the working image.

[0152] The blowing device refers to the air pump and air port used to blow away the covering material. The models of the air pump and air port are selected by the staff according to the actual situation and are not detailed here. After the blowing intensity is matched, the blowing device is controlled to blow up the covering material according to the blowing intensity and the working image is updated.

[0153] Step 54: Issue an alarm when there is no change in the working image.

[0154] If the updated working image does not change after blowing, an alarm will be issued.

[0155] The air blowing removal method comprises the following steps: Step 440: Determine the completion time point of the finished product feature in the working image and the appearance time point of the preset blank feature in the working image based on the processing time, and calculate the stage efficiency of the user based on the completion time point and the appearance time point.

[0156] The completion time point refers to the time point when the finished product features are completed. The completion time point can be obtained by querying the work images within the recorded processing time, and determining the time when the finished product features appear, which is the completion time point.

[0157] The blank features refer to the features of the workpiece at the beginning. The blank features are manually input by the staff and will not be described in detail here.

[0158] The appearance time point refers to the time point at which the blank appears. The appearance time point can be determined by querying the time point at which the blank feature appears within the processing time from the recorded processing time.

[0159] The production time is calculated based on the completion time and the appearance time. The production time is the stage efficiency. For example, if the start time is 12:00 and the appearance time is 12:10, the stage efficiency is 10 minutes.

[0160] Step 441: When a processing feature is detected, the number of detections is recorded. When the number of detections reaches a preset benchmark number, the initial time point and completion time point of each processing time are determined, and the stage efficiency of each processing time when processing different processing features is calculated to determine the most efficient stage.

[0161] The detection number refers to the number of times the machining feature is detected, and the detection number is incremented by 1 when the machining feature is detected. The reference number refers to the number of times used to detect the machining efficiency, and the reference number is manually input by the staff and will not be described here.

[0162] When the detection number reaches the preset reference number, the initial time point and the completion time point in each machining time are extracted, and the production efficiency of each time is determined according to these time points. The production efficiencies are sorted, and a selected number of production efficiencies are selected to determine the highest production efficiency group.

[0163] Step 4420: If the highest efficiency stage is different, the work images of each highest efficiency stage are extracted, and the common features in the work images are recorded.

[0164] The highest efficiency stage is different, which means that part of the production efficiency is the same when producing different machining features. When determining the recorded machining time, the machining time corresponding to the number is given according to the machining feature to calculate the machining efficiency of the corresponding feature. The common features refer to the features that appear in these work images. The common features can be determined by extracting the features that appear in the work images of each highest efficiency stage. The common features in this embodiment refer to sound features.

[0165] When different highest efficiency stages are obtained, the work images of each highest stage efficiency are extracted, and the common features appearing in the work images are recorded.

[0166] For example, a finished product has A, B, C, D, and E, five different machining features, and the production time of the five machining features is the same, but part of the process is different. After completing the five finished products, there are 25 recorded times, A1 to A5, B1 to B5, C1 to C5, D1 to D5, and E1 to E5. Calculate the machining efficiency of the 25 recorded times, and find that the highest production efficiency group is A1, B2, and C4. It means that when producing A1, B2, and C4, the production efficiency is the highest. At this time, since the stages are different, the common features in this group of production efficiencies are recorded.

[0167] Step 44200: If the common features appear in the remaining work images, check the sorting order. If the sorting order is greater than the work images without common features, record it as a beneficial feature.

[0168] The beneficial feature is a feature that is beneficial to the production of the order user. The beneficial feature can be determined by substituting the common features into the remaining work images to query the work images with common features. The arrangement order of the work time of these work images is queried, and it is checked which common features appear in the high ranking and which features appear in the low ranking. The common features in the high ranking are the beneficial features.

[0169] For example, common features include music, rain, footsteps, and car noise. Extract all working images with music features and check their processing efficiency ranking. If the ranking of processed images with music features is greater than 50%, record it as a beneficial feature; if not, record it as a non-useful feature.

[0170] Step 44201: Determine a simulation method based on beneficial features and output simulation instructions.

[0171] A simulation method is a method used to simulate beneficial characteristics. This method is determined by extracting beneficial characteristics and inserting them into a simulation data table. The simulation data table contains simulation methods corresponding to different beneficial characteristics, which have been previously determined through sound. After a simulation method is matched, a simulation instruction is output.

[0172] Step 44202: Based on the simulation instructions, prompt the customer to simulate the production environment using a simulation method.

[0173] After issuing the simulation instructions, the customer is prompted by sound to perform the simulation according to the simulation method.

[0174] Step 4421: If the stages of highest production efficiency are the same, the task station is determined based on the highest production efficiency.

[0175] The stage of highest production efficiency is the same, which means that when the user processes the product, he is only good at processing a single processing feature. The task station refers to a job position that processes a certain processing feature alone, which will not be elaborated here.

[0176] At this time, the processing feature that the machine is good at is determined based on its highest production efficiency, and the task station is determined based on this processing feature.

[0177] Step 44210: Prompt the order-receiving user to arrive at the task station based on the task station; supervise the order-receiving user after he arrives at the task station.

[0178] After matching the task station that needs to be reached, the user who receives the order will be prompted to arrive at the corresponding processing station through sound. After the user who receives the order arrives at the task station, production will continue.

[0179] Based on the same inventive concept, an embodiment of the present invention provides a task scheduling system based on a gig system, including: The acquisition module is used to obtain task release instructions, acceptance instructions, identity information, features to be processed, warning distance, comparison images, light intensity, removal instructions, and simulation instructions; A memory, configured to store any one of the aforementioned task scheduling methods based on a gig-job system; The processor is configured to load, execute, and implement the program stored in the memory.

[0180] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiment. All technical solutions based on the concept of the present invention are within the scope of protection of the present invention. It should be noted that for those skilled in the art, various improvements and modifications that do not depart from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A task scheduling method based on a gig system, characterized in that: include: S1: Receive a task release instruction, and retrieve a task level and task type based on the task release instruction; S2: Determine the ability level based on the task type and the task level, and query whether the preset ability information of the order-receiving user is less than the ability level; if so, filter the task for the order-receiving user; if not, grant the order-receiving user permission to accept the task; S3: Based on the open task acceptance authority, receiving an acceptance instruction from the order-accepting user, generating a task package based on the acceptance instruction, and removing the task from the shelf; S4: Collect identity information based on the task package. After the identity information is consistent with the preset order-taking user information, match the supervision method and perform supervision.

2. A task scheduling method based on a gig system according to claim 1, characterized in that: The regulatory approaches include: S40: matching a supervision instruction based on the receiving instruction, and collecting a working image based on the supervision instruction; S41: selecting a processed image from the working image based on a preset finished product feature, matching a corresponding finished product image based on the processed image, and matching the processing feature based on the finished product image; S42: determining a detection area in the processed image based on a comparison relationship between the processed features and the finished product image; S43: collecting features to be processed in the detection area; S44: Determining a production score based on the comparison between the to-be-processed feature and the processing feature; S45: If the production score is lower than the preset required score, a modification reminder is issued, and a modification method of the processing feature is determined based on the processing feature; if the production score is not lower than the preset required score, the inspection is continued and the processing time is recorded.

3. A task scheduling method based on a gig system according to claim 2, characterized in that: A cheating detection method is also included between step S41 and step S42, and the cheating detection method includes: S410: Identifying and marking the arm from the working image based on preset arm features, and calculating intersection points on extension lines of adjacent positions; S411: In a preset unit time, sequentially connect the intersection points of the same part to obtain a movement trend; and determine a moving direction based on the movement trend; S412: When the moving direction is consistent with the preset approach direction, collecting a warning distance between the intersection point and a preset warning line; S413: When the warning distance is less than a preset reference distance, issuing a warning; when the work image does not contain the finished product feature, accumulating the disappearance time; S4140: If the disappearance time is not greater than the preset production time, continue testing; S4141: If the disappearance time is greater than the preset production time, an alarm is issued to prompt the order-taking user to display the product and capture a display image, and the detection time is recorded; S415: Determine an image change area based on the display image and the working image before the cumulative disappearance time; S416: Collecting a number of cost features in the image change area and defining it as a newly added number, and calculating a newly added efficiency based on the newly added number and the detection time; S4170: If the newly added efficiency is greater than the preset production efficiency, an alarm is issued and recorded; S4171: If the newly added efficiency is not greater than the preset production efficiency, a prompt is given and the number of detections is recorded. If the number of detections is greater than the preset maximum number of detections or cheating production is detected, the task is stopped and the user is prompted to stop the task.

4. The task scheduling method based on the gig system according to claim 3, characterized in that: A cover detection method is also included between step S410 and step S411, and the cover detection method includes: S4100: Selecting arm features in the image based on the working image and defining the image as a detection image; S4101: Determine a frame selection area based on the detection image; S4102: When the frame selection area reaches a preset alert area, determining the covering position of the detection image in the working image; determining a gap image in the working image based on the covering position, and determining an avoidance vector based on the gap image; and determining a preset rotation angle of the camera device (4) based on the avoidance vector; S4103: performing an evasive movement according to the evasive vector, rotating according to the rotation angle, and updating the working image; S4104: If the frame selection area of ​​the updated working image is the same as the frame selection area of ​​the original image, supervision is performed using a preset auxiliary supervision method.

5. The task scheduling method based on the gig system according to claim 4, characterized in that: The auxiliary supervision method includes: S41040: Segmenting the gap image using a preset segmentation line to obtain segmented regions; selecting a segmented region with a large area from the segmented regions and defining it as a rotation region; and determining a rotation angle according to the rotation region. S41041: determining a preset movable reflector (1) based on the rotation angle, and determining a reflection angle of the movable reflector (1) based on a preset reflection position and a processed image; S41042: Controlling the movable reflector (1) to rotate based on the reflection angle, and updating the working image according to the rotation angle.

6. The task scheduling method based on the gig system according to claim 2, characterized in that: After step S40, a verification method is also included, and the verification method includes: S400: Acquiring a contrast image on a preset contrast area in the working image; S401: determining a contrast color difference based on the comparison image and a preset reference contrast feature; S403: When the contrast color difference is less than a preset maximum contrast color difference value, issuing a replacement instruction; S404: Based on the replacement instruction, the preset protective film (4) is replaced and the working image is updated.

7. The task scheduling method based on the gig system according to claim 6, characterized in that: After step S404, a cleaning method is also included, and the cleaning method includes: S405: outputting a cleaning instruction during replacement, controlling a preset light-emitting device to illuminate the protective film (4) in response to the cleaning instruction, and collecting the intensity of light passing through the protective film (4); S406: Selecting a cleaning area based on the light intensity and a preset light transmittance, and determining a cleaning intensity according to the light intensity; S407: Based on the cleaning area, a preset cleaning device is controlled to align with the cleaning area, and the protective film (4) is cleaned according to the cleaning intensity, and the light-emitting device is controlled to emit light; S408: Stop cleaning when the light intensity is not less than the preset light transmission intensity.

8. The task scheduling method based on the gig system according to claim 1, characterized in that: Also included is an air blowing removal method, the air blowing removal method comprising: S50: selecting a coverage area from the working image frame based on a preset coverage feature, defining it as a coverage image, and outputting a removal instruction; S51: Based on the removal instruction, retrieve historical working images within a preset unit time, select a close image showing a magnification trend based on the historical working images, and match the covering material based on the close image; S52: Matching the blowing intensity based on the covering material; S53: Controlling a preset blowing device to blow up the covering material according to the blowing intensity, and updating the working image; S54: issuing an alarm when the working image does not change.

9. The task scheduling method based on the gig system according to claim 2, characterized in that: When the production score is not lower than the preset required score in step S45, an efficiency improvement method is also included, and the efficiency improvement method includes: S440: Determine, based on the processing time, a completion time point of the finished product feature in the working image and an appearance time point of a preset roughcast feature in the working image, and calculate the user's stage efficiency based on the completion time point and the appearance time point; S441: When a processing feature is detected, the number of detections is recorded. When the number of detections reaches a preset reference number, the initial time point and the completion time point of each processing time are determined, and the stage efficiency of each processing time when processing different processing features is calculated to determine the most efficient stage. S4420: If the highest efficiency stages are different, extracting the working images of the respective highest efficiency stages and recording common features in the working images; S44200: If the common feature appears in the other working images, check its sorting order. If the sorting order is greater than that of the working images without the common feature, record it as a beneficial feature; S44201: Determine a simulation method based on the beneficial features and output a simulation instruction; S44202: Based on the simulation instruction, prompt the customer to simulate the production environment using the simulation method; S4421: If the stages of the highest production efficiency are the same, determining the task station according to the highest production efficiency; S44210: Prompt the order-receiving user to arrive at the task station based on the task station; and supervise the order-receiving user after the user arrives at the task station.

10. A task scheduling system based on a gig system, characterized in that: include: The acquisition module is used to obtain task release instructions, acceptance instructions, and identity information; A memory, configured to store a task scheduling method based on a gig-job system according to any one of claims 1 to 9; The processor is configured to load, execute, and implement the program stored in the memory.

Images