A method and system for location-triggered photographing of a pre-processed appearance detection device

Abstract

The application relates to the field of industrial visual detection, and discloses a position trigger photographing method and system of a pretreatment appearance detection device, which comprises the following steps: establishing a detection object photographing task when a detection object enters a detection area; generating a target photographing position list according to device configuration data and a conveying reference position; writing the target photographing position list into a to-be-triggered queue according to a conveying direction; cyclically reading a conveying current position, generating a camera trigger request; updating a trigger state of a corresponding target photographing position; and updating a task state of the detection object photographing task according to the trigger state and an image binding state. The conveying position of the detection object when the detection object enters the detection area is taken as the conveying reference position, and the target photographing position is managed in a queue according to the conveying direction, so that the correlation management among the detection object photographing task, the target photographing position, a camera trigger event and a camera collected image is realized.

Description

Technical Field

[0001] This invention relates to the field of industrial visual inspection technology, specifically to a location-triggered imaging method and system for pre-processing appearance inspection equipment. Background Technology

[0002] With the development of machine vision and industrial automation technologies, appearance inspection equipment is widely used for surface quality inspection of glass, films, sheets, electronic components, and other industrial products. Appearance inspection equipment typically uses a conveyor mechanism to move the object to be inspected sequentially through multiple image acquisition stations. Cameras at each station acquire images of different areas of the object, and image processing programs then identify contamination, scratches, foreign objects, edge defects, and localized appearance anomalies. To cover different inspection areas, multiple images are usually acquired continuously at the same camera station during the object's movement.

[0003] Existing position-triggered photography methods for pre-processing appearance inspection equipment typically start a delay timer after detecting the object entering the equipment, and trigger the corresponding camera to take a picture when the preset delay time is reached. For multiple images from the same camera station, trigger signals are continuously output according to a preset time interval. However, the time delay and time interval-based photography triggering method makes it difficult to keep the multiple photography trigger positions of multiple camera stations corresponding to the actual conveying position of the object when the conveyor stops or the conveying speed changes. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a position-triggered photography method and system for pre-processing appearance inspection equipment, which solves the problem that existing position-triggered photography methods for pre-processing appearance inspection equipment are difficult to ensure that the trigger positions of multiple photography at multiple camera stations correspond to the actual transport position of the object being inspected.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a position-triggered image capture method for a pre-processing appearance inspection device, comprising the following steps: Acquire the feed detection signal and the conveying position signal, establish a photo capture task for the detected object when it enters the detection area, and record the conveying reference position; A target photo location list is generated based on the equipment configuration data and the conveying reference position. The target photo location list includes camera station information, image sequence number information, target photo location information, and trigger status information. Write the list of target image locations into the queue to be triggered according to the conveying direction; The system continuously reads the current location and generates a camera trigger request based on the arrival relationship between the current location and the target image location information. The camera trigger signal is output according to the camera trigger request, and the trigger status of the corresponding target shooting position is updated; The images captured by the camera are bound to the camera station information, image sequence information, and trigger location information in the object detection photography task, and the task status of the object detection photography task is updated according to the trigger status and image binding status.

[0006] By adopting the above technical solution, the transport position when the object enters the inspection area is used as the transport reference position. A list of target shooting positions with camera station information, image sequence information, and trigger status information is generated according to the equipment configuration data. The target shooting positions are queued according to the transport direction. Thus, the trigger signal of the corresponding camera can be output according to the arrival relationship between the current transport position and the target shooting position. This realizes the association management between the object shooting task, the target shooting position, the camera trigger event, and the camera acquisition image. This solves the problem that the position trigger shooting method of the existing pre-processing appearance inspection equipment is difficult to keep the multiple shooting trigger positions of multiple camera stations corresponding to the actual transport position of the object.

[0007] Preferably, the step of establishing a photographing task for the object being inspected when it enters the inspection area includes: reading the feeding detection signal within the control cycle of the preprocessing appearance inspection equipment; reading the current conveying position when the feeding detection signal meets the valid feeding edge condition; writing the read current conveying position into the photographing task for the object being inspected as the conveying reference position; assigning an object number to the photographing task for the object being inspected; and writing the object number, the conveying reference position, the conveying direction identifier, the task status identifier, and the image attribution record into the task buffer.

[0008] Preferably, the equipment configuration data includes the number of camera stations, the mapping relationship of camera trigger output points, the camera station installation offset, the number of photos taken per station, the image position offset, the trigger compensation amount, and the position judgment tolerance.

[0009] Preferably, generating the target image capture position list based on the equipment configuration data and the conveying reference position includes: using the conveying reference position as the position reference for the current object image capture task; reading the equipment configuration data associated with the object image capture task; reading the camera station installation offset according to the camera station order; reading the image position offset and trigger compensation amount according to the image sequence number; forming target image capture position information in the same conveying coordinate system by combining the camera station installation offset, the image position offset, and the trigger compensation amount according to the conveying direction identifier; and combining the target image capture position information with the object detection number, camera station information, image sequence number information, camera trigger output point information, and trigger status information to form the target image capture position list.

[0010] Preferably, the step of reading the image position offset and trigger compensation amount in sequence according to the image number includes: when acquiring images at a fixed position interval at the same camera station, generating an image position offset by calling the position interval sequence associated with the camera station according to the image number; when acquiring images at unequal position intervals at the same camera station, generating an image position offset by calling the preset position point table associated with the camera station according to the image number; writing the image position offset and the corresponding image number into the target point record; and writing the trigger compensation amount and the corresponding target point record into the same target shooting position list.

[0011] Preferably, the step of writing the target image location list into the trigger queue according to the conveying direction includes: reading the conveying direction identifier in the detection object image task; when the conveying direction identifier is forward conveying, writing the target image location list into the trigger queue in ascending order of conveying coordinates; when the conveying direction identifier is reverse conveying, writing the target image location list into the trigger queue in descending order of conveying coordinates; and when there are two or more detection object image tasks, writing the untriggered target image locations in each detection object image task into the global trigger queue according to the conveying direction and task establishment order.

[0012] Preferably, generating a camera trigger request includes: reading the current transport position in each control cycle; reading the target image position in the pending trigger state from the pending trigger queue; determining whether the current transport position has reached or passed the target image position along the transport direction based on the transport direction identifier; generating a camera trigger request including the detection object number, camera station information, image sequence number information, and camera trigger output point information when the current transport position meets the position judgment tolerance and the target image position is in the pending trigger state; and maintaining the target image position in the pending trigger state when the current transport position does not meet the position judgment tolerance.

[0013] Preferably, updating the trigger status of the corresponding target image capture position includes: parsing the camera station information in the camera trigger request; determining the target camera trigger output point according to the camera trigger output point mapping relationship; reading the trigger permission status of the target camera station; when the trigger permission status is "allow triggering", controlling the target camera trigger output point to output a trigger pulse; after the trigger pulse is output, updating the trigger status of the corresponding target image capture position to "triggered"; and recording the actual trigger position, trigger time marker, camera station information, image sequence number information, and camera trigger output point information.

[0014] Preferably, updating the task status of the detection object photography task based on the triggering state and image binding state includes: maintaining the task status of the untriggered target photography position, the pending trigger queue, and the detection object photography task when the current transport position is in a stopped state or a speed change state; continuing to generate a camera trigger request based on the arrival relationship between the current transport position and the target photography position information after the current transport position resumes changing along the transport direction; writing the camera-acquired image into the image ownership record of the corresponding detection object photography task after the camera returns the acquired image; marking the detection object photography task as completed when all target photography positions in the same detection object photography task are in a triggered state and the image ownership record meets the task end condition; and marking the detection object photography task as abnormal when there is a position feedback anomaly, a target photography position exceeding the boundary, a trigger output anomaly, or an image return anomaly.

[0015] A position-triggered image capture system for a pre-processing appearance inspection device, applied to the above-mentioned method, includes: The signal acquisition module is used to acquire the feed detection signal and the conveying position signal. When the object to be detected enters the detection area, a photo capture task is set up and the conveying reference position is recorded. The target location generation module is used to generate a target photo location list based on the equipment configuration data and the conveying reference location. The target photo location list includes camera station information, image sequence number information, target photo location information, and trigger status information. The queue management module is used to write the list of target photo locations into the queue to be triggered according to the conveying direction; The location triggering judgment module is used to cyclically read the current location being transmitted and generate a camera triggering request based on the arrival relationship between the current location being transmitted and the target image capture location information; The trigger output module is used to output a camera trigger signal according to the camera trigger request and update the trigger status of the corresponding target shooting position; The task status management module is used to bind the images acquired by the camera with the camera position information, image sequence information and trigger position information in the detection object photography task, and update the task status of the detection object photography task according to the trigger status and image binding status.

[0016] This invention provides a position-triggered imaging method and system for a pre-processing appearance inspection device. It has the following beneficial effects: 1. This invention uses the transport position of the object entering the inspection area as the transport reference position, generates a target photo-taking position list with camera station information, image sequence number information, and trigger status information based on equipment configuration data, and manages the target photo-taking positions in a queue according to the transport direction. This enables the output of the corresponding camera trigger signal based on the arrival relationship between the current transport position and the target photo-taking position, realizing the association management between the object photo-taking task, the target photo-taking position, the camera trigger event, and the camera acquired image. This solves the problem that the position-triggered photo-taking method of existing pre-processing appearance inspection equipment is difficult to keep the multiple photo-taking trigger positions of multiple camera stations corresponding to the actual transport position of the object.

[0017] 2. This invention establishes an independent image capture task for each detection object and associates the target image capture location with camera station information, image sequence number information, trigger status information, and image attribution information. This enables unified management of multiple camera stations and images acquired by multiple cameras according to the detection object. In the case of continuous feeding and multiple image capture tasks for detection objects existing in parallel, the triggering order of each target image capture location and the attribution relationship of each camera image can be clearly defined, reducing duplicate triggering, missed triggering, and image attribution confusion.

[0018] 3. This invention sets the camera station installation offset, the number of photos taken per station, the image position offset, the trigger compensation amount, and the camera trigger output point mapping relationship as equipment configuration data, so that the target photo position can be configured according to the equipment structure and the inspection area. When the camera station layout, the number of photos taken, or the image acquisition area are adjusted, a new list of target photo positions can be formed by updating the corresponding equipment configuration data, which is convenient for adapting to different pre-processing appearance inspection equipment and inspection tasks.

[0019] 4. This invention uses the actual transport position of the object to be detected as the basis for camera photo capture triggering, and converts the photo capture position of each camera station into the corresponding target photo capture position for management. This makes the photo capture triggering process no longer dependent on fixed time delays and fixed time intervals. When the conveying mechanism stops, decelerates, accelerates or restarts, the target photo capture position that has not yet been reached can remain in a pending triggering state, and the triggering is completed after the object to be detected continues to move to the corresponding position. This ensures that the multiple photo capture positions of multiple camera stations correspond to the actual transport position of the object to be detected. Attached Figure Description

[0020] Figure 1 This is a flowchart of a position-triggered image capture method for a pre-processing appearance inspection device proposed in this invention; Figure 2 This is an architecture diagram of a position-triggered image capture system for a pre-processing appearance inspection device proposed in this invention; Figure 3This is a schematic diagram of the application structure of the location-triggered image capture algorithm proposed in an embodiment of the present invention; Figure 4 This is a target triggering position planning diagram for multiple images from five camera stations proposed in an embodiment of the present invention; Figure 5 This is a schematic diagram illustrating the difference between the time-triggered method and the position-triggered method proposed in the embodiments of the present invention. Detailed Implementation

[0021] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Example 1: In a first embodiment of the present invention, the present invention provides a location-triggered image capture method for a pre-processed appearance inspection device, such as... Figure 1 As shown, it includes the following steps: Please see the appendix Figure 1 Appendix Figure 3 To be continued Figure 5 It acquires the feed detection signal and the conveying position signal, establishes a photo capture task for the detected object when the detected object enters the detection area, and records the conveying reference position; Furthermore, when the object enters the inspection area, an object photography task is established, including: reading the feeding detection signal within the control cycle of the pre-processing appearance inspection equipment; reading the current conveying position when the feeding detection signal meets the effective feeding edge condition; writing the read current conveying position into the object photography task as the conveying reference position; assigning an object number to the object photography task; and writing the object number, conveying reference position, conveying direction identifier, task status identifier, and image attribution record into the task buffer.

[0023] Specifically, this step is used to determine the establishment time of the photo-taking task of the object after it enters the pre-processing appearance inspection equipment based on the feeding detection signal, and to solidify the conveying position corresponding to that time as the position reference for the subsequent target photo-taking position; by establishing task records for different objects, the subsequent camera station information, image sequence number information, trigger status information and image attribution information can all be associated with the corresponding object.

[0024] In this embodiment, the control program reads the feed detection signal and the conveying position signal according to a preset control cycle. The current sampling state of the feed detection signal is recorded as follows: The sampling state of the previous control cycle is recorded as The effective edge marking of the feed is recorded as The method of determination is as follows ,in, This indicates the valid status of the feed detection unit output during the current control cycle. This indicates the recognition result of the detected object entering the detection area from the feed detection position. In this embodiment, the control period is preferably 10 milliseconds, and the feed detection signal is continuously valid for 30 milliseconds before participating in edge judgment.

[0025] when When in an active state, the control program reads the current position count value of the encoder. And convert it to convey the current position. : ,in, Indicates the reference transport position. Indicates the reference encoder count value. This represents the conveying displacement corresponding to a unit encoder count; in this embodiment, The preferred count is 0.01 mm per count. The control program writes the current transport position into the object detection imaging task and uses it as the transport reference position. .

[0026] The control program assigns a unique detection object number to each newly established detection object photography task, and writes the detection object number, transport reference position, transport direction indicator, task status indicator, and image attribution record into the task buffer. The task status indicator is initially set to "pending planning," and the image attribution record is initially set to "empty."

[0027] Generally, the task buffer stores multiple image capture tasks for each detected object in the order they enter the detection area. The task buffer preferably allows for the simultaneous storage of twenty image capture tasks. When the conveying direction changes, the conveying direction identifier is updated accordingly, but the already recorded conveying reference position remains unchanged. This process provides a unified data entry point for subsequent equipment configuration data retrieval and the generation of a target image capture location list.

[0028] A target photo location list is generated based on the equipment configuration data and the conveying reference position. The target photo location list includes camera station information, image sequence number information, target photo location information, and trigger status information. Furthermore, the equipment configuration data includes the number of camera stations, the mapping relationship of camera trigger output points, the installation offset of camera stations, the number of photos taken per station, the image position offset, the trigger compensation amount, and the position judgment tolerance.

[0029] Furthermore, a target photo location list is generated based on the equipment configuration data and the conveying reference position, including: using the conveying reference position as the position reference for the current photo-taking task of the detected object; reading the equipment configuration data associated with the photo-taking task of the detected object; reading the camera station installation offset according to the camera station sequence; reading the image position offset and trigger compensation amount according to the image sequence number; forming target photo location information under the same conveying coordinate system by combining the camera station installation offset, image position offset, and trigger compensation amount according to the conveying direction indicator; and combining the target photo location information with the detected object number, camera station information, image sequence number information, camera trigger output point information, and trigger status information to form a target photo location list.

[0030] Furthermore, the image position offset and trigger compensation amount are read in sequence according to the image number, including: when acquiring images at a fixed position interval at the same camera station, the image position offset is generated by calling the position interval sequence associated with that camera station according to the image number; when acquiring images at unequal position intervals at the same camera station, the image position offset is generated by calling the preset position point table associated with that camera station according to the image number; the image position offset and the corresponding image number are written into the target point record; and the trigger compensation amount and the corresponding target point record are written into the same target shooting position list.

[0031] Specifically, after the object photography task is established and the transport reference position is obtained, the control program calls the equipment configuration data corresponding to the current equipment and the object photography task, and maps the spatial installation relationship of each camera station, the acquisition position relationship of multiple images, and the position compensation during the camera triggering process to the transport coordinate system, thereby forming a target photography position list that corresponds one-to-one with the current object. This target photography position list serves as the data source for subsequent queue sorting and position trigger judgment, so that each target photography position can maintain a clear camera station affiliation, image order, and triggering status.

[0032] In this embodiment, the control program transmits the reference position. This serves as the location reference for the current object being photographed, and the number of camera stations is read from the equipment configuration data. Camera trigger output point mapping relationship, camera workstation installation offset Number of photos taken per workstation Image position offset Trigger compensation amount and position judgment tolerance ;in, Indicates the camera workstation number. This indicates the image number within the corresponding camera station.

[0033] No. The first camera station The target location corresponding to the image. Determined according to the following formula ,in, This indicates the conveyor position recorded when the object being inspected passes the feed inspection position. This indicates the direction of transport; the value is 1 for forward transport and -1 for reverse transport. Indicates the feed detection position up to the [number]. The conveying distance of the camera station's photo reference point Indicates the first The image is offset from the transport position relative to the corresponding camera station's photo reference point. This indicates the position correction amount used to compensate for trigger output delay, camera response delay, and installation deviation.

[0034] At the same camera station, images can be acquired at fixed intervals; in this case, the image position shifts. Generate according to the following formula ,in, Indicates the first The transport position interval between two adjacent images at a single camera station; the same camera station can use unequal distance positions for image acquisition; the control program retrieves images from a preset position table according to the image sequence number. Read the corresponding position value from the image and determine the image position offset. ,in, Indicates the relationship with the first A table of preset location points associated with each camera station. This indicates that the location in the point table is related to the first... The corresponding transport position offset for each image is recorded in the preset position table as 0 mm, 40 mm, 95 mm, 155 mm and 220 mm.

[0035] The control program generates target point records sequentially according to the camera station order and image sequence number, and writes the detection object number, camera station information, image sequence number information, target image capture position information, camera trigger output point information, and trigger status information into the target point record. The initial trigger state is set to the pending trigger state. The camera trigger output point is read from the camera trigger output point mapping relationship according to the camera station information. In some embodiments, the five camera stations are mapped to output points Y20, Y21, Y22, Y23, and Y24, respectively. After all target point records are generated, the control program summarizes them into a target image capture position list corresponding to the current detection object image capture task, for subsequent establishment of a pending trigger queue according to the conveying direction.

[0036] Write the list of target photo locations into the queue to be triggered according to the direction of transport; Furthermore, the target image location list is written into the trigger queue according to the conveying direction, including: reading the conveying direction identifier in the detection object image task; when the conveying direction identifier is forward conveying, writing the target image location list into the trigger queue in ascending order of conveying coordinates; when the conveying direction identifier is reverse conveying, writing the target image location list into the trigger queue in descending order of conveying coordinates; when there are two or more detection object image tasks, writing the untriggered target image locations in each detection object image task into the global trigger queue according to the conveying direction and task creation order.

[0037] Specifically, after the target image location list is generated, the control program sorts the target image locations uniformly according to the transport direction identifier recorded in the object image capture task, and converts the sorting result into a queue to be triggered that can be read item by item by the position trigger judgment program, so that the arrangement direction of the target image locations is consistent with the actual movement direction of the object being detected.

[0038] In this embodiment, the control program reads the conveying direction indicator. and for the first Generate a queue sorting value for each target's photographed location. : ,in, Indicates the first The target point's location in the transport coordinate system for taking the picture. Indicates the conveying direction of the object being detected; when conveying in the forward direction... Preferably 1, during reverse conveying The preferred value is -1. This indicates the queue sorting value of the target point along the current transport direction.

[0039] Specifically, the control procedure follows The images are written into the queue to be triggered in ascending order. When the image is conveyed in the forward direction, the target image position corresponding to the processing is arranged in ascending order according to the conveying coordinates. When the image is conveyed in the reverse direction, the target image position corresponding to the processing is arranged in descending order according to the conveying coordinates. Each queue record includes the detection object number, camera station information, image sequence number information, target image position information, camera trigger output point information, and trigger status information.

[0040] Two or more detection object photography tasks can exist simultaneously in the same device; the control program extracts the target points in the pending trigger state in each detection object photography task and writes them into the global pending trigger queue according to the queue sorting value and task creation order. When the queue sorting values ​​of two target points are the same, the target point with the earlier task creation time is written first; in this embodiment, the maximum number of records in the global pending trigger queue is preferably 1000.

[0041] Triggered target points are no longer written to the global pending trigger queue. After a new detection object photo task is created, only the pending target points in that task are inserted into the corresponding sorting position, thus providing a continuous and clear target point reading order for subsequent current position judgments.

[0042] The system continuously reads the current location and generates a camera trigger request based on the arrival relationship between the current location and the target image location information. Furthermore, generating a camera trigger request includes: reading the current position of the conveyor in each control cycle; reading the target image position in the pending trigger state from the pending trigger queue; determining whether the current position of the conveyor has reached or passed the target image position along the conveyor direction based on the conveyor direction identifier; generating a camera trigger request including the detection object number, camera station information, image sequence number information, and camera trigger output point information when the current position of the conveyor meets the position judgment tolerance and the target image position is in the pending trigger state; and keeping the target image position in the pending trigger state when the current position of the conveyor does not meet the position judgment tolerance.

[0043] Specifically, after the trigger queue is established, the control program continuously acquires the actual delivery position of the object to be detected according to the control cycle, and compares the actual delivery position with the target photo position in the queue one by one to determine whether the object to be detected has entered the photo triggering range of the corresponding camera station, and provides request data for subsequent camera trigger signal output.

[0044] In this embodiment, the control program reads and transmits the current position in each control cycle. And retrieve the first one that is in the pending triggering state from the pending triggering queue. Target photo location Control cycle The preferred time is 10 milliseconds. The control program is based on the conveying direction indicator. Determine the arrival distance of the current location relative to the target image location. : ,in, This indicates the transport position read in the current control cycle. Indicates the first The target location corresponding to each target point. This indicates the direction of transport; the value is 1 for forward transport and -1 for reverse transport. This indicates the degree to which the object being detected has reached the target image position along the transport direction.

[0045] When satisfied And the first When a target's image capture location is in a pending trigger state, the control program generates a camera trigger request; among which... This indicates the position determination tolerance, used to absorb position deviations caused by position register refreshes and encoder count discrepancies. In this embodiment, The preferred size is 1 mm; the camera trigger request writes the detection object number, camera station information, image sequence number information, target image capture location information, and camera trigger output point information.

[0046] When the arrival conditions are not met, the control program maintains the pending trigger state of the corresponding target photo position and continues to judge in the next control cycle. When the current position passes through multiple target photo positions within a control cycle, the control program generates camera trigger requests in sequence according to the order in the pending trigger queue, thereby avoiding the omission of the reached target point.

[0047] Output a camera trigger signal according to the camera trigger request, and update the trigger status of the corresponding target shooting position; Furthermore, updating the trigger status of the corresponding target image capture position includes: parsing the camera station information in the camera trigger request; determining the target camera trigger output point according to the camera trigger output point mapping relationship; reading the trigger permission status of the target camera station; when the trigger permission status is "allowed to trigger", controlling the target camera trigger output point to output a trigger pulse; after the trigger pulse output is completed, updating the trigger status of the corresponding target image capture position to "triggered"; and recording the actual trigger position, trigger time marker, camera station information, image sequence number information, and camera trigger output point information.

[0048] Specifically, after a camera trigger request is generated, the control program calls the corresponding camera trigger output point based on the camera station information, and outputs a trigger pulse with a preset duration when the target camera station meets the trigger conditions. At the same time, the trigger event is saved for subsequent image binding and task status judgment.

[0049] In this embodiment, the control program parses the camera trigger request to obtain the detection object number, camera station information, image sequence number information, and camera trigger output point information, and determines the target camera trigger output point based on the camera trigger output point mapping relationship. The camera station number is... At that time, the corresponding target camera trigger output point is recorded as .

[0050] The control program reads the trigger permission status of the target camera station. The trigger output state is determined according to the following formula. : ,in, This indicates whether the target camera station is in a trigger-enabled state; the value is 1 when triggering is enabled and 0 when triggering is disabled. This indicates the status of a camera trigger request; the value is 1 when a valid camera trigger request exists. Indicates whether to output a trigger pulse to the target camera's trigger output point.

[0051] when When the value is 1, the control program keeps the target camera's trigger output point valid for a duration equal to the trigger pulse width. The state is then restored to an invalid state; in this embodiment, the trigger pulse width is... The preferred value is 20 milliseconds; the preferred target camera trigger output points for the five camera stations are Y20 to Y24.

[0052] After the trigger pulse is output, the control program updates the trigger status of the corresponding target image location to "triggered" and records the actual trigger location. The system includes trigger time markers, camera position information, image sequence number information, and camera trigger output point information. Once a target image is triggered, the target image position will no longer be included in subsequent position determinations. When the target camera position is in a prohibited triggering state, the control program will keep the corresponding target image position in a pending triggering state and reread the trigger permission state in the next control cycle.

[0053] The camera-captured images are bound to the camera position information, image sequence information, and trigger position information in the object detection photography task, and the task status of the object detection photography task is updated according to the trigger status and image binding status.

[0054] Furthermore, the task status of the detection object photography task is updated based on the trigger status and image binding status, including: when the current transport position is in a stopped state or a speed change state, maintaining the task status of the untriggered target photography position, the pending trigger queue, and the detection object photography task; after the current transport position resumes changing along the transport direction, continuing to generate camera trigger requests based on the arrival relationship between the current transport position and the target photography position information; after the camera returns the acquired image, writing the acquired image into the image attribution record of the corresponding detection object photography task; when all target photography positions in the same detection object photography task are in a triggered state and the image attribution record meets the task termination condition, marking the detection object photography task as completed; when the detection object photography task has position feedback anomalies, target photography position out of bounds, trigger output anomalies, or image return anomalies, marking the detection object photography task as abnormal.

[0055] Specifically, after a camera trigger event is generated, the control program uses the detection object number, camera station information, image sequence number information, and actual trigger position stored in the trigger event to establish an image receiving index, and writes the image returned by the camera into the corresponding detection object's image capture task, thereby forming a correspondence between the target image capture position, the camera trigger event, and the camera's captured image; the control program also checks the trigger status and image binding status of each target image capture position to determine whether the detection object's image capture task continues to wait, completes, or enters an abnormal processing state.

[0056] In this embodiment, when the camera returns an image, it carries camera position information and an image reception timestamp. The control program extracts trigger events from the trigger event record that are not yet bound to images and correspond to the camera position information, determines the corresponding image sequence number according to the order of trigger time, and writes the image identifier, storage address, camera position information, image sequence number information, and actual trigger location of the image acquired by the camera into the image ownership record.

[0057] The first The trigger state of each target's photo location is denoted as Record the corresponding image binding state as When the trigger has been completed. A value of 1 indicates that image binding is complete. The value is 1. This indicates the completion status of the object capture task. Determined according to the following formula ,in, This indicates the total number of target image locations included in the current object detection image capture task. This indicates whether all target imaging locations of the detected object have been triggered and the corresponding images have been acquired. When When the value is 1, the control program marks the object detection and image capture task as completed. In this embodiment, when five camera stations each capture five images, The preferred value is 25.

[0058] The control program identifies the conveying status based on the change in conveying position within adjacent control cycles. The change in conveying position is denoted as... ,in, This indicates the current position of the current control cycle. This indicates the current position of the transport in the previous control cycle. When The absolute value is not greater than the stopping threshold. At the same time, maintain the current state of the untriggered target image capture location, the pending trigger queue, and the detection object image capture task; in this embodiment, stop judging the threshold. The preferred value is 0.1 mm. After the current position of the conveyor resumes its change along the conveying direction, the control program continues to execute the target image position arrival judgment without needing to re-establish the object image task.

[0059] If the camera fails to receive the corresponding image after the image waiting time has elapsed following the camera trigger, an image return error is generated. In this embodiment, the image waiting time is preferably 3000 milliseconds. When a position feedback interruption, a target image capture position exceeding the effective delivery area, an incomplete trigger output, or an image return error is detected, the control program marks the detected object image capture task as an abnormal state and retains the corresponding abnormal type and target point record for subsequent processing.

[0060] Example 2: In a second embodiment of the present invention, the present invention provides a position-triggered imaging system for a pre-processing appearance inspection device, such as... Figure 2 As shown, it includes the following modules: The signal acquisition module is used to acquire the feed detection signal and the conveying position signal. When the object to be detected enters the detection area, a photo capture task is set up and the conveying reference position is recorded. The target location generation module is used to generate a list of target photo locations based on the equipment configuration data and the transport reference location. The target photo location list includes camera station information, image sequence number information, target photo location information, and trigger status information. The queue management module is used to write the list of target photo locations into the queue to be triggered according to the conveying direction; The location trigger judgment module is used to continuously read the current location and generate a camera trigger request based on the arrival relationship between the current location and the target image location information; The trigger output module is used to output a camera trigger signal according to the camera trigger request and update the trigger status of the corresponding target shooting position; The task status management module is used to bind the images acquired by the camera with the camera position information, image sequence information and trigger position information in the object detection photography task, and update the task status of the object detection photography task according to the trigger status and image binding status.

[0061] When glass objects continuously enter the pre-processing appearance inspection equipment, multiple camera stations set along the conveying direction need to complete multiple image acquisitions at their respective inspection positions. Because the conveying mechanism may be affected by downstream process waiting, equipment interlocking, manual pauses, and speed adjustments during operation, using a fixed time delay for triggering image capture can easily lead to inconsistencies between the triggering time and the actual position of the object being inspected. Furthermore, continuous feeding results in multiple image capture tasks existing in parallel, requiring unified data management for each target's image capture position, camera station, image sequence number, and image attribution. To address these issues, a position-triggered image capture system for the pre-processing appearance inspection equipment, as provided in this invention, is adopted. Its architecture is as follows: Figure 2 As shown. The specific implementation process of this system is as follows: First, the signal acquisition module continuously acquires the feeding detection signal and the conveying position signal. When the object enters the detection area, it establishes a detection object photo task and records the conveying reference position to provide a position reference for the generation of the target photo position. Subsequently, the target location generation module calls the equipment configuration data and generates a target photo location list containing camera station information, image sequence number information, target photo location information and trigger status information based on the transport reference position; Next, the queue management module arranges the target photo location list according to the conveying direction and writes the target photo locations that are in the pending triggering state into the pending triggering queue to determine the triggering order of each target photo location; Then, the position triggering judgment module reads the current position of the transmission in a loop and compares the current position of the transmission with the target photo position in the queue to be triggered. When the current position of the transmission reaches or crosses the target photo position, a camera triggering request is generated. The trigger output module determines the corresponding camera trigger output point based on the camera trigger request, outputs the camera trigger signal, and updates the trigger status of the corresponding target shooting position to the triggered status. Finally, the task status management module binds the images captured by the camera with the camera station information, image sequence information, and trigger position information. When the conveying stops or the speed changes, it maintains the current state of the unfinished task and updates the image capture task of the detected object to a waiting state, a completed state, or an abnormal state according to the trigger state and the image binding state. Through the collaborative processing of the above modules, the image capture triggering and task management based on the actual conveying position of the detected object are realized.

[0062] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A position-triggered image capture method for a pre-processing appearance inspection device, characterized in that, Includes the following steps: Acquire the feed detection signal and the conveying position signal, establish a photo capture task for the detected object when it enters the detection area, and record the conveying reference position; A target photo location list is generated based on the equipment configuration data and the conveying reference position. The target photo location list includes camera station information, image sequence number information, target photo location information, and trigger status information. Write the list of target image locations into the queue to be triggered according to the conveying direction; The system continuously reads the current location and generates a camera trigger request based on the arrival relationship between the current location and the target image location information. The camera trigger signal is output according to the camera trigger request, and the trigger status of the corresponding target shooting position is updated; The images captured by the camera are bound to the camera station information, image sequence information, and trigger location information in the object detection photography task, and the task status of the object detection photography task is updated according to the trigger status and image binding status.

2. The position-triggered imaging method for a pre-processing appearance inspection device according to claim 1, characterized in that: The step of establishing a photographing task for the object being inspected when it enters the inspection area includes: reading the feeding detection signal within the control cycle of the pre-processing appearance inspection equipment; reading the current conveying position when the feeding detection signal meets the effective feeding edge condition; writing the read current conveying position into the photographing task for the object being inspected as the conveying reference position; assigning an object number to the photographing task for the object being inspected; and writing the object number, the conveying reference position, the conveying direction identifier, the task status identifier, and the image attribution record into the task buffer.

3. The position-triggered imaging method for a pre-processing appearance inspection device according to claim 1, characterized in that: The equipment configuration data includes the number of camera stations, the mapping relationship of camera trigger output points, the installation offset of camera stations, the number of photos taken per station, the image position offset, the trigger compensation amount, and the position judgment tolerance.

4. The position-triggered imaging method for a pre-processing appearance inspection device according to claim 1, characterized in that: The step of generating a target image capture location list based on equipment configuration data and the conveying reference position includes: using the conveying reference position as the position reference for the current object image capture task; reading the equipment configuration data associated with the object image capture task; reading the camera station installation offset according to the camera station order; reading the image position offset and trigger compensation amount according to the image sequence number; forming target image capture location information in the same conveying coordinate system based on the conveying direction identifier, using the camera station installation offset, the image position offset, and the trigger compensation amount; and combining the target image capture location information with the object detection number, camera station information, image sequence number information, camera trigger output point information, and trigger status information to form a target image capture location list.

5. The position-triggered imaging method for a pre-processing appearance inspection device according to claim 4, characterized in that: The step of reading the image position offset and trigger compensation amount in sequence according to the image number includes: when acquiring images at a fixed position interval at the same camera station, generating an image position offset by calling the position interval sequence associated with the camera station according to the image number; when acquiring images at unequal position intervals at the same camera station, generating an image position offset by calling the preset position point table associated with the camera station according to the image number; writing the image position offset and the corresponding image number into the target point record; and writing the trigger compensation amount and the corresponding target point record into the same target shooting position list.

6. The position-triggered imaging method for a pre-processing appearance inspection device according to claim 1, characterized in that: The step of writing the target image location list into the trigger queue according to the conveying direction includes: reading the conveying direction identifier in the detection object image task; when the conveying direction identifier is forward conveying, writing the target image location list into the trigger queue in ascending order of conveying coordinates; when the conveying direction identifier is reverse conveying, writing the target image location list into the trigger queue in descending order of conveying coordinates; and when there are two or more detection object image tasks, writing the untriggered target image locations in each detection object image task into the global trigger queue according to the conveying direction and task establishment order.

7. The position-triggered imaging method for a pre-processing appearance inspection device according to claim 1, characterized in that: The process of generating a camera trigger request includes: reading the current transport position in each control cycle; reading the target image position in the pending trigger state from the pending trigger queue; determining whether the current transport position has reached or passed the target image position along the transport direction based on the transport direction identifier; generating a camera trigger request including the detection object number, camera station information, image sequence number information, and camera trigger output point information when the current transport position meets the position judgment tolerance and the target image position is in the pending trigger state; and maintaining the target image position in the pending trigger state when the current transport position does not meet the position judgment tolerance.

8. The position-triggered imaging method for a pre-processing appearance inspection device according to claim 1, characterized in that: The step of updating the trigger status of the corresponding target image capture position includes: parsing the camera station information in the camera trigger request; determining the target camera trigger output point according to the camera trigger output point mapping relationship; reading the trigger permission status of the target camera station; when the trigger permission status is "allow triggering", controlling the target camera trigger output point to output a trigger pulse; after the trigger pulse is output, updating the trigger status of the corresponding target image capture position to "triggered"; and recording the actual trigger position, trigger time marker, camera station information, image sequence number information, and camera trigger output point information.

9. The position-triggered imaging method for a pre-processing appearance inspection device according to claim 1, characterized in that: The step of updating the task status of the detection object photography task based on the trigger status and image binding status includes: maintaining the task status of the untriggered target photography position, the pending trigger queue, and the detection object photography task when the current transport position is in a stopped state or a speed change state; continuing to generate a camera trigger request based on the arrival relationship between the current transport position and the target photography position information after the current transport position resumes changing along the transport direction; writing the camera-acquired image into the image ownership record of the corresponding detection object photography task after the camera returns the acquired image; marking the detection object photography task as completed when all target photography positions in the same detection object photography task are in a triggered state and the image ownership record meets the task end condition; and marking the detection object photography task as abnormal when there is a position feedback anomaly, a target photography position exceeding the boundary, a trigger output anomaly, or an image return anomaly.

10. A position-triggered image capture system for a pre-processing appearance inspection device, characterized in that: The method applied to any one of claims 1 to 9 includes: The signal acquisition module is used to acquire the feed detection signal and the conveying position signal. When the object to be detected enters the detection area, a photo capture task is set up and the conveying reference position is recorded. The target location generation module is used to generate a target photo location list based on the equipment configuration data and the conveying reference location. The target photo location list includes camera station information, image sequence number information, target photo location information, and trigger status information. The queue management module is used to write the list of target photo locations into the queue to be triggered according to the conveying direction; The location triggering judgment module is used to cyclically read the current location being transmitted and generate a camera triggering request based on the arrival relationship between the current location being transmitted and the target image capture location information; The trigger output module is used to output a camera trigger signal according to the camera trigger request and update the trigger status of the corresponding target shooting position; The task status management module is used to bind the images acquired by the camera with the camera position information, image sequence information and trigger position information in the detection object photography task, and update the task status of the detection object photography task according to the trigger status and image binding status.

Images