Vision-based feed device, method and program product

Abstract

Embodiments of the present disclosure disclose a feeding device, method and program product based on visual detection. A specific embodiment of the device comprises a control unit, a feeding mechanism, a cutting mechanism, a visual detection module, a reagent strip conveying mechanism and a feeding mechanism. The feeding mechanism comprises a reagent strip feeding mechanism, which comprises a whole plate feeding mechanism, a whole plate storage mechanism and a plate feeding servo mechanism. The visual detection module comprises a reagent strip visual detection module. The reagent strip conveying mechanism is located on one side of the reagent strip visual detection module. One end of the reagent strip feeding conveyor belt is adjacent to the cutting mechanism. The embodiment can improve the qualified rate and feeding efficiency of the reagent strip.

Description

TECHNICAL FIELD

[0001] Embodiments of the present disclosure relate to the technical field of visual detection, and in particular, to a feeding device, method and program product based on visual detection. BACKGROUND

[0002] A reagent strip is a paper strip with dry chemical reagents, which can be used to detect chemical components in a liquid sample. It is widely used in various technical fields such as biology and medicine due to its convenience, reliability, low cost and other advantages. Most related reagent strips are whole plate reagent strips, which need to be cut by a cutter and then manually picked up for sorting to obtain qualified reagent strips.

[0003] However, the inventors have found that when the above feeding method is used to operate the reagent strip, the following technical problems often exist:

[0004] First, for some special reagent strips, manual picking up is easy to cause contamination of the reagent strip, thereby causing damage to the reagent strip. For some small-sized reagent strips, manual sorting is easy to ignore the quality problem of small size, and the sorting efficiency is low, thereby resulting in low qualified rate of reagent strips and slow feeding speed.

[0005] Second, the conveying speed and cutting speed are not determined according to the length of the reagent strip, which is easy to further cause low feeding efficiency and low qualified rate of reagent strips.

[0006] Third, it is not possible to accurately determine whether the reagent strip is qualified, resulting in a low qualified rate of the obtained reagent strips. SUMMARY

[0007] The summary part of the present disclosure is used to introduce the concepts in a brief form, which will be described in detail in the specific embodiments part. The summary part of the present disclosure is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions. Some embodiments of the present disclosure propose a feeding device, method and computer program product based on visual detection to solve one or more of the technical problems mentioned in the background part.

[0008] In a first aspect, some embodiments of the present disclosure provide a vision detection-based feeding device, comprising: a control unit, a feeding mechanism, a cutting mechanism, a vision detection module, a reagent strip conveying mechanism, and a feeding mechanism, wherein the feeding mechanism is in communication with the control unit, the feeding mechanism comprises a reagent strip feeding mechanism, the reagent strip feeding mechanism comprises a whole plate feeding mechanism, a whole plate storage mechanism, and a plate conveying servo mechanism, the control unit is configured to control the whole plate feeding mechanism to grab the whole plate reagent strip stored in the whole plate storage mechanism to the plate conveying servo mechanism, and control the plate conveying servo mechanism to deliver the carried whole plate reagent strip to the cutting mechanism; the cutting mechanism is located at one end of the plate conveying servo mechanism, the cutting mechanism is in communication with the control unit, and the control unit is further configured to control the cutting mechanism to cut the whole plate reagent strip located below the cutting mechanism according to a target length to obtain at least one target reagent strip; the vision detection module is in communication with the control unit, the vision detection module comprises a reagent strip vision detection module, the reagent strip vision detection module is located on one side of the cutting mechanism, the reagent strip vision detection module comprises a reagent strip camera, the control unit is further configured to control the reagent strip camera to take pictures of the target reagent strip to obtain a target reagent strip image, and to perform defect detection on the target reagent strip image to obtain a defect detection result; the reagent strip conveying mechanism is in communication with the control unit, and the reagent strip conveying mechanism is located on one side of the reagent strip vision detection module; the feeding mechanism comprises a reagent strip feeding conveyor belt, one end of the reagent strip feeding conveyor belt is adjacent to the cutting mechanism, the feeding mechanism is in communication with the control unit, and the control unit is further configured to control the reagent strip conveying mechanism to grab the target reagent strip corresponding to the defect detection result that does not exist defects to the reagent strip feeding conveyor belt.

[0009] Optionally, the control unit comprises an artificial intelligence chip, and the control unit is further configured to control the plate conveying servo mechanism to deliver the carried whole plate reagent strip to the cutting mechanism at a target feeding speed, and control the cutting mechanism to cut the received whole plate reagent strip according to the target length at a target cutting speed to obtain at least one target reagent strip, wherein the target feeding speed and the target cutting speed are obtained by analyzing the length of the target reagent strip, the feeding speed range of the plate conveying servo mechanism, and the cutting speed range of the cutting mechanism through the artificial intelligence chip included in the control unit, and the machine learning model carried by the artificial intelligence chip is trained through a training sample set.

[0010] In a second aspect, some embodiments of the present disclosure provide a feeding method based on visual detection, applied to the feeding device based on visual detection as described in the first aspect, wherein the feeding device comprises a feeding mechanism, a cutting mechanism, a visual detection module and a control unit, the feeding mechanism is in communication connection with the control unit, the feeding mechanism comprises a reagent strip feeding mechanism, the reagent strip feeding mechanism comprises a whole plate feeding mechanism, a whole plate storage mechanism and a plate feeding servo mechanism, the visual detection module comprises a reagent strip visual detection module, the reagent strip visual detection module comprises a reagent strip camera, and the method comprises: grabbing the whole plate reagent strip stored in the whole plate storage mechanism to the plate feeding servo mechanism through the whole plate feeding mechanism; conveying the whole plate reagent strip to the cutting mechanism through the plate feeding servo mechanism; cutting the whole plate reagent strip according to a target length through the cutting mechanism to obtain at least one target reagent strip; photographing the target reagent strip through the reagent strip camera to obtain a target reagent strip image; and detecting defects of the target reagent strip image.

[0011] In a third aspect, some embodiments of the present disclosure provide a computer program product comprising a computer program which, when executed by a processor, implements the method described in any implementation manner of the second aspect.

[0012] The above various embodiments of the present disclosure have the following beneficial effects: the feeding device based on visual detection of some embodiments of the present disclosure can improve the qualified rate and feeding efficiency of reagent strips. Specifically, the reason for the low qualified rate and slow feeding speed of reagent strips is that manual taking is easy to cause contamination of some special reagent strips, thereby causing damage to the reagent strips. For some small-sized reagent strips, manual sorting is easy to ignore the quality problems of small size, and the sorting efficiency is low. Based on this, the feeding device based on visual detection of some embodiments of the present disclosure includes a control unit, a feeding mechanism, a cutting mechanism, a visual detection module, a reagent strip handling mechanism and a feeding mechanism, wherein the feeding mechanism is in communication connection with the control unit, the feeding mechanism includes a reagent strip feeding mechanism, the reagent strip feeding mechanism includes a whole plate feeding mechanism, a whole plate storage mechanism and a plate feeding servo mechanism, the control unit is configured to control the whole plate feeding mechanism to grab the whole plate reagent strip stored in the whole plate storage mechanism to the plate feeding servo mechanism, and control the plate feeding servo mechanism to deliver the whole plate reagent strip carried to the cutting mechanism. The cutting mechanism is located at one end of the plate feeding servo mechanism, the cutting mechanism is in communication connection with the control unit, and the control unit is further configured to control the cutting mechanism to cut the whole plate reagent strip located below the cutting mechanism according to the target length to obtain at least one target reagent strip. The visual detection module is in communication connection with the control unit, the visual detection module includes a reagent strip visual detection module, the reagent strip visual detection module is located on one side of the cutting mechanism, the reagent strip visual detection module includes a reagent strip camera, and the control unit is further configured to control the reagent strip camera to shoot the target reagent strip to obtain a target reagent strip image, and to detect defects of the target reagent strip image to obtain a defect detection result. The reagent strip handling mechanism is in communication connection with the control unit, and the reagent strip handling mechanism is located on one side of the reagent strip visual detection module. The feeding mechanism includes a reagent strip feeding conveyor belt, one end of the reagent strip feeding conveyor belt is adjacent to the cutting mechanism, the feeding mechanism is in communication connection with the control unit, and the control unit is further configured to control the reagent strip handling mechanism to grab the target reagent strip corresponding to the defect detection result which does not exist defects to the reagent strip feeding conveyor belt. Because the feeding device based on visual detection is operated by a machine, manual contact with the reagent strip is avoided. Therefore, the probability of damage to the reagent strip can be reduced. Also, by automatic detection through the reagent strip visual detection module and taking the reagent strip by the reagent strip robot, the quality problems can be reduced and the sorting efficiency can be improved. Therefore, the feeding device based on visual detection of some embodiments of the present disclosure can improve the qualified rate and feeding efficiency of reagent strips. BRIEF DESCRIPTION OF DRAWINGS

[0013] The above and other features, aspects and advantages of the present disclosure will become more apparent after a reading of the following detailed description together with the accompanying drawings. Throughout the drawings, similar or same reference numerals are used to denote similar or same elements. It should be understood that the drawings are schematic and elements and features do not necessarily appear to scale.

[0014] Figure 1 is a structural schematic diagram of some embodiments of a vision detection based feeding device according to the present disclosure;

[0015] Figure 2 is a flow chart of some embodiments of a vision detection based feeding method according to the present disclosure. DETAILED DESCRIPTION

[0016] Embodiments of the present disclosure will be described in detail with reference to the drawings, wherein the same or similar components are denoted by the same or similar reference numerals, and thus repeated description is omitted. While the present disclosure is shown and described in connection with certain embodiments, it is not intended to be limited to the presented implementations, but instead, the present disclosure can be practiced within a wide and equivalent range of embodiments. Indeed, these embodiments are presented by way of example only and various changes might be made by those skilled in the art to the embodiments described without departing from the scope of the disclosure. Accordingly, the present disclosure is not limited to the embodiments described herein, but instead has broad applicability.

[0017] It is further noted that, for the sake of brevity, only some of the relevant details are shown in the drawings and described below. Embodiments of the present disclosure and features of embodiments can be combined with each other as technically feasible and without departing from the scope of the present disclosure.

[0018] It is to be noted that the terms "first", "second", and the like in the present disclosure are used only to distinguish different devices, modules or units, and do not imply the order of execution or the sequence of occurrence of the functions executed by these devices, modules or units.

[0019] It is to be noted that the terms "one", "multiple", etc. in the present disclosure are illustrative and not restrictive, and those skilled in the art should understand that "one" or "multiple" should be understood as "one or more" unless otherwise explicitly stated in the context.

[0020] The names of the messages or information exchanged between the devices in the embodiments of the present disclosure are used only for illustrative purposes, and are not intended to limit the scope of the messages or information.

[0021] The present disclosure will be described in detail with reference to the drawings and in conjunction with embodiments.

[0022] Figure 1 is a structural schematic diagram of some embodiments of a vision detection based feeding device according to the present disclosure. Figure 1The feeding mechanism 1, the cutting mechanism 2, the visual detection module 3, the reagent strip conveying mechanism 4 and the feeding mechanism 5 are included. The feeding mechanism 1 includes the whole plate storage mechanism 101, the whole plate feeding mechanism 102 and the plate conveying servo mechanism 103.

[0023] In some embodiments, the feeding device based on visual detection can include a control unit, a feeding mechanism 1, a cutting mechanism 2, a visual detection module 3, a reagent strip conveying mechanism 4 and a feeding mechanism 5. The control unit can be a unit for controlling each module included in the feeding device. For example, the control unit can include, but is not limited to, at least one of a PLC (Programmable Logic Controller), a SoC (System on Chip), an MCU (Microcontroller Unit) and a DSP (Digital Signal Processor). The feeding mechanism 1 can be a mechanism for conveying materials. The cutting mechanism 2 can be a mechanism for cutting whole plate reagent strips. The whole plate reagent strip can be a reagent strip that has not been cut. For example, the cutting mechanism 2 can include a cutter mechanism and a cut reagent strip storage mechanism. The visual detection module 3 can be a module for visually detecting the surface of a material. The feeding mechanism 1 can be in communication with the control unit. The reagent strip conveying mechanism 4 can be a robot for grabbing reagent strips. The feeding mechanism 5 can be a mechanism for conveying qualified reagent strips. For example, the feeding mechanism 5 can be a conveyor belt. The feeding mechanism 1 can include a reagent strip feeding mechanism. The feeding mechanism 1 can be a mechanism for conveying whole plate reagent strips. The reagent strip feeding mechanism can include a whole plate storage mechanism 101, a whole plate feeding mechanism 102 and a plate conveying servo mechanism 103. The whole plate feeding mechanism 102 can be a mechanism for grabbing whole plate reagent strips stored in the whole plate storage mechanism 101 to the plate conveying servo mechanism 103. For example, the whole plate feeding mechanism 102 can include a moving cylinder and a grabbing module. The whole plate storage mechanism 101 can be a storage bin for storing whole plate reagent strips. The plate conveying servo mechanism 103 can be a mechanism for conveying whole plate reagent strips. For example, the plate conveying servo mechanism 103 can include a servo motor and a conveying belt. The control unit can be configured to control the whole plate feeding mechanism 102 to grab whole plate reagent strips stored in the whole plate storage mechanism 101 to the plate conveying servo mechanism 103, and control the plate conveying servo mechanism 103 to convey the carried whole plate reagent strips to the cutting mechanism 2. Thus, the stored whole plate reagent strips can be conveyed to the lower part of the cutting mechanism 2.

[0024] In some embodiments, the cutting mechanism 2 can be located at one end of the plate feeding servo mechanism 103. Specifically, the cutting mechanism 2 can be located at the end of the plate feeding servo mechanism 103 away from the whole plate feeding mechanism 102. The cutting mechanism 2 can be in communication with the control unit. The control unit can be further configured to control the cutting mechanism 2 to cut the whole plate reagent strip located below the cutting mechanism 2 into at least one target reagent strip according to a target length. The value of the target length can be set by the worker. Thus, the whole plate reagent strip can be automatically cut into a reagent strip of a desired length by the cutting mechanism 2.

[0025] In some embodiments, the visual detection module 3 can be in communication with the control unit. The visual detection module 3 can include a reagent strip visual detection module. The reagent strip visual detection module can be a module for detecting the surface of a reagent strip. The reagent strip visual detection module can be located at one side of the cutting mechanism 2. Specifically, the reagent strip visual detection module can be located at the side of the cutting mechanism 2 away from the plate feeding servo mechanism 103. The reagent strip visual detection module can include a reagent strip camera. The reagent strip camera can be a device for photographing a target reagent strip. For example, the reagent strip camera can be a CCD (Charge Coupled Device) camera. The control unit can be further configured to control the reagent strip camera to photograph the target reagent strip to obtain a target reagent strip image, and to perform defect detection on the target reagent strip image to obtain a defect detection result. The defect detection result can represent whether the target reagent strip has a defect or does not have a defect. Specifically, the defect detection can include, but is not limited to, at least one of the following: length detection, width detection, damage detection. Thus, it can be determined whether the cut reagent strip has a defect.

[0026] In some embodiments, the reagent strip conveying mechanism 4 can be in communication with the control unit. The reagent strip conveying mechanism 4 can be located at one side of the reagent strip visual detection module.

[0027] In some embodiments, the feeding mechanism 5 can include a reagent strip feeding conveyor. One end of the reagent strip feeding conveyor can be adjacent to the cutting mechanism 2. The feeding mechanism 5 can be in communication with the control unit. The control unit can be further configured to control the reagent strip conveying mechanism 4 to grasp the target reagent strip corresponding to the defect detection result representing no defect to the reagent strip feeding conveyor. Thus, the target reagent strip without a defect can be supplied.

[0028] Optionally, the feeding device can include two sets of plate feeding servo mechanisms, two sets of cutting mechanisms, two sets of reagent strip visual inspection modules, and two sets of feeding mechanisms. In this way, the feeding efficiency can be improved.

[0029] Optionally, the feeding mechanism 1 can further include an upper cover feeding mechanism and a lower cover feeding mechanism. The upper cover feeding mechanism can be a mechanism for transporting material upper covers. The lower cover feeding mechanism can be a mechanism for transporting material lower covers. For example, the upper cover feeding mechanism and the lower cover feeding mechanism can both be conveyors. The material upper cover can be an upper cover of a packaging box for packaging reagent strips. The material lower cover can be a lower cover of a packaging box for packaging reagent strips. The visual inspection module 3 can further include an upper cover visual inspection module and a lower cover visual inspection module. The upper cover visual inspection module can include an upper cover camera. The lower cover visual inspection module can include a lower cover camera. The upper cover visual inspection module can be located above the upper cover feeding mechanism. The lower cover visual inspection module can be located above the lower cover feeding mechanism.

[0030] Optionally, the control unit can be further configured to: control the upper cover camera to capture the material upper cover carried on the upper cover feeding mechanism to obtain a material upper cover image; control the lower cover camera to capture the material lower cover carried on the lower cover feeding mechanism to obtain a material lower cover image; perform object pose detection on the material upper cover image to obtain an upper cover object pose detection result; perform object pose detection on the material lower cover image to obtain a lower cover object pose detection result. In practice, the control unit can perform front and back detection on the material upper cover image and the material lower cover image to determine whether the material upper cover and the material lower cover are front-up. As an example, the control unit can input the material upper cover image into a pre-trained upper cover object pose detection model to obtain the upper cover object pose detection result. And input the material lower cover image into a pre-trained lower cover object pose detection model to obtain the lower cover object pose detection result. The upper cover object pose detection model can be a machine learning model that takes the material upper cover image as input and outputs the upper cover object pose detection result. The lower cover object pose detection model can be a machine learning model that takes the material lower cover image as input and outputs the lower cover object pose detection result. As an example, the machine learning model can be a BP (Back Propagation) neural network. In this way, it can be determined whether the material upper cover and the material lower cover are front-up, so as to facilitate subsequent turning over of the material upper cover and the material lower cover that are back-up.

[0031] Optionally, the feeding device may further include a gripping mechanism. The gripping mechanism may be communicatively connected to the control unit. The gripping mechanism may include an upper cover gripping mechanism and a lower cover gripping mechanism. The upper cover gripping mechanism may be located on one side of the upper cover feeding mechanism. The lower cover gripping mechanism may be located on one side of the lower cover feeding mechanism. The upper cover gripping mechanism may be a mechanism for gripping the upper cover of the material. The lower cover gripping mechanism may be a mechanism for gripping the lower cover of the material. For example, both the upper cover gripping mechanism and the lower cover gripping mechanism may be robots. Specifically, both the upper cover gripping mechanism and the lower cover gripping mechanism may be four-axis robots. The feeding device may further include a flipping mechanism. The flipping mechanism may be communicatively connected to the control unit. The flipping mechanism may include an upper cover flipping mechanism and a lower cover flipping mechanism. The upper cover flipping mechanism may be located on one side of the upper cover feeding mechanism. The lower cover flipping mechanism may be located on one side of the lower cover feeding mechanism. The upper cover flipping mechanism may be a mechanism for flipping the upper cover of the material between the front and back directions. The lower cover flip mechanism can be a mechanism for flipping the material lower cover in the forward and reverse directions. As an example, the upper cover flip mechanism and the lower cover flip mechanism can both be 180° flip mechanisms.

[0032] Optionally, the control unit may be further configured to: in response to the upper cover posture detection result indicating that the reverse side of the material upper cover is facing upward, control the upper cover grasping mechanism to grasp the material upper cover to the upper cover flipping mechanism. Control the upper cover flipping mechanism to flip the material upper cover. In response to the lower cover posture detection result indicating that the reverse side of the material lower cover is facing upward, control the lower cover grasping mechanism to grasp the material lower cover to the lower cover flipping mechanism. In practice, in response to the upper cover posture detection result indicating that the reverse side of the material upper cover is facing upward, the control unit may determine the position of the material upper cover corresponding to the material upper cover image based on the material upper cover image, control the upper cover grasping mechanism to grasp the material upper cover to the upper cover flipping mechanism based on the determined position of the material upper cover, and control the upper cover grasping mechanism to flip the material upper cover to face the front side upward. In response to the lower cover posture detection result indicating that the reverse side of the lower cover is facing upward, the control unit can determine the position of the lower cover corresponding to the image of the lower cover based on the image of the lower cover, control the lower cover grasping mechanism to grasp the lower cover to the lower cover flipping mechanism based on the determined position of the lower cover, and control the lower cover grasping mechanism to flip the lower cover so that the front side faces upward. In this way, the reversed lower cover and the reversed upper cover can be flipped, thereby facilitating subsequent assembly.

[0033] Optionally, the control unit may be further configured to:

[0034] In a first step, a cover clamping portion position of the material upper cover image is detected to obtain cover clamping portion position information. The cover clamping portion position information can be information representing a position of a portion of the material upper cover corresponding to the material upper cover image, which is used to clamp the material lower cover. For example, the cover clamping portion position information can be a coordinate pair. In practice, the control unit can input the material upper cover image into a pre-trained cover clamping portion position generation model to obtain the cover clamping portion position information. The cover clamping portion position generation model can be a machine learning model taking the material upper cover image as input and outputting the cover clamping portion position information. For example, the machine learning model can be a convolutional neural network model.

[0035] In a second step, it is determined whether the cover clamping portion position information satisfies a preset cover clamping portion position condition. The preset cover clamping portion position condition can be that the position represented by the cover clamping portion position information is within a preset position range.

[0036] In a third step, a lower cover clamping portion position of the material lower cover image is detected to obtain lower cover clamping portion position information. The lower cover clamping portion position information can be information representing a position of a portion of the material lower cover corresponding to the material lower cover image, which is used to clamp the material upper cover. For example, the lower cover clamping portion position information can be a coordinate pair. In practice, the control unit can input the material lower cover image into a pre-trained lower cover clamping portion position generation model to obtain the lower cover clamping portion position information. The lower cover clamping portion position generation model can be a machine learning model taking the material lower cover image as input and outputting the lower cover clamping portion position information. For example, the machine learning model can be a convolutional neural network model.

[0037] In a fourth step, it is determined whether the lower cover clamping portion position information satisfies a preset lower cover clamping portion position condition. The preset lower cover clamping portion position condition can be that the position represented by the lower cover clamping portion position information is within a preset position range.

[0038] Optionally, the control unit can be further configured to perform damage detection on the material upper cover image to obtain an upper cover damage result. The upper cover damage result can be a result indicating that the material upper cover corresponding to the material upper cover image is damaged. For example, the control unit can perform damage detection on the material upper cover image by visual comparison to obtain the upper cover damage result. For another example, the control unit can input the material upper cover image into a pre-trained upper cover damage result generation model to obtain the upper cover damage result. The upper cover damage result generation model can be a machine learning model taking the material upper cover image as input and outputting the upper cover damage result. For example, the machine learning model can be a convolutional neural network model. The control unit can also perform damage detection on the material lower cover image to obtain a lower cover damage result. The lower cover damage result can be a result indicating that the material lower cover corresponding to the material lower cover image is damaged. For example, the control unit can perform damage detection on the material lower cover image by visual comparison to obtain the lower cover damage result. For another example, the control unit can input the material lower cover image into a pre-trained lower cover damage result generation model to obtain the lower cover damage result. The lower cover damage result generation model can be a machine learning model taking the material lower cover image as input and outputting the lower cover damage result. For example, the machine learning model can be a convolutional neural network model. In this way, it can be determined whether the material upper cover and the material lower cover are damaged.

[0039] Optionally, the feeding device can further include an upper cover waste bin. The upper cover waste bin can be located on one side of the upper cover grabbing mechanism. The upper cover waste bin can be a storage bin for carrying unqualified material upper covers. The control unit can be further configured to control the upper cover grabbing mechanism to grab the material upper cover into the upper cover waste bin in response to the upper cover clamping portion position information not satisfying the upper cover clamping portion position condition and / or the upper cover damage result indicating that the material upper cover is contaminated. In this way, unqualified material upper covers can be placed in the waste bin, thereby ensuring that the final material upper cover meets the requirements.

[0040] Optionally, the feeding device can further include a lower cover waste bin. The lower cover waste bin can be located on one side of the lower cover grabbing mechanism. The lower cover waste bin can be a storage bin for carrying unqualified material lower covers. The control unit can be further configured to control the lower cover grabbing mechanism to grab the material lower cover into the lower cover waste bin in response to the lower cover clamping portion position information not satisfying the lower cover clamping portion position condition and / or the lower cover damage result indicating that the material lower cover is contaminated. In this way, unqualified material lower covers can be placed in the waste bin, thereby ensuring that the final material lower cover meets the requirements.

[0041] Optionally, the control unit can be further configured to perform length detection on the target reagent strip image to obtain a target reagent strip length value. It is determined whether the target reagent strip length value meets a preset reagent strip length condition. The reagent strip length condition can be that the target reagent strip length value is not within a preset reagent strip length range. Width detection is performed on the target reagent strip image to obtain a target reagent strip width value. It is determined whether the target reagent strip width value meets a preset reagent strip width condition. The reagent strip width condition can be that the target reagent strip width value is not within a preset reagent strip width range. Dirt detection is performed on the target reagent strip image to obtain a dirt detection result. The dirt detection result can represent whether the target reagent strip corresponding to the target reagent strip image has dirt. In practice, the control unit can perform dirt detection on the target reagent strip image in various ways to obtain the dirt detection result. As an example, the control unit can input the target reagent strip image into a pre-trained reagent strip dirt detection model to obtain the dirt detection result. The reagent strip dirt detection model can be a neural network model that takes a target reagent strip image as input and outputs a dirt detection result. For example, the neural network model can be a convolutional neural network model. In this way, it can be determined whether the cut reagent strip is qualified.

[0042] Optionally, the feeding device can further include a reagent strip waste box. The reagent strip waste box can be located on one side of the reagent strip conveying mechanism 4. The reagent strip waste box can be a storage box for storing unqualified reagent strips. The control unit can be further configured to, after the defect detection on the target reagent strip image, control the reagent strip conveying mechanism 4 to grasp the target reagent strip into the reagent strip waste box in response to at least one of the following: the target reagent strip length value does not meet the preset reagent strip length condition, the target reagent strip width value does not meet the preset reagent strip width condition, and the dirt detection result represents that the target reagent strip has dirt. In this way, unqualified reagent strips can be put into the waste box, thereby ensuring that the finally obtained reagent strips meet the requirements.

[0043] Optionally, the upper cover feeding mechanism can include an upper cover hopper, an upper cover elevator, and an upper cover conveyor belt. The bottom end of the upper cover elevator can be connected to the upper cover hopper. The top end of the upper cover elevator can be connected to the upper cover conveyor belt. The lower cover feeding mechanism can include a lower cover hopper, a lower cover elevator, and a lower cover conveyor belt. The bottom end of the lower cover elevator can be connected to the lower cover hopper. The top end of the lower cover elevator can be connected to the lower cover conveyor belt.

[0044] Optionally, the control unit can further comprise an artificial intelligence chip. The control unit can be further configured to control the plate feeding servo mechanism 103 to feed the received whole reagent strip to the cutting mechanism 2 at a target feeding speed, and control the cutting mechanism 2 to cut the received whole reagent strip into at least one target reagent strip at a target cutting speed, wherein the target feeding speed can be a running speed of the plate feeding servo mechanism 103 at which the qualified rate of the obtained reagent strip is the highest. The target cutting speed can be a running speed of the cutting mechanism 2 at which the qualified rate of the obtained reagent strip is the highest. The target feeding speed and the target cutting speed are obtained by the artificial intelligence chip of the control unit analyzing the length of the target reagent strip, the feeding speed range of the plate feeding servo mechanism 103 and the cutting speed range of the cutting mechanism 2. The feeding speed range can be a speed range in which the plate feeding servo mechanism 103 works. The cutting speed range can be a speed range in which the cutting mechanism 2 works. The machine learning model carried by the artificial intelligence chip is obtained by training a training sample set. The training sample set comprises a sample feeding speed range of a sample feeding mechanism, a sample cutting speed range of a sample cutting mechanism, a length of a sample reagent strip, a sample feeding speed and a sample cutting speed. The machine learning model is trained by taking the sample feeding speed range of the sample plate feeding servo mechanism, the sample cutting speed range of the sample cutting mechanism and the length of the sample reagent strip as input, and taking the sample feeding speed and the sample cutting speed as expected output.

[0045] The optional content above is an inventive point of an embodiment of the present disclosure, which solves the second technical problem mentioned in the background that the conveying speed and the cutting speed are not determined according to the length of the reagent strip, which easily further leads to low feeding efficiency and low qualification rate of the reagent strip. The reasons for further leading to low feeding efficiency and low qualification rate of the reagent strip are as follows: the conveying speed and the cutting speed are not determined according to the length of the reagent strip. If the above factors are solved, the feeding efficiency and the qualification rate of the reagent strip can be further improved. In order to achieve this effect, the control unit of the present disclosure includes an artificial intelligence chip. And the control unit is further configured to control the plate feeding servo mechanism to convey the carried whole plate reagent strip to the cutting mechanism according to a target feeding speed, and control the cutting mechanism to cut the received whole plate reagent strip according to a target cutting speed according to a target length to obtain at least one target reagent strip, wherein the target feeding speed and the target cutting speed are obtained by analyzing the length of the target reagent strip, the feeding speed range of the plate feeding servo mechanism and the cutting speed range of the cutting mechanism by the artificial intelligence chip included in the control unit, and the machine learning model carried by the artificial intelligence chip is obtained by training a set of training samples. Therefore, the target feeding speed and the target cutting speed are determined by the length of the target reagent strip, the feeding speed range of the plate feeding servo mechanism and the cutting speed range of the cutting mechanism, so that the plate feeding servo mechanism and the cutting mechanism can cooperate to obtain the optimal running speed, and the feeding efficiency and the qualification rate of the reagent strip can be further improved.

[0046] Optionally, the reagent strip visual detection module can include a reagent strip light source. The reagent strip light source can be arranged on the reagent strip camera.

[0047] Optionally, the control unit may be further configured to: determine the distance between the target reagent strip and the reagent strip imaging device; determine the feeding speed of the plate feeding servo mechanism; based on the distance and the feeding speed, determine the focal length corresponding to the reagent strip imaging device and the light intensity corresponding to the reagent strip light source, and obtain a target focal length corresponding to the reagent strip imaging device and a target light intensity corresponding to the reagent strip light source. The target focal length may be a focal length set by the reagent strip imaging device that makes the captured image of the target reagent strip clearest. The target light intensity may be a light intensity set by the reagent strip light source that makes the captured image of the target reagent strip clearest. In practice, the target focal length and the target light intensity are obtained by analyzing the distance between the target reagent strip and the reagent strip imaging device and the feeding speed of the plate feeding servo mechanism by the artificial intelligence chip included in the control unit. The machine learning model carried by the artificial intelligence chip is trained using a training sample set. The above-mentioned training sample set includes the distance between the sample target reagent strip and the sample reagent strip camera device, the feeding speed of the sample feeding plate servo mechanism, the sample focal length and the sample light intensity. The above-mentioned machine learning model is trained with the distance between the sample target reagent strip and the sample reagent strip camera device and the feeding speed of the sample feeding plate servo mechanism as input and with the sample focal length and the sample light intensity as the expected output.

[0048] The above optional content, as an inventive feature of an embodiment of the present disclosure, solves the third technical problem mentioned in the background technology: "the inability to accurately determine whether the reagent strip is qualified, resulting in a low qualified rate of the obtained reagent strips." The low qualified rate of the obtained reagent strips is due to the following: the inability to accurately determine whether the reagent strip is qualified. If the above factors are resolved, the qualified rate of the obtained reagent strips can be further improved. To achieve this effect, the above-mentioned reagent strip visual inspection module of the present disclosure includes a reagent strip light source. The above-mentioned reagent strip light source is disposed on the above-mentioned reagent strip camera device. The above-mentioned control unit is further configured to: determine the distance between the target reagent strip and the above-mentioned reagent strip camera device; determine the feeding speed of the above-mentioned feeding servo mechanism; and determine the focal length corresponding to the above-mentioned reagent strip camera device and the light intensity corresponding to the above-mentioned reagent strip light source based on the above-mentioned distance and the above-mentioned feeding speed, thereby obtaining a target focal length corresponding to the above-mentioned reagent strip camera device and a target light intensity corresponding to the above-mentioned reagent strip light source. Thus, the focal length corresponding to the above-mentioned reagent strip camera device and the light intensity corresponding to the above-mentioned reagent strip light source are determined based on the distance between the target reagent strip and the above-mentioned reagent strip camera device and the feeding speed. The clarity of the obtained image can be improved, so that whether the reagent strip has defects can be determined more accurately, thereby improving the qualified rate of the reagent strip.

[0049] The above various embodiments of the present disclosure have the following beneficial effects: the feeding device based on visual detection of some embodiments of the present disclosure can improve the qualified rate and feeding efficiency of reagent strips. Specifically, the reason for the low qualified rate and slow feeding speed of reagent strips is that manual taking is easy to cause contamination of some special reagent strips, thereby causing damage to the reagent strips. For some small-sized reagent strips, manual sorting is easy to ignore the quality problems of small size, and the sorting efficiency is low. Based on this, the feeding device based on visual detection of some embodiments of the present disclosure includes a control unit, a feeding mechanism, a cutting mechanism, a visual detection module, a reagent strip handling mechanism and a feeding mechanism, wherein the feeding mechanism is in communication connection with the control unit, the feeding mechanism includes a reagent strip feeding mechanism, the reagent strip feeding mechanism includes a whole plate feeding mechanism, a whole plate storage mechanism and a plate feeding servo mechanism, the control unit is configured to control the whole plate feeding mechanism to grab the whole plate reagent strip stored in the whole plate storage mechanism to the plate feeding servo mechanism, and control the plate feeding servo mechanism to deliver the whole plate reagent strip carried to the cutting mechanism. The cutting mechanism is located at one end of the plate feeding servo mechanism, the cutting mechanism is in communication connection with the control unit, and the control unit is further configured to control the cutting mechanism to cut the whole plate reagent strip located below the cutting mechanism according to the target length to obtain at least one target reagent strip. The visual detection module is in communication connection with the control unit, the visual detection module includes a reagent strip visual detection module, the reagent strip visual detection module is located on one side of the cutting mechanism, the reagent strip visual detection module includes a reagent strip camera, and the control unit is further configured to control the reagent strip camera to shoot the target reagent strip to obtain a target reagent strip image, and to detect defects of the target reagent strip image to obtain a defect detection result. The reagent strip handling mechanism is in communication connection with the control unit, and the reagent strip handling mechanism is located on one side of the reagent strip visual detection module. The feeding mechanism includes a reagent strip feeding conveyor belt, one end of the reagent strip feeding conveyor belt is adjacent to the cutting mechanism, the feeding mechanism is in communication connection with the control unit, and the control unit is further configured to control the reagent strip handling mechanism to grab the target reagent strip corresponding to the defect detection result which does not exist defects to the reagent strip feeding conveyor belt. Because the feeding device based on visual detection is operated by a machine, manual contact with the reagent strip is avoided. Therefore, the probability of damage to the reagent strip can be reduced. Also, by automatically detecting through the reagent strip visual detection module and taking the reagent strip through the reagent strip robot, the quality problems can be reduced and the sorting efficiency can be improved. Therefore, the feeding device based on visual detection of some embodiments of the present disclosure can improve the qualified rate and feeding efficiency of reagent strips.

[0050] With continued reference to Figure 2, shows a flow 200 of some embodiments of the vision detection based feeding method according to the present disclosure. The vision detection based feeding method comprises the following steps:

[0051] Step 201, grabbing the whole plate reagent strip stored in the whole plate storage mechanism to the plate feeding servo mechanism by the whole plate feeding mechanism.

[0052] In some embodiments, the execution subject of the vision detection based feeding method (for example Figure 1 The vision detection based feeding device shown can grab the whole plate reagent strip stored in the whole plate storage mechanism to the plate feeding servo mechanism by the whole plate feeding mechanism. Wherein, the feeding device comprises a control unit, a feeding mechanism, a cutting mechanism, a vision detection module, a reagent strip handling mechanism and a feeding mechanism. The feeding mechanism comprises a reagent strip feeding mechanism. The reagent strip feeding mechanism comprises a whole plate feeding mechanism, a whole plate storage mechanism and a plate feeding servo mechanism. The vision detection module comprises a reagent strip vision detection module. The reagent strip vision detection module comprises a reagent strip camera.

[0053] Step 202, feeding the whole plate reagent strip to the cutting mechanism by the plate feeding servo mechanism.

[0054] In some embodiments, the execution subject can feed the whole plate reagent strip to the cutting mechanism by the plate feeding servo mechanism. Thus, the stored whole plate reagent strip can be fed to the lower part of the cutting mechanism.

[0055] Step 203, cutting the whole plate reagent strip according to the target length by the cutting mechanism to obtain at least one target reagent strip.

[0056] In some embodiments, the execution subject can cut the whole plate reagent strip according to the target length by the cutting mechanism to obtain at least one target reagent strip. Thus, the whole plate reagent strip can be automatically cut into the reagent strip of the required length by the cutting mechanism.

[0057] Step 204, taking a photo of each target reagent strip in the at least one target reagent strip by the reagent strip camera to obtain at least one target reagent strip image.

[0058] In some embodiments, the execution subject can take a photo of each target reagent strip in the at least one target reagent strip by the reagent strip camera to obtain at least one target reagent strip image.

[0059] Step 205, performing defect detection on each target reagent strip image in the at least one target reagent strip image to obtain a set of defect detection results.

[0060] In some embodiments, the execution subject can perform defect detection on each of the at least one target reagent strip image to obtain a set of defect detection results. In this way, it can be determined whether the cut target reagent strip has defects.

[0061] At step 206, the corresponding defect detection result is indicative of the target reagent strip without defects, and the reagent strip conveying mechanism is used to pick up the target reagent strip to the reagent strip feeding conveyor.

[0062] In some embodiments, the execution subject can perform defect detection on each of the at least one target reagent strip image to obtain a set of defect detection results. In this way, it can be determined whether the cut target reagent strip has defects.

[0063] The above-mentioned embodiments of the present disclosure have the following beneficial effects: the feeding method based on visual detection of some embodiments of the present disclosure can improve the qualified rate and feeding efficiency of the reagent strip. Specifically, the reason for the low qualified rate and slow feeding speed of the reagent strip is that manual picking up is easy to cause contamination of the reagent strip, thereby causing damage to the reagent strip. For some small-sized reagent strips, manual sorting is easy to ignore the quality problems of small size, and the sorting efficiency is low. Based on this, the feeding method based on visual detection of some embodiments of the present disclosure first uses the whole plate feeding mechanism to pick up the whole plate reagent strip stored in the whole plate storage mechanism to the plate feeding servo mechanism. Second, the plate feeding servo mechanism is used to transport the whole plate reagent strip to the cutting mechanism. In this way, the stored whole plate reagent strip can be transported below the cutting mechanism. Then, the cutting mechanism is used to cut the whole plate reagent strip according to the target length to obtain at least one target reagent strip. In this way, the cutting mechanism can automatically cut the whole plate reagent strip into a reagent strip of a desired length. Then, the reagent strip camera is used to take a picture of each of the at least one target reagent strip to obtain at least one target reagent strip image. Then, defect detection is performed on each of the at least one target reagent strip image to obtain a set of defect detection results. In this way, it can be determined whether the cut reagent strip has defects. Finally, the reagent strip conveying mechanism is used to pick up the target reagent strip corresponding to the defect detection result without defects to the reagent strip feeding conveyor. In this way, the target reagent strip without defects can be supplied. In this way, the feeding method based on visual detection of some embodiments of the present disclosure can improve the qualified rate and feeding efficiency of the reagent strip.

[0064] The above description is merely exemplary of some of the many possible embodiments of the present disclosure and of the principles thereof. It is to be understood that those skilled in the art will be able to devise various embodiments of the present disclosure without departing from the scope of the present disclosure as disclosed in the above description and attached claims, and that the scope of the present disclosure is not limited to the specific technical features described above. For example, the technical features described above can be replaced with other technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) to form other technical solutions.

Claims

1. A feeding device based on visual detection, comprising a control unit, a feeding mechanism, a cutting mechanism, a visual detection module, a reagent strip conveying mechanism and a feeding mechanism, wherein, the feeding mechanism is in communication connection with the control unit, the feeding mechanism comprises a reagent strip feeding mechanism, the reagent strip feeding mechanism comprises a whole plate feeding mechanism, a whole plate storage mechanism and a plate conveying servo mechanism, the control unit is configured to control the whole plate feeding mechanism to grab the whole plate reagent strip stored in the whole plate storage mechanism to the plate conveying servo mechanism, and control the plate conveying servo mechanism to deliver the carried whole plate reagent strip to the cutting mechanism; the cutting mechanism is located at one end of the plate conveying servo mechanism, the cutting mechanism is in communication connection with the control unit, and the control unit is further configured to control the cutting mechanism to cut the whole plate reagent strip located below the cutting mechanism according to a target length to obtain at least one target reagent strip, the value of the target length is set by a worker, and the whole plate reagent strip is automatically cut into a reagent strip with a required length by the cutting mechanism; the visual detection module is in communication connection with the control unit, the visual detection module comprises a reagent strip visual detection module, the reagent strip visual detection module is located on one side of the cutting mechanism, the reagent strip visual detection module comprises a reagent strip camera device, and the control unit is further configured to control the reagent strip camera device to take a picture of the target reagent strip to obtain a target reagent strip image, and to perform defect detection on the target reagent strip image to obtain a defect detection result; the reagent strip conveying mechanism is in communication connection with the control unit, and the reagent strip conveying mechanism is located on one side of the reagent strip visual detection module; the feeding mechanism comprises a reagent strip feeding conveyor belt, one end of the reagent strip feeding conveyor belt is adjacent to the cutting mechanism, the feeding mechanism is in communication connection with the control unit, and the control unit is further configured to control the reagent strip conveying mechanism to grab the target reagent strip corresponding to the defect detection result which represents that there is no defect to the reagent strip feeding conveyor belt; the feeding mechanism further comprises an upper cover feeding mechanism and a lower cover feeding mechanism; the visual detection module further comprises an upper cover visual detection module and a lower cover visual detection module, the upper cover visual detection module comprises an upper cover camera device, the lower cover visual detection module comprises a lower cover camera device, the upper cover visual detection module is located above the upper cover feeding mechanism, and the lower cover visual detection module is located above the lower cover feeding mechanism; and the control unit is further configured to: control the upper cover camera device to take a picture of the material upper cover carried on the upper cover feeding mechanism to obtain a material upper cover image; control the lower cover camera device to take a picture of the material lower cover carried on the lower cover feeding mechanism to obtain a material lower cover image; the control unit comprises at least one of PLC, SoC, MCU, DSP and artificial intelligence chip; and the control unit is further configured to: The upper cover clamping part position information of the material upper cover image is detected to obtain upper cover clamping part position information, wherein the upper cover clamping part position information is information representing the position of the part on the material upper cover corresponding to the material upper cover image for clamping with the material lower cover, and the upper cover clamping part position information is a coordinate pair; The control unit inputs the material upper cover image into a pre-trained upper cover clamping part position generation model to obtain upper cover clamping part position information, wherein the upper cover clamping part position generation model is a machine learning model taking the material upper cover image as input and outputting the upper cover clamping part position information, and the machine learning model is a convolutional neural network model; It is determined whether the upper cover clamping part position information meets a preset upper cover clamping part position condition; The lower cover clamping part position information of the material lower cover image is detected to obtain lower cover clamping part position information, wherein the lower cover clamping part position information is information representing the position of the part on the material lower cover corresponding to the material lower cover image for clamping with the material upper cover, and the lower cover clamping part position information is a coordinate pair; It is determined whether the lower cover clamping part position information meets a preset lower cover clamping part position condition; The control unit is further configured to control the plate feeding servo mechanism to deliver the whole plate reagent strip carried to the cutting mechanism at a target feeding speed, and control the cutting mechanism to cut the received whole plate reagent strip into at least one target reagent strip at a target cutting speed, wherein the target feeding speed is an operating speed of the plate feeding servo mechanism at which the qualified rate of the obtained reagent strip is the highest, the target cutting speed is an operating speed of the cutting mechanism at which the qualified rate of the obtained reagent strip is the highest, and the target feeding speed and the target cutting speed are obtained by analyzing the length of the target reagent strip, the feeding speed range of the plate feeding servo mechanism and the cutting speed range of the cutting mechanism by an artificial intelligence chip included in the control unit, the feeding speed range is a speed range in which the plate feeding servo mechanism works, and the cutting speed range is a speed range in which the cutting mechanism works.

2. The feeding device based on visual detection according to claim 1, wherein The material upper cover image is subjected to object posture detection to obtain upper cover object posture detection results; The material lower cover image is subjected to object posture detection to obtain lower cover object posture detection results; The feeding device further comprises a grabbing mechanism, which is in communication connection with the control unit, and the grabbing mechanism comprises an upper cover grabbing mechanism and a lower cover grabbing mechanism, the upper cover grabbing mechanism is located on one side of the upper cover feeding mechanism, and the lower cover grabbing mechanism is located on one side of the lower cover feeding mechanism; The feeding device further comprises a turnover mechanism, which is in communication connection with the control unit, and the turnover mechanism comprises an upper cover turnover mechanism and a lower cover turnover mechanism, the upper cover turnover mechanism is located on one side of the upper cover grabbing mechanism, and the lower cover turnover mechanism is located on one side of the lower cover grabbing mechanism; and The control unit is further configured to: In response to the upper cover posture detection result representing that the back of the material upper cover faces upward, the upper cover grabbing mechanism is controlled to grab the material upper cover to an upper cover overturning mechanism; The upper cover overturning mechanism is controlled to overturn the material upper cover; In response to the lower cover posture detection result representing that the back of the material lower cover faces upward, the lower cover grabbing mechanism is controlled to grab the material lower cover to a lower cover overturning mechanism; The lower cover overturning mechanism is controlled to overturn the material lower cover.

3. The vision detection based feed device of claim 2, wherein, The control unit is further configured to perform damage detection on the material upper cover image to obtain an upper cover damage result, and perform damage detection on the material lower cover image to obtain a lower cover damage result.

4. The vision detection based feed device of claim 3, wherein, The feeding device further comprises an upper cover waste box located on one side of the upper cover grabbing mechanism; and The control unit is further configured to, in response to the upper cover clamping part position information not satisfying the preset upper cover clamping part position condition and / or the upper cover damage result representing that the material upper cover is contaminated, control the upper cover grabbing mechanism to grab the material upper cover to the upper cover waste box. The feeding device further comprises a lower cover waste box located on one side of the lower cover grabbing mechanism; and The control unit is further configured to, in response to the lower cover clamping part position information not satisfying the preset lower cover clamping part position condition and / or the lower cover damage result representing that the material lower cover is contaminated, control the lower cover grabbing mechanism to grab the material lower cover to the lower cover waste box.

5. The vision detection based feed device of claim 1, wherein, The control unit is further configured to: perform length detection on the target reagent strip image to obtain a target reagent strip length value; determine whether the target reagent strip length value satisfies a preset reagent strip length condition; perform width detection on the target reagent strip image to obtain a target reagent strip width value; determine whether the target reagent strip width value satisfies a preset reagent strip width condition; perform contamination detection on the target reagent strip image to obtain a contamination detection result.

6. The vision detection based feed device of claim 5, wherein, The feeding device further comprises a reagent strip waste box located on one side of the reagent strip conveying mechanism; and The control unit is further configured to, after the defect detection on the target reagent strip image, in response to at least one of the following: the target reagent strip length value does not satisfy the preset reagent strip length condition, the target reagent strip width value does not satisfy the preset reagent strip width condition, and the contamination detection result represents that the target reagent strip is contaminated, control the reagent strip conveying mechanism to grab the target reagent strip to the reagent strip waste box.

7. The vision detection based feed device of claim 1, wherein, The upper cover feeding mechanism comprises an upper cover hopper, an upper cover elevator, and an upper cover conveyor belt, a bottom end of the upper cover elevator is connected to the upper cover hopper, and a top end of the upper cover elevator is connected to the upper cover conveyor belt; The lower cover feeding mechanism comprises a lower cover hopper, a lower cover elevator, and a lower cover conveyor belt, a bottom end of the lower cover elevator is connected to the lower cover hopper, and a top end of the lower cover elevator is connected to the lower cover conveyor belt.

8. A vision-based feeding method applied to the vision-based feeding apparatus of any one of claims 1 to 7, wherein, The feeding device comprises a control unit, a feeding mechanism, a cutting mechanism, a visual detection module, a reagent strip conveying mechanism and a feeding mechanism, the feeding mechanism comprises a reagent strip feeding mechanism, the reagent strip feeding mechanism comprises a whole plate feeding mechanism, a whole plate storage mechanism and a plate conveying servo mechanism, the visual detection module comprises a reagent strip visual detection module, the reagent strip visual detection module comprises a reagent strip camera; and the method comprises: grasping the whole plate reagent strip stored in the whole plate storage mechanism to the plate conveying servo mechanism through the whole plate feeding mechanism; delivering the whole plate reagent strip to the cutting mechanism through the plate conveying servo mechanism; cutting the whole plate reagent strip according to the target length through the cutting mechanism to obtain at least one target reagent strip; photographing each target reagent strip in the at least one target reagent strip through the reagent strip camera to obtain at least one target reagent strip image; detecting defects of each target reagent strip image in the at least one target reagent strip image to obtain a set of defect detection results; grasping the target reagent strip corresponding to the defect detection result which is not defective to the reagent strip feeding conveyor belt through the reagent strip conveying mechanism.

9. A computer program product comprising a computer program which, when executed by a processor, implements the method according to claim 8.

Images