Automatic product quality detection method applied to carton processing

Abstract

The invention discloses an automatic product quality detection method applied to carton processing, and relates to the technical field of carton quality detection.Monitoring equipment is arranged on a carton processing production line and used for obtaining processing working condition videos of a plurality of cartons on the carton processing production line; after video frame labeling is conducted on the processing working condition video of each carton, a working condition data set corresponding to each carton is obtained, the processing completion quality of the cartons is evaluated according to the working condition data set of each carton, and the cartons with the processing completion quality not meeting the expected processing quality are recycled. Conveying the cartons with the processing completion quality meeting the expected processing quality to a carton packaging area through a conveying belt for packaging, and performing finished product quality evaluation on the packaged finished product cartons to obtain a quality evaluation coefficient corresponding to each finished product carton, and dividing the finished product cartons into different quality categories according to the quality evaluation coefficients of the finished product cartons, and carrying out corresponding processing operation on the finished product cartons of different quality categories.

Description

Technical Field

[0001] The present invention relates to the technical field of carton quality detection, and in particular to an automated product quality detection method applied to carton processing. Background Art

[0002] As an important product in the packaging industry, the quality of cartons directly affects the packaging effect and product safety. During the carton production process, due to the influence of factors such as raw materials, production processes and equipment, the quality of carton products may fluctuate to a certain extent.

[0003] Traditional quality inspection methods for carton products often have relatively low inspection efficiency and accuracy, and the classification of carton quality inspection is often limited to two categories: qualified quality and unqualified quality. Cartons cannot be classified into different quality categories more finely. For cartons that do not meet the quality requirements, there is a lack of subsequent processing methods to convert them into quality that meets the production requirements. These problems are objectively existing in the current field of carton quality inspection. Summary of the invention

[0004] In order to solve the above problems, the purpose of the present invention is to provide an automated product quality detection method for carton processing.

[0005] The purpose of the present invention can be achieved by the following technical solution: an automated product quality detection method applied to carton processing, comprising the following steps:

[0006] Step S1: arrange monitoring equipment on the carton processing production line, and obtain processing condition videos of several cartons on the carton processing production line through the monitoring equipment, and annotate the video frames of the processing condition video of each carton, so as to obtain the corresponding working condition data set of each carton;

[0007] Step S2: Evaluate the processing quality of the corresponding carton according to the working condition data set of each carton, recycle the carton whose processing quality does not meet the preset expected processing quality, and transport the carton whose processing quality meets the expected processing quality to the carton packaging area via a conveyor belt;

[0008] Step S3: packing a number of cartons in a carton packing area, and performing a finished product quality assessment on the finished carton after packing, thereby obtaining a quality assessment coefficient corresponding to each finished carton;

[0009] Step S4: The finished cartons are divided into different quality categories according to the quality assessment coefficients of the finished cartons, and corresponding processing operations are performed on the finished cartons of different quality categories.

[0010] Furthermore, the process of arranging monitoring equipment on the carton processing production line and obtaining processing condition videos of several cartons on the carton processing production line through the monitoring equipment includes:

[0011] Select a number of monitoring points on the carton processing production line, number the monitoring points in sequence, and record them as i, i=1, 2, 3, ..., n, n is a natural number greater than 0, and arrange monitoring equipment at each monitoring point on the carton processing production line;

[0012] Get the monitoring coverage area of ​​each monitoring device, record the monitoring coverage area corresponding to the monitoring device numbered i as Area[i], and get the total working coverage area of ​​the carton processing production line for carton processing, and record it as Area 工作 , the monitoring coverage areas of several monitoring devices arranged on the carton processing production line are spliced ​​to obtain the total monitoring coverage area, which is recorded as Area 监控 ;

[0013] When Area 监控 <Area 工作 When the number of monitoring points is increased on the carton processing production line, corresponding monitoring equipment is arranged at the monitoring points;

[0014] When Area 监控 ≥Area 工作 When the carton processing line is started, several cartons are processed and produced by the carton processing line, and all monitoring equipment obtains the processing condition video of several cartons on the carton processing line.

[0015] Furthermore, the process of annotating the video frames of the processing condition video of each carton and then obtaining the corresponding working condition data set for each carton includes:

[0016] Set the standard frame and obtain the frame length corresponding to the standard frame, and record the frame length as L 标准帧 , obtain the video stream length of the processing condition video corresponding to each carton, obtain the carton code corresponding to each carton, and use the carton code as the identity of the processing condition video corresponding to the carton, record the carton code as Sign, and the carton code corresponding to each carton is unique;

[0017] The length of the video stream of the processing video of the carton coded as Sign is recorded as L 视频流 [Sign], and then according to L 视频流 [Sign] and L 标准帧 The processing video of each carton is divided into a number of initial unit video frames, and the number of initial unit video frames is recorded as N, then N = Up (L 视频流 [Sign] / L标准帧 ), where Up() is a round-up function used to round the value in () upwards;

[0018] After the processing condition video of each carton is divided into a number of initial unit video frames, the video frames are labeled by obtaining the frame information of the processing condition video corresponding to each initial unit video frame, thereby obtaining the processing condition data set corresponding to each carton.

[0019] Furthermore, the process of video frame annotation includes:

[0020] The frame information corresponding to each initial unit video frame is used to record the actual processing action when processing the carton, and the standard frame information is set to record each standard processing action of processing the carton;

[0021] The frame information corresponding to each initial unit video frame and each standard frame information are composed of a number of frame positioning points. The frame positioning points of the frame information and the standard frame information are respectively used for the action positioning of the corresponding actual processing action and the standard processing action. The action completion degree of the actual processing action recorded corresponding to each initial unit video frame is obtained by comparing the frame positioning points of the frame information and the standard frame information. The action completion degree is the ratio of the number of frame positioning points of the frame information and the standard frame information at the same position to the total number of frame positioning points in the frame information. The action completion degree of the actual processing action is recorded as λ, 0<λ≤1;

[0022] The action completion degree λ is marked on the corresponding unit video frame. When all the unit video frames of the processing condition video corresponding to each carton are marked with the corresponding action completion degree, all the unit video frames of the processing condition video corresponding to each carton are integrated to generate the corresponding condition data set.

[0023] Furthermore, the process of evaluating the processing quality of the corresponding carton according to the working condition data set of each carton and recycling the carton whose processing quality does not meet the preset expected processing quality includes:

[0024] The working condition data set corresponding to each carton includes several actual processing actions for processing the current carton, and the action completion degree λ marked on each actual processing action. The action completion degrees λ corresponding to each actual processing action under the working condition data set of the same carton are accumulated and the average value is taken. The average value is used as the processing evaluation coefficient for evaluating the processing completion quality of the current carton and is recorded as JG 评估 ;

[0025] Set the processing completion threshold and record it as JG 阈值 ;

[0026] When JG 评估 ≥JG阈值 When evaluating the processing quality of the corresponding carton, the processing quality meets the expected processing quality;

[0027] When JG 评估 <JG 阈值 When evaluating the processing quality of the corresponding cartons, it is evaluated whether the processing quality does not meet the expected processing quality, and the cartons whose processing quality does not meet the expected processing quality are recycled.

[0028] Furthermore, the process of evaluating the quality of the finished cartons after packaging and obtaining the quality evaluation coefficient corresponding to each finished carton includes:

[0029] The cartons that meet the expected processing quality are transferred to the conveyor belt connected to the carton processing production line, and a carton packing area is set up. The conveyor belt transports each carton that meets the expected processing quality to the carton packing area, and several cartons are packed in the carton packing area, and then several corresponding finished carton boxes are generated after the packing is completed;

[0030] The finished product quality assessment of the finished cartons after packaging includes structural strength assessment, size and shape assessment, carton appearance defect assessment and carton compressive strength assessment, and then respectively obtain the corresponding first assessment sub-coefficient, second assessment sub-coefficient, third assessment sub-coefficient and fourth assessment sub-coefficient of the finished carton, set weights a, b and c, and perform weighted operations on the first assessment sub-coefficient, the second assessment sub-coefficient, the third assessment sub-coefficient and the fourth assessment sub-coefficient through weights a, b and c, and then obtain the quality assessment coefficient corresponding to the finished carton, and record the quality assessment coefficient as μ.

[0031] Furthermore, the structural strength assessment, size and shape assessment, carton appearance defect assessment and carton compressive strength assessment in the finished product quality assessment are carried out by relevant professionals based on the corresponding carton specification structural strength, carton standard size and shape, carton specification surface appearance and carton specification compressive strength, and the actual structural strength, size and shape, surface appearance and compressive strength of the current finished carton. The resulting first assessment sub-coefficient, second assessment sub-coefficient, third assessment sub-coefficient and fourth assessment sub-coefficient are respectively used to reflect the proportion of the actual structural strength, size and shape, surface appearance and compressive strength of the current finished carton that meets the corresponding carton specification structural strength, carton standard size and shape, carton specification surface appearance and carton specification compressive strength.

[0032] Furthermore, the process of classifying the finished cartons into different quality categories according to the quality assessment coefficients of the finished cartons and performing corresponding processing operations on the finished cartons of different quality categories includes:

[0033] Set the quality definition intervals corresponding to different quality categories, the quality categories include superior products, good products, qualified products and defective products, and the quality definition intervals include the first quality interval, the second quality interval, the third quality interval and the fourth quality interval, which are respectively denoted as Ω1, Ω2, Ω3 and Ω4, Ω1 = [0.9, 1], Ω2 = [0.7, 0.9), Ω3 = [0.6, 0.7), Ω4 = [0, 0.6);

[0034] Classify the quality category of all finished cartons of μ∈Ω1 as superior;

[0035] Classify the quality category of all finished cartons in μ∈Ω2 as good products;

[0036] Classify the quality categories of all finished cartons of μ∈Ω3 into qualified products;

[0037] Classify the quality category of all finished cartons of μ∈Ω4 as defective;

[0038] The processing operations for finished cartons of different quality categories include delivery operations, secondary processing operations, and destruction operations;

[0039] The finished cartons of the superior and good quality categories are delivered to the third party that has demand for the finished cartons through the delivery operation;

[0040] Perform secondary processing operations on finished cartons that are qualified products, and reprocess the corresponding finished cartons through secondary processing operations until the qualified finished cartons are converted into superior or good finished cartons;

[0041] Finished cartons that are defective in quality category are destroyed, and all defective finished cartons are restored to corresponding carton raw materials through the destruction operation.

[0042] Compared with the prior art, the beneficial effects of the present invention are as follows: by arranging monitoring equipment on a carton processing production line to obtain processing condition videos of several cartons, obtaining a condition data set of each carton after annotating the video frames of the processing condition videos, and evaluating the processing quality of the corresponding carton based on the condition data set, recycling cartons whose processing quality does not meet the expected processing quality, transporting cartons whose processing quality meets the expected processing quality to a carton packing area through a conveyor belt, packing several cartons in the carton packing area, and evaluating the quality of the finished cartons after packing, thereby obtaining a quality evaluation coefficient corresponding to each finished carton, dividing the finished cartons into different quality categories according to the quality evaluation coefficients of the finished cartons, and performing corresponding processing operations on the finished cartons of different quality categories, thereby improving the efficiency and accuracy of detecting the processing quality of the cartons to a certain extent, and more specifically dividing the finished cartons into different quality categories according to the quality evaluation coefficients, and performing subsequent different processing operations on cartons of different quality categories, so as to ensure the production of more finished cartons that meet the requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0044] like Figure 1 As shown, the automated product quality detection method applied to carton processing includes the following steps:

[0045] Step S1: arrange monitoring equipment on the carton processing production line, and obtain processing condition videos of several cartons on the carton processing production line through the monitoring equipment, and annotate the video frames of the processing condition video of each carton, so as to obtain the corresponding working condition data set of each carton;

[0046] Step S2: Evaluate the processing quality of the corresponding carton according to the working condition data set of each carton, recycle the carton whose processing quality does not meet the preset expected processing quality, and transport the carton whose processing quality meets the expected processing quality to the carton packaging area via a conveyor belt;

[0047] Step S3: packing a number of cartons in a carton packing area, and performing a finished product quality assessment on the finished carton after packing, thereby obtaining a quality assessment coefficient corresponding to each finished carton;

[0048] Step S4: The finished cartons are divided into different quality categories according to the quality assessment coefficients of the finished cartons, and corresponding processing operations are performed on the finished cartons of different quality categories.

[0049] It should be further explained that, in the specific implementation process, the process of arranging monitoring equipment on the carton processing production line and obtaining the processing condition video of several cartons on the carton processing production line through the monitoring equipment includes:

[0050] Select several monitoring points on the carton processing production line, number them in sequence, and record the number as i, then i = 1, 2, 3, ..., n, where n is a natural number greater than 0, and arrange monitoring equipment at each monitoring point on the carton processing production line;

[0051] Get the monitoring coverage area of ​​each monitoring device, and record the monitoring coverage area corresponding to the monitoring device numbered i as Area[i], and get the total working coverage area of ​​the carton processing production line for carton processing, and record it as Area 工作 ;

[0052] The monitoring coverage areas of several monitoring devices arranged on the carton processing production line are spliced ​​to obtain the total monitoring coverage area, which is recorded as Area 监控 , then Area 监控 The expression formula is as follows:

[0053]

[0054] When Area 监控 <Area 工作 When the number of monitoring points is increased on the carton processing production line, corresponding monitoring equipment is arranged at the monitoring points;

[0055] When Area 监控 ≥Area 工作 When , it indicates that the monitoring equipment arranged on the current carton processing production line has met the monitoring coverage requirements for the current carton processing, the carton processing production line is started, and then the carton processing production line processes and produces several cartons, and all the monitoring equipment obtains the processing condition videos of several cartons on the carton processing production line.

[0056] It should be noted that each monitoring device located on the carton processing production line obtains the working condition video of one shooting position of the carton, and splices the working condition videos of all shooting positions of each carton to obtain the processing condition video of each carton.

[0057] It should be further explained that, in the specific implementation process, the process of annotating the video frames of the processing condition video of each carton and then obtaining the corresponding working condition data set for each carton includes:

[0058] Set the standard frame and obtain the frame length corresponding to the standard frame, and record the frame length as L 标准帧, obtain the video stream length of the processing condition video corresponding to each carton, obtain the carton code corresponding to each carton, and use the carton code as the identity of the processing condition video corresponding to the carton, record the carton code as Sign, and the carton code corresponding to each carton is unique;

[0059] The length of the video stream of the processing video of the carton coded as Sign is recorded as L 视频流 [Sign], and then according to L 视频流 [Sign] and L 标准帧 The processing video of each carton is divided into a number of initial unit video frames, and the number of the unit video frames is recorded as N, then N = Up (L 视频流 [Sign] / L 标准帧 ), where Up() is a round-up function used to round the value in () upwards;

[0060] After each carton processing condition video is divided into a number of initial unit video frames, the video frame annotation is performed by obtaining the frame information of each initial unit video frame corresponding to the processing condition video. The content of the video frame annotation is as follows:

[0061] The frame information corresponding to each initial unit video frame is used to record the actual processing action when processing the carton, and the standard frame information is set, and the standard frame information is used to record each standard processing action of processing the carton;

[0062] The frame information corresponding to each initial unit video frame and each standard frame information are composed of a number of frame positioning points. The frame positioning points of the frame information and the standard frame information are respectively used for the action positioning of the corresponding actual processing action and the standard processing action. The action completion degree of the actual processing action recorded corresponding to each initial unit video frame is obtained by comparing the frame positioning points of the frame information and the standard frame information. The action completion degree is the ratio of the number of frame positioning points of the frame information and the standard frame information at the same position to the total number of frame positioning points in the frame information. The action completion degree of the actual processing action is recorded as λ, 0<λ≤1;

[0063] The action completion degree λ is marked on the corresponding unit video frame. When all the unit video frames of the processing condition video corresponding to each carton are marked with the corresponding action completion degree, all the unit video frames of the processing condition video corresponding to each carton are integrated to generate the corresponding condition data set for each carton.

[0064] It should be further explained that, in the specific implementation process, the process of evaluating the processing quality of the corresponding carton according to the working condition data set of each carton, and recycling the carton whose processing quality does not meet the preset expected processing quality includes:

[0065] The working condition data set corresponding to each carton includes several actual processing actions corresponding to the processing of the current carton, and the action completion degree λ marked on each actual processing action. The action completion degrees λ corresponding to each actual processing action under the working condition data set of the same carton are accumulated and the average value is taken. The average value is used as the processing evaluation coefficient for evaluating the processing completion quality of the current carton, and the processing evaluation coefficient is recorded as JG 评估 ;

[0066] Set the processing completion threshold and record it as JG 阈值 The processing completion threshold is the minimum value of the processing evaluation coefficient when the processing completion quality of the carton meets the expected processing quality. 评估 With JG 阈值 The size dependency evaluation determines whether the processing quality of each carton meets the preset expected processing quality;

[0067] When JG 评估 ≥JG 阈值 When evaluating the processing quality of the corresponding carton, the processing quality meets the expected processing quality;

[0068] When JG 评估 <JG 阈值 When evaluating the processing quality of the corresponding cartons, it is evaluated whether the processing quality does not meet the expected processing quality, and the cartons whose processing quality does not meet the expected processing quality are recycled.

[0069] It should be further explained that, in the specific implementation process, the process of evaluating the quality of the finished cartons after packaging and then obtaining the quality evaluation coefficient corresponding to each finished carton includes:

[0070] The cartons whose processing quality meets the expected processing quality are transferred to a conveyor belt connected to the carton processing production line, a carton packing area is set up, each carton that meets the expected processing quality is transported to the carton packing area by the conveyor belt, and several cartons are packed in the carton packing area, and then several corresponding finished carton boxes are generated after the packing is completed;

[0071] The finished product quality assessment of the finished cartons after packaging includes structural strength assessment, size and shape assessment, carton appearance defect assessment and carton compressive strength assessment, and then the corresponding first assessment sub-coefficient, second assessment sub-coefficient, third assessment sub-coefficient and fourth assessment sub-coefficient of the finished carton are obtained, and are recorded as η1, η2, η3 and η4 respectively;

[0072] Wherein, η1, η2, η3 and η4 are all decimals between 0 and 1;

[0073] Set weights a, b and c, where 0<a<1, 0<b<1, 0<c<1, 0<a+b+c<1, and perform weighted operations on the first evaluation sub-coefficient, the second evaluation sub-coefficient, the third evaluation sub-coefficient and the fourth evaluation sub-coefficient through weights a, b and c, and then obtain the quality evaluation coefficient corresponding to the finished carton, and record the quality evaluation coefficient as μ, then the expression formula of μ is as follows:

[0074] μ=[eta1×a+eta2×b+eta3×c+eta4×(1-abc)] / 4;

[0075] Among them, 0<μ<1;

[0076] It should be noted that the structural strength assessment, size and shape assessment, carton appearance defect assessment and carton compressive strength assessment in the finished product quality assessment are carried out by relevant professionals based on the corresponding carton specification structural strength, carton standard size and shape, carton specification surface appearance and carton specification compressive strength, and the actual structural strength, size and shape, surface appearance and compressive strength of the current finished product carton. The resulting first assessment sub-coefficient, second assessment sub-coefficient, third assessment sub-coefficient and fourth assessment sub-coefficient are respectively used to reflect the proportion of the actual structural strength, size and shape, surface appearance and compressive strength of the current finished product carton that meets the corresponding carton specification structural strength, carton standard size and shape, carton specification surface appearance and carton specification compressive strength.

[0077] It should be further explained that, in the specific implementation process, the finished cartons are classified into different quality categories according to the quality assessment coefficients of the finished cartons, and the process of performing corresponding processing operations on the finished cartons of different quality categories includes:

[0078] Set quality definition intervals corresponding to different quality categories, the quality categories corresponding to the finished carton include superior products, good products, qualified products and defective products, and the quality definition intervals include a first quality interval, a second quality interval, a third quality interval and a fourth quality interval;

[0079] The first quality interval, the second quality interval, the third quality interval and the fourth quality interval are respectively denoted as Ω1, Ω2, Ω3 and Ω4. The corresponding relationship between the quality category of the finished carton and the quality definition interval is as follows:

[0080] Superior product—first quality interval, first quality interval Ω1=[0.9, 1];

[0081] Good product—second quality interval, second quality interval Ω2=[0.7, 0.9);

[0082] Qualified products—third quality interval, third quality interval Ω3=[0.6, 0.7);

[0083] Defective products—the fourth quality interval, the fourth quality interval Ω4=[0, 0.6);

[0084] According to the subordinate relationship between μ and Ω1, Ω2, Ω3 and Ω4, several finished cartons are divided into corresponding quality categories, as follows:

[0085] Classify the quality category of all finished cartons of μ∈Ω1 as superior;

[0086] Classify the quality category of all finished cartons in μ∈Ω2 as good products;

[0087] Classify the quality categories of all finished cartons of μ∈Ω3 into qualified products;

[0088] Classify the quality category of all finished cartons of μ∈Ω4 as defective;

[0089] The processing operations for finished cartons of different quality categories include delivery operations, secondary processing operations and destruction operations, as follows:

[0090] The finished cartons of the superior and good quality categories are delivered to the third party that has demand for the finished cartons through the delivery operation;

[0091] Perform secondary processing operations on finished cartons that are qualified products, and reprocess the corresponding finished cartons through secondary processing operations until the qualified finished cartons are converted into superior or good finished cartons;

[0092] Finished cartons that are defective in quality category are destroyed, and all defective finished cartons are restored to corresponding carton raw materials through the destruction operation.

[0093] It should be noted that the carton raw materials restored after the destruction operation are transferred to the carton processing production line, and continue to be processed through a series of processing methods to generate new finished cartons.

[0094] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An automated product quality detection method for carton processing, characterized in that: The following steps are involved: Step S1: arrange monitoring equipment on the carton processing production line, and obtain processing condition videos of several cartons on the carton processing production line through the monitoring equipment, and annotate the video frames of the processing condition video of each carton, so as to obtain the corresponding working condition data set of each carton; Step S2: Evaluate the processing quality of the corresponding carton according to the working condition data set of each carton, recycle the carton whose processing quality does not meet the preset expected processing quality, and transport the carton whose processing quality meets the expected processing quality to the carton packaging area via a conveyor belt; Step S3: packing a number of cartons in a carton packing area, and performing a finished product quality assessment on the finished carton after packing, thereby obtaining a quality assessment coefficient corresponding to each finished carton; Step S4: The finished cartons are divided into different quality categories according to the quality assessment coefficients of the finished cartons, and corresponding processing operations are performed on the finished cartons of different quality categories.

2. The automated product quality detection method for carton processing according to claim 1 is characterized in that: The process of arranging monitoring equipment on a carton processing production line and obtaining processing condition videos of several cartons on the carton processing production line through the monitoring equipment includes: Select a number of monitoring points on the carton processing production line, number the monitoring points in sequence, and record them as i, i=1, 2, 3, ..., n, n is a natural number greater than 0, and arrange monitoring equipment at each monitoring point on the carton processing production line; Get the monitoring coverage area of ​​each monitoring device, record the monitoring coverage area corresponding to the monitoring device numbered i as Area[i], and get the total working coverage area of ​​the carton processing production line for carton processing, and record it as Area 工作 , the monitoring coverage areas of several monitoring devices arranged on the carton processing production line are spliced ​​to obtain the total monitoring coverage area, which is recorded as Area 监控 ; When Area 监控 <Area 工作 When the number of monitoring points is increased on the carton processing production line, corresponding monitoring equipment is arranged at the monitoring points; When Area 监控 ≥Area 工作 When the carton processing line is started, several cartons are processed and produced by the carton processing line, and all monitoring equipment obtains the processing condition video of several cartons on the carton processing line.

3. The automated product quality detection method for carton processing according to claim 2 is characterized in that: The process of annotating the video frames of the processing condition video of each carton and then obtaining the corresponding working condition data set for each carton includes: Set the standard frame and obtain the frame length corresponding to the standard frame, and record the frame length as L 标准帧 , obtain the video stream length of the processing condition video corresponding to each carton, obtain the carton code corresponding to each carton, and use the carton code as the identity of the processing condition video corresponding to the carton, record the carton code as Sign, and the carton code corresponding to each carton is unique; The length of the video stream of the processing video of the carton coded as Sign is recorded as L 视频流 [Sign], and then according to L 视频流 [Sign] and L 标准帧 The processing video of each carton is divided into a number of initial unit video frames, and the number of initial unit video frames is recorded as N, then N = Up (L 视频流 [Sign] / L 标准帧 ), where Up() is a round-up function used to round the value in () upwards; After the processing condition video of each carton is divided into a number of initial unit video frames, the video frames are labeled by obtaining the frame information of the processing condition video corresponding to each initial unit video frame, thereby obtaining the processing condition data set corresponding to each carton.

4. The automated product quality detection method for carton processing according to claim 3 is characterized in that: The process of video frame annotation includes: The frame information corresponding to each initial unit video frame is used to record the actual processing action when processing the carton, and the standard frame information is set to record each standard processing action of processing the carton; The frame information corresponding to each initial unit video frame and each standard frame information are composed of a number of frame positioning points. The frame positioning points of the frame information and the standard frame information are respectively used for the action positioning of the corresponding actual processing action and the standard processing action. The action completion degree of the actual processing action recorded corresponding to each initial unit video frame is obtained by comparing the frame positioning points of the frame information and the standard frame information. The action completion degree is the ratio of the number of frame positioning points of the frame information and the standard frame information at the same position to the total number of frame positioning points in the frame information. The action completion degree of the actual processing action is recorded as λ, 0<λ≤1; The action completion degree λ is marked on the corresponding unit video frame. When all the unit video frames of the processing condition video corresponding to each carton are marked with the corresponding action completion degree, all the unit video frames of the processing condition video corresponding to each carton are integrated to generate the corresponding condition data set.

5. The automated product quality detection method for carton processing according to claim 4 is characterized in that: The process of evaluating the processing quality of the corresponding carton according to the working condition data set of each carton and recycling the carton whose processing quality does not meet the preset expected processing quality includes: The working condition data set corresponding to each carton includes several actual processing actions for processing the current carton, and the action completion degree λ marked on each actual processing action. The action completion degrees λ corresponding to each actual processing action under the working condition data set of the same carton are accumulated and the average value is taken. The average value is used as the processing evaluation coefficient for evaluating the processing completion quality of the current carton and is recorded as JG 评估 ; Set the processing completion threshold and record it as JG 阈值 ; When JG 评估 ≥JG 阈值 When evaluating the processing quality of the corresponding carton, the processing quality meets the expected processing quality; When JG 评估 <JG 阈值 When evaluating the processing quality of the corresponding cartons, it is evaluated whether the processing quality does not meet the expected processing quality, and the cartons whose processing quality does not meet the expected processing quality are recycled.

6. The automated product quality detection method for carton processing according to claim 5 is characterized in that: The process of evaluating the quality of the finished cartons after packaging and then obtaining the quality evaluation coefficient corresponding to each finished carton includes: The cartons that meet the expected processing quality are transferred to the conveyor belt connected to the carton processing production line, and a carton packing area is set up. The conveyor belt transports each carton that meets the expected processing quality to the carton packing area, and several cartons are packed in the carton packing area, and then several corresponding finished carton boxes are generated after the packing is completed; The finished product quality assessment of the finished cartons after packaging includes structural strength assessment, size and shape assessment, carton appearance defect assessment and carton compressive strength assessment, and then respectively obtain the corresponding first assessment sub-coefficient, second assessment sub-coefficient, third assessment sub-coefficient and fourth assessment sub-coefficient of the finished carton, set weights a, b and c, and perform weighted operations on the first assessment sub-coefficient, the second assessment sub-coefficient, the third assessment sub-coefficient and the fourth assessment sub-coefficient through weights a, b and c, and then obtain the quality assessment coefficient corresponding to the finished carton, and record the quality assessment coefficient as μ.

7. The automated product quality detection method for carton processing according to claim 6 is characterized in that: The structural strength assessment, size and shape assessment, carton appearance defect assessment and carton compressive strength assessment in the finished product quality assessment are evaluated by relevant professionals based on the corresponding carton specification structural strength, carton standard size and shape, carton specification surface appearance and carton specification compressive strength, and the actual structural strength, size and shape, surface appearance and compressive strength of the current finished product carton. The resulting first assessment sub-coefficient, second assessment sub-coefficient, third assessment sub-coefficient and fourth assessment sub-coefficient are respectively used to reflect the proportion of the actual structural strength, size and shape, surface appearance and compressive strength of the current finished product carton that meets the corresponding carton specification structural strength, carton standard size and shape, carton specification surface appearance and carton specification compressive strength.

8. The automated product quality detection method for carton processing according to claim 7 is characterized in that: The process of classifying finished cartons into different quality categories according to the quality assessment coefficients of the finished cartons and performing corresponding processing operations on the finished cartons of different quality categories includes: Set the quality definition intervals corresponding to different quality categories, the quality categories include superior products, good products, qualified products and defective products, and the quality definition intervals include the first quality interval, the second quality interval, the third quality interval and the fourth quality interval, which are respectively denoted as Ω1, Ω2, Ω3 and Ω4, Ω1 = [0.9, 1], Ω2 = [0.7, 0.9), Ω3 = [0.6, 0.7), Ω4 = [0, 0.6); Classify the quality category of all finished cartons of μ∈Ω1 as superior; Classify the quality category of all finished cartons in μ∈Ω2 as good products; Classify the quality categories of all finished cartons of μ∈Ω3 into qualified products; Classify the quality category of all finished cartons of μ∈Ω4 as defective; The processing operations for finished cartons of different quality categories include delivery operations, secondary processing operations, and destruction operations; The finished cartons of the superior and good quality categories are delivered to the third party that has demand for the finished cartons through the delivery operation; Perform secondary processing operations on finished cartons that are qualified products, and reprocess the corresponding finished cartons through secondary processing operations until the qualified finished cartons are converted into superior or good finished cartons; Finished cartons that are defective in quality category are destroyed, and all defective finished cartons are restored to corresponding carton raw materials through the destruction operation.

Images