Varnish spraying and baking operation supervision method based on visual identification

Through visual recognition technology and analysis algorithms, real-time monitoring of paint spraying and baking operations in the automotive repair industry has been solved, and the existing technology cannot be effectively monitored and managed, improving the efficiency and accuracy of environmental protection supervision and reducing costs.

CN119047688AActive Publication Date: 2024-11-29Guangzhou Yuanshi Technology Co., Ltd.

Patent Information

Application Number
CN202411053702.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-02
Publication Date
2024-11-29
Estimated Expiration
2044-08-02

AI Technical Summary

Technical Problem

The existing technology cannot effectively monitor and manage paint spraying operations in the automobile maintenance industry in real time, resulting in unorganized emission problems. The existing regulatory methods are inefficient and costly, and it is impossible to detect and correct violations in a timely manner.

Method used

The spraying and baking operation supervision method based on visual recognition is adopted. Through image data acquisition and analysis, object status analysis, operation status analysis and violation analysis algorithms are used to identify and record violations in real time, and issue notifications to correct them in a timely manner.

Benefits of technology

Real-time monitoring and analysis of the spraying and baking operation process is achieved, the efficiency and accuracy of environmental protection supervision is improved, the supervision costs are reduced, and the violations are reduced, and it is suitable for different spraying operation environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119047688A_ABST
    Figure CN119047688A_ABST
Patent Text Reader

Abstract

The invention discloses a paint spraying and baking operation supervision method based on visual identification, and relates to the technical field of environmental protection monitoring, and the method comprises the following steps: 1, system initialization: a system stores a plurality of algorithms, and predefines a plurality of compliance indexes; associating the compliance index with a violation analysis algorithm; step 2, image data acquisition: acquiring image data by using a data acquisition terminal and uploading the image data to a system; 3, monitoring object state analysis: analyzing the object state of the monitoring object through the image data; step 4, operation state analysis: according to object state analysis, identifying the operation state of the supervision object; step 5, violation analysis: carrying out violation analysis according to the object state and the operation state to obtain an analysis result; and step 6, violation recording and notification. According to the scheme, real-time monitoring and analysis of the paint spraying and baking operation process can be realized through a visual identification technology, the efficiency and accuracy of environmental protection supervision are effectively improved, and the cost is lower.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring, and in particular to a method for supervising spray painting operations based on visual recognition. Background Art

[0002] With the continuous progress of society, environmental protection has become a common focus of global concern. Various industries are actively taking measures to reduce the negative impact on the environment. In the automobile repair industry, spray painting is the main link that produces volatile organic compounds (VOCs), which poses a significant threat to the environment.

[0003] In order to ensure the quality of spray painting operations while effectively reducing their impact on the environment, environmental management departments have issued a series of regulatory requirements and emission standards. These regulations are intended to guide and supervise auto repair companies to adopt more environmentally friendly operating methods when performing spray painting operations.

[0004] It is particularly noteworthy that non-standard spray painting operations will lead to serious unorganized emissions, which is a difficult problem that needs to be solved urgently in the current environmental protection work. In order to meet this challenge and ensure that the pollution emissions from spray painting operations meet the prescribed standards, the environmental management department needs to strictly supervise and manage the spray painting operations of auto repair companies. Through supervision and management, auto repair companies can be encouraged to improve the standardization of spray painting operations and reduce pollutant emissions, thereby ensuring the quality of automobile repairs while making positive contributions to environmental protection.

[0005] At present, the main supervision methods for spray painting operations in auto repair enterprises are:

[0006] 1. Environmental management departments regularly conduct supervisory monitoring of auto repair companies, including organized emission monitoring and unorganized emission monitoring, and evaluate whether the company's emissions meet the emission standards through monitoring results. Supervisory monitoring is a periodic monitoring, which has the problem of too long a time period and cannot fully reflect the actual situation. Moreover, this monitoring method is a terminal result monitoring method, which cannot supervise the production process of the company, and therefore cannot make targeted requirements or improvement suggestions.

[0007] Second, the environmental management department will also supervise and manage the process and working environment of the spray paint operation of auto repair enterprises through manual inspection. However, the manual inspection method is inefficient and prone to omissions, and it is impossible to fully and effectively monitor the process of spray paint operation.

[0008] 3. Environmental management departments require enterprises to install online monitoring equipment such as VOCs online analyzers to continuously monitor their pollution emissions. This method solves the problem of long supervisory monitoring cycles, but its main problem is that the purchase and installation costs of online monitoring equipment are high, and it requires long-term operation and maintenance services such as regular calibration and replacement of consumables. The operating costs are also high, which is not affordable for all auto repair companies, so it is difficult to promote.

[0009] In addition, existing technologies are unable to identify and analyze the status of spray painting operations in real time, and are unable to detect and correct violations in a timely manner. Summary of the invention

[0010] In view of the problems existing in the prior art, the present invention provides a method for supervising spray painting operations based on visual recognition, which can effectively improve the real-time supervision and management of the spray painting operation process and reduce the supervision cost.

[0011] The technical solution of the present invention is achieved in this way:

[0012] A method for supervising spray paint operations based on visual recognition, wherein data of an operation site is collected and submitted to a system, and the system receives, identifies, and analyzes the data, and outputs the violation status of the operation site; the operation site is a spray paint room; the spray paint room includes a number of supervision objects;

[0013] The method comprises the following steps:

[0014] Step 1, system initialization: the system stores a plurality of algorithms, including an object status analysis algorithm, a job status analysis algorithm and a violation analysis algorithm; the system predefines a number of compliance indicators; and associates the compliance indicators with the violation analysis algorithm;

[0015] Step 2, image data acquisition: using a data acquisition terminal to acquire image data of the work site, and uploading the image data to the system;

[0016] Step 3: Supervision object state analysis: the system analyzes the object state of the supervision object through the object state analysis algorithm and the image data;

[0017] Step 4, operation status analysis: the system uses the operation status analysis algorithm to analyze the operation actions and operation duration of the operation site according to the object status, and identifies the operation status of the supervised object according to the operation actions and the operation duration;

[0018] The operation actions include: starting the operation and ending the operation;

[0019] The operation status includes: in operation W1, not in operation W0;

[0020] When the job status analysis is that the job is in progress, execute step 5; otherwise, execute step 3;

[0021] Step 5, violation analysis: the system uses the violation analysis algorithm to perform violation analysis according to the object status and the operation status to obtain analysis results; the analysis results include violations and compliance;

[0022] When the analysis result is a violation, execute step 6; otherwise, execute step 3;

[0023] Step 6: Violation record and notification: The system records the violation data and issues a violation notification; go to step 3.

[0024] This solution uses visual recognition technology to identify the status of spray paint operations, which can achieve real-time monitoring and analysis of the spray paint operation process. When illegal operations are identified, timely response and processing can be carried out to reduce the negative effects of illegal spray paint operations on the environment, effectively improving the efficiency and accuracy of environmental protection supervision, and at a lower cost.

[0025] As a further optimization of the above scheme, the spray paint room is provided with a main door and a side door; the spray paint room is provided with a lighting unit and an exhaust gas treatment facility; the exhaust gas treatment facility collects and treats the exhaust gas generated in the spray paint operation; the exhaust gas treatment facility includes a fan;

[0026] Waste gas treatment facilities are a comprehensive system that includes a variety of equipment and technologies for the collection, treatment and discharge of waste gas. These systems are designed to reduce or eliminate harmful gas emissions generated during industrial production processes to protect the environment and human health.

[0027] The supervised objects include the main door, the side door, the lighting unit and the exhaust gas treatment facility.

[0028] The main door is mainly used for vehicles and parts or other equipment or parts that need to be sprayed and painted to enter and exit the spray paint room; the side door is mainly used for workers to enter and exit the spray paint room; the lighting unit includes conventional lighting lamps for illuminating the spray paint room.

[0029] As a further optimization of the above scheme, the object states include gate closed D0, gate open D1, gate half open D2, side door closed S0, side door open S1, lighting unit off L0, lighting unit open L1, exhaust gas treatment facility stopped F0, exhaust gas treatment facility fully operating F1, and exhaust gas treatment facility only fan operating F2.

[0030] As a further optimization of the above solution, the object state analysis algorithm includes a visual recognition algorithm and a power consumption analysis algorithm;

[0031] The visual recognition algorithm includes an object detection algorithm and an image classification algorithm;

[0032] The spray paint room is also provided with an electricity consumption monitoring device for collecting electricity consumption data of the exhaust gas treatment facility; the electricity consumption analysis algorithm analyzes the object status of the exhaust gas treatment facility based on the electricity consumption data.

[0033] The object detection algorithm is used to identify different regulatory objects in the framed captured images; the image classification algorithm is used to classify the framed images according to the status of the regulatory objects.

[0034] As a further optimization of the above solution, in the operation status analysis algorithm, the analysis process of the operation action is expressed as:

[0035]

[0036] If the object status of the exhaust gas treatment facility changes from the stop status F0 to the full operation status F1 or only the fan operation status F2, and the gate status changes from open D1 to closed D0, and the lighting unit status is open L1, it is considered to have started the operation;

[0037] Or if the object state of the exhaust gas treatment facility is the stopped state F0, and the state of the lighting unit is open L1, and the gate state changes from open D1 to closed D0, and the duration of state D0 is I, it is considered to be the start of the operation;

[0038]

[0039] If the object status of the exhaust gas treatment facility changes from the fully operating state F1 or the only fan operating state F2 to the stopped state F0, and the gate state changes from closed D0 to open D1, and the state of the lighting unit is open L1, the operation is considered to be completed;

[0040] Or if the object state of the exhaust gas treatment facility is the stopped state F0, and the state of the lighting unit is the open state L1, and the gate state changes from the closed state D0 to the open state D1, and the duration of the state D1 is I, the operation is deemed to be completed;

[0041] Among them, C s Indicates the start of the operation, C e Indicates the end of the job;

[0042] D(t) represents the object state of the gate at time t;

[0043] L(t) represents the object state of the lighting unit at time t;

[0044] F(t) represents the object state of the exhaust gas treatment facility at time t;

[0045] t-1 represents the previous time node at time t; I represents the duration in seconds; t+I represents the duration of I seconds starting from time t.

[0046] As a further optimization of the above scheme, the calculation of the operation duration is expressed as: P t =T e -T s ; Among them, P t Indicates the duration of the operation in seconds, T e Indicates the end time of the operation, T s Indicates the time when the job starts.

[0047] As a further optimization of the above solution, the operation status analysis also includes abnormal operation analysis; the operation status also includes abnormal operation;

[0048] The system stores a first preset time; the abnormal operation analysis analyzes whether it is an abnormal operation according to the operation duration and the first preset time;

[0049] When the operation duration exceeds the preset range of the first preset time, it is determined to be an abnormal operation;

[0050] When the operation status is an abnormal operation, the system performs abnormal processing.

[0051] In this solution, abnormal operations are mainly targeted at those that occur during the idle hours of the company performing spray paint operations or when the start or end time of the operation is close to the company's off-get off work time. This is manifested as the failure to end the operation in time after a normal operation is actually completed, resulting in the operation duration in the analysis result being far longer than the actual duration.

[0052] As a further optimization of the above solution, the exception processing includes a self-correction process, and the system corrects the operation duration through the self-correction process;

[0053] The system is provided with a second preset time and ratio threshold;

[0054] when hour,

[0055] in, Indicates the first preset time, Indicates the second preset time, P t represents the job duration, and θ represents the ratio threshold.

[0056] As a further optimization of the above scheme, the violation analysis algorithm includes analysis of side door not being closed for a long time, analysis of the number of side door openings and closings, analysis of main door not being closed for a long time, analysis of fan not being turned on, analysis of exhaust gas treatment facilities not being in operation, and analysis of abnormal number of spraying and baking paint operations;

[0057] The analysis of the side door not being closed for a long time is that when the operation state is in operation W1, and the object state of the side door is S1, and the duration t1 ≥ the first threshold θ1, it is determined to be a violation;

[0058] The side door opening and closing times analysis is to count the opening and closing times N1 of the side door; when the operation state is in operation W1, and N1 ≥ θ2, it is determined to be a violation; θ2 represents the second threshold;

[0059] The main door is not closed for a long time. When the operation state is in operation W1, and the object state of the gate is D2, and the duration t3 ≥ the third threshold θ3, it is determined to be a violation;

[0060] The analysis of the fan not being turned on is that when the operation state is in operation W1, and the object state of the exhaust gas treatment facility is F0, and the duration t4 ≥ the fourth threshold θ4, it is determined to be a violation;

[0061] The exhaust gas treatment facility is not in operation analysis: when the operation state is in operation W1, and the object state of the exhaust gas treatment facility is F0 or F2, and the duration t5 ≥ the fifth threshold θ5, it is determined to be a violation;

[0062] The analysis results also include suspected violations; suspected violations indicate that abnormalities occurred during the spray paint operation, which can be treated as violations and handled, or other abnormal handling mechanisms can be adopted to handle them.

[0063] The abnormal analysis of the number of spray painting operations is that the system counts the total number of spray painting operations N2 on the day and the average number of spray painting operations per day when It is considered as suspected violation.

[0064] Wherein, θ6 represents the sixth threshold; N2(d) represents the total number of spray painting operations on the dth day; and i represents the number of continuous days.

[0065] As a further optimization of the above scheme, the system also calculates and records the enterprise production activity level index for each day. The ratio of the enterprise production activity level index to the total number of spraying and baking paint operations on the same day is the proportional coefficient ρ; it also includes the mean of the proportional coefficient

[0066] The abnormal analysis of the number of spray paint operations also includes: When , it is determined to be a suspected violation; where θ7 represents the seventh threshold; ρ(d) represents the proportional coefficient of the dth day.

[0067] The enterprise production activity level indicators can be calculated by collecting enterprise production activity level data, such as vehicle inlet and outlet volume, raw and auxiliary material usage, etc.

[0068] As a further optimization of the above solution, the violation data includes the compliance index, violation start time, violation end time and violation duration.

[0069] As a further optimization of the above solution, the violation notification includes any combination of one or more of the information notification within the system, WeChat notification, SMS notification, telephone voice notification, email notification and sound and light alarm notification at the work site, and the staff is notified through messages to respond and handle abnormal phenomena in a timely manner.

[0070] Compared with the prior art, the present invention achieves the following beneficial effects:

[0071] (1) This technical solution analyzes the violations of each working state during the operation, and then analyzes the start, end and duration of the operation. It determines whether there is a violation based on the analysis results, and records and issues alarms in a timely manner when a violation is found. This can effectively reduce the violation of the operation, thereby reducing the impact of the spray paint operation on the environment;

[0072] (2) This technical solution uses visual recognition technology and violation analysis algorithms to identify the status of spray painting operations, which can achieve real-time monitoring and analysis of the spray painting process, and can complete automatic identification without manual intervention, effectively improving the efficiency and accuracy of environmental protection supervision;

[0073] (3) This technical solution only requires the installation of image data acquisition terminals such as cameras and power consumption monitoring equipment to monitor and analyze violations of spray painting operations, without the need to use more expensive online monitoring equipment such as VOCs online analyzers. The installation and maintenance costs are low, which can effectively reduce the cost burden of enterprises in environmental protection work;

[0074] (4) This technical solution can be flexibly deployed in different spray paint operation environments according to actual needs, without the need for large-scale hardware modification, and has low implementation difficulty. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 It is a flow chart of a method for supervising a spray painting operation based on visual recognition provided by an embodiment of the present invention;

[0076] Figure 2 It is a data flow diagram of a spray paint operation supervision method based on visual recognition provided by an embodiment of the present invention;

[0077] Figure 3 It is a schematic diagram of the layout of a spray paint room provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0078] In order to make the purpose, technical solution and advantages of the present invention more clear, the technical solution in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0079] like Figures 1 to 3 As shown, this embodiment provides a method for supervising spray paint operations based on visual recognition, by collecting data from the work site and submitting it to the system, the system receives the data through a data collection terminal; the system identifies and analyzes the data, and outputs the violation status of the work site; the work site is a spray paint room; the spray paint room includes several supervision objects.

[0080] In this embodiment, the spray paint room is provided with a main door and a side door; the spray paint room is provided with a lighting unit and a waste gas treatment facility; the waste gas treatment facility collects and treats the waste gas generated during the spray paint operation; the waste gas treatment facility includes a fan; the supervision objects include the main door, the side door, the lighting unit and the waste gas treatment facility.

[0081] The main door is mainly used for vehicles and parts or other equipment or parts that need to be sprayed and painted to enter and exit the spray paint room; the side door is mainly used for workers to enter and exit the spray paint room; the lighting unit includes conventional lighting lamps for illuminating the spray paint room.

[0082] The method comprises the following steps:

[0083] Step 1: System initialization: The system stores a variety of algorithms, including object status analysis algorithm, job status analysis algorithm and violation analysis algorithm; the system predefines several compliance indicators; and associates the compliance indicators with the violation analysis algorithm.

[0084] In this embodiment, the compliance indicators include:

[0085] The manual entry and exit door of the spray booth is open or half-open (R1).

[0086] Frequent door opening during spray painting operation (R2),

[0087] Failure to close the door in time during operation (R3),

[0088] The fan was not turned on in time during the spraying operation (R4),

[0089] Treatment facilities were not turned on during the operation (R5),

[0090] No spray painting was carried out in the spray paint room (R6).

[0091] The violation analysis algorithms include the side door not being closed for a long time analysis A1, the side door opening and closing times analysis A2, the main door not being closed for a long time analysis A3, the fan not being turned on analysis A4, the exhaust gas treatment facility not being in operation analysis A5, and the abnormal number of spray paint operations analysis A6.

[0092] The correlation between compliance indicators and violation analysis algorithms is as follows:

[0093] R1 and A1 are aimed at regulating side doors;

[0094] R2 and A2 are aimed at regulating side doors;

[0095] R3 and A3, the supervision object is the gate;

[0096] R4 and A4 are regulated by waste gas treatment facilities;

[0097] R5 and A5 are aimed at regulating waste gas treatment facilities;

[0098] R6 and A6 are designed to supervise the entire spray booth.

[0099] Step 2, image data acquisition: The data acquisition terminal is connected to the data acquisition interface of the system through the Internet, and the image data of the work site is collected in real time by the data acquisition terminal, and the image data is continuously uploaded to the system. The system receives the image data and keeps a record of it. In this solution, a high-definition camera is used as a data acquisition terminal to monitor and collect videos of the spray paint room, and a monitoring video stream is obtained as image data. The system keeps a record of the collected video. In addition, the connection method between the data acquisition terminal and the system is not limited to the Internet, and any other form of networking can also be used for connection.

[0100] In this embodiment, the high-definition camera is installed on the front side of the spray paint booth to ensure that all monitored objects are within the monitoring range of the high-definition camera. At the same time, it is installed outside the spray paint booth to avoid being affected by the paint mist in the spray paint booth, which greatly reduces the subsequent maintenance cost. At this time, since the lighting unit is set in the spray paint booth, the high-definition camera identifies whether the internal lighting unit is turned on by shooting the light emitted from the spray paint booth. For example, through the transparent glass or window set in the spray paint booth, it can be identified whether there is light / bright light in the spray paint booth, and then whether the lighting unit is turned on.

[0101] Step 3: Supervision object status analysis: The system analyzes the object status of the supervision object through the object status analysis algorithm and image data.

[0102] In this embodiment, the object states include gate closed D0, gate open D1, gate half open D2, side door closed S0, side door open S1, lighting unit off L0, lighting unit on L1, exhaust gas treatment facility stopped F0, exhaust gas treatment facility fully running F1, and exhaust gas treatment facility only fan running F2.

[0103] In this embodiment, the object state analysis algorithm includes a visual recognition algorithm and a power consumption analysis algorithm;

[0104] The visual recognition algorithm includes an object detection algorithm and an image classification algorithm; in this solution, the two visual recognition algorithms are used collaboratively.

[0105] The object detection algorithm is used to identify different regulatory objects in the framed captured images; the image classification algorithm is used to classify the framed images according to the status of the regulatory objects.

[0106] The spray booth is also equipped with power consumption monitoring equipment to collect power consumption data of the exhaust gas treatment facility; the power consumption analysis algorithm analyzes the object status of the exhaust gas treatment facility based on the power consumption data. The power consumption monitoring equipment is installed on the power supply circuit of the exhaust gas treatment facility and is connected to the data acquisition interface of the system.

[0107] In addition, whether to install electricity monitoring equipment is an optional option. If electricity monitoring equipment is not installed, corresponding visual recognition and analysis algorithms can be added in steps 3 and 4.

[0108] The system stores and records the object status obtained after identification.

[0109] Step 4: Operation status analysis: The system uses the operation status analysis algorithm to analyze the operation actions and operation duration at the operation site according to the object status, and identifies the operation status of the supervised object based on the operation actions and operation duration;

[0110] The operation actions include: starting the operation and ending the operation;

[0111] The job status includes: working W1 and not working W0.

[0112] In this embodiment, in the operation status analysis algorithm, the analysis process of the operation action is expressed as:

[0113]

[0114] If the object status of the exhaust gas treatment facility changes from the stop status F0 to the full operation status F1 or only the fan operation status F2, and the gate status changes from open D1 to closed D0, and the lighting unit status is open L1, it is considered to have started the operation;

[0115] Or if the object state of the exhaust gas treatment facility is the stop state F0, and the state of the lighting unit is open L1, and the gate state changes from open D1 to closed D0, and the duration of state D0 is I, it is considered to be the start of the operation.

[0116] In this embodiment, if the power consumption monitoring device is not installed, and the state of the lighting unit is open L1, and the state of the door changes from open D1 to closed D0, and the duration of state D0 is I, it is considered that the operation has started.

[0117]

[0118] If the object status of the exhaust gas treatment facility changes from the fully operating state F1 or the only fan operating state F2 to the stopped state F0, and the gate state changes from closed D0 to open D1, and the state of the lighting unit is open L1, the operation is considered to be completed;

[0119] Or if the object state of the exhaust gas treatment facility is the stop state F0, and the state of the lighting unit is open L1, and the gate state changes from closed D0 to open D1, and the duration of state D1 is I, the operation is deemed to be completed.

[0120] In this embodiment, the value of I is 120 seconds.

[0121] In this embodiment, if the power consumption monitoring device is not installed, and the state of the lighting unit is open L1, and the door state changes from closed D0 to open D1, and the duration of state D1 is I, it is considered that the operation is completed.

[0122] Among them, C s Indicates the start of the operation, C e Indicates the end of the job;

[0123] D(t) represents the object state of the gate at time t;

[0124] L(t) represents the object state of the lighting unit at time t;

[0125] F(t) represents the object state of the waste gas treatment facility at time t;

[0126] t-1 represents the previous time node at time t; I represents the duration in seconds; t+I represents the duration of I seconds starting from time t.

[0127] When the operation action changes, record the end time or start time of the operation.

[0128] In this embodiment, the calculation of the operation duration is expressed as: t =T e -T s ; Among them, P t Indicates the duration of the operation in seconds, Te Indicates the end time of the operation, T s Indicates the time when the job starts.

[0129] In this embodiment, the operation status analysis also includes abnormal operation analysis; the operation status also includes abnormal operation;

[0130] The system stores a first preset time; the abnormal operation analysis analyzes whether it is an abnormal operation according to the operation duration and the first preset time;

[0131] When the operation duration exceeds the preset range of the first preset time, it is determined to be an abnormal operation;

[0132] When the job status is abnormal, the system performs exception processing.

[0133] In this embodiment, the exception processing includes a self-correction process, through which the system corrects the operation duration;

[0134] The system is provided with a second preset time and ratio threshold;

[0135] when hour,

[0136] in, Indicates the first preset time, Indicates the second preset time, P t represents the duration of the operation, and θ represents the ratio threshold. In this embodiment, θ is set to 1.

[0137] In this solution, abnormal operations are mainly targeted at those that occur during the idle hours of the company performing spray paint operations or when the start or end time of the operation is close to the company's off-get off work time. This is manifested as the failure to end the operation in time after a normal operation is actually completed, resulting in the operation duration in the analysis result being far longer than the actual duration.

[0138] The system records the historical spray paint operation data of all enterprises, including the number of operations per day, the duration of each operation, etc.; based on these data, the average spray paint operation duration of all enterprises can be calculated as the first preset time; the system counts the spray paint operation duration of the current enterprise each day, and calculates the daily average spray paint operation duration as the second preset time.

[0139] When the job status is analyzed as being in progress, execute step 5; otherwise, execute step 3.

[0140] The system stores and records the job status obtained after identification.

[0141] Step 5: Violation analysis: The system uses a violation analysis algorithm to perform violation analysis based on the object status and job status to obtain analysis results; the analysis results include violations, suspected violations, and compliance.

[0142] In this embodiment, the analysis of the side door not being closed for a long time is that when the operation state is in operation W1, and the object state of the side door is S1, and the duration t1 ≥ the first threshold θ1, it is determined to be a violation; in this embodiment, θ1 is taken as 60 seconds.

[0143] The side door opening and closing times are analyzed as follows: the side door opening and closing times N1 are counted; when the operation state is in operation W1, and N1 ≥ θ2, it is determined to be a violation; θ2 represents the second threshold; in this embodiment, θ2 takes a value of 15 times.

[0144] Analysis of the main door not being closed for a long time: when the operating state is in operation W1, and the object state of the gate is D2, and the duration t3 ≥ the third threshold θ3, it is determined to be a violation; in this embodiment, θ3 is 60 seconds.

[0145] The analysis of the fan not being turned on is that when the operating state is in operation W1, and the object state of the exhaust gas treatment facility is F0, and the duration t4 ≥ the fourth threshold θ4, it is determined to be a violation; in this embodiment, θ4 is taken as 60 seconds.

[0146] The analysis of the exhaust gas treatment facility not being in operation is that when the operating state is in operation W1, and the object state of the exhaust gas treatment facility is F0 or F2, and the duration t5 ≥ the fifth threshold θ5, it is judged as a violation; in this embodiment, θ5 is taken as 60 seconds.

[0147] The abnormal analysis of the number of spray painting operations is that the system counts the total number of spray painting operations N2 on the day and the average number of spray painting operations per day when It is considered as suspected violation.

[0148] Wherein, θ6 represents the sixth threshold value; N2(d) represents the total number of spraying and baking paint operations on the dth day; i represents the number of continuous days. In this embodiment, i is 5, and θ6 is 0.2.

[0149] In this embodiment, the system also calculates and records the enterprise production activity level index for each day. The ratio of the enterprise production activity level index to the total number of spraying and baking paint operations on the same day is the proportional coefficient ρ; it also includes the proportional coefficient mean

[0150] The abnormal analysis of spray painting operation times also includes: When θ7 is the seventh threshold, ρ(d) is the proportional coefficient of the dth day. In this embodiment, the number of spraying and baking paint operations is abnormal and is treated as illegal data for processing.

[0151] The enterprise production activity level indicators can be calculated by collecting enterprise production activity level data, such as vehicle inlet and outlet volume, raw and auxiliary material usage, etc.

[0152] If the analysis result is a violation, go to step 6; otherwise, go to step 3.

[0153] Step 6, violation records and notifications: The system records the violation data, forms historical violation data, and issues violation notifications. In this embodiment, the violation data includes compliance indicators, violation start time, violation end time, and violation duration. Violation notifications include system internal information notifications, WeChat notifications, SMS notifications, telephone voice notifications, email notifications, and sound and light alarm notifications at the work site. Through the message, the staff, such as environmental supervisors or spray paint operation site workers, are notified to respond and handle abnormal phenomena in a timely manner. At the same time, the system also saves the message sending records.

[0154] Continue to step 3.

[0155] This solution uses visual recognition technology to identify the status of spray paint operations, which can achieve real-time monitoring and analysis of the spray paint operation process. When illegal operations are identified, timely response and processing can be carried out to reduce the negative effects of illegal spray paint operations on the environment, effectively improving the efficiency and accuracy of environmental protection supervision, and at a lower cost.

[0156] According to the disclosure and teaching of the above description, those skilled in the art to which the present invention belongs may also make changes and modifications to the above embodiments. Therefore, the present invention is not limited to the specific embodiments disclosed and described above, and some modifications and changes to the present invention should also fall within the scope of protection of the claims of the present invention. In addition, although some specific terms are used in this specification, these terms are only for the convenience of description and do not constitute any limitation to the present invention.

Claims

1. A method for supervising spray painting operations based on visual recognition, by collecting data from the operation site and submitting it to a system, the system receives, identifies, and analyzes the data, and outputs the violation status of the operation site; characterized in that: The work site is a spray paint room; the spray paint room includes several supervision objects; The method comprises the following steps: Step 1, system initialization: the system stores a plurality of algorithms, including an object status analysis algorithm, a job status analysis algorithm and a violation analysis algorithm; the system predefines a number of compliance indicators; and associates the compliance indicators with the violation analysis algorithm; Step 2, image data acquisition: using a data acquisition terminal to acquire image data of the work site, and uploading the image data to the system; Step 3: Supervision object state analysis: the system analyzes the object state of the supervision object through the object state analysis algorithm and the image data; Step 4, operation status analysis: the system uses the operation status analysis algorithm to analyze the operation actions and operation duration of the operation site according to the object status, and identifies the operation status of the supervised object according to the operation actions and the operation duration; The operation actions include: starting the operation and ending the operation; The operation status includes: in operation W1, not in operation W0; When the job status analysis is that the job is in progress, execute step 5; otherwise, execute step 3; Step 5, violation analysis: the system uses the violation analysis algorithm to perform violation analysis according to the object status and the operation status to obtain analysis results; the analysis results include violations and compliance; When the analysis result is a violation, execute step 6; otherwise, execute step 3; Step 6: Violation record and notification: The system records the violation data and issues a violation notification; go to step 3.

2. A method for monitoring spray painting operations based on visual recognition according to claim 1, characterized in that: The spray paint room is provided with a main door and a side door; the spray paint room is provided with a lighting unit and an exhaust gas treatment facility; the exhaust gas treatment facility collects and treats the exhaust gas generated during the spray paint operation; the exhaust gas treatment facility includes a fan; The supervised objects include the main door, the side door, the lighting unit and the exhaust gas treatment facility.

3. The method for monitoring spray painting operations based on visual recognition according to claim 2, characterized in that: The object states include gate closed D0, gate open D1, gate half open D2, side door closed S0, side door open S1, lighting unit off L0, lighting unit on L1, exhaust gas treatment facility stopped F0, exhaust gas treatment facility fully running F1, and exhaust gas treatment facility only fan running F2.

4. The method for monitoring spray painting operations based on visual recognition according to claim 2, characterized in that: The object state analysis algorithm includes a visual recognition algorithm and a power consumption analysis algorithm; The visual recognition algorithm includes an object detection algorithm and an image classification algorithm; The spray paint room is also provided with an electricity consumption monitoring device for collecting electricity consumption data of the exhaust gas treatment facility; the electricity consumption analysis algorithm analyzes the object status of the exhaust gas treatment facility based on the electricity consumption data.

5. The method for monitoring spray painting operations based on visual recognition according to claim 3 is characterized in that: In the operation status analysis algorithm, the analysis process of the operation action is expressed as: Among them, C s Indicates the start of the operation, C e Indicates the end of the operation; D(t) represents the object state of the gate at time t; L(t) represents the object state of the lighting unit at time t; F(t) represents the object state of the exhaust gas treatment facility at time t; t-1 represents the previous time node at time t; I represents the duration in seconds; t+I represents the duration of I seconds starting from time t.

6. The method for monitoring spray painting operations based on visual recognition according to claim 1, characterized in that: The calculation of the operation duration is expressed as: P t =T e -T s ; Among them, P t Indicates the duration of the operation in seconds, T e Indicates the end time of the operation, T s Indicates the time when the job starts.

7. The method for monitoring spray painting operations based on visual recognition according to claim 1, characterized in that: The operation status analysis also includes abnormal operation analysis; the operation status also includes abnormal operation; The system stores a first preset time; the abnormal operation analysis analyzes whether it is an abnormal operation according to the operation duration and the first preset time; When the operation duration exceeds the preset range of the first preset time, it is determined to be an abnormal operation; When the operation status is an abnormal operation, the system performs abnormal processing.

8. The method for monitoring spray painting operations based on visual recognition according to claim 7, characterized in that: The abnormal processing includes a self-correction process, and the system corrects the operation duration through the self-correction process; The system is provided with a second preset time and ratio threshold; when hour, in, Indicates the first preset time, Indicates the second preset time, P t represents the job duration, and θ represents the ratio threshold.

9. The method for monitoring spray painting operations based on visual recognition according to claim 3, characterized in that: The violation analysis algorithm includes analysis of side door not being closed for a long time, analysis of the number of side door openings and closings, analysis of main door not being closed for a long time, analysis of fan not being turned on, analysis of exhaust gas treatment facilities not being in operation, and analysis of abnormal number of spraying and baking paint operations; The side door not being closed for a long time is analyzed as follows: when the operation state is in operation W1, and the object state of the side door is S1, and the duration t1 ≥ the first threshold θ1, it is determined to be a violation; The side door opening and closing times analysis is to count the opening and closing times N1 of the side door; when the operation state is in operation W1, and N1 ≥ θ2, it is determined to be a violation; θ2 represents the second threshold; The main door is not closed for a long time. When the operation state is in operation W1, and the object state of the gate is D2, and the duration t3 ≥ the third threshold θ3, it is determined to be a violation; The analysis of the fan not being turned on is that when the operating state is in operation W1, and the object state of the exhaust gas treatment facility is F0, and the duration t4 ≥ the fourth threshold θ4, it is determined to be a violation; The exhaust gas treatment facility is not in operation analysis: when the operation state is in operation W1, and the object state of the exhaust gas treatment facility is F0 or F2, and the duration t5 ≥ the fifth threshold θ5, it is determined to be a violation; The analysis results also include suspected violations; The abnormal analysis of the number of spray painting operations is that the system counts the total number of spray painting operations N2 on the day and the average number of spray painting operations per day when It is considered as suspected violation. Wherein, θ6 represents the sixth threshold; N2(d) represents the total number of spray painting operations on the dth day; and i represents the number of continuous days.

10. The method for monitoring spray painting operations based on visual recognition according to claim 9, characterized in that: The system also calculates and records the enterprise production activity level index for each day, and the ratio of the enterprise production activity level index to the total number of spraying and baking paint operations on the same day is a proportional coefficient ρ; Also includes the mean of the proportionality coefficient The abnormal analysis of the number of spray paint operations also includes: When , it is determined to be a suspected violation; where θ7 represents the seventh threshold; ρ(d) represents the proportional coefficient of the dth day.

Citation Information

Patent Citations

  • Method for monitoring illegal production of pollution source enterprise based on power consumption data

    CN113298422A

  • Paint spray booth based on Internet of Things

    CN115055322A

  • Violation type alarm reminding collection system based on image recognition algorithm

    CN116091974A

  • Intelligent kitchen supervision system and method for personnel operation state based on visual analysis

    CN116863401A

  • Wafer cleaning apparatus and cleaning method thereof

    KR1020230159179A

Cited By

  • Business state analysis method and system based on computer vision and signal processing

    CN121438181A