A method for safety monitoring of reinforced concrete silo construction based on machine vision

Abstract

The present invention relates to the field of smart construction, and in particular to a method for monitoring the construction safety of a reinforced concrete silo based on machine vision, comprising periodically acquiring a personnel change rate and a personnel quantity reference value in each monitoring area, and determining a personnel feature state corresponding to each monitoring area according to the personnel change rate and the personnel quantity reference value; determining a corresponding personnel analysis method according to the personnel feature state of each monitoring area; in single-frame analysis, determining a type I personnel and a type II personnel according to the head feature difference of personnel in a target frame image; in multi-frame analysis, determining a shooting state according to the personnel distance uniformity and a personnel distance reference value of a number of monitoring image frames corresponding to a current monitoring period, and determining an image selection method according to the shooting state; determining a final selected route according to a risk factor and / or a vibration impact value of a transportation route; the present invention improves the safety of route planning of transportation vehicles by improving the safety monitoring efficiency of a construction site.

Description

Technical Field

[0001] The present invention relates to the field of smart construction, and in particular to a reinforced concrete silo construction safety monitoring method based on machine vision. Background Art

[0002] With the discovery of science and technology, the integration of traditional scenes with artificial intelligence has become the current trend of technological development. Construction safety monitoring based on machine vision is an innovative solution that combines artificial intelligence and computer vision technology. It aims to improve the efficiency and accuracy of construction site safety management. By using artificial intelligence machine vision technology, a real-time detection and early warning platform can be built to automatically issue early warnings to prevent construction accidents and provide safety warnings for some construction management. For example, in the silo construction site, in addition to the silo, it also includes working towers, substations, fire pump stations, loading buildings, and other structures and road yards. The corresponding devices used are also diverse. The transportation routes of vehicles on traditional construction sites are completely selected by the driver at will. Therefore, the safety factor is difficult to guarantee. In order to achieve safe and civilized construction, how to provide artificial intelligence-assisted route selection is a technical problem that technicians in this field need to solve urgently.

[0003] Chinese patent publication number CN112652166B discloses a construction site transportation scheduling system, which includes the following: a model building unit; a simulation training unit; a time calculation unit: used to estimate the time that each transport vehicle will need to travel on the transport route according to the current wet or dry road conditions, and then, combined with the remaining parking spaces at the construction site and the unloading time t5 of the transport vehicle, obtain the time when the next transport vehicle enters the construction site; a positioning monitoring unit; and a scheduling unit. In addition, Chinese patent publication number CN117876971 B discloses a construction safety monitoring and early warning method based on machine vision. The method combines the initial grayscale value and displacement change degree of each pixel in each frame image, clusters all pixels, and obtains multiple construction site element clusters; obtains the enhancement weight of the pixels contained in each construction site element cluster in each frame image; combines the initial grayscale value of each pixel in each frame image and the overall grayscale difference between the construction site element cluster and other construction site element clusters, obtains the enhancement coefficient of each pixel in each frame image; obtains the enhanced grayscale value of each pixel in each frame image, obtains the enhanced video frame image, obtains the enhanced monitoring video; and monitors and warns the construction site. It can be seen that the above technical solutions respectively disclose the technical means of using machine vision for safety analysis and analyzing the construction site to select the transportation route, but there are the following problems: it is impossible to extract the images that need to be analyzed according to the movement characteristics of the personnel, and it is impossible to effectively select the transportation route according to the actual collected personnel status, resulting in poor route selection accuracy. Summary of the invention

[0004] To this end, the present invention provides a reinforced concrete silo construction safety monitoring method based on machine vision, so as to overcome the problems in the prior art that it is impossible to extract images that need to be analyzed according to the movement characteristics of personnel in a targeted manner, and it is impossible to effectively select the transportation route according to the actual collected personnel status, resulting in poor route selection accuracy.

[0005] To achieve the above object, the present invention provides a reinforced concrete silo construction safety monitoring method based on machine vision, comprising:

[0006] Periodically obtain the personnel change rate and personnel quantity reference value of each monitoring area, and determine the personnel characteristic state corresponding to each monitoring area according to the personnel change rate and personnel quantity reference value;

[0007] Determine the corresponding personnel analysis method according to the personnel characteristic status in each monitoring area, including multi-frame analysis or single-frame analysis;

[0008] In single-frame analysis, the first-class personnel and the second-class personnel are determined based on the difference of the head features of the personnel in the target frame image;

[0009] In the multi-frame analysis, the shooting state is determined according to the uniformity of the personnel distance of several monitoring image frames corresponding to the current monitoring period and the personnel distance reference value, and the image selection method is determined according to the shooting state, including image selection according to the environmental impact coefficient, or image selection according to the personnel distance;

[0010] Determine the evaluation coefficient corresponding to the monitoring area according to the number of Category I personnel and Category II personnel in the monitoring area, and determine the risk coefficient of the transportation route according to the evaluation coefficient of the monitoring area corresponding to the transportation route of the transport vehicle and the number of personnel in each monitoring area;

[0011] Determine whether to perform vibration impact value analysis according to the number of transport routes whose corresponding risk factors of transport vehicles are less than a preset risk factor;

[0012] The final selected route is determined based on the hazard factor and / or vibration impact value of the transportation route.

[0013] Further, the personnel analysis method is determined according to the characteristic status of the personnel in the monitoring area;

[0014] If the personnel characteristic state is the first preset personnel characteristic state, the personnel analysis method is multi-frame analysis;

[0015] If the personnel feature state is the second preset personnel feature state, the personnel analysis method is single-frame analysis.

[0016] Further, the characteristic status of personnel in a single monitoring area is determined based on the rate of change of personnel in the monitoring area and a reference value of the number of personnel;

[0017] If the rate of change of personnel in the monitoring area is greater than the preset rate of change or the reference value of the number of personnel is greater than the preset reference value of the number of personnel, the personnel characteristic state is the first preset personnel characteristic state;

[0018] If the personnel change rate is less than or equal to the preset change rate and the personnel quantity reference value is less than or equal to the preset personnel quantity reference value, the personnel characteristic state is the second preset personnel characteristic state.

[0019] Furthermore, the multi-frame analysis includes:

[0020] Acquire a number of monitoring image frames corresponding to the current monitoring period based on a preset extraction criterion;

[0021] Determine the image selection method according to the shooting state of a plurality of monitoring image frames;

[0022] Acquire a number of valid monitoring image frames, and determine the personnel risk factor based on the valid monitoring image frames.

[0023] Furthermore, the shooting status is determined based on the personnel distance uniformity and the personnel distance reference value, and the shooting status includes that the personnel distance uniformity is less than the preset personnel distance uniformity or the personnel distance reference value is greater than the preset personnel distance reference value, or that the personnel distance uniformity is greater than or equal to the preset personnel distance uniformity and the personnel distance reference value is less than or equal to the preset personnel distance reference value.

[0024] Furthermore, the image selection method includes:

[0025] If the uniformity of the personnel distance is less than the preset uniformity of the personnel distance or the reference value of the personnel distance is greater than the preset reference value of the personnel distance, the image is selected according to the environmental impact coefficient;

[0026] If the personnel distance uniformity is greater than or equal to the preset personnel distance uniformity and the personnel distance reference value is less than or equal to the preset personnel distance reference value, image selection is performed based on the personnel distance.

[0027] Furthermore, for a single monitoring area, the evaluation coefficient may be determined by determining it based on the number of a type of personnel;

[0028] One type of personnel is identified based on the difference in head features or the similarity in dangerous actions;

[0029] The evaluation coefficient is positively correlated with the number of the first category of personnel.

[0030] Furthermore, a risk factor assessment is performed for the transport routes of transport vehicles. The risk factor of a single transport route is related to the assessment factor of each monitoring area corresponding to the transport route and the number of personnel in each monitoring area.

[0031] The risk factor of the transport route is positively correlated with the evaluation coefficient of each monitoring area corresponding to the transport route and the number of personnel in each monitoring area.

[0032] Furthermore, when evaluating the risk factor of the transport route of the transport vehicle, if there are at least two transport routes with risk factors less than the preset risk factor, a vibration impact value analysis is performed and the

[0033] The transportation route with the smallest vibration impact value is recorded as the final selected route.

[0034] The vibration impact value is related to the number of associated construction projects and the weight of the vehicle.

[0035] Furthermore, if there is only one transport route for the transport vehicle whose risk factor is less than a preset risk factor, then this transport route is the final selected route.

[0036] Compared with the prior art, the beneficial effect of the present invention lies in that, in the technical scheme of the present invention, the personnel analysis method is determined according to the characteristic status of the personnel in the monitoring area, so that the selection of the personnel analysis method is more in line with the actual work scene, thereby avoiding the problem in the prior art that multiple monitoring images cannot be effectively screened according to the personnel status, and, in the multi-frame analysis of the present invention, the image selection method is determined according to the shooting status of several monitoring image frames, taking into account the influence of the personnel distance on the image analysis, thereby making the image selected by the image selection method more beneficial to the subsequent analysis and judgment, and in addition, the present invention takes into account the vibration effect of the vehicle movement process on the construction project, further improving the construction safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of a reinforced concrete silo construction safety monitoring method based on machine vision of the present invention;

[0038] Figure 2 This is a flow chart of the present invention for determining a personnel analysis method according to the characteristic status of personnel in a monitored area;

[0039] Figure 3 The figure is a flow chart of selecting an image selection method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0041] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0042] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0043] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0044] See also Figures 1 to 3 As shown, the present invention provides a reinforced concrete silo construction safety monitoring method based on machine vision, comprising:

[0045] Periodically obtain the personnel change rate and personnel quantity reference value of each monitoring area, and determine the personnel characteristic state corresponding to each monitoring area according to the personnel change rate and personnel quantity reference value;

[0046] Determine the corresponding personnel analysis method according to the personnel characteristic status in each monitoring area, including multi-frame analysis or single-frame analysis;

[0047] In single-frame analysis, the first-class personnel and the second-class personnel are determined based on the difference of the head features of the personnel in the target frame image;

[0048] In the multi-frame analysis, the shooting state is determined according to the uniformity of the personnel distance of several monitoring image frames corresponding to the current monitoring period and the personnel distance reference value, and the image selection method is determined according to the shooting state, including image selection according to the environmental impact coefficient, or image selection according to the personnel distance;

[0049] Determine the evaluation coefficient corresponding to the monitoring area according to the number of Category I personnel and Category II personnel in the monitoring area, and determine the risk coefficient of the transportation route according to the evaluation coefficient of the monitoring area corresponding to the transportation route of the transport vehicle and the number of personnel in each monitoring area;

[0050] Determine whether to perform vibration impact value analysis according to the number of transport routes whose corresponding risk factors of transport vehicles are less than a preset risk factor;

[0051] The final selected route is determined based on the hazard factor and / or vibration impact value of the transportation route.

[0052] The application scenario of the present invention is the vehicle transportation route control at the reinforced concrete silo construction site. The vehicles at the construction site include but are not limited to concrete tankers, concrete pump trucks, full-rotation full-casing drilling rigs, rotary drilling rigs, crawler cranes and dump trucks. A number of industrial visual monitoring devices are set up at the construction site, and each monitoring device corresponds to a monitoring area.

[0053] Specifically, the personnel analysis method is determined according to the characteristic status of the personnel in the monitored area;

[0054] If the personnel characteristic state is the first preset personnel characteristic state, the personnel analysis method is multi-frame analysis;

[0055] If the personnel feature state is the second preset personnel feature state, the personnel analysis method is single-frame analysis.

[0056] Specifically, the personnel characteristic status of a single monitoring area is determined based on the personnel change rate and the personnel quantity reference value in the monitoring area;

[0057] If the rate of change of personnel in the monitoring area is greater than the preset rate of change or the reference value of the number of personnel is greater than the preset reference value of the number of personnel, the personnel characteristic state is the first preset personnel characteristic state;

[0058] If the personnel change rate is less than or equal to the preset change rate and the personnel quantity reference value is less than or equal to the preset personnel quantity reference value, the personnel characteristic state is the second preset personnel characteristic state.

[0059] The present invention is provided with a cyclic monitoring cycle, and the duration of a single monitoring cycle is related to the user's monitoring accuracy requirement. The greater the user's monitoring accuracy requirement, the shorter the duration of the monitoring cycle, and a value of the duration of a monitoring cycle is provided. The duration of the monitoring cycle is 15 minutes. The method for confirming the personnel change rate and the reference value of the number of personnel is that within a single monitoring cycle, the number of personnel in the monitoring area is recorded once every 1 minute, and the maximum number of personnel minus the minimum number of personnel is recorded as the personnel change rate, and the average number of personnel recorded every 1 minute is recorded as the reference value of the number of personnel; the values ​​of the preset change rate and the preset reference value of the number of personnel can be set by the user according to the actual application scenario. The greater the user's demand for construction safety, the smaller the preset change rate and the preset reference value of the number of personnel are. The values ​​of the preset change rate and the preset reference value of the number of personnel are provided, and the preset change rate is 5, and the preset reference value of the number of personnel is 10 people.

[0060] Specifically, multi-frame analysis includes:

[0061] Acquire a number of monitoring image frames corresponding to the current monitoring period based on a preset extraction criterion;

[0062] Determine the image selection method according to the shooting state of a plurality of monitoring image frames;

[0063] Acquire a number of valid monitoring image frames, and determine the personnel risk factor based on the valid monitoring image frames.

[0064] Specifically, the shooting state is determined according to the uniformity of the personnel distance and the reference value of the personnel distance, and the shooting state includes:

[0065] A first preset shooting state in which the uniformity of the personnel distance is less than a preset uniformity of the personnel distance or the reference value of the personnel distance is greater than a preset reference value of the personnel distance;

[0066] A second preset shooting state in which the personnel distance uniformity is greater than or equal to a preset personnel distance uniformity and the personnel distance reference value is less than or equal to a preset personnel distance reference value.

[0067] The preset extraction criterion is to record a monitoring image frame every minute. The method for confirming the uniformity of personnel distance and the reference value of personnel distance is to detect the personnel distance corresponding to each monitoring image frame. The personnel distance is the average value of the distance between each person and the monitoring device in the monitoring area at the corresponding moment of the monitoring image frame. The uniformity of personnel distance = 1 / (the maximum value of the personnel distance corresponding to each monitoring image frame - the minimum value of the personnel distance corresponding to each monitoring image frame). The reference value of personnel distance is the average value of the sum of the personnel distances corresponding to all monitoring image frames. The user can set the values ​​of the preset personnel distance uniformity and the preset personnel distance reference value according to historical experience and actual application scenarios. A preset personnel distance uniformity and a preset personnel distance reference value are provided. The preset personnel distance uniformity is 20% of the maximum value of the personnel distance corresponding to each monitoring image frame. The preset personnel distance reference value = (the maximum value of the personnel distance corresponding to each monitoring image frame + the minimum value of the personnel distance corresponding to each monitoring image frame) / 2.

[0068] Specifically, the image selection method includes:

[0069] If the uniformity of the personnel distance is less than the preset uniformity of the personnel distance or the reference value of the personnel distance is greater than the preset reference value of the personnel distance, the image is selected according to the environmental impact coefficient;

[0070] If the personnel distance uniformity is greater than or equal to the preset personnel distance uniformity and the personnel distance reference value is less than or equal to the preset personnel distance reference value, image selection is performed based on the personnel distance.

[0071] Specifically, for a single monitoring area, the evaluation coefficient is confirmed by the ratio of the number of Category 1 personnel to the number of Category 2 personnel, where the number of Category 1 personnel is the sum of the number of Category 1 personnel in each valid image frame, and the number of Category 2 personnel is the sum of the number of Category 2 personnel in each valid image frame.

[0072] In single-frame analysis, the first-class personnel and the second-class personnel are determined based on the difference of the head features of the personnel in the target frame image;

[0073] In multi-frame analysis, valid image frames are first selected, and then a type of person is determined based on the similarity of dangerous actions.

[0074] One type of personnel is identified based on the difference in head features or the similarity in dangerous actions;

[0075] The evaluation coefficient is positively correlated with the number of the first category of personnel.

[0076] When selecting images according to the environmental impact coefficient, the monitoring image frames whose environmental impact coefficient is less than the preset environmental impact coefficient are recorded as valid monitoring image frames, the environmental impact coefficient = the number of environmental vehicles × 0.4 + environmental brightness × 0.3 + environmental smoke value × 0.3, the environmental smoke value is the value detected by the smoke sensor, and the preset environmental impact coefficient is determined by the user according to the use effect of the environmental impact coefficient in the historical records. It can be understood that the higher the user's demand for the monitoring effect, the smaller the environmental impact coefficient. Among them, the first type of personnel is the personnel whose head feature difference is greater than the preset head feature difference, and the second type of personnel is the personnel whose head feature difference is less than or equal to the preset head feature difference. The head feature difference = 1-the similarity between the person's head image and the preset head image, and the preset head feature difference is 0.5. It can be understood that the preset head image includes several head images of people wearing helmets. The similarity between the person's head image and the preset head image is obtained by a deep learning convolutional neural network and normalized so that the similarity ranges from 0 to 1. This is content that is easy for technical personnel in this field to understand and will not be elaborated here.

[0077] When image selection is performed according to the personnel distance, the monitoring image frames whose personnel distance is less than the average value of the personnel distances of each monitoring image frame are recorded as valid monitoring image frames.

[0078] Category 1 personnel are those whose dangerous action similarity is greater than the preset dangerous action similarity, and Category 2 personnel are those whose dangerous action similarity is less than or equal to the dangerous action similarity. The dangerous action similarity is determined by a deep learning convolutional neural network and normalized so that the similarity ranges from 0 to 1. This is easy for technical personnel in this field to understand and will not be elaborated here.

[0079] Specifically, the risk factor of the transport route of the transport vehicle is evaluated. The risk factor of a single transport route is related to the evaluation factor of each monitoring area corresponding to the transport route and the number of personnel in each monitoring area;

[0080] The risk factor of the transport route is positively correlated with the evaluation coefficient of each monitoring area corresponding to the transport route and the number of personnel in each monitoring area.

[0081] The risk coefficient of the transport route = the average value of the evaluation coefficients of the monitoring areas corresponding to the transport route + the average value of the number of personnel in the monitoring areas corresponding to the transport route.

[0082] The monitoring area corresponding to the transportation route is the monitoring area through which the transportation route passes.

[0083] Specifically, when evaluating the risk factor of the transport routes of the transport vehicles, if there are at least two transport routes for the transport vehicles whose risk factors are less than the preset risk factors, a vibration impact value analysis is performed, and the transport route with the smallest vibration impact value is recorded as the final selected route.

[0084] The vibration impact value is related to the number of associated construction projects and the weight of the vehicle.

[0085] The preset risk factor is determined by the user based on his own needs and historical experience. The average value of the risk factor corresponding to the transportation route that meets the user's needs in the historical process is recorded as the preset risk factor. The vibration impact value = the number of related construction projects × 0.5 + vehicle weight × 0.5;

[0086] The number of associated construction projects is the number of construction projects that are under construction on the transport route and marked as susceptible to vibration. The construction projects susceptible to vibration are determined by the user. It is understandable that some construction projects need to avoid the impact of environmental vibration to improve construction efficiency and quality, such as drilling and concrete pouring during the construction of silo cast-in-place piles.

[0087] Specifically, if there is only one transport route for the transport vehicle with a risk factor less than a preset risk factor, then this transport route will be the final selected route.

[0088] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0089] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A reinforced concrete silo construction safety monitoring method based on machine vision, characterized in that: include: Periodically obtain the personnel change rate and personnel quantity reference value of each monitoring area, and determine the personnel characteristic state corresponding to each monitoring area according to the personnel change rate and personnel quantity reference value; Determine the corresponding personnel analysis method according to the personnel characteristic status in each monitoring area, including multi-frame analysis or single-frame analysis; In single-frame analysis, the first-class personnel and the second-class personnel are determined based on the difference of the head features of the personnel in the target frame image; In the multi-frame analysis, the shooting state is determined according to the uniformity of the personnel distance of several monitoring image frames corresponding to the current monitoring period and the personnel distance reference value, and the image selection method is determined according to the shooting state, including image selection according to the environmental impact coefficient, or image selection according to the personnel distance; Determine the evaluation coefficient corresponding to the monitoring area according to the number of Category I personnel and Category II personnel in the monitoring area, and determine the risk coefficient of the transportation route according to the evaluation coefficient of the monitoring area corresponding to the transportation route of the transport vehicle and the number of personnel in each monitoring area; Determine whether to perform vibration impact value analysis according to the number of transport routes whose corresponding risk factors of transport vehicles are less than a preset risk factor; Determine the final route based on the hazard factor and / or vibration impact value of the transportation route; In a single monitoring cycle, the number of people in the monitoring area is recorded every 1 minute, and the maximum number of people minus the minimum number of people is recorded as the personnel change rate; Detect the personnel distance corresponding to each monitoring image frame, where the personnel distance is the average value of the distance between each person and the monitoring device in the monitoring area at the time corresponding to the monitoring image frame, and the uniformity of the personnel distance = 1 / (the maximum value of the personnel distance corresponding to each monitoring image frame - the minimum value of the personnel distance corresponding to each monitoring image frame); For a single monitoring area, the evaluation coefficient may be determined based on the number of people in one category; One type of personnel is identified based on the difference in head features or the similarity in dangerous actions; The evaluation coefficient is positively correlated with the number of the first category of personnel; Vibration impact value = number of related construction projects × 0.5 + vehicle weight × 0.5; The number of associated construction projects is the number of construction projects on the transport route that are under construction and marked as susceptible to vibration.

2. The method for monitoring the construction safety of reinforced concrete silos based on machine vision according to claim 1 is characterized in that: Determine the personnel analysis method based on the personnel characteristics and status in the monitored area; If the personnel characteristic state is the first preset personnel characteristic state, the personnel analysis method is multi-frame analysis; If the personnel feature state is the second preset personnel feature state, the personnel analysis method is single-frame analysis.

3. The method for monitoring the construction safety of reinforced concrete silos based on machine vision according to claim 2 is characterized in that: The personnel characteristic status of a single monitoring area is determined based on the personnel change rate and the personnel quantity reference value in the monitoring area; If the rate of change of personnel in the monitoring area is greater than the preset rate of change or the reference value of the number of personnel is greater than the preset reference value of the number of personnel, the personnel characteristic state is the first preset personnel characteristic state; If the personnel change rate is less than or equal to the preset change rate and the personnel quantity reference value is less than or equal to the preset personnel quantity reference value, the personnel characteristic state is the second preset personnel characteristic state.

4. The method for monitoring the construction safety of reinforced concrete silos based on machine vision according to claim 3 is characterized in that: Multi-frame analysis includes: Acquire a number of monitoring image frames corresponding to the current monitoring period based on a preset extraction criterion; Determine the image selection method according to the shooting state of a plurality of monitoring image frames; Acquire a number of valid monitoring image frames, and determine the personnel risk factor based on the valid monitoring image frames.

5. The method for monitoring the construction safety of reinforced concrete silos based on machine vision according to claim 4 is characterized in that: The shooting status is determined according to the personnel distance uniformity and the personnel distance reference value, and the shooting status includes that the personnel distance uniformity is less than the preset personnel distance uniformity or the personnel distance reference value is greater than the preset personnel distance reference value, or, the personnel distance uniformity is greater than or equal to the preset personnel distance uniformity and the personnel distance reference value is less than or equal to the preset personnel distance reference value.

6. The method for monitoring the construction safety of reinforced concrete silos based on machine vision according to claim 5 is characterized in that: The image selection method includes: If the uniformity of the personnel distance is less than the preset uniformity of the personnel distance or the reference value of the personnel distance is greater than the preset reference value of the personnel distance, the image is selected according to the environmental impact coefficient; If the personnel distance uniformity is greater than or equal to the preset personnel distance uniformity and the personnel distance reference value is less than or equal to the preset personnel distance reference value, image selection is performed based on the personnel distance.

7. The method for monitoring the construction safety of reinforced concrete silos based on machine vision according to claim 6 is characterized in that: The risk factor of the transport route of the transport vehicle is evaluated. The risk factor of a single transport route is related to the evaluation factor of each monitoring area corresponding to the transport route and the number of personnel in each monitoring area; The risk factor of the transport route is positively correlated with the evaluation coefficient of each monitoring area corresponding to the transport route and the number of personnel in each monitoring area.

8. The method for monitoring the construction safety of reinforced concrete silos based on machine vision according to claim 7 is characterized in that: When evaluating the risk factor of the transport routes of transport vehicles, if there are at least two transport routes with risk factors less than the preset risk factors, a vibration impact value analysis is performed, and the transport route with the smallest vibration impact value is recorded as the final selected route; The vibration impact value is related to the number of associated construction projects and the weight of the vehicle.

9. The method for monitoring the construction safety of reinforced concrete silos based on machine vision according to claim 8 is characterized in that: If there is only one transport route for the transport vehicle with a risk factor less than the preset risk factor, then this transport route will be the final selected route.

Images