Production behavior detection method, system and device

Abstract

The invention relates to a production behavior detection method, system and device, and relates to the technical field of production behavior detection, and the method comprises the steps: obtaining a production image; judging whether a person exists in the image according to the production image; when a person exists in the image, determining the posture of the person according to the production image; determining the personnel gravity center according to the personnel attitude; determining a stable interval according to the personnel posture; when the gravity center of the person exceeds the stable interval, determining a falling position according to the production image; and controlling a preset warning device to give a fall warning according to the fall position. The method and the device have the effects of improving the accuracy of a production behavior detection technology and reducing detection result errors.

Description

Technical Field

[0001] The present invention relates to the field of production behavior detection, and in particular to a production behavior detection method, system and device. Background Art

[0002] Production behavior detection refers to the technology of monitoring and analyzing the behavior of personnel in the production process.

[0003] In the prior art, production behavior detection technology is often used in factories to ensure the safety, efficiency and quality of production activities. Production behavior detection technology generally obtains images of people through cameras, and then judges the behavior of people from the images, so as to urgently stop the operation of equipment in the factory when people perform dangerous behaviors to ensure the safety of personnel.

[0004] The factory environment is relatively complex. When people are affected by the factory environment and their body shapes change, it is easy for production behavior detection technology to detect that people are engaging in dangerous behavior, causing the equipment in the factory to stop urgently, which in turn reduces the service life of the equipment or even causes it to be damaged. Summary of the invention

[0005] In order to improve the accuracy of production behavior detection technology and reduce the possibility of errors in detection results, the present invention provides a production behavior detection method, system and device.

[0006] In a first aspect, the present invention provides a production behavior detection method, which adopts the following technical solution: A production behavior detection method, comprising: Get production images; Judging whether there is a person in the image based on the production image; When there is a person in the image, the person's posture is determined based on the production image; Determine the center of gravity of the person based on their posture; Determine the stable interval according to the personnel's posture; When the center of gravity of the person exceeds the stable range, the fall position is determined based on the production image; A preset warning device is controlled according to the falling position to issue a fall warning.

[0007] By adopting the above technical solution, the center of gravity of the person is determined from the picture, and when the center of gravity of the person is shifted, the management personnel is notified through the warning device that a person has fallen, thereby facilitating the management personnel to deal with the fallen person and the cause of the fall in a timely manner, and reducing the situation where the production behavior detection technology detects that the person has performed dangerous behavior.

[0008] Optionally, also include: When the center of gravity of the person exceeds the stable range, the swing amplitude of the person's limbs is determined according to the person's posture; Determine the degree of deviation of the center of gravity of the person according to the swing amplitude; When the degree of deviation exceeds the preset fall threshold, the personnel number is determined based on the production image; Control the preset detection device to obtain health parameters according to the personnel number; Determine whether the physiological state of a person is abnormal based on health parameters; When a person's physiological state is abnormal, the preset warning device is controlled according to the person's number to issue a health warning.

[0009] By adopting the above technical solution, when a person falls, the maximum deviation of the person's center of gravity can be determined by the swing amplitude of the person's limbs, so that the person's pulse, body temperature and other vital signs can be read when the deviation is large, so as to determine whether the person has physiological problems, and notify the management personnel through the warning device when the person has physiological problems.

[0010] Optionally, an obstacle handling method is also included, and the obstacle handling method includes: When the person's physiological state is normal, determine the inspection path based on the fall location; Control the preset drone to fly to the fall location according to the inspection path and obtain the fall image; Judging whether there is an obstacle at the falling position according to the falling image; When there is an obstacle at the falling position, the obstacle height is determined based on the obstacle; When the obstacle height is lower than the preset attention threshold, the preset warning device is controlled according to the fall position to issue an obstacle warning.

[0011] By adopting the above technical solution, when a person falls, the drone can further explore the environment where the person falls, so as to determine whether there is an obstacle where the person falls. If there is an obstacle where the person falls, it can be determined that the person fell due to tripping over the obstacle, so as to promptly notify the management personnel of the existence of obstacles in the factory through the warning device, thereby facilitating the management personnel to deal with the obstacles in time.

[0012] Optionally, the obstacle handling method further includes: When the obstacle height is lower than the preset attention threshold, it is determined whether there is a wall around the obstacle based on the fall image; When there is a wall around the obstacle, the prompt position is determined based on the fall image and the preset attention threshold; When there is no wall around the obstacle, the prompt position is determined based on the obstacle; Based on the situation of determining the prompt location, determining the installation path according to the prompt location; According to the installation path, the preset drone is controlled to adhere the preset prompt device to the prompt position, and the preset prompt device is controlled to emit a light prompt.

[0013] By adopting the above technical solution, when the height of the obstacle is low, it is easy for people to find it difficult to notice the obstacle, which may cause people to be tripped by the obstacle. The prompt device can emit a light prompt at the lower obstacle to remind people of the existence of the obstacle, thereby reducing the situation where people are tripped by the obstacle.

[0014] Optionally, it also includes a method for treating accumulated water, and the method for treating accumulated water includes: When there is no obstacle at the falling position, determine whether there is water accumulation at the falling position based on the falling image; When there is water accumulation at the fall location, the center position of the water accumulation is determined based on the fall image; Determine the suction path based on the center position; The preset drone is controlled to fly to the center position according to the absorption path, and the water accumulation area is determined based on the fall image; Determine the volume of accumulated water based on the accumulated water area; Determine the absorption time according to the volume of accumulated water; The suction device preset on the drone is controlled to suck the accumulated water according to the suction time.

[0015] By adopting the above technical solution, when a person falls, the drone can be used to further explore the environment where the person falls, so as to determine whether there is water accumulation where the person falls, and when there is water accumulation where the person falls, it can be determined that the person fell due to slipping due to the accumulated water, so that the accumulated water can be absorbed in time through the suction device, thereby reducing the number of people falling.

[0016] Optionally, the accumulated water treatment method further comprises: Determine the central thickness of the water accumulation according to the water accumulation area; Determine the degree of slippage of the ponding water based on the center thickness; When the slip degree is lower than the offset degree, the presence of impurities in the accumulated water is determined based on the fall image; When impurities are present in the accumulated water, the preset drone is controlled to add a preset filter to the preset suction device, and the suction device preset on the drone is controlled to suck the accumulated water according to the suction time; Determine the cleaning path based on the center location; The preset dust suction device is controlled according to the cleaning path to absorb impurities.

[0017] By adopting the above technical solution, when there are impurities in the stagnant water, it is easy to cause the friction between the stagnant water and the ground to increase. By comparing the center of gravity offset of the personnel, it is judged whether there are impurities in the stagnant water, and the impurities are inspected for the second time through images. Therefore, when there are impurities in the stagnant water, a filter is installed on the suction device to reduce the situation where impurities enter the suction device and cause damage to the suction device, and the impurities are absorbed by the dust suction device.

[0018] Optionally, a method for treating coma is also included, and the method for treating coma includes: Determine whether a person is drowsy based on their posture; When the person is asleep, obtain the carbon dioxide concentration; When the carbon dioxide concentration is higher than the preset hypoxia threshold, the difference between the carbon dioxide concentration and the preset hypoxia threshold is calculated and defined as the concentration difference; Determine the ventilation growth rate based on the concentration difference; The preset ventilation device is controlled to increase the ventilation speed according to the ventilation growth rate.

[0019] By adopting the above technical solution, when the ventilation speed in the factory is slow, it is easy to cause the oxygen content in the factory to decrease, which in turn causes people to fall asleep. Whether the people are drowsy can be judged through images, so that the ventilation speed of the ventilation device can be adjusted according to the oxygen content in the factory when the people are drowsy.

[0020] Optionally, the method for treating drowsiness further comprises: When personnel are asleep, determine their location based on production images; Determine the fan number based on the personnel position; Determine the fan location based on the fan number; Determine the blowing angle according to the fan position and personnel position; According to the fan number, the preset blowing device is controlled to blow air toward the personnel according to the blowing angle.

[0021] By adopting the above technical solution, when a person is drowsy, air is blown to the person through the blowing device to increase the air flow speed at the person's location, thereby increasing the oxygen content at the person's location, and the airflow prompts the person to concentrate, thereby reducing the drowsiness of the person.

[0022] In the second aspect, the present application provides a production behavior detection system, which adopts the following technical solution: A production behavior detection system, comprising: an acquisition module for acquiring production images, health parameters, fall images, and carbon dioxide concentrations; A memory, used to store a program of any of the above production behavior detection methods; The program in the memory can be loaded and executed by the processor.

[0023] In a third aspect, the present application provides a production behavior detection device, which adopts the following technical solution: A production behavior detection device comprises a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute any of the above-mentioned production behavior detection methods.

[0024] By adopting the above technical solution, the center of gravity of the person is determined from the picture, and when the center of gravity of the person is shifted, the management personnel is notified through the warning device that a person has fallen, thereby facilitating the management personnel to deal with the fallen person and the cause of the fall in a timely manner, and reducing the situation where the production behavior detection technology detects that the person has performed dangerous behavior.

[0025] In summary, the present application includes at least one of the following beneficial technical effects: 1. Determine the center of gravity of the personnel from the picture, and notify the management personnel through the warning device when the center of gravity of the personnel is offset that there is a fall of the personnel, so that the management personnel can deal with the fallen personnel and the reasons for the fall in time, and reduce the situation where the production behavior detection technology detects that the personnel are performing dangerous behaviors; 2. When a person falls, the maximum deviation of the person's center of gravity is determined by the swing amplitude of the person's limbs. When the deviation is large, the person's pulse, body temperature and other vital signs are read to determine whether the person has physiological problems. If a person has physiological problems, the management personnel are notified through the warning device; 3. When a person falls, the drone is used to further explore the environment where the person falls, so as to determine whether there is an obstacle where the person falls. If there is an obstacle where the person falls, it is determined that the person fell due to tripping over the obstacle, so that the management personnel can be notified of the existence of obstacles in the factory through the warning device in time, so as to facilitate the management personnel to deal with the obstacles in time. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 Is a process for producing behavioral detection methods Figure 1 ; Figure 2 Is a process for producing behavioral detection methods Figure 2 ; Figure 3 The process of obstacle handling method Figure 1 ; Figure 4 The process of obstacle handling method Figure 2 ; Figure 5 The process of treating waterlogging Figure 1 ; Figure 6 The process of treating waterlogging Figure 2 ; Figure 7 This is the process of dealing with coma Figure 1 ; Figure 8 This is the process of dealing with coma Figure 2 . DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and 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.

[0028] The embodiment of the present invention discloses a production behavior detection method, which is used to determine whether a person's body shape changes due to environmental factors in the factory based on a person's image, thereby reducing the situation where the production behavior detection technology determines that the person is in a dangerous situation.

[0029] Reference Figure 1 , a production behavior detection method, comprising: Step 100: Acquire a production image.

[0030] Production images refer to pictures of factories that are monitored using production behavior detection technology. Production images can be obtained through cameras. The method for obtaining production images is selected by the staff based on actual conditions and will not be elaborated here.

[0031] Step 101: Determine whether there is a person in the production image.

[0032] Image recognition technology can be used to determine whether there are people in the production image. The method of determining whether there are people is common knowledge among people in this field and will not be elaborated here.

[0033] Step 102: When there is a person in the image, determine the person's posture according to the production image.

[0034] Personnel posture refers to the movement of personnel in the production image, and the posture of personnel can be identified by human posture recognition technology. The method of identifying the posture of personnel is common knowledge in this field and will not be elaborated here.

[0035] Step 103: Determine the center of gravity of the person according to the person's posture.

[0036] The center of gravity of a person refers to the position of the center of gravity of the person, and the center of gravity of the person can be extracted from the posture of the person. The method of extracting the center of gravity of the person is common knowledge to people in this field and will not be elaborated here.

[0037] Step 104: Determine a stable interval according to the posture of the person.

[0038] The stable interval refers to the range of the center of gravity position in which a person can stand steadily in the current posture. Generally, the position of the person's feet is extracted from the person's posture, and the person's feet and the area between them are taken as the stable interval. The method of extracting the stable interval is common knowledge among people in this field and will not be elaborated here.

[0039] Step 105: When the center of gravity of the person exceeds the stable range, the fall position is determined according to the production image.

[0040] If the center of gravity of a person exceeds the stable range, it means that the person cannot stand steadily in the current posture, that is, the person is at risk of falling. The fall position refers to the position where the person is at risk of falling. The fall position can be identified from the production image through image recognition technology. The method for determining the fall position is common knowledge among people in this field and will not be elaborated here.

[0041] Step 106: Control a preset warning device to issue a fall warning according to the fall position.

[0042] The warning device is a device used to notify factory managers of accidents. The warning device is selected by the staff according to the actual situation and will not be described in detail here. The center of gravity of the personnel is determined from the picture, and when the center of gravity of the personnel is offset, the warning device is used to notify the management personnel that a person has fallen, so that the management personnel can deal with the fallen personnel and the reasons for the fall in a timely manner, reducing the number of situations where the production behavior detection technology detects that the personnel have engaged in dangerous behavior.

[0043] Reference Figure 2 , a production behavior detection method, further comprising: Step 107: When the center of gravity of the person exceeds the stable range, the swing amplitude of the person's limbs is determined according to the person's posture.

[0044] The swing amplitude refers to the amplitude of the swing of the limbs of a person to restore the center of gravity. The swing amplitude can be extracted from the posture of the person. The method of extracting the swing amplitude is common knowledge among people in this field and will not be elaborated here.

[0045] Step 108: Determine the degree of deviation of the center of gravity of the person according to the swing amplitude.

[0046] The degree of deviation refers to an artificially set numerical value used to show the degree of deviation of a person's center of gravity. The higher the degree of deviation, the more likely the person is to fall. The degree of deviation can be obtained from a deviation degree table. The deviation degree table refers to a relationship table that records the swing amplitude obtained through experiments and its corresponding degree of deviation in advance.

[0047] Step 109: When the deviation degree exceeds the preset fall threshold, the personnel number is determined according to the production image.

[0048] The fall threshold refers to the minimum value of the deviation degree at which a person is prone to fall. The fall threshold is selected by the staff according to the actual situation and will not be described in detail here. The deviation degree exceeding the fall threshold means that the person is prone to fall at this time. The personnel number refers to the number set in advance to distinguish personnel. Each personnel number corresponds to a person one by one. The personnel number is generally printed on the work clothes of the factory. The personnel number can be obtained by identifying the production image through image recognition technology. The identification method of the personnel number is common knowledge known to people in this field and will not be described in detail here.

[0049] Step 110: Control the preset detection device to obtain health parameters according to the personnel number.

[0050] The detection device refers to the equipment worn by each person in the factory to obtain vital signs such as pulse and body temperature. The detection device is selected by the staff according to the actual situation and will not be described here. Health parameters are vital signs such as pulse and body temperature obtained by the detection device.

[0051] Step 111: Determine whether the physiological state of the person is abnormal based on the health parameters.

[0052] The pulse, body temperature and other vital signs information of the person in the health parameters can be compared with the pre-set healthy range of the pulse, body temperature and other vital signs information, and when the vital signs information in the health parameters does not fall within the healthy range, it is judged that the physiological state of the person is abnormal.

[0053] Step 112: When the physiological state of a person is abnormal, a preset warning device is controlled according to the person number to issue a health warning.

[0054] When a person has a disease or health problem, it is easy for him to fall. The physiological state of the person is judged by monitoring the person's health parameters, and when the physiological state of the person is poor, the alarm device is used to notify the management personnel to handle it.

[0055] Reference Figure 3 , obstacle handling methods include: Step 200: When the physiological state of the person is not abnormal, determine the inspection path according to the fall location.

[0056] The drone refers to a device used to further explore the factory environment. The drone is selected by the staff according to the actual situation and will not be described in detail here. The inspection path refers to the route that the drone flies to the fall location. The inspection path can be automatically generated according to the fall location through the control program of the drone. The method for determining the inspection path is common knowledge among people in this field and will not be described in detail here.

[0057] Step 201: Control a preset drone to fly to the fall location according to the inspection path, and obtain a fall image.

[0058] The fall image is a picture of the ground obtained by the drone flying to the fall location. The fall image can be obtained through the camera on the drone. The method for obtaining the fall image is selected by the staff according to the actual situation and will not be elaborated here.

[0059] Step 202: Determine whether there is an obstacle at the fall location based on the fall image.

[0060] Obstacles refer to objects on the ground such as piled raw materials, waste generated during the production process, garbage, etc. Image recognition technology can be used to determine whether there are obstacles on the ground at the fall location. The method of determining obstacles is common knowledge among people in this field and will not be elaborated here.

[0061] Step 203: When there is an obstacle at the falling position, determine the obstacle height based on the obstacle.

[0062] The presence of an obstacle at the falling position indicates that the obstacle is the cause of the person's fall. The obstacle height refers to the height of the highest point of the obstacle from the ground. The obstacle height can be obtained through image recognition technology. The method for identifying the obstacle height is common knowledge among people in this field and will not be elaborated here.

[0063] Step 204: When the obstacle height is lower than a preset attention threshold, a preset warning device is controlled to issue an obstacle warning according to the fall position.

[0064] The attention threshold refers to the lowest height that people can easily notice. The attention threshold is selected by the staff according to the actual situation and will not be elaborated here. If the obstacle height is lower than the attention threshold, it means that the obstacle is too low, which may cause people to fail to notice the obstacle and trip over it. At this time, the warning device will notify the management personnel of the existence of the obstacle so that the obstacle can be handled in time to reduce the number of people falling.

[0065] Reference Figure 4 , the obstacle handling method also includes: Step 205: When the obstacle height is lower than the preset attention threshold, determine whether there is a wall around the obstacle based on the fall image.

[0066] Image recognition technology can be used to determine whether there is a wall within a range with a preset detection distance as a radius centered on the obstacle, where the detection distance is selected by the staff based on actual conditions. A wall refers to an object that is higher than the height of the obstacle. The method for determining a wall is common knowledge among people in this field and will not be elaborated here.

[0067] Step 206: When there is a wall around the obstacle, determine the prompt position according to the fall image and a preset attention threshold.

[0068] The prompt device refers to a device used to remind people of the existence of obstacles. The prompt device generally uses a light-emitting diode. The bottom of the prompt device is generally provided with rubber clay for fixing. The prompt device is fixed by pressing the prompt device through the rubber clay. The prompt device is selected by the staff according to the actual situation and will not be described here. The prompt position refers to the position for installing the prompt device. The existence of a wall around the obstacle means that there is an object higher than the obstacle. At this time, the height of the wall at the attention threshold is taken as the prompt position. When the height of the wall is lower than the attention threshold, the highest point on the wall is taken as the prompt position. When there are multiple prompt positions, the point closest to the obstacle is selected as the prompt position.

[0069] Step 207: When there is no wall around the obstacle, determine the prompt position based on the obstacle.

[0070] If there is no wall around the obstacle, the highest point on the obstacle is used as the prompt position. When there are multiple prompt positions, the midpoint of the prompt positions is used as the prompt position.

[0071] Step 208: Based on the determination of the prompt location, determine the installation path according to the prompt location.

[0072] The installation path refers to the path for installing the prompt device to the prompt position through the drone. The installation path can be automatically generated according to the prompt position through the control program of the drone. The method for determining the installation path is common knowledge among people in this field and will not be elaborated here.

[0073] Step 209: Control the preset drone according to the installation path to attach the preset prompt device to the prompt position, and control the preset prompt device to emit a light prompt.

[0074] When the height of the obstacle is relatively low, the prompt device is pressed to the prompt position by the drone so that the prompt device is adhered to the prompt position through the rubber mud at the bottom of the prompt device, so that the prompt device emits a light prompt at the lower obstacle to remind people of the existence of the obstacle, thereby reducing the situation where people are tripped by the obstacle.

[0075] Reference Figure 5 , water treatment methods include: Step 300: When there is no obstacle at the falling position, determine whether there is water accumulation at the falling position according to the falling image.

[0076] Puddles refer to liquid on the ground. Image recognition technology can be used to determine whether there is puddles on the ground at the fall location. The method for determining puddles is common knowledge among people in this field and will not be elaborated here.

[0077] Step 301: When there is water accumulation at the fall location, determine the center position of the water accumulation according to the fall image.

[0078] The center position refers to the center point of the accumulated water. The center position can be obtained by image recognition technology. The method for identifying the center position is common knowledge among people in this field and will not be elaborated here.

[0079] Step 302: Determine the suction path according to the center position.

[0080] The absorption path refers to the path that the UAV flies to the center position. The absorption path can be automatically generated according to the center position through the control program of the UAV. The method for determining the absorption path is common knowledge among people in this field and will not be elaborated here.

[0081] Step 303: Control the preset drone to fly to the center position according to the absorption path, and determine the water accumulation area according to the fall image.

[0082] The waterlogging area refers to the area value of the waterlogging. The waterlogging area can be obtained by using image recognition technology. The method for identifying the waterlogging area is common knowledge among people in this field and will not be elaborated here.

[0083] Step 304: Determine the volume of accumulated water according to the accumulated water area.

[0084] The accumulated water volume refers to the total volume of the liquid on the ground. The thickness of the water layer on flat ground is related to the hydrophilicity of the ground. When the ground material is the same, the average thickness of the water layer is estimated based on the hydrophilicity of the ground, and then the product of the accumulated water area and the average thickness is calculated as the accumulated water volume. The method for estimating the average thickness is common knowledge among people in this field and will not be elaborated here.

[0085] Step 305: Determine the absorption time according to the volume of accumulated water.

[0086] The suction device refers to a device installed on the drone for absorbing accumulated water. The suction device is selected by the staff according to the actual situation and will not be described in detail here. The suction time refers to the time required for the suction device to suck up the accumulated water. The suction time can be obtained by querying the suction schedule. The suction schedule refers to a relationship table that records different volumes of accumulated water and their corresponding suction times.

[0087] Step 306: Control the suction device preset on the drone to suck the accumulated water according to the suction time.

[0088] When a person falls, the drone can be used to further explore the environment where the person falls, so as to determine whether there is water accumulation where the person falls. If there is water accumulation where the person falls, it can be determined that the person fell due to slipping due to the accumulated water, so that the accumulated water can be absorbed in time through the suction device, thereby reducing the number of people falling.

[0089] Reference Figure 6 , the water treatment methods also include: Step 307: Determine the central thickness of the water accumulation according to the water accumulation area.

[0090] The central thickness refers to the thickness of the thickest part of the water layer in the accumulated water. When the water layer is on flat ground, the surface of the water layer is generally arc-shaped, and the curvature of the arc is related to the hydrophilicity of the ground. The central thickness of the accumulated water is estimated based on the accumulated water area and the hydrophilicity of the ground. The estimation method of the central thickness is common knowledge among people in this field and will not be elaborated here.

[0091] Step 308: Determine the degree of slippage of the accumulated water according to the center thickness.

[0092] The degree of slip refers to the maximum degree of displacement that can be caused by accumulated water. Accumulated water can easily reduce the friction between people and the ground. The thicker the accumulated water, the greater the reduction in friction. The greater the reduction in friction, the greater the degree of displacement of the center of gravity of the people. That is, the greater the center thickness, the higher the degree of slip. The degree of slip can be obtained by querying the slip degree table. The slip degree table refers to a relationship table that records different center thicknesses obtained in advance through experiments and their corresponding slip degrees.

[0093] Step 309: When the slip degree is lower than the offset degree, determine whether there are impurities in the accumulated water based on the fall image.

[0094] Impurities refer to dust, sand, algae and other substances that can easily further reduce the friction between people and the ground. The degree of slippage is lower than the degree of displacement, which means that the degree of displacement of the center of gravity of the people is higher than that caused by accumulated water, that is, there may be impurities in the accumulated water. Image recognition technology can be used to determine whether there are impurities in the accumulated water. The method of determining impurities is common knowledge among people in this field and will not be elaborated here.

[0095] Step 310: When impurities exist in the accumulated water, the preset drone is controlled to add a preset filter to the preset suction device, and the suction device preset on the drone is controlled to suck the accumulated water according to the suction time.

[0096] The filter tip refers to a device that is matched with the suction port of the suction device and is used to reduce the amount of impurities that enter the suction device. The filter tip is selected by the staff according to the actual situation and will not be described in detail here. When there are impurities in the stagnant water, the filter tip is used to reduce the amount of impurities that enter the suction device along with the stagnant water when the suction device absorbs the stagnant water, causing damage to the suction device.

[0097] Step 311: Determine a cleaning path according to the center position.

[0098] The dust suction device refers to a device used to absorb impurities. The dust suction device is selected by the staff according to the actual situation and will not be described in detail here. The cleaning path refers to the route for the dust suction device to absorb impurities. The cleaning path can be automatically generated according to the center position by the control program of the dust suction device. The method for determining the cleaning path is common knowledge among people in the field and will not be described in detail here.

[0099] Step 312: Control a preset dust collecting device to absorb impurities according to the cleaning path.

[0100] When there are impurities in the stagnant water, it is easy to increase the friction between the stagnant water and the ground. By comparing the center of gravity offset of the personnel, it is judged whether there are impurities in the stagnant water, and the impurities are inspected again through images. When there are impurities in the stagnant water, a filter is installed on the suction device to reduce the situation where impurities enter the suction device and cause damage to the suction device, and the impurities are absorbed by the dust suction device.

[0101] Reference Figure 7 , treatments for coma include: Step 400: Determine whether the person is drowsy based on the person's posture.

[0102] Whether a person is drowsy can be determined by human posture recognition technology. The method of identifying whether a person is drowsy is common knowledge among people in the field and will not be elaborated here.

[0103] Step 401: When the person is asleep, obtain the carbon dioxide concentration.

[0104] The carbon dioxide concentration refers to the proportion of carbon dioxide in the air in the factory. The carbon dioxide concentration can be obtained through a carbon dioxide sensor. The method for obtaining the carbon dioxide concentration is common knowledge among people in this field and will not be elaborated here.

[0105] Step 402: When the carbon dioxide concentration is higher than the preset hypoxia threshold, the difference between the carbon dioxide concentration and the preset hypoxia threshold is calculated and defined as the concentration difference.

[0106] The hypoxia threshold refers to the highest carbon dioxide concentration at which the human body can function normally. The oxygen content in the air is negatively correlated with the carbon dioxide content. The higher the carbon dioxide content, the lower the oxygen content. The hypoxia threshold is the carbon dioxide concentration corresponding to the lowest oxygen content at which the human body can function normally. The hypoxia threshold is selected by the staff based on actual conditions and will not be elaborated here.

[0107] A carbon dioxide concentration higher than the hypoxia threshold means that the oxygen content in the air is low, which can easily cause people in this environment to become drowsy. The concentration difference is the difference between the carbon dioxide concentration and the hypoxia threshold, and the concentration difference is used to show the degree of hypoxia in the air.

[0108] Step 403: Determine the ventilation growth rate according to the concentration difference.

[0109] The ventilation device refers to the equipment used for air exchange inside and outside the factory. The ventilation device generally uses a fan. The ventilation device is selected by the staff according to the actual situation, so it will not be described here. The ventilation speed increase is to speed up the air exchange speed inside and outside the factory through the ventilation device so that the oxygen content of the air in the factory can be restored to normal. The ventilation speed increase can be obtained by querying the ventilation speed increase table. The ventilation speed increase table refers to the relationship table that records different concentration differences and their corresponding ventilation speed increases.

[0110] Step 404: Control the preset ventilation device to increase the ventilation speed according to the ventilation speed increase rate.

[0111] When the ventilation speed in a factory is slow, the oxygen content in the factory is likely to decrease, which in turn causes people to fall asleep. The system uses images to determine whether a person is drowsy, and adjusts the ventilation speed of the ventilation device according to the oxygen content in the factory when the person is drowsy.

[0112] Reference Figure 8 , treatment methods for coma also include: Step 405: When the personnel is asleep, determine the personnel's position based on the production image.

[0113] The personnel position refers to the position of the sleeping person, and the personnel position can be obtained through image recognition technology. The method for identifying the personnel position is common knowledge among people in this field and will not be elaborated here.

[0114] Step 406: Determine the fan number according to the personnel position.

[0115] The blowing device refers to a device installed in the factory to speed up the air circulation speed in the factory. The blowing device is selected by the staff according to the actual situation, so it will not be described here. The fan number refers to the number set manually to distinguish the blowing device. Each blowing device corresponds to the fan number one by one. The fan number can be obtained by querying the fan installation table. The fan installation table records the relationship table of all the blowing device numbers and their corresponding installation positions. From the fan installation table, find the fan number of the blowing device whose installation position is closest to the personnel position.

[0116] Step 407: Determine the fan position according to the fan number.

[0117] The fan position is the installation position of the air blowing device corresponding to the fan number found in the fan installation table.

[0118] Step 408: Determine the blowing angle according to the fan position and the personnel position.

[0119] The blowing angle refers to the angle value required for the blowing device to blow air from the fan position to the personnel position. The method for determining the blowing angle is common knowledge among people in this field and will not be elaborated here.

[0120] Step 409: Control the preset blowing device according to the fan number to blow air toward the personnel at a blowing angle.

[0121] When a person is drowsy, air is blown toward the person through a blowing device to increase the air flow speed at the person's location, thereby increasing the oxygen content at the person's location, and the airflow prompts the person to concentrate, thereby reducing the person's drowsiness.

[0122] Based on the same inventive concept, an embodiment of the present invention provides a production behavior detection system, including: an acquisition module for acquiring production images, health parameters, fall images, and carbon dioxide concentrations; A memory, used to store a program of any of the above production behavior detection methods; The program in the memory can be loaded and executed by the processor.

[0123] Based on the same inventive concept, an embodiment of the present invention provides a production behavior detection device, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute any of the above-mentioned production behavior detection methods.

[0124] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0125] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A production behavior detection method, characterized in that: include: Get production images; Judging whether there is a person in the image based on the production image; When there is a person in the image, the person's posture is determined based on the production image; Determine the center of gravity of the person based on their posture; Determine the stability interval based on the personnel's posture; When the center of gravity of the person exceeds the stable range, the fall position is determined based on the production image; A preset warning device is controlled according to the falling position to issue a fall warning.

2. A production behavior detection method according to claim 1, characterized in that: Also includes: When the center of gravity of the person exceeds the stable range, the swing amplitude of the person's limbs is determined according to the person's posture; Determine the degree of deviation of the center of gravity of the person according to the swing amplitude; When the degree of deviation exceeds the preset fall threshold, the personnel number is determined based on the production image; Control the preset detection device to obtain health parameters according to the personnel number; Determine whether the physiological state of a person is abnormal based on health parameters; When a person's physiological state is abnormal, the preset warning device is controlled according to the person's number to issue a health warning.

3. A production behavior detection method according to claim 2, characterized in that: Also included is an obstacle handling method, the obstacle handling method comprising: When the person's physiological state is normal, determine the inspection path based on the fall location; Control the preset drone to fly to the fall location according to the inspection path and obtain the fall image; Judging whether there is an obstacle at the falling position according to the falling image; When there is an obstacle at the falling position, the obstacle height is determined based on the obstacle; When the obstacle height is lower than the preset attention threshold, the preset warning device is controlled according to the fall position to issue an obstacle warning.

4. A production behavior detection method according to claim 3, characterized in that: The obstacle handling method also includes: When the obstacle height is lower than the preset attention threshold, it is determined whether there is a wall around the obstacle based on the fall image; When there is a wall around the obstacle, the prompt position is determined based on the fall image and the preset attention threshold; When there is no wall around the obstacle, the prompt position is determined based on the obstacle; Based on the situation of determining the prompt location, determining the installation path according to the prompt location; According to the installation path, the preset drone is controlled to adhere the preset prompt device to the prompt position, and the preset prompt device is controlled to emit a light prompt.

5. A production behavior detection method according to claim 4, characterized in that: Also included is a method for treating accumulated water, the method comprising: When there is no obstacle at the falling position, determine whether there is water accumulation at the falling position based on the falling image; When there is water accumulation at the fall location, the center position of the water accumulation is determined based on the fall image; Determine the suction path based on the center position; The preset drone is controlled to fly to the center position according to the absorption path, and the water accumulation area is determined based on the fall image; Determine the volume of accumulated water based on the accumulated water area; Determine the absorption time according to the volume of accumulated water; The suction device preset on the drone is controlled to suck the accumulated water according to the suction time.

6. A production behavior detection method according to claim 5, characterized in that: The method for treating accumulated water also includes: Determine the central thickness of the water accumulation according to the water accumulation area; Determine the degree of slippage of the ponding water based on the center thickness; When the slip degree is lower than the offset degree, the presence of impurities in the accumulated water is determined based on the fall image; When impurities are present in the accumulated water, the preset drone is controlled to add a preset filter to the preset suction device, and the suction device preset on the drone is controlled to suck the accumulated water according to the suction time; Determine the cleaning path based on the center location; The preset dust suction device is controlled according to the cleaning path to absorb impurities.

7. A production behavior detection method according to claim 1, characterized in that: Also included is a method for treating lethargy, the method comprising: Determine whether a person is drowsy based on their posture; When the person is asleep, obtain the carbon dioxide concentration; When the carbon dioxide concentration is higher than the preset hypoxia threshold, the difference between the carbon dioxide concentration and the preset hypoxia threshold is calculated and defined as the concentration difference; Determine the ventilation growth rate based on the concentration difference; The preset ventilation device is controlled to increase the ventilation speed according to the ventilation growth rate.

8. A production behavior detection method according to claim 7, characterized in that: The method for treating lethargy also includes: When personnel are asleep, determine their location based on production images; Determine the fan number based on the personnel position; Determine the fan location based on the fan number; Determine the blowing angle according to the fan position and personnel position; According to the fan number, the preset blowing device is controlled to blow air toward the personnel according to the blowing angle.

9. A production behavior detection system, characterized in that: include: an acquisition module for acquiring production images, health parameters, fall images, and carbon dioxide concentrations; A memory for storing a program of a production behavior detection method according to any one of claims 1 to 8; The program in the memory can be loaded and executed by the processor.

10. A production behavior detection device, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute a production behavior detection method as claimed in any one of claims 1 to 8.

Images