Dust detection method, device, electronic device and storage medium

Abstract

Embodiments of the present application disclose a dust detection method, device, electronic device, and storage medium. This solution can obtain at least one image of a detection area, determine a target area in the image, calculate the image features of the target area, extract human key point information and vehicle key point information in the image, calculate human features and vehicle features respectively according to the human key point information and vehicle key point information, and determine whether dust is generated in the detection area according to the image features, human features, and vehicle features. Embodiments of the present application comprehensively judge whether a dust event occurs in the detection area by combining human features and vehicle features on the basis of image features, reduce the interference degree of other factors on the detection result, and are beneficial to improving the monitoring efficiency and accuracy of dust events.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular, to a dust detection method, device, electronic device, and storage medium. Background Art

[0002] In recent years, with the development and construction of society, a large amount of dust has been generated. Dust is an open pollution source that enters the atmosphere due to the dust on the ground being carried by wind, human activities, or other means. It is an important component of total suspended particulate matter in ambient air. Among them, construction sites, demolition sites, mining areas, and other engineering operation areas are high-incidence areas for generating dust and are also the main pollution sources of PM2.5. Therefore, in order to ensure the health of the human respiratory system, it is necessary to monitor and identify dust.

[0003] Currently, the common dust monitoring devices on the market mainly use monitoring principles such as the beta-ray method and the light scattering method. However, in the actual use process, the applicant found that when detecting dust through the existing technology, the detection accuracy is easily affected by environmental factors, resulting in a large error in the final detection result. Summary of the Invention

[0004] The embodiments of this application provide a dust detection method, device, electronic device, and storage medium, which can quickly and accurately identify whether a dust event occurs in the detection area, reduce the interference degree of other factors on the detection result, and is beneficial to improving the monitoring efficiency and accuracy of dust events.

[0005] The embodiments of this application provide a dust detection method, including:

[0006] Obtain at least one image of the detection area;

[0007] Determine a target area in the image and calculate the image features of the target area;

[0008] Extract the human key point information and vehicle key point information in the image, and calculate the human features and vehicle features respectively according to the human key point information and vehicle key point information;

[0009] Judge whether dust occurs in the detection area according to the image features, human features, and vehicle features.

[0010] In one embodiment, the determining a target area in the image and calculating the image features of the target area includes:

[0011] Determine the dust area in the image according to the color information and clarity information of the image;

[0012] Calculate the dust emission probability of the dust-emitting area and use the dust emission probability as the image feature of the target area.

[0013] In one embodiment, the determining whether dust is emitted in the detection area according to the image feature, human feature, and vehicle feature includes:

[0014] Adjust the probability threshold according to the human feature and vehicle feature;

[0015] Compare the dust emission probability with the adjusted probability threshold to determine whether dust is emitted in the detection area according to the comparison result.

[0016] In one embodiment, the adjusting the probability threshold according to the human feature and vehicle feature includes:

[0017] Obtain the weight values corresponding to the human feature and vehicle feature respectively;

[0018] Adjust the probability threshold according to the weight values corresponding to the human feature and vehicle feature respectively.

[0019] In one embodiment, the extracting the human key point information in the image includes:

[0020] Extract the human bone point features of the person in the image and determine the hand position and preset facial feature positions of the person according to the human bone point features;

[0021] The calculating the human feature according to the human key point information includes:

[0022] Calculate the overlapping ratio of the area corresponding to the hand position and the area corresponding to the preset facial feature positions as the human feature.

[0023] In one embodiment, the extracting the vehicle key point information in the image includes:

[0024] Identify the steering information of the vehicle in the image and determine the target vehicle area according to the steering information;

[0025] The calculating the vehicle feature according to the vehicle key point information includes:

[0026] Judge whether there is sediment dropping in the target vehicle area and use the judgment result as the vehicle feature.

[0027] In one embodiment, after determining the target area in the image, the calculating the image feature of the target area further includes:

[0028] Divide the target area into two sub-image areas, perform image enhancement processing on the two sub-image areas respectively, and calculate the image features of the target area according to the processed images.

[0029] An embodiment of the present application further provides a dust detection device, including:

[0030] An acquisition module, configured to acquire at least one image of a detection area;

[0031] A determination module, configured to determine a target area in the image and calculate the image features of the target area;

[0032] An extraction module, configured to extract human key point information and vehicle key point information in the image, and calculate human features and vehicle features respectively according to the human key point information and the vehicle key point information;

[0033] A judgment module, configured to judge whether dust is generated in the detection area according to the image features, human features, and vehicle features.

[0034] An embodiment of the present application further provides an electronic device, characterized in that the electronic device includes a memory and a processor, a computer program is stored in the memory, and the processor executes the steps in any one of the dust detection methods provided by the embodiments of the present application by calling the computer program stored in the memory.

[0035] An embodiment of the present application further provides a storage medium, characterized in that the storage medium stores a computer program, and the computer program is suitable for being loaded by a processor to execute the steps in any one of the dust detection methods provided by the embodiments of the present application.

[0036] The dust detection method provided by the embodiments of the present application can acquire at least one image of a detection area, determine a target area in the image, calculate the image features of the target area, extract human key point information and vehicle key point information in the image, and calculate human features and vehicle features respectively according to the human key point information and the vehicle key point information, and judge whether dust is generated in the detection area according to the image features, human features, and vehicle features. In the embodiments of the present application, combined with human features and vehicle features on the basis of image features, it is comprehensively judged whether a dust event occurs in the detection area, reducing the interference degree of other factors on the detection result, and being beneficial to improving the monitoring efficiency and accuracy of dust events. Description of the Drawings

[0037] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those skilled in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0038] Figure 1 is the first flow diagram of the dust detection method provided by the embodiments of the present application;

[0039] Figure 2 is the second flow diagram of the dust detection method provided by the embodiments of the present application;

[0040] Figure 3 is the first structural diagram of the dust detection device provided by the embodiments of the present application;

[0041] Figure 4 is the second structural diagram of the dust detection device provided by the embodiments of the present application;

[0042] Figure 5 is the structural diagram of the terminal provided by the embodiments of the present application. Detailed Embodiments

[0043] Here, the exemplary embodiments will be described in detail, and the examples are shown in the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0044] It should be noted that in this document, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including that element. In addition, components, features, and elements with the same name in different embodiments of the present application may have the same meaning or different meanings, and their specific meanings need to be determined based on their explanations in the specific embodiments or further combined with the context in the specific embodiments.

[0045] It should be understood that although the steps in the flowcharts in the embodiments of the present application are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order restriction, and they can be executed in other orders. Moreover, at least a part of the steps in the figure may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same moment, but can be executed at different moments, and their execution order is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or sub-steps or stages of other steps.

[0046] It should be noted that in this article, step codes such as 101 and 102 are used. The purpose is to more clearly and briefly express the corresponding content and do not constitute a substantial restriction on the order. Those skilled in the art may execute 102 first and then 101 during specific implementation, etc., but these should all be within the protection scope of the present application.

[0047] The mention of "embodiment" in this article means that the specific features, structures or characteristics described in combination with the embodiment may be included in at least one embodiment of the present application. The appearance of this phrase in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0048] The embodiments of the present application provide a dust detection method. The execution subject of this dust detection method can be the dust detection device provided by the embodiments of the present application, or a server, an electronic device, etc. that integrates this dust detection device, where the dust detection device can be implemented in a hardware or software manner.

[0049] As Figure 1 shown, Figure 1 is the first flow schematic diagram of the dust detection method provided by the embodiments of the present application. The specific process of this dust detection method can be as follows:

[0050] 101. Obtain at least one image of the detection area.

[0051] In the embodiments of the present application, the above-mentioned detection area can be an area where dust events often occur, such as a construction area, a mining area, and a desert area, etc. Specifically, at least one camera can be set in the detection area for real-time shooting of the image corresponding to this area. It should be noted that the above-mentioned at least one image can be obtained by a single camera shooting multiple times, or can be obtained by multiple cameras shooting simultaneously.

[0052] In one embodiment, the camera may be a 360° panoramic camera that can collect 360° surround data. The camera may include several ultra-wide-angle cameras that can be used to shoot the current detection area in real time.

[0053] In one embodiment, considering that if the current air quality is good, the probability of dust emission in the detection area is small, the camera does not need to shoot to avoid wasting equipment resources. Therefore, it is possible to first determine whether the current particle concentration (such as PM2.5) is higher than the preset value. If not, there is no need to shoot. If the current particle concentration is higher than the preset value, the step of obtaining at least one image of the detection area can be further performed. That is, the step of obtaining at least one image of the detection area may include: determining whether the current particle concentration is higher than the preset value. If so, continuously shooting the detection area to obtain at least one image of the detection area. Among them, the above-mentioned PM2.5 can be queried from the local meteorological website through the network, or a sensor can be integrated inside the camera used to shoot images for measurement.

[0054] There are many methods for measuring particle concentration. In one embodiment, the particle concentration detection method is divided into two categories: sampling method and non-sampling method. With the advancement of science and technology, especially the rapid development of optical technology and sensor industry, the advantages of using light scattering method for concentration measurement in non-sampling method are more prominent. In the non-sampling method, the concentration is monitored according to the principle of light scattering, that is, after a certain intensity of light is irradiated on the particles in the air, it will scatter scattered light to its surroundings. The intensity of the scattered light is in a certain proportional relationship with the particle concentration. The particle concentration is measured by collecting the scattered light and measuring the size of the electrical signal generated by the scattered light irradiating the photoelectric sensor.

[0055] 102. Determine a target area in the image, and calculate image features of the target area.

[0056] In one embodiment, the target area is a dust area in the image, so that when calculating the image features later, only the image features of the dust area need to be calculated, avoiding interference of the non-dust area on the calculation results, which can further improve the accuracy of dust detection. Specifically, the target area in the image can be determined based on the color information of the image, such as dividing the image into multiple sub-areas, and calculating the pixel density of a preset color in each of the multiple sub-areas. The preset color can be a common dust color, such as gray, brown, yellow, etc. The preset color can be represented by an RGB value. When the pixel density of the preset color is greater than the preset value, the sub-area can be determined as the target area where dust occurs.

[0057] In one embodiment, the target area in the image can also be determined according to the clarity information of the image. For example, the image can be divided into multiple sub-areas, and the clarity of each sub-area can be calculated respectively, so as to determine the sub-areas with clarity less than the preset value as the target area. Optionally, in other embodiments, the above-mentioned target area can also be determined according to the color information and clarity information of the image respectively. By comprehensively considering the color information and clarity information, the accuracy of determining the target area where dust is raised can be further improved.

[0058] In one embodiment, after determining the target area in the image, the corresponding image features can be calculated for this target area. For example, the image corresponding to the target area can be analyzed through a convolutional neural network, so as to obtain the image features of the image corresponding to the target area. Among them, the above-mentioned image features can include shape information, color information, brightness information, texture information, etc.

[0059] In one embodiment, before calculating the image features of the target area, image enhancement processing can also be performed on the image. For example, the pixel size of the image collected by the camera is 1920x1080. In this embodiment, the image can be segmented into small image blocks of 224x224 pixels. By performing image segmentation, the recognition efficiency can be improved. At the same time, by performing data enhancement processing such as scaling and rotating on the segmented image, the number of samples can be increased, thereby improving the accuracy of the output result of the above-mentioned convolutional neural network. For another example, the original image can also be cropped first, and then the global feature map and the local feature map can be obtained through multiple convolutional layers respectively. The above-mentioned global feature map and local feature map can be obtained through different convolutional kernels and fully connected layers, and then a new feature map, that is, an enhanced image, can be obtained through image fusion.

[0060] 103. Extract the human key point information and vehicle key point information in the image, and calculate the human features and vehicle features respectively according to the human key point information and vehicle key point information.

[0061] In one embodiment, in order to further improve the accuracy of dust-raising event detection, on the basis of judging whether dust is raised in the detection area according to the image features, the human features and vehicle features corresponding to the people and vehicles in the image can also be further referred to. For example, when it is detected that pedestrians have abnormal behaviors such as covering their faces or mouths and noses, the probability of a dust-raising event in the detection area can be increased. Correspondingly, when it is detected that dust is raised due to the falling of the sediment carried by dump trucks, trucks, etc., the probability of a dust-raising event in the detection area can be reduced.

[0062] In one embodiment, when a person is detected in an image, the hand position and the preset facial feature positions of the person can be further detected to determine whether the person has actions such as covering the face or covering the mouth and nose. Specifically, multiple human bone point features of the person can be determined from the above-mentioned image. For example, the human bone point features may include the nose, mouth, left and right eyes, left and right ears, left and right shoulders, left and right elbows, left and right wrists, left and right hips, left and right knees, and left and right ankles, a total of 18 bone point features. Based on the above human bone point features, the hand position and the nose and mouth positions of the person in the image can be determined, so as to further determine whether the person has actions such as covering the face or covering the mouth and nose. Specifically, the overlapping ratio of the area corresponding to the hand position and the area corresponding to the nose and mouth positions can be calculated. For example, when the overlapping ratio of the area corresponding to the hand position and the area corresponding to the nose and mouth positions exceeds 80%, it can be determined that the current person has the action of covering the mouth and nose. Therefore, in the embodiment of the present application, the steps of extracting the human key point information in the image may include: extracting the human bone point features of the person in the image, and determining the hand position and the preset facial feature positions of the person according to the human bone point features. The steps of calculating the human features according to the human key point information may include: calculating the overlapping ratio of the area corresponding to the hand position and the area corresponding to the preset facial feature positions as the human features.

[0063] In one embodiment, the posture of the person can also be determined from the above human bone point features. By analyzing the posture, the actions made by the person can be determined, and further determine whether the person makes actions such as covering the face or covering the mouth and nose. It should be noted that when there are multiple persons in the above image, the human key point information and human features of each person can be detected separately.

[0064] In one embodiment, when a vehicle is detected in an image, it can be further determined whether there is sediment falling from the vehicle. Considering that when the vehicle makes different turns, the areas where sediment is likely to fall are also different. Therefore, to further improve the detection efficiency, the embodiment of the present application can also first identify the steering information of the vehicle and determine the target vehicle area according to the steering information, so that it is only necessary to detect whether there is sediment falling in the target vehicle area. For example, different turns of the vehicle can include different angles such as straight ahead, right side, left side, straight back, 45 degrees to the left front, etc. For different angles, in this embodiment, different detection areas are defined to detect whether there is dust raising caused by sediment falling on the vehicle. For example, for the left and right angles, we can simulate the positions of the baffle and wheels in the vehicle for key monitoring. For the straight ahead and straight back directions, the detection area can be expanded horizontally to detect the sediment falling situation of the carriage. Therefore, in the embodiment of the present application, the step of extracting the vehicle key point information in the image can include: identifying the steering information of the vehicle in the image and determining the target vehicle area according to the steering information. The step of calculating the vehicle characteristics according to the vehicle key point information can include: judging whether there is sediment falling in the target vehicle area and using the judgment result as the vehicle characteristics.

[0065] In the actual application process, considering that small vehicles such as cars and tricycles do not have problems such as sediment falling, the embodiment of the present application can also judge whether the vehicle model in the image is a preset model before extracting the vehicle key point information in the image. If not, only the human key point information in the image needs to be extracted. If so, the step of extracting the vehicle key point information in the image can be continued. Among them, the above preset model can be large trucks, muck trucks, mixer trucks, etc. with a preset shape, and the present application will not elaborate further on this.

[0066] 104. Judge whether dust raising occurs in the detection area according to the image characteristics, human characteristics, and vehicle characteristics.

[0067] After obtaining the above image characteristics, human characteristics, and vehicle characteristics, it is possible to comprehensively judge whether a dust raising event occurs in the detection area. For example, when the image characteristics indicate that there is a high probability (such as higher than 60%) of a dust raising event in this area, and at the same time the human characteristics indicate that the people in the image make actions such as covering the face or covering the mouth and nose, and the vehicle characteristics indicate that there is no sediment falling, etc., it can be determined that a dust raising event has occurred in this detection area. Another example is that when the image characteristics indicate that there is a low probability (such as lower than 50%) of a dust raising event in this area, and at the same time the human characteristics indicate that the people in the image do not make actions such as covering the face or covering the mouth and nose, and the vehicle characteristics indicate that there is sediment falling from the truck, etc., it can be determined that a dust raising event has not occurred in this detection area.

[0068] As described above, the dust emission detection method provided by the embodiments of the present application can obtain at least one image of the detection area, determine the target area in the image, calculate the image features of the target area, extract the human key point information and vehicle key point information in the image, and calculate the human features and vehicle features respectively according to the human key point information and vehicle key point information. Whether dust emission occurs in the detection area is judged according to the image features, human features and vehicle features. The embodiments of the present application comprehensively judge whether a dust emission event occurs in the detection area by combining human features and vehicle features on the basis of image features, reduce the interference degree of other factors on the detection result, and are beneficial to improving the monitoring efficiency and accuracy of dust emission events.

[0069] The method described in the previous embodiments will be further described in detail below.

[0070] Please refer to Figure 2 , Figure 2 which is the second process schematic diagram of the dust emission detection method provided by the embodiments of the present application. The method includes:

[0071] 201. Obtain at least one image of the detection area.

[0072] In the embodiments of the present application, the above detection area can be an area where dust emission events often occur, such as a construction area, a mining area, and a desert area. Specifically, at least one camera can be set in the detection area to capture at least one image corresponding to the area in real time.

[0073] In one embodiment, after obtaining at least one image of the detection area, the image can also be preprocessed first. The preprocessing can include image cropping, image denoising, contrast enhancement, etc. Among them, image cropping can remove the useless areas in the captured image, such as large blank areas or sky areas in the image, to avoid interference and reduce the calculation amount. Image denoising can improve the accuracy of subsequent detection, and contrast enhancement can deepen the contour of the target vehicle to facilitate subsequent edge detection steps.

[0074] Taking the enhancement of image contrast as an example, in one embodiment, the contrast of an image can be enhanced by performing projective transformation on multiple images and then performing upsampling processing on the images after projective transformation by trilinear interpolation. In other embodiments, the contrast can also be enhanced by combining the top-hat and bottom-hat transformations of morphology and the method of gray-scale stretching. Among them, the top-hat transformation is to subtract the image after opening operation from the original image to extract the brighter gray-scale regions in the image, where the opening operation can compensate for the uneven background brightness; the bottom-hat transformation is to subtract the image after closing operation from the original image to extract the regions with lower gray-scale in the image. The original image is added with the image after top-hat transformation and then subtracted by the image after bottom-hat transformation, making the bright regions in the image brighter and the dark regions darker. In another embodiment, the image contrast can also be enhanced based on the conical non-uniform stretching algorithm of the gray-scale histogram. The basic idea of this method is to first interpolate the uniform gray-scale axis of the image histogram unevenly according to the gray-scale distribution, that is, more interpolation is performed in the regions with high gray-scale distribution and less interpolation is performed in the regions with low gray-scale distribution, and then the interpolated gray-scale axis is uniformized according to the interpolation points, so as to achieve non-uniform stretching of the histogram.

[0075] 202. Determine the dust-raising area in the image according to the color information and clarity information of the image.

[0076] In one embodiment, the above color information can be represented by the pixel point density of a preset color, and the clarity information can be represented by transparency. For example, the image can be divided into multiple sub-regions, and the pixel point density of the preset color and the transparency are calculated in the multiple sub-regions respectively. The preset color can be the color of common dust, such as gray, brown, yellow, etc. The above preset color can be represented by RGB values. When the pixel point density of the preset color is greater than the preset value and the clarity is less than the preset value, it can be determined that the sub-region is the dust-raising area in the image.

[0077] 203. Calculate the dust-raising probability of the dust-raising area and use the dust-raising probability as the image feature of the target area.

[0078] In one embodiment, after determining the dust-raising area, the dust-raising probability can be further calculated according to the pixel point density of the preset color and the transparency of the dust-raising area. Optionally, in other embodiments, the image corresponding to the above dust-raising area can also be input into a BP neural network model to predict the dust-raising probability. For example, the RGB image corresponding to the dust-raising area can be first converted in color mode to the LAB color space to obtain the converted image, and then the converted image is input into the BP neural network model to obtain the predicted value of the occurrence of the dust-raising event.

[0079] In one embodiment, considering that the occurrence time of dust problems may be short, and in addition, the areas of dust are not uniform and do not have a fixed form, it is relatively difficult to collect samples for dust detection. To solve this problem, the embodiments of the present application can also adopt a data augmentation method. Specifically, a point in the dust area can be cropped into two parts, A and B, and different data augmentation methods are performed on the two parts respectively. For example, light changes are made to area A, and stretching changes are made to area B, etc. That is, the step of calculating the image features of the target area may further include: dividing the target area into two sub-image areas, performing image enhancement processing on the two sub-image areas respectively, and calculating the image features of the target area according to the processed images.

[0080] 204. Extract the human key point information and vehicle key point information in the image, and calculate the human features and vehicle features respectively according to the human key point information and vehicle key point information.

[0081] In one embodiment, when a person is detected in the image, the hand position and preset facial feature positions of the person can be further detected to determine whether the person has behaviors such as covering the face or covering the mouth and nose. Specifically, multiple human bone point features of the person can be determined through the above image. Through the above human bone point features, the hand position and the nose and mouth positions of the person in the image can be determined, so as to further determine whether the person has behaviors such as covering the face or covering the mouth and nose.

[0082] Correspondingly, when a vehicle is detected in the image, it can be further determined whether there is sediment falling from the vehicle. Considering that when the vehicle makes different turns, the areas where sediment is likely to fall are also different. Therefore, to further improve the detection efficiency, the embodiments of the present application can also first identify the steering information of the vehicle, and determine the target vehicle area according to the steering information, so that it is only necessary to detect whether there is sediment falling in the target vehicle area.

[0083] 205. Adjust the probability threshold according to the human features and vehicle features.

[0084] For example, when the above human features represent that the person in the image makes actions such as covering the face or covering the mouth and nose, the initial probability threshold can be reduced. Correspondingly, when the vehicle features represent situations such as sediment falling, the initial probability threshold can be increased.

[0085] In one embodiment, when adjusting the probability threshold according to the human features and vehicle features, different weight ratios can also be set for the human features and vehicle features. That is, the step of adjusting the probability threshold according to the human features and vehicle features may include: obtaining the weight values corresponding to the human features and vehicle features respectively, and adjusting the probability threshold according to the weight values corresponding to the human features and vehicle features.

[0086] 206. Compare the dust emission probability with the adjusted probability threshold to determine whether dust emission occurs in the detection area according to the comparison result.

[0087] For example, when the dust emission probability of the detection area characterized by the image features is 60%, if the initial probability threshold of 50% is adjusted to 70% according to the human body features and vehicle features, at this time the dust emission probability is less than the adjusted probability threshold, so it can be determined that no dust emission event occurs in the detection area. If the initial probability threshold of 50% is adjusted to 55% according to the human body features and vehicle features, at this time the dust emission probability is greater than the adjusted probability threshold, so it can be determined that a dust emission event occurs in the detection area.

[0088] As described above, the dust emission detection method proposed in the embodiment of the present application can obtain at least one image of the detection area, determine the dust emission area in the image according to the color information and clarity information of the image, calculate the dust emission probability of the dust emission area, and use the dust emission probability as the image feature of the target area, extract the human key point information and vehicle key point information in the image, and calculate the human body features and vehicle features respectively according to the human key point information and vehicle key point information, adjust the probability threshold according to the human body features and vehicle features, and compare the dust emission probability with the adjusted probability threshold to determine whether dust emission occurs in the detection area according to the comparison result. The embodiment of the present application combines the human body features and vehicle features on the basis of the image features to comprehensively judge whether a dust emission event occurs in the detection area, reduces the interference degree of other factors on the detection result, and is beneficial to improving the monitoring efficiency and accuracy of the dust emission event.

[0089] To implement the above method, the embodiment of the present application also provides a dust emission detection device, which can be specifically integrated in terminal devices such as mobile phones and tablet computers.

[0090] For example, as Figure 3 shown, it is the first structural schematic diagram of the dust emission detection device provided by the embodiment of the present application. The dust emission detection device may include:

[0091] An acquisition module 301, configured to acquire at least one image of the detection area;

[0092] A determination module 302, configured to determine a target area in the image and calculate the image feature of the target area;

[0093] An extraction module 303, configured to extract the human key point information and vehicle key point information in the image, and calculate the human body features and vehicle features respectively according to the human key point information and vehicle key point information;

[0094] A judgment module 304, configured to judge whether dust emission occurs in the detection area according to the image feature, human body feature, and vehicle feature.

[0095] In one embodiment, please further refer to Figure 4 , wherein the determination module 302 may specifically include:

[0096] A determination sub-module 3021, configured to determine a dust-raising area in the image according to the color information and clarity information of the image;

[0097] A calculation sub-module 3022, configured to calculate the dust-raising probability of the dust-raising area and use the dust-raising probability as the image feature of the target area.

[0098] In one embodiment, the judgment module 304 may specifically include:

[0099] An adjustment sub-module 3041, configured to adjust a probability threshold according to the human body feature and the vehicle feature;

[0100] A judgment sub-module 3042, configured to compare the dust-raising probability with the adjusted probability threshold to judge whether dust-raising occurs in the detection area according to the comparison result.

[0101] In one embodiment, the adjustment sub-module 3041 is specifically configured to obtain weight values corresponding to the human body feature and the vehicle feature respectively, and adjust the probability threshold according to the weight values corresponding to the human body feature and the vehicle feature respectively.

[0102] All the above technical solutions can be combined arbitrarily to form alternative embodiments of the present application, which will not be elaborated herein one by one.

[0103] As can be seen from the above, the dust-raising detection device proposed in the embodiment of the present application can obtain at least one image of a detection area, determine a target area in the image, calculate the image feature of the target area, extract human key point information and vehicle key point information in the image, and calculate the human body feature and the vehicle feature respectively according to the human key point information and the vehicle key point information, and judge whether dust-raising occurs in the detection area according to the image feature, the human body feature and the vehicle feature. The embodiment of the present application combines the human body feature and the vehicle feature on the basis of the image feature to comprehensively judge whether a dust-raising event occurs in the detection area, reduces the interference degree of other factors on the detection result, and is beneficial to improving the monitoring efficiency and accuracy of the dust-raising event.

[0104] Correspondingly, the embodiment of the present application further provides an electronic device, which may be a terminal or a server. The terminal may be a terminal device such as a smart phone, a tablet computer, a notebook computer, a touch screen, a game console, a personal computer (PC), a personal digital assistant (PDA), etc. As Figure 5 shown, Figure 5The figure is a schematic structural diagram of an electronic device provided by an embodiment of the present application. The electronic device 400 includes a processor 401 having one or more processing cores, a memory 402 having one or more computer-readable storage media, and a computer program stored in the memory 402 and executable on the processor. Among them, the processor 401 is electrically connected to the memory 402. Those skilled in the art can understand that the structural diagram of the electronic device shown in the figure does not constitute a limitation on the electronic device, and it may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0105] The processor 401 is the control center of the electronic device 400, connecting various parts of the entire electronic device 400 through various interfaces and lines. By running or loading software programs and / or modules stored in the memory 402, and calling data stored in the memory 402, it executes various functions of the electronic device 400 and processes data, thereby monitoring the entire electronic device 400.

[0106] In the embodiment of the present application, the processor 401 in the electronic device 400 will load the instructions corresponding to the processes of one or more application programs into the memory 402 according to the following steps, and the processor 401 will run the application programs stored in the memory 402 to implement various functions:

[0107] Obtain at least one image of the detection area;

[0108] Determine a target area in the image and calculate the image features of the target area;

[0109] Extract the human key point information and vehicle key point information in the image, and calculate the human features and vehicle features respectively according to the human key point information and vehicle key point information;

[0110] Judge whether dust is raised in the detection area according to the image features, human features, and vehicle features.

[0111] For the specific implementation of each of the above operations, reference may be made to the previous embodiments, which will not be elaborated here.

[0112] Optionally, as Figure 5 shown, the electronic device 400 further includes: a touch display screen 403, a radio frequency circuit 404, an audio circuit 405, an input unit 406, and a power supply 407. Among them, the processor 401 is electrically connected to the touch display screen 403, the radio frequency circuit 404, the audio circuit 405, the input unit 406, and the power supply 407 respectively. Those skilled in the art can understand, Figure 5The electronic device structure shown does not limit the electronic device, which may include more or fewer components than shown, or combine certain components, or have a different component arrangement.

[0113] The touch display screen 403 can be used to display a graphical user interface and receive operation instructions generated by a user acting on the graphical user interface. The touch display screen 403 may include a display panel and a touch panel. Among them, the display panel can be used to display information input by the user or information provided to the user, as well as various graphical user interfaces of the electronic device. These graphical user interfaces can be composed of graphics, text, icons, videos, and any combination thereof. Optionally, the display panel can be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), etc. The touch panel can be used to collect touch operations of the user thereon or nearby (such as operations of the user using a finger, a stylus, or any suitable object or accessory on or near the touch panel), and generate corresponding operation instructions, and the operation instructions execute corresponding programs. Optionally, the touch panel can include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the touch orientation of the user, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into contact coordinates, and then sends it to the processor 401, and can receive and execute commands sent by the processor 401. The touch panel can cover the display panel. When the touch panel detects a touch operation thereon or nearby, it is transmitted to the processor 401 to determine the type of touch event. Subsequently, the processor 401 provides a corresponding visual output on the display panel according to the type of touch event. In the embodiments of the present application, the touch panel and the display panel can be integrated into the touch display screen 403 to implement input and output functions. However, in some embodiments, the touch panel and the touch panel can be implemented as two independent components to implement input and output functions. That is, the touch display screen 403 can also be used as part of the input unit 406 to implement the input function.

[0114] In the embodiments of the present application, the processor 401 executes an application program to generate a graphical user interface on the touch display screen 403. The touch display screen 403 is used to present the graphical user interface and receive operation instructions generated by a user acting on the graphical user interface.

[0115] The radio frequency circuit 404 can be used to receive and transmit radio frequency signals to establish wireless communication with a network device or other electronic devices, and to receive and transmit signals between the network device or other electronic devices. In the embodiment of the present application, the above-mentioned electronic device can establish communication with a camera installed on a target vehicle or a parking space through the radio frequency circuit. Thus, when the target vehicle in the vehicle area photographed by the camera is scratched, the electronic device can give a timely reminder, and the image photographed by the camera can be quickly viewed through the electronic device, thereby avoiding or reducing the losses caused by the car being scratched.

[0116] The audio circuit 405 can be used to provide an audio interface between the user and the electronic device through a speaker and a microphone. The audio circuit 405 can transmit the electrical signal converted from the received audio data to the speaker, and the speaker converts it into a sound signal for output; on the other hand, the microphone converts the collected sound signal into an electrical signal, which is received by the audio circuit 405 and then converted into audio data. After the audio data is output to the processor 401 for processing, it is sent through the radio frequency circuit 404 to, for example, another electronic device, or the audio data is output to the memory 402 for further processing. The audio circuit 405 may also include an earphone jack to provide communication between the peripheral earphone and the electronic device.

[0117] The input unit 406 can be used to receive input digital, character information or user characteristic information (such as fingerprint, iris, facial information, etc.), and to generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function controls.

[0118] The power supply 407 is used to supply power to each component of the electronic device 400. Optionally, the power supply 407 can be logically connected to the processor 401 through a power management system, so as to realize functions such as management of charging, discharging, and power consumption management through the power management system. The power supply 407 can also include any components such as one or more DC or AC power supplies, a recharge system, a power failure detection circuit, a power converter or inverter, and a power status indicator.

[0119] Although Figure 5 not shown in the figure, the electronic device 400 may also include a camera, a sensor, a Wi-Fi module, a Bluetooth module, etc., which will not be elaborated here.

[0120] In the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0121] As can be seen from the above, the electronic device 400 provided in this embodiment obtains at least one image of a detection area, determines a target area in the image, calculates the image features of the target area, extracts the human key point information and vehicle key point information in the image, calculates the human features and vehicle features respectively according to the human key point information and vehicle key point information, and determines whether dust is raised in the detection area according to the image features, human features and vehicle features. In the embodiment of the present application, based on the image features, the human features and vehicle features are combined to comprehensively judge whether a dust raising event occurs in the detection area, reducing the interference degree of other factors on the detection result, which is beneficial to improving the monitoring efficiency and accuracy of the dust raising event.

[0122] Those of ordinary skill in the art can understand that all or part of the steps in the above various methods can be completed by instructions or by controlling related hardware through instructions. The instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0123] For this reason, the embodiment of the present application provides a computer-readable storage medium, in which multiple computer programs are stored. The computer programs can be loaded by a processor to execute the steps in any dust detection method provided by the embodiment of the present application. For example, the computer program can execute the following steps:

[0124] Obtain at least one image of a detection area;

[0125] Determine a target area in the image and calculate the image features of the target area;

[0126] Extract the human key point information and vehicle key point information in the image, and calculate the human features and vehicle features respectively according to the human key point information and vehicle key point information;

[0127] Judge whether dust is raised in the detection area according to the image features, human features and vehicle features.

[0128] For the specific implementation of each of the above operations, reference can be made to the previous embodiments and will not be elaborated here.

[0129] Among them, the storage medium may include: read-only memory (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disc, etc.

[0130] Since the computer programs stored in the storage medium can execute the steps in any dust detection method provided by the embodiment of the present application, the beneficial effects that can be achieved by any dust detection method provided by the embodiment of the present application can be realized. For details, reference can be made to the previous embodiments and will not be elaborated here.

[0131] The above has introduced in detail a dust detection method, device, electronic device and storage medium provided by the embodiments of the present application. Specific examples are used in this article to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those skilled in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A dust detection method, characterized in that, it includes: Obtaining at least one image of the detection area; Determining a target area in the image, calculating the image features of the target area, wherein, according to the color information and clarity information of the image, determining the dust area in the image, calculating the dust probability of the dust area, and using the dust probability as the image feature of the target area; Extracting the human key point information and vehicle key point information in the image, and calculating the human feature and vehicle feature respectively according to the human key point information and vehicle key point information; Obtaining the weight values corresponding to the human feature and vehicle feature respectively, adjusting the probability threshold according to the weight values corresponding to the human feature and vehicle feature respectively, comparing the dust probability with the adjusted probability threshold, so as to judge whether dust is generated in the detection area according to the comparison result.

2. The dust detection method according to claim 1, characterized in that, the extracting the human key point information in the image includes: Extracting the human bone point features of the person in the image, and determining the hand position and preset facial feature positions of the person according to the human bone point features; the calculating the human feature according to the human key point information includes: Calculating the coincidence ratio of the areas corresponding to the hand position and the preset facial feature positions as the human feature.

3. The dust detection method according to claim 1, characterized in that, the extracting the vehicle key point information in the image includes: Identifying the steering information of the vehicle in the image, and determining the target vehicle area according to the steering information; the calculating the vehicle feature according to the vehicle key point information includes: Judging whether there is sediment falling in the target vehicle area, and using the judgment result as the vehicle feature.

4. The dust detection method according to any one of claims 1 to 3, characterized in that, after determining the target area in the image, calculating the image features of the target area, further includes: Dividing the target area into two sub-image areas, respectively performing image enhancement processing on the two sub-image areas, and calculating the image features of the target area according to the processed image.

5. A dust detection device, characterized in that, it includes: An acquisition module for acquiring at least one image of the detection area; A determination module for determining a target area in the image, calculating the image features of the target area, wherein, according to the color information and clarity information of the image, determining the dust area in the image, calculating the dust probability of the dust area, and using the dust probability as the image feature of the target area; An extraction module for extracting the human key point information and vehicle key point information in the image, and calculating the human feature and vehicle feature respectively according to the human key point information and vehicle key point information; A judgment module, configured to obtain the weight values corresponding to the human body features and vehicle features respectively, adjust a probability threshold according to the weight values corresponding to the human body features and vehicle features respectively, and compare the dust emission probability with the adjusted probability threshold, so as to judge whether dust emission occurs in the detection area according to the comparison result.

6. An electronic device, characterized in that the electronic device includes a memory and a processor, a computer program is stored in the memory, and the processor executes the steps in the dust emission detection method according to any one of claims 1 to 4 by calling the computer program stored in the memory.

7. A storage medium, characterized in that the storage medium stores a computer program, and the computer program is suitable for being loaded by a processor to execute the steps in the dust emission detection method according to any one of claims 1 to 4.

Images