A new energy vehicle charging pile safety monitoring system
In the safety monitoring system of the charging pile of new energy vehicles, the search window and the affected degree analysis module are used to determine the heat source and the affected area, and the thermal image is enhanced, and the problem of false alarms in the thermal image of infrared cameras is solved, and the new energy is realized Accurate safety monitoring of car charging piles.
Patent Information
- Application Number
- CN202310673243.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-08
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2043-06-08
AI Technical Summary
In the prior art, due to the range of the heat source, the thermal images collected by infrared cameras cannot accurately detect abnormal charging piles, resulting in false alarms and false detection phenomena, affecting the detection accuracy of the safety monitoring system.
The charging pile image acquisition module, the heating source positioning module, the heat source affecting area positioning module, the affected degree analysis module and the image enhancement module are used to traverse the thermal image through the search window, determine the heating source area and the heat source affecting area, and analyze the affected charging pile. and enhance the thermal image to form an enhanced thermal image with obvious contrast characteristics for safety monitoring.
By accurately positioning the heat source and the affected area of the heat source, analyzing the degree of impact of the charging piles, improving the comparison characteristics between the charging piles, and achieving accurate and safe identification and monitoring of the charging piles of new energy vehicles, reducing false alarm phenomena.
Smart Images

Figure CN116797995B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of image processing, and in particular to a new energy vehicle charging pile safety monitoring system. Background Art
[0002] Charging piles are a core deployment project for new energy vehicles and are developing rapidly. Safety issues arising from charging piles are an important part of the charging pile deployment project. When charging new energy vehicles with charging piles, abnormal temperatures caused by abnormal charging conditions can cause major safety accidents such as fires. Therefore, it is necessary to monitor the temperature of the charging piles in real time, and to provide timely feedback on control, alarm and other commands to ensure the safety of the charging area.
[0003] In the prior art, in order to save costs and improve detection efficiency, infrared cameras can be used to collect thermal images of the charging area, and abnormal charging piles can be determined based on the temperature information fed back from the thermal images. However, the heating of the charging piles will have a wide range of effects, so that two or more charging piles in the thermal image are affected by a heat source, and the affected charging piles will all produce a certain abnormal temperature, resulting in false alarms and false detections, affecting the detection accuracy of the safety monitoring system. Summary of the invention
[0004] In order to solve the technical problem that abnormal charging piles cannot be accurately detected due to the influence of heat sources in thermal images, the purpose of the present invention is to provide a new energy vehicle charging pile safety monitoring system, and the technical solutions adopted are as follows:
[0005] The present invention proposes a new energy vehicle charging pile safety monitoring system, the system comprising:
[0006] A charging pile image acquisition module, used to acquire a thermal image of a charging area; the thermal image includes a plurality of charging pile areas;
[0007] A heat source positioning module is used to traverse the entire thermal image according to a search window of a preset size, obtain the heat source probability of the search window according to the thermal value fluctuation characteristics and the thermal value size in the search window; and determine the heat source area according to the heat source probability;
[0008] A heat source affected area positioning module is used to screen the area corresponding to the search window according to the position change of the search window relative to the heat source area and the change of the thermal value to obtain the heat source affected area;
[0009] An impact degree analysis module, used to obtain the overlapped area between each charging pile area and the heat source impacted area; and obtain the impact degree of each charging pile area according to the area and thermal value distribution of the overlapped area;
[0010] An image enhancement module, used to enhance the charging pile area according to the degree of impact to obtain an enhanced thermal image;
[0011] A safety monitoring module is used to perform safety monitoring based on the thermal values in the enhanced thermal images of consecutive frames.
[0012] Furthermore, the method for acquiring the charging pile area includes:
[0013] Obtain an RGB image corresponding to the thermal image; in the RGB image, obtain an initial charging pile area through a pre-trained semantic segmentation network; map the initial charging pile area to the thermal image to obtain the charging pile area.
[0014] Furthermore, the method for obtaining the probability of a heat source includes:
[0015] The standard deviation of the thermal value in the search window is used as the thermal value fluctuation feature; the thermal value fluctuation feature is negatively correlated and normalized to obtain the first heat source probability;
[0016] The average thermal value in the search window is used as a representative thermal value; a thermal value difference between the representative thermal value and the average thermal value of the thermal image is obtained, and the thermal value difference is negatively correlated to obtain a second heat source probability;
[0017] The first heat source probability and the second heat source probability are multiplied and then normalized to obtain the heat source probability.
[0018] Further, determining the heat source area according to the heat source probability includes:
[0019] If the probability of the heat source in the search window is greater than a preset probability threshold, the area corresponding to the search window is used as the initial heat source area; all the initial heat source areas are clustered according to the positions of the initial heat source areas, and the area corresponding to each cluster cluster is used as the heat source area.
[0020] Furthermore, the method for obtaining the heat source affected area includes:
[0021] The average thermal value corresponding to each search window area of the heat source area in the preset direction is counted, and the distance between each search window area and the heat source area in the preset direction is used as the horizontal coordinate, and the average thermal value in each search window area is used as the vertical coordinate to construct a two-dimensional coordinate system, and obtain a position-thermal value change curve in each preset direction in the two-dimensional coordinate system;
[0022] The tangent slope of each position in the position-thermal value change curve is counted, and if the tangent slope is greater than a preset slope threshold, the search window area at the position corresponding to the tangent slope is used as the heat source influence area.
[0023] Furthermore, the method for obtaining the degree of impact includes:
[0024] Obtain the area ratio of the total area of the overlapping area to the area of the corresponding charging pile area, obtain the thermal value range and the thermal value maximum in the overlapping area, use the ratio of the thermal value maximum to the thermal value range as the thermal influence weight, and use the product of the thermal influence weight and the area ratio as the degree of influence.
[0025] Furthermore, the method for acquiring the enhanced thermal image includes:
[0026] The negative correlation of the affected degree is mapped and normalized to obtain an adjustment coefficient, and the product of the adjustment coefficient and the original thermal value of each pixel point in the corresponding charging pile area is used as the initial adjusted thermal value of each pixel point;
[0027] Obtaining the ambient temperature thermal value, if the initial adjustment thermal value is less than the ambient temperature thermal value, using the ambient temperature thermal value as the adjustment thermal value of the corresponding pixel point; if the initial adjustment thermal value is not less than the ambient temperature thermal value, using the initial adjustment thermal value as the adjustment thermal value of the corresponding pixel point;
[0028] The adjusted thermal values of all pixels in all charging pile areas are obtained, and the adjusted thermal values are replaced with the original thermal values to obtain the enhanced thermal image.
[0029] Further, the safety monitoring according to the thermal values in the enhanced thermal images of the continuous frames includes:
[0030] The area consisting of pixels whose thermal values are greater than a preset dangerous thermal value in the enhanced thermal image is used as the monitoring area; if the monitoring area appears in the enhanced thermal image of a preset number of consecutive frames, an alarm signal is fed back.
[0031] The present invention has the following beneficial effects:
[0032] The present invention takes into account the temperature distribution characteristics of the heat source itself, and uses a search window to determine the accurate heat source area in the thermal image. Further taking into account the temperature radiation characteristics of the area surrounding the heat source, the heat source influence area is determined according to the position change and calorific value change of the search window relative to the heat source area. The determination of the heat source influence area can be used as a reference for analyzing the impact of the charging pile. The area and temperature of the charging pile in the heat source influence area can reflect the degree to which the charging pile is affected by the heat source. Therefore, the charging pile area can be enhanced by the degree of impact of the charging pile area, and the thermal value of the charging pile area can be adjusted to obtain an enhanced thermal image with obvious characteristics. In the enhanced thermal image, because it is enhanced based on the degree of impact, the contrast features between the charging piles are obvious, and accurate safety monitoring can be performed using the enhanced thermal image. The present invention accurately locates the heat source area and the heat source impact area, analyzes the degree of impact of the charging pile area, and then processes the thermal image, and accurately identifies safety through the enhanced thermal image with obvious contrast features of the charging pile area. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0034] Figure 1 A block diagram of a new energy vehicle charging pile safety monitoring system provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0035] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation method, structure, characteristics and effects of a new energy vehicle charging pile safety monitoring system proposed by the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.
[0036] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
[0037] The specific scheme of a new energy vehicle charging pile safety monitoring system provided by the present invention is described in detail below with reference to the accompanying drawings.
[0038] See also Figure 1, which shows a block diagram of a new energy vehicle charging pile safety monitoring system provided by an embodiment of the present invention, the system includes a charging pile image acquisition module 101, a heat source positioning module 102, a heat source impact area positioning module 103, an impact degree analysis module 104, an image enhancement module 105 and a safety monitoring module 106.
[0039] The charging pile image acquisition module 101 is used to acquire a thermal image of a charging area; the thermal image includes a plurality of charging pile areas.
[0040] In one embodiment of the present invention, the deployment position of the infrared camera should be able to capture as many charging piles as possible. Safety monitoring through thermal images collected by the infrared camera is less costly than traditional sensor data safety monitoring. During the information collection stage, one infrared camera can be used to collect temperature information of multiple charging piles or information of all charging piles in the entire charging area.
[0041] In order to analyze each charging pile area later, it is first necessary to determine the position of the charging pile area in the thermal image. Preferably, in one embodiment of the present invention, the method for obtaining the charging pile area includes: considering that the thermal image contains less image detail information and can only represent thermal distribution information, in order to accurately identify the charging pile area, it is necessary to combine the RGB image. Therefore, the RGB image corresponding to the thermal image is obtained, that is, the camera that collects the RGB image should be in the same position as the infrared camera that collects the thermal image, and the camera field of view should be consistent with the infrared camera field of view, so that the RGB image corresponds to the thermal image. Because the RGB image contains rich color and image detail information, the initial charging pile area can be obtained through a pre-trained semantic segmentation network, that is, the initial charging pile area is the area in the RGB image, so the initial charging pile area is mapped to the thermal image, and the charging pile area in the thermal image can be obtained.
[0042] It should be noted that the semantic segmentation network is a technical means well known to those skilled in the art, and the specific network structure and internal details are not described here in detail. Only a brief training process of the semantic segmentation network in one embodiment of the present invention is briefly described:
[0043] The specific training methods of semantic segmentation network include:
[0044] (1) Use images containing charging piles as training data. Label the charging pile pixels as 1 and other pixels as 0 to obtain label data.
[0045] (2) The semantic segmentation network adopts an encoding-decoding structure, and the training data and label data are normalized and then input into the network.
[0046] The semantic segmentation encoder is used to extract the features of the input data and obtain the feature map. The semantic segmentation decoder samples and transforms the feature map and outputs the semantic segmentation result.
[0047] (3) The cross entropy loss function is used to train the network.
[0048] The heat source positioning module 102 is used to traverse the entire thermal image according to a search window of a preset size, obtain the heat source probability of the search window according to the thermal value fluctuation characteristics and the thermal value size in the search window, and determine the heat source area according to the heat source probability.
[0049] Because the heating of charging piles is a range of temperature changes, due to the angle problem of the infrared camera, the distance between the charging pile areas in the thermal image is very close, so that the abnormal temperature heat source area affects two or more charging pile areas. In order to avoid misidentification during safety monitoring, the location of the heat source should be determined first. According to prior knowledge, the temperature is always transferred from the high temperature area to the low temperature area. Therefore, for the non-heat source area, there is a more obvious temperature change inside the area, while the heat source, as a stable temperature output, does not have an obvious temperature change inside the area, and the heat source area will maintain a relatively stable and high temperature distribution. Therefore, based on this principle, it is possible to determine whether the area is a heat source area by the thermal value fluctuation characteristics and thermal value size in a region. In order to accurately determine the location of the heat source area, the entire thermal image can be traversed according to the search window of a preset size, and a local analysis can be performed in the search window area. The probability of the heat source in the search window area is obtained according to the thermal value fluctuation characteristics and thermal value size in the window, and then the heat source area can be determined according to the heat source probability.
[0050] In one embodiment of the present invention, the size of the search window is set to 3×3.
[0051] Preferably, in one embodiment of the present invention, a specific method for obtaining the probability of a heat source includes:
[0052] The standard deviation of the thermal value in the search window is used as the thermal value fluctuation feature; the thermal value fluctuation feature is negatively correlated and normalized to obtain the first heat source probability; the average thermal value in the search window is used as the representative thermal value; the thermal value difference between the representative thermal value and the average thermal value of the thermal image is obtained, and the thermal value difference is negatively correlated to obtain the second heat source probability; the first heat source probability and the second heat source probability are multiplied and normalized to obtain the heat source probability. In one embodiment of the present invention, the formula of the heat source probability is expressed as:
[0053]
[0054] Where D is the probability of the heat source, σ is the standard deviation of the thermal value in the search window, represents the thermal value, is the average thermal value of the thermal image, exp() is an exponential function with a natural constant as the base, Norm() is a normalization function, sigmoid() is a sigmoid function, exp(-σ) is the probability of the first heat source, is the probability of the second heat source. In one embodiment of the present invention, Norm() may use a maximum and minimum value normalization algorithm, and other normalization methods may also be used in other embodiments, which will not be described in detail here.
[0055] In the heat source probability formula, the standard deviation, that is, the thermal value fluctuation characteristics, is negatively correlated and normalized using an exponential function with a natural constant as the base. The larger the thermal value fluctuation characteristics, the less likely the corresponding area is to be a heat source area, and the smaller the corresponding heat source probability. The larger the difference between the representative thermal value and the average thermal value of the thermal image, the higher the thermal value in the search window area, the higher the temperature in the area, the more likely the corresponding area is to be a heat source area, and the greater the probability of a heat source. The sigmod function is further used to normalize the results to obtain a heat source probability with a numerical range between 0 and 1.
[0056] Preferably, in one embodiment of the present invention, determining the heat source area according to the heat source probability includes:
[0057] If the probability of the heat source in the search window is greater than the preset probability threshold, the area corresponding to the search window is used as the initial heat source area; taking into account the possible influence of abnormal noise, which may lead to inaccurate heat source area judgment, because the area corresponding to the search window is a smaller local area, the heat source area should be a large area clustered by the initial heat source area, so all the initial heat source areas can be clustered according to the position of the initial heat source area, and the area corresponding to each cluster is used as the heat source area, and the discrete initial heat source area that is not clustered is the area where the noise influence is generated. In an embodiment of the present invention, the initial heat source area is clustered by the DBSCAN clustering algorithm, and the initial parameters of the specific algorithm can be set according to the implementer's tolerance for errors. Because the DBSCAN clustering algorithm is a technical means well known to those skilled in the art, it is not limited or elaborated here. In one embodiment of the present invention, the probability threshold is set to 0.8.
[0058] The heat source influence area positioning module 103 is used to screen the area corresponding to the search window according to the position change and thermal value change of the search window relative to the heat source area to obtain the heat source influence area.
[0059] The influence of the heat source on the surrounding temperature will slowly decrease as the distance increases, that is, the farther away from the heat source, the slower the temperature changes. For the affected area of the heat source, the affected area must be close to the heat source and have obvious temperature diffusion characteristics. Therefore, the area corresponding to the search window can be screened according to the position change of the search window relative to the heat source area and the change of thermal value to obtain the heat affected area.
[0060] Preferably, in one embodiment of the present invention, the method for obtaining the heat source influence area specifically includes: counting the average thermal value corresponding to each search window area of the heat source area in the preset direction, taking the distance between each search window area and the heat source area in the preset direction as the horizontal coordinate, and the average thermal value in each search window area as the vertical coordinate, constructing a two-dimensional coordinate system, and obtaining the position-thermal value change curve in each preset direction in the two-dimensional coordinate system; counting the tangent slope of each position in the position-thermal value change curve, the tangent slope can characterize the speed of temperature change, and if the tangent slope is greater than the preset slope threshold, the search window area corresponding to the position of the tangent slope is taken as the heat source influence area. In one embodiment of the present invention, the preset directions are set to 360, that is, with the heat source area as the center, each angle corresponds to a direction; the slope threshold is set to 0.3.
[0061] The impact degree analysis module 104 is used to obtain the overlapped area between each charging pile area and the heat source impact area; and obtain the impact degree of each charging pile area according to the area and thermal value distribution of the overlapped area.
[0062] The relative position of the heat source affected area and the charging pile area determines whether the corresponding charging pile is affected by the heat source or the degree of the impact. Therefore, the overlap area between each charging pile area and the heat source affected area can be obtained. The larger the overlap area, the higher the temperature in the area, which means that the corresponding charging pile is more affected by the heat source. Therefore, the degree of impact of each charging pile area can be obtained based on the area and thermal value distribution of the overlap area.
[0063] Preferably, in one embodiment of the present invention, the method for obtaining the degree of influence includes:
[0064] The area ratio of the total area of the overlapped area to the area of the corresponding charging pile area is obtained, and the thermal value range and the maximum thermal value in the overlapped area are obtained. The ratio of the maximum thermal value to the thermal value range is used as the thermal influence weight, and the product of the thermal influence weight and the area ratio is used as the degree of influence. The degree of influence is expressed by the formula:
[0065]
[0066] in, is the degree of impact, s is the total area of the overlapping area, S is the area of the charging pile area, and wmax is the maximum thermal value in the overlapped area, w min is the minimum thermal value in the overlapped area, w max -w min The thermal value in the overlapping area is extremely poor.
[0067] It can be seen from the influence degree formula that the larger the area ratio is, the greater the influence of the charging pile by the heat source is, and the greater the influence degree is; the larger the maximum thermal value in the overlapping area is, the closer the overlapping area is to the heat source, the more susceptible it is to temperature diffusion, and the greater the influence degree is; the thermal value range as the denominator of the maximum thermal value can be used as the effectiveness of the maximum thermal value to adjust the maximum thermal value, that is, the larger the range is, the greater the temperature change in the overlapping area is, the farther the overlapping area is from the heat source, and the smaller the influence degree of the corresponding charging pile is.
[0068] The image enhancement module 105 is used to enhance the charging pile area according to the degree of impact to obtain an enhanced thermal image.
[0069] The greater the degree of impact on the charging pile area, the more abnormal temperature in the charging pile area may not be caused by itself. Therefore, it is necessary to adjust the thermal value of each charging pile area in the thermal image to enhance the thermal value contrast between the heat source and the charging pile, and then obtain an enhanced thermal image with obvious contrast features.
[0070] Preferably, in one embodiment of the present invention, the method for acquiring enhanced thermal images includes:
[0071] The negative correlation of the affected degree is mapped and normalized to obtain the adjustment coefficient, and the product of the adjustment coefficient and the original thermal value of each pixel in the corresponding charging pile area is used as the initial adjustment thermal value of each pixel. In one embodiment of the present invention, the initial adjustment thermal value is expressed by the formula:
[0072]
[0073] Among them, V is the initial adjustment thermal value, is the normalized degree of influence, and w is the original thermal value. That is, in the initial adjustment thermal value formula, the degree of influence is first normalized, and its value range is limited to between 0 and 1. The normalized degree of influence is further subtracted from the constant 1 to obtain the negative correlation mapping and normalized degree of influence. The greater the degree of influence, the more the corresponding original thermal value needs to be adjusted down.
[0074] Obtain the ambient temperature thermal value. If the initial adjustment thermal value is less than the ambient temperature thermal value, it means that the initial adjustment thermal value is over-adjusted. In order to avoid the impact of over-adjustment, the ambient temperature thermal value is used as the adjustment thermal value of the corresponding pixel point; if the initial adjustment thermal value is not less than the ambient temperature thermal value, the initial adjustment thermal value is used as the adjustment thermal value of the corresponding pixel point. It should be noted that the ambient temperature thermal value can be obtained through the thermal value in the background area of the thermal value non-charging pile area, which will not be elaborated here.
[0075] The adjusted thermal values of all pixels in all charging pile areas are obtained, and the adjusted thermal values are replaced with the original thermal values to obtain an enhanced thermal image.
[0076] The safety monitoring module 106 is used to perform safety monitoring based on the thermal values in the enhanced thermal images of consecutive frames.
[0077] According to prior knowledge, new energy vehicles will definitely generate heat during the charging process using charging piles, but the heat generated under normal charging conditions will vary within a fixed temperature range. When the temperature exceeds the normal range, it may be an abnormal charging state caused by a fault. Therefore, the abnormal temperature can be safely monitored based on the thermal value characteristics shown in the enhanced thermal value image. In addition, because the enhanced thermal value image has obvious contrast features, it is possible to accurately locate the abnormal temperature area, and then accurately locate the abnormal charging pile, thereby realizing safe monitoring of the new energy vehicle charging piles.
[0078] Preferably, in one embodiment of the present invention, performing safety monitoring based on thermal values in enhanced thermal images of consecutive frames includes:
[0079] The area consisting of pixels whose thermal values are greater than a preset dangerous thermal value in the enhanced thermal image is used as the monitoring area; if the monitoring area appears in the enhanced thermal images of a preset number of consecutive frames, an alarm signal is fed back.
[0080] In one embodiment of the present invention, considering that the temperature of the charging pile of a new energy vehicle under normal charging state will not exceed 50 degrees Celsius, 50 degrees Celsius is converted into a thermal value through existing technology to obtain a dangerous thermal value; in one embodiment of the present invention, the frame rate of the image collected by the infrared camera is 30, that is, 30 frames of enhanced thermal images can be obtained in one second, and the number of consecutive frames is set to 1800, that is, if the monitoring area exists continuously for one minute, it means that the current charging pile is in an abnormal charging state, and an alarm signal is fed back.
[0081] In summary, the embodiment of the present invention sets a search window in the thermal image, and determines the heat source area according to the fluctuation characteristics and size of the thermal values in the search window. The heat-affected area is determined according to the change in thermal values caused by the position change of the search window relative to the heat source. The degree of influence of each charging pile area is obtained using the area and thermal value information of the overlapping area of the heat-affected area and the charging pile area. The charging pile area is enhanced according to the degree of influence to obtain an enhanced thermal image. The charging pile is safety tested using the enhanced thermal image. The present invention obtains an enhanced thermal image with obvious contrast features between the charging piles by analyzing the degree of influence of the charging piles on the heat source during the charging process, thereby achieving accurate safety testing of charging piles for new energy vehicles.
[0082] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0083] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.
Claims
1. A new energy vehicle charging pile safety monitoring system, characterized in that: The system comprises: A charging pile image acquisition module, used to acquire a thermal image of a charging area; the thermal image includes a plurality of charging pile areas; A heat source positioning module is used to traverse the entire thermal image according to a search window of a preset size, obtain the heat source probability of the search window according to the thermal value fluctuation characteristics and the thermal value size in the search window; and determine the heat source area according to the heat source probability; A heat source affected area positioning module is used to screen the area corresponding to the search window according to the position change of the search window relative to the heat source area and the change of the thermal value to obtain the heat source affected area; An impact degree analysis module, used to obtain the overlap area between each charging pile area and the heat source impact area; and obtain the impact degree of each charging pile area according to the area and thermal value distribution of the overlap area; An image enhancement module, used to enhance the charging pile area according to the degree of impact to obtain an enhanced thermal image; A safety monitoring module, used for performing safety monitoring according to the thermal values in the enhanced thermal images of consecutive frames; The method for acquiring the enhanced thermal image comprises: The negative correlation of the affected degree is mapped and normalized to obtain an adjustment coefficient, and the product of the adjustment coefficient and the original thermal value of each pixel point in the corresponding charging pile area is used as the initial adjusted thermal value of each pixel point; Obtaining the ambient temperature thermal value, if the initial adjustment thermal value is less than the ambient temperature thermal value, using the ambient temperature thermal value as the adjustment thermal value of the corresponding pixel point; if the initial adjustment thermal value is not less than the ambient temperature thermal value, using the initial adjustment thermal value as the adjustment thermal value of the corresponding pixel point; The adjusted thermal values of all pixels in all charging pile areas are obtained, and the adjusted thermal values are replaced with the original thermal values to obtain the enhanced thermal image.
2. A new energy vehicle charging pile safety monitoring system according to claim 1, characterized in that: The method for acquiring the charging pile area includes: Obtain an RGB image corresponding to the thermal image; in the RGB image, obtain an initial charging pile area through a pre-trained semantic segmentation network; map the initial charging pile area to the thermal image to obtain the charging pile area.
3. A new energy vehicle charging pile safety monitoring system according to claim 1, characterized in that: The method for obtaining the probability of a heat source includes: The standard deviation of the thermal value in the search window is used as the thermal value fluctuation feature; the thermal value fluctuation feature is negatively correlated and normalized to obtain the first heat source probability; The average thermal value in the search window is used as a representative thermal value; a thermal value difference between the representative thermal value and the average thermal value of the thermal image is obtained, and the thermal value difference is negatively correlated to obtain a second heat source probability; The first heat source probability and the second heat source probability are multiplied and then normalized to obtain the heat source probability.
4. A new energy vehicle charging pile safety monitoring system according to claim 1, characterized in that: Determining the heat source area according to the heat source probability includes: If the probability of the heat source in the search window is greater than a preset probability threshold, the area corresponding to the search window is used as the initial heat source area; all the initial heat source areas are clustered according to the positions of the initial heat source areas, and the area corresponding to each cluster cluster is used as the heat source area.
5. A new energy vehicle charging pile safety monitoring system according to claim 1, characterized in that: The method for obtaining the heat source affected area includes: The average thermal value corresponding to each search window area of the heat source area in the preset direction is counted, and the distance between each search window area and the heat source area in the preset direction is used as the horizontal coordinate, and the average thermal value in each search window area is used as the vertical coordinate to construct a two-dimensional coordinate system, and obtain a position-thermal value change curve in each preset direction in the two-dimensional coordinate system; The tangent slope of each position in the position-thermal value change curve is counted, and if the tangent slope is greater than a preset slope threshold, the search window area at the position corresponding to the tangent slope is used as the heat source influence area.
6. A new energy vehicle charging pile safety monitoring system according to claim 1, characterized in that: The method for obtaining the degree of impact includes: Obtain the area ratio of the total area of the overlapping area to the area of the corresponding charging pile area, obtain the thermal value range and the thermal value maximum in the overlapping area, use the ratio of the thermal value maximum to the thermal value range as the thermal influence weight, and use the product of the thermal influence weight and the area ratio as the degree of influence.
7. A new energy vehicle charging pile safety monitoring system according to claim 1, characterized in that: The safety monitoring according to the thermal values in the enhanced thermal images of the continuous frames comprises: The area consisting of pixels whose thermal values are greater than a preset dangerous thermal value in the enhanced thermal image is used as the monitoring area; if the monitoring area appears in the enhanced thermal image of a preset number of consecutive frames, an alarm signal is fed back.
