Illumination method based on direct current illumination system

Abstract

The invention relates to the technical field of direct-current lighting, in particular to a lighting method based on a direct-current lighting system, and the method comprises the steps: collecting a lighting area image of each street lamp at each moment in a rainy day and an ambient light intensity value at each moment; identifying umbrellas in the illumination area image to obtain umbrella areas in the illumination area image; key points of the umbrella areas are extracted based on the shape features of the umbrellas; determining detail richness and pattern contrast of each key point; constructing a feature descriptor of each key point, and determining a pedestrian flow evaluation value of each street lamp at each moment; determining the reflection intensity of each key point; the illumination area image at each moment is equally divided into image blocks, and the rain silk possibility of each image block is determined; screening feature image blocks in all the image blocks at each moment; determining the rain vigor evaluation intensity at each moment; and the brightness of the street lamp is adjusted by using a PWM (Pulse Width Modulation) dimming technology. The flexibility of street lamp illumination brightness adjustment is improved.

Description

Technical Field

[0001] The present application relates to the field of direct current lighting technology, and in particular to a lighting method based on a direct current lighting system. Background Art

[0002] The DC lighting system converts AC power into DC power through a DC centralized drive power supply cabinet, providing a stable DC power supply for the entire lighting system. Each DC LED street light is connected in parallel, and the DC power supply method provides a more stable voltage, extending the service life of LED lamps. It has the advantages of centralized power supply, easy management and maintenance, and energy saving and high efficiency.

[0003] Traditional lighting methods have poor flexibility and cannot be effectively adjusted according to environmental changes and pedestrian activities, especially in rainy weather, when raindrops are suspended in the air and interact with light, causing light scattering and refraction, affecting lighting effects or causing energy waste. Summary of the invention

[0004] In order to solve the above technical problems, the present application provides a lighting method based on a DC lighting system to solve the existing problems.

[0005] The lighting method based on the DC lighting system of the present application adopts the following technical solutions:

[0006] An embodiment of the present application provides a lighting method based on a DC lighting system, the method comprising the following steps:

[0007] Collect the lighting area images of each street lamp at each time on rainy days and the ambient light intensity value at each time;

[0008] Identify umbrellas in the lighting area image to obtain each umbrella area in the lighting area image; extract key points of each umbrella area based on the shape features of the umbrella; analyze the distribution of corner points in the local neighborhood of each key point to determine the detail richness of each key point;

[0009] Obtain each peak point in the grayscale histogram of all pixel points in the local neighborhood of each key point, and determine the pattern contrast of each key point based on the position distribution relationship and grayscale difference between all any two peak points in the local neighborhood of each key point;

[0010] Based on the detail richness, pattern contrast, and gradient amplitude of each key point, the feature descriptor of each key point is determined. Combined with the lighting area images at adjacent moments, the moving direction of each umbrella is obtained to determine the pedestrian flow evaluation value of each street lamp at each moment.

[0011] Analyze the grayscale difference between the pixels in the local neighborhood of each key point and all the pixels in the umbrella area, as well as the difference between the pattern contrasts of the key points, to determine the reflection intensity of each key point; obtain the optical flow direction and optical flow intensity of each pixel in the image of the illuminated area at each moment, divide the image of the illuminated area at each moment into image blocks, and determine the possibility of raindrops in each image block based on the distribution of the optical flow direction and optical flow intensity of the pixels in each image block;

[0012] Threshold segmentation is performed based on the possibility of raindrops to determine the characteristic image blocks in all image blocks at each moment; the distribution of the optical flow intensity of the pixels in all the characteristic image blocks is analyzed, and the rain intensity assessment intensity at each moment is determined in combination with the reflection intensity; the pedestrian flow assessment value at each moment, the rain intensity assessment intensity, and the ambient light intensity value are comprehensively considered to determine the illuminated surface brightness coefficient at each moment, and the brightness of the street lamps is adjusted using PWM dimming technology.

[0013] In one embodiment, extracting key points of each umbrella area based on the shape feature of the umbrella includes:

[0014] Obtain the perpendicular bisectors of each boundary line of the umbrella region, determine the pixel point where the perpendicular bisectors of the umbrella region intersect the most as the umbrella tail feature point, determine the central pixel point of the line connecting the central pixel point of each boundary line and the umbrella tail feature point as the umbrella surface feature point, determine the central pixel point connecting the endpoint of each boundary line and the umbrella tail pixel point as the umbrella strut feature point, and determine the endpoint of each boundary line as the bead tail feature point;

[0015] The tail feature points, umbrella surface feature points, umbrella support feature points, and bead tail feature points of the umbrella area are all recorded as key points.

[0016] In one embodiment, the detail richness is the ratio of the number of corner points in the local neighborhood of each key point to the maximum number of corner points in the local neighborhood of all key points in the corresponding umbrella area.

[0017] In one embodiment, the process of determining the pattern contrast is as follows:

[0018] In the local neighborhood of each key point, for the grayscale value a corresponding to any peak point and the grayscale value b corresponding to any remaining peak point, the probability of the pixel point with grayscale value b in the neighborhood of the pixel point with grayscale value a is counted, and the average of the probabilities of all the pixel points with grayscale value a in the local neighborhood of each key point is taken as the co-occurrence probability of grayscale values ​​a and b;

[0019] Calculate the difference between the grayscale value a and the grayscale value b, record it as the first difference, calculate the product of the first difference and the co-occurrence probability, calculate the sum of the product of the peak point corresponding to the grayscale value a and all the remaining peak points, normalize the accumulated result of the sum of all the peak points in the local neighborhood of each key point, and obtain the pattern contrast.

[0020] In one embodiment, the pedestrian flow evaluation value is determined as follows:

[0021] Based on the feature descriptors of each key point, a matching algorithm is used to obtain the same umbrella area in the lighting area image at adjacent moments, so as to determine the moving direction of each umbrella at each moment; the lighting area image at each moment is divided into a preset number of area blocks, and the umbrellas whose moving directions deviate from the lighting area image and are biased towards the direction of the adjacent street lamp in the area blocks close to the boundary of the lighting area image are determined as the pedestrian flow prediction values ​​of the adjacent street lamp;

[0022] The sum of the number of umbrella areas in the lighting area image of each street lamp at each moment and the pedestrian flow prediction value is used as the pedestrian flow evaluation value.

[0023] In one embodiment, the calculation method of the reflected light intensity is:

[0024] In the formula, E i is the reflection intensity of the i-th key point, C max is the maximum value of the pattern contrast of all key points in the illumination area image where the i-th key point is located, C i is the pattern contrast of the i-th key point, l i is the grayscale mean of all pixels in the local neighborhood of the i-th key point, is the grayscale mean of all pixels in the umbrella area corresponding to the i-th key point.

[0025] In one embodiment, the rain possibility is the ratio of a normalized value of an average level of optical flow intensity of all pixels in each image block to a normalized value of a discrete degree of optical flow directions of all pixels.

[0026] In one embodiment, the determination process of the rain assessment intensity is as follows:

[0027] For the illumination area image at each moment, the rain assessment intensity is the product of the average optical flow intensity of all pixels in all feature image blocks and the average reflection intensity of all key points in all umbrella areas.

[0028] In one embodiment, the process of determining the illumination surface brightness coefficient is as follows:

[0029] For each street lamp, the multiplication result of the pedestrian flow assessment value at each moment and the rain assessment intensity is calculated, and the ratio of the multiplication result to the ambient light intensity value is determined as the illumination surface brightness coefficient at each moment.

[0030] In one embodiment, the brightness of the street lamp is adjusted by using PWM dimming technology, including:

[0031] The PWM signal correction duty cycle at each moment is determined based on the lighting brightness coefficient, and the calculation method is:

[0032] In the formula, R g，j is the duty cycle of the PWM signal correction of the g-th street lamp at the j-th moment, R g，j-1 is the duty cycle of the PWM signal correction of the g-th street lamp at the j-1th moment, S g，j is the lighting brightness coefficient of the g-th street lamp at the j-th moment, S g，j-1 is the lighting brightness coefficient of the g-th street lamp at the j-1th moment.

[0033] This application has at least the following beneficial effects:

[0034] The present application collects the illumination area images of each street lamp at each time on a rainy day and the ambient light intensity value at each time; identifies the umbrellas in the illumination area images to obtain the umbrella areas in the illumination area images; extracts the key points of each umbrella area based on the shape features of the umbrellas; analyzes the distribution of corner points in the local neighborhood of each key point to determine the detail richness of each key point; the detail richness reflects the pattern complexity of the corresponding umbrella in the local neighborhood of each key point, and uses this as a feature to improve the accuracy of identifying the same target umbrella at different times; obtains the grayscale histogram of all pixel points in the local neighborhood of each key point The pattern contrast of each key point is determined based on the position distribution relationship between any two peak points in the local neighborhood of each key point and the grayscale difference; the pattern contrast reflects the brightness difference in the local neighborhood of each key point, embodies the detail clarity in the local neighborhood of each key point, and improves the accuracy of image detail information extraction; based on the detail richness, pattern contrast, and gradient amplitude of each key point, the feature descriptor of each key point is determined, and the moving direction of each umbrella is obtained in combination with the lighting area image at adjacent moments, and the pedestrian flow evaluation value of each street lamp at each moment is determined; pedestrian flow evaluation The value reflects the pedestrian flow prediction result of each street lamp at each moment, which improves the reliability of street lamp brightness adjustment; the grayscale difference between the pixels in the local neighborhood of each key point and all the pixels in the umbrella area, as well as the difference between the pattern contrasts of the key points, are analyzed to determine the reflection intensity of each key point; the optical flow direction and optical flow intensity of each pixel in the image of the illuminated area at each moment are obtained, and the image of the illuminated area at each moment is divided into image blocks, and the possibility of raindrops in each image block is determined based on the distribution of the optical flow direction and optical flow intensity of the pixels in each image block; the reflection intensity and the possibility of raindrops both reflect the rain intensity of each street lamp at each moment. The size provides a reliable reference standard for adjusting the brightness of street lamps; threshold segmentation is performed based on the possibility of raindrops to determine the characteristic image blocks in all image blocks at each moment; the distribution of the optical flow intensity of pixels in all characteristic image blocks is analyzed, and the rain intensity assessment intensity at each moment is determined in combination with the reflection intensity; the pedestrian flow assessment value at each moment, the rain intensity assessment intensity, and the ambient light intensity value are combined to determine the illumination brightness coefficient at each moment, and the brightness of street lamps is adjusted using PWM dimming technology; on the premise of ensuring the safety of pedestrians walking in the park on rainy days, unnecessary lighting losses are reduced to save energy and protect the environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 A flowchart of the steps of the lighting method based on the DC lighting system provided in this application;

[0037] Figure 2 This is a schematic diagram of the key points of the umbrella area;

[0038] Figure 3 It is a schematic diagram of the image division of the lighting area;

[0039] Figure 4 Corrected duty cycle determination flow chart for PWM signal. DETAILED DESCRIPTION

[0040] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the lighting method based on the DC lighting system proposed by the present application, its specific implementation, structure, features and effects 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.

[0041] 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 application belongs.

[0042] The specific scheme of the lighting method based on the DC lighting system provided by the present application is described in detail below with reference to the accompanying drawings.

[0043] An embodiment of the present application provides a lighting method based on a DC lighting system. Specifically, the following lighting method based on a DC lighting system is provided. Figure 1 , the method comprises the following steps:

[0044] Step S001, collecting the illumination area images of each street lamp at each time on a rainy day and the ambient light intensity value at each time, and performing preprocessing.

[0045] In this embodiment, a collection device is installed on each LED street lamp. The collection device includes a photosensitive sensor and a camera. The photosensitive sensor is used to collect the light intensity value of the ambient light at each moment, which is recorded as the ambient light intensity value at each moment. The camera is used to collect the image of the lighting area irradiated by the LED street lamp to the ground at each moment. At the same time, in order to reduce the energy consumption and computing resources of the high-frequency operation of the collection device, the collection interval of the collection device in this embodiment is set to 1s. The implementer can set it according to the actual situation, and this embodiment does not make any restrictions here.

[0046] This embodiment uses a bilateral filtering algorithm to perform image denoising on the collected illumination area images to eliminate noise interference in the camera acquisition process. The denoised illumination area images are then subjected to histogram equalization to adjust the histogram distribution of the illumination area images to enhance the contrast of the illumination area images and improve the visual effect of the illumination area images, so as to facilitate subsequent analysis of rain conditions and park pedestrian activities. The bilateral filtering algorithm and histogram equalization are both well-known technologies, and the specific process will not be repeated here. The implementer can select other existing feasible filtering algorithms and image enhancement algorithms according to actual conditions. Finally, the illumination area images at each moment are converted into grayscale images for subsequent analysis.

[0047] Step S002, identifying umbrellas in the lighting area image to obtain each umbrella area in the lighting area image; extracting each key point of each umbrella area based on the shape features of the umbrella; analyzing the distribution of corner points in the local neighborhood of each key point to determine the detail richness of each key point.

[0048] The flow of people in parks at night is usually significantly random, which places extremely high flexibility requirements on the lighting system. In rainy scenes in parks, traditional lighting methods cannot effectively and energy-savingly control LED street lights. However, it is extremely necessary to adaptively adjust the lighting brightness of LED street lights according to rain conditions and pedestrian activities in parks. This can reduce unnecessary lighting losses while ensuring the safety of pedestrians walking in parks on rainy days.

[0049] In park scenes, rainy weather is common, and pedestrians usually hold umbrellas to visit the park. The camera is installed on the LED street lamp. Due to the obstruction of the umbrella, it is difficult to directly visually detect the pedestrians in the park. Therefore, a recognition model is pre-built to recognize the umbrella in the illuminated area image and extract the umbrella boundary box. Among them, the recognition model described in this embodiment is Mask R-CNN, which is an existing well-known technology. The implementer can select other existing feasible recognition models according to actual conditions, such as Faster R-CNN, YOlOv5, etc., which are not limited in this embodiment.

[0050] In a rainy park scene at night, multiple umbrellas may appear simultaneously in an illuminated area image, that is, there are multiple umbrella bounding boxes. The activity trajectory of each umbrella can represent the activity trajectory of a park pedestrian. Therefore, it is necessary to analyze each umbrella in the illuminated area image.

[0051] In the structure of an umbrella, the umbrella surface is usually supported by multiple umbrella ribs. The umbrella surface between any two adjacent umbrella ribs is called an umbrella surface sector. The end of the umbrella surface sector is an umbrella edge. In the illumination area image, the umbrella edge is a smooth edge straight line close to the umbrella boundary box. In this embodiment, according to the coordinates of the umbrella boundary box, multiple umbrella areas are extracted from the illumination area image, and the Canny edge detection algorithm is used for any umbrella area to obtain an edge binarization map, and the edge binarization map is used as the input of the contour search algorithm. In this embodiment, the contour search algorithm specifically uses the findContours function in OpenCV to output the outer contour of the umbrella area, and uses the outer contour as the input of the Hough line detection algorithm to output the umbrella edges of the umbrella area, which are recorded as boundary lines. The Canny edge detection algorithm and the contour search algorithm are both existing well-known technologies, and this embodiment will not be described in detail here.

[0052] The umbrella structure corresponding to the umbrella area includes an umbrella surface, umbrella ribs, an umbrella tail and a bead tail. The umbrella strut is the joint point in the middle of the umbrella ribs that can be folded and opened. This embodiment first obtains the perpendicular bisectors of each boundary line of the umbrella area, and determines the pixel point where the perpendicular bisectors of the umbrella area intersect the most as the umbrella tail feature point, and determines the center pixel point of the line connecting the center pixel point of each boundary line and the umbrella tail feature point as the umbrella surface feature point, and determines the center pixel point connecting the endpoint of each boundary line with the umbrella tail pixel point as the umbrella strut feature point, and determines the endpoint of each boundary line as the bead tail feature point, and all the umbrella tail feature points, umbrella surface feature points, umbrella strut feature points, and bead tail feature points of each umbrella area are recorded as key points. The schematic diagram of the key points of the umbrella area is shown in the figure below. Figure 2 As shown, Figure 2 The pentagonal symbol in the middle is the characteristic point of the umbrella tail, the diamond symbol is the characteristic point of the umbrella surface, the hexagonal symbol is the characteristic point of the umbrella support, and the circular symbol is the characteristic point of the bead tail.

[0053] Since the umbrellas held by pedestrians on rainy days are mostly different, the umbrella surface of each umbrella often has a unique texture and is highly distinguishable. In the image of the illuminated area, the image area outside the outer contour of the umbrella area is usually the background part such as the road and park landscape, which changes with the activities of pedestrians in the park and is not constant. Therefore, this embodiment sets a 7*7 local window centered on each key point, and uses the image area within the outer contour of the umbrella area in the local window as the local neighborhood of each key point. The implementer can set the size of the local window according to the actual situation, and this embodiment does not limit it.

[0054] The illuminated area image is used as the input of the Harris corner detection algorithm, and all corner points on the illuminated area image are output. The corner points can represent the structural details of the image. The number of corner points in the local neighborhood of each key point in each umbrella area is counted, and the maximum value of the number of corner points of all key points in each umbrella area is obtained. The ratio of the number of corner points in the local neighborhood of each key point in each umbrella area to the maximum value is used as the detail richness of each key point in each umbrella area. The Harris corner detection algorithm is a well-known technology, and the specific process will not be described in detail.

[0055] When the number of corner points is greater, it means that the details of the umbrella surface pattern structure in the local neighborhood of the umbrella key point are richer, and therefore, the richness of the details is greater.

[0056] Step S003, obtaining each peak point in the grayscale histogram of all pixels in the local neighborhood of each key point, and determining the pattern contrast of each key point based on the position distribution relationship and grayscale difference between all arbitrary two peak points in the local neighborhood of each key point.

[0057] The grayscale histogram of all pixels in the local neighborhood of each key point in each umbrella area is obtained, and the grayscale histogram is used as the input of the peak search AMPD (Automatic multiscale-based peak detection) algorithm to output each peak point in the grayscale histogram, and each peak point corresponds to a grayscale value. In this embodiment, the grayscale values ​​a and b corresponding to the peak points x and y are taken as examples. In the local neighborhood of the key point, the probability η that the grayscale value of the pixel point in the eight neighborhoods of any pixel point with a grayscale value of a is b is calculated. a，b , each pixel with gray value a has a probability η a，b , calculate the corresponding probability η of all pixels with gray value a a，b The mean of is recorded as the co-occurrence probability of gray values ​​a and b. Among them, the probability η a，b Specifically, it is the ratio of the number of pixels with gray value b in the eight neighborhoods to the number of all pixels in the eight neighborhoods. The AMPD algorithm is a well-known technology, and the specific process will not be described in detail.

[0058] Calculate the difference between the grayscale value a and the grayscale value b, record it as the first difference, calculate the product of the first difference and the co-occurrence probability, calculate the sum of the product of the peak point x and all the remaining peak points in the local neighborhood, normalize the accumulated result of the sum of all the peak points in the local neighborhood of each key point, and use the normalized result as the pattern contrast of each key point in each umbrella area.

[0059] It should be noted that the difference represents the degree of difference between two variables, and can be calculated specifically by using the absolute value of the difference, the square of the difference, the ratio, etc. This embodiment uses the absolute value of the difference as the method for calculating the difference. In the calculation process of the pattern contrast, the normalization result is obtained by using the Sigmoid function. The implementer can choose other existing feasible normalization methods according to actual conditions, and this embodiment does not limit this.

[0060] It should be understood that, the more times the grayscale value a and the grayscale value b appear adjacent to each other in the local neighborhood of the key point, that is, the greater the co-occurrence probability, the more frequent the transition between the grayscale value a and the grayscale value b, and the greater the first difference, the more significant the details of the umbrella pattern in the local neighborhood of the umbrella key point, and the greater the pattern contrast.

[0061] Step S004, based on the detail richness, pattern contrast, and gradient amplitude of each key point, determine the feature descriptor of each key point, combine the lighting area images at adjacent moments, obtain the moving direction of each umbrella, and determine the pedestrian flow evaluation value of each street lamp at each moment.

[0062] In this embodiment, the Sobel operator is used to obtain the gradient amplitude of each key point, and the detail richness, pattern contrast and gradient amplitude of each key point in each umbrella area are sequentially formed into vectors as feature descriptors of each key point. Among them, using the Sobel operator to obtain the gradient amplitude is an existing known technology, and the implementer can select other existing feasible gradient calculation algorithms.

[0063] The lighting area image of the previous acquisition moment of each moment is obtained, and each umbrella area, each key point in the umbrella area and the feature descriptor of each key point of the lighting area image of the previous acquisition moment are obtained by the same method. Each key point and the feature descriptor of each key point of the lighting area image of each moment and the previous moment are used as the input of the FLANN (Fast_Library_for_Approximate_Nearest_Neighbors) fast nearest neighbor search package algorithm, wherein the parameters trees of the FLANN algorithm take the permanent value of 5, and checks take the permanent value of 50, and the feature matching results of the key points are output. When the number of feature matching points of the umbrella area in the lighting area image of each moment and the lighting area image of the previous moment is greater than the first threshold, it is determined that the two umbrella areas are the same target umbrella at different moments, wherein the first threshold is 80% of the number of all key points in the corresponding umbrella area in the lighting area image of each moment, and the implementer can set the first threshold according to the actual situation, and this embodiment is not limited here. The FLANN algorithm is an existing well-known technology, and the specific process is not described in detail.

[0064] In a rainy park scene, the greater the pedestrian flow, the brighter the street lights should be in order to allow pedestrians to see potential dangers more clearly and prevent accidents. This embodiment uses umbrella areas to represent pedestrians, records the center coordinates of the same target umbrella in the umbrella boundary box at each moment and the previous moment, points the center coordinates of the previous moment to the center coordinates of each moment, and obtains the moving direction of the target umbrella. It should be noted that: if there is an umbrella area at each moment that has no corresponding umbrella area at the previous moment, the moving direction of the umbrella is set to the horizontal direction of the position of the umbrella area relative to the image boundary. According to the above steps, the moving direction of each umbrella area in the lighting area image at each moment is obtained.

[0065] To evaluate pedestrian flow, in addition to the number of umbrella areas in the lighting area image at each moment, it is also necessary to predict whether there will be new pedestrians or umbrellas in the future. This application divides the lighting area image under each LED street light into three area blocks in the horizontal direction, from left to right, namely the left row area, the middle row area and the right row area. The lighting area image division diagram is shown in the figure below. Figure 3 As shown, Figure 3 In the figure, 301 is the left row area, 302 is the middle row area, 303 is the right row area, the triangle symbol is the center coordinate of the umbrella area within the umbrella boundary box, and the straight arrow of the triangle symbol is the moving direction of the umbrella area.

[0066] When the center coordinates of the umbrella boundary box of the g-th LED street lamp at each moment fall in the left-traffic area and the moving direction is also to the left, it means that the pedestrian is very likely to move to the LED street lamp lighting area to the left of the g-th LED street lamp. When the moving direction is right, it means that the pedestrian is very likely to still be in the LED street lamp area at the next moment. The number of umbrella areas with a moving direction to the left in the left-traffic area is counted as the pedestrian flow prediction value of the adjacent LED street lamp to the left of the g-th LED street lamp.

[0067] When the center coordinates of the umbrella boundary box fall in the middle traffic area and the moving direction is left or right, the pedestrian is very likely to be in the illumination area of ​​this LED street light in a short period of time. When the center coordinates of the umbrella boundary box fall in the right traffic area and the moving direction is also to the right, it means that the pedestrian is very likely to move to the LED street light illumination area to the right of the g-th LED street light. When the moving direction is left, it means that the pedestrian is very likely to still be in the LED street light area at the next moment. Count the number of umbrella areas with the right moving direction in the right traffic area as the pedestrian flow prediction value of the adjacent LED street light to the right of the g-th LED street light.

[0068] It should be noted that the DC lighting system can manage all the LED street lights in the park and encode them in sequence according to their positions. Taking the g-th LED street light as an example, the g-1th LED street light is used as the left adjacent LED street light of the g-th LED street light, and the g+1th LED street light is used as the right adjacent LED street light of the g-th LED street light.

[0069] The number of umbrella areas in the image of the lighting area of ​​each LED street lamp at each moment is counted as the pedestrian flow value, and the cumulative sum of the pedestrian flow value of each LED street lamp at each moment and all pedestrian flow prediction values ​​is calculated as the pedestrian flow evaluation value of each LED street lamp at each moment. The larger the pedestrian flow evaluation value, the more pedestrians there are in the LED street lamp lighting area at each moment and at subsequent moments. For pedestrian safety, the brightness of the LED street lamp should be increased.

[0070] Step S005, analyzing the grayscale differences between the pixels in the local neighborhood of each key point and all the pixels in the umbrella area, as well as the differences between the pattern contrasts of the key points, to determine the reflection intensity of each key point; obtaining the optical flow direction and optical flow intensity of each pixel in the image of the illuminated area at each moment, dividing the image of the illuminated area at each moment into image blocks, and determining the possibility of raindrops in each image block based on the distribution of the optical flow direction and optical flow intensity of the pixels in each image block.

[0071] In the rainy scene in the park at night, the rain intensity will affect the lighting effect of the LED street lamp. It is necessary to evaluate the rain intensity in the park and adjust the brightness of the LED street lamp according to the evaluation results. Some rainwater remains on the umbrella surface, forming water stains. Under the illumination of the LED street lamp, the smooth surface of the umbrella makes it easier to reflect light, forming a strong reflective area. The texture blur and brightness in the reflective area of ​​the umbrella surface water stains are high. Calculate the reflective intensity of each key point in each umbrella area. The specific calculation method is:

[0072] In the formula, E i is the reflection intensity of the i-th key point, C max is the maximum value of the pattern contrast of all key points in the illumination area image where the i-th key point is located, C i is the pattern contrast of the i-th key point, l i is the grayscale mean of all pixels in the local neighborhood of the i-th key point, is the grayscale mean of all pixels in the umbrella area corresponding to the i-th key point.

[0073] It should be understood that when (C max -C i ) is larger, indicating that the edge significance in the local neighborhood of the umbrella key point is lower and the texture fuzziness is higher. The larger it is, the higher the brightness of the pixels in the local neighborhood of the umbrella key point is, and the more likely the local neighborhood of the umbrella key point is to be the reflective area of ​​water stains on the umbrella surface, and the greater the reflective intensity of the water stains on the umbrella surface is.

[0074] When it rains in the park at night with light rainfall, rainwater is more likely to remain on the umbrella surface, forming a reflective area of ​​water stains on the umbrella surface. Therefore, the reflection intensity can reflect the rainfall when the rainfall is relatively light; when the rainfall is heavy, the raindrops are usually larger and denser, and have a stronger ability to reflect and refract light, making it easy to observe the phenomenon of raindrops with significant movement trends. This embodiment obtains the illumination area image of each street lamp at each moment and the previous moment, and uses the RAFT (Recurrent All-Pairs Field Transforms for Optical Flow) model to estimate the optical flow, and outputs the optical flow direction and optical flow intensity of each pixel point in the illumination area image of each street lamp at each moment. The illumination area image is divided into 256 image blocks. The RAFT model is an existing well-known technology, and the specific process is not described in detail. The possibility of raindrops in each image block is calculated, and the specific calculation method is:

[0075] Where τ is the probability of rain in each image block, μ is the mean value of the optical flow intensity of all pixels in each image block, and σ 2 is the degree of discreteness of the optical flow direction of all pixels in each image block, and Norm() is a normalization function. The discreteness can be calculated by variance, standard deviation, coefficient of variation, etc. This embodiment uses variance as the calculation method for discreteness.

[0076] It should be understood that when μ is larger, it means that the image block is not a static area in the illuminated area image and has a significant motion trend. 2 The smaller it is, the more concentrated the movement direction of the pixels in the image block is, the more likely the image block is to contain raindrop pixels, and the higher the raindrop possibility τ is.

[0077] Step S006, performing threshold segmentation based on the possibility of raindrops to determine the characteristic image blocks in all image blocks at each moment; analyzing the distribution of the optical flow intensity of the pixels in all the characteristic image blocks, and combining the reflection intensity to determine the rain intensity assessment intensity at each moment; comprehensively considering the pedestrian flow assessment value at each moment, the rain intensity assessment intensity, and the ambient light intensity value, to determine the illumination brightness coefficient at each moment, and using PWM dimming technology to adjust the brightness of the street lamps.

[0078] The raindrop probability of all image blocks in the image of the lighting area of ​​each street lamp at each time is obtained and used as the input of the Ostu algorithm. The segmentation threshold is output and the image blocks with the raindrop probability greater than the segmentation threshold are recorded as feature image blocks. The feature image blocks are very likely to contain a large number of raindrop pixels. The raindrop assessment intensity of each street lamp at each time is calculated as follows:

[0079] Where P is the rain intensity assessment of each street lamp at each time. is the mean value of the optical flow intensity of all pixels in all feature image blocks in the image of the illuminated area of ​​each street lamp at each moment, and V is the mean value of the reflection intensity of all key points in all umbrella areas in the image of the illuminated area of ​​each street lamp at each moment.

[0080] It should be understood that in a scene with light rain, the larger V is, the greater the reflective intensity of the water stains on the umbrella surface is, indicating that the rainfall is greater at this time. At the same time, in a scene with heavy rain, The larger it is, the more significant the movement trend of the raindrop pixel is. The faster the raindrop speed is, the greater the rainfall is at this time. The greater the rain assessment intensity P is, the higher the brightness of the LED street lamp should be adjusted to ensure the safety of pedestrians walking in the park on rainy days.

[0081] According to the pedestrian flow information, rain information and ambient light information at each moment in the LED street lamp lighting area, the lighting brightness coefficient of the LED street lamp is obtained, wherein specifically: for each street lamp, the multiplication result of the pedestrian flow assessment value at each moment and the rain assessment intensity is calculated, and the ratio of the multiplication result to the ambient light intensity value is determined as the lighting brightness coefficient at each moment. The larger the lighting brightness coefficient, the higher the brightness of the LED street lamp should be to ensure the safety of pedestrians walking in the park on rainy days.

[0082] LED street lamps use PWM (Pulse Width Modulation) dimming technology to adjust brightness. The PWM signal correction duty cycle of each street lamp at each moment is determined based on the lighting brightness coefficient of each street lamp at each moment. The specific calculation method is:

[0083] In the formula, R g，j is the duty cycle of the PWM signal correction of the g-th street lamp at the j-th moment, R g，j-1 is the corrected duty cycle of the PWM signal of the g-th street lamp at the j-1th moment, S g，j is the lighting brightness coefficient of the g-th street lamp at the j-th moment, S g，j-1 is the lighting brightness coefficient of the g-th street lamp at the j-1th moment. The PWM signal correction duty cycle determination flow chart is as follows: Figure 4 It should be noted that, in order to ensure the calculation of the duty cycle of the PWM signal correction, the brightness of the LED street lamp is not adjusted at the initial moment.

[0084] It should be understood that when The larger the value, the heavier the rain in the park, the more pedestrian traffic, and the weaker the ambient light. In order to ensure the safety of pedestrians walking in the park on rainy days, the brightness of the LED street lamps should be increased. The smaller the value is, the brightness of the LED street lamp should be appropriately reduced to achieve energy-saving control of the LED street lamp and adopt a lower PWM signal duty cycle.

[0085] At this point, the PWM signal correction duty cycle of each LED street lamp at the current moment is obtained. Through the single-lamp controller, the PWM signal correction duty cycle is used as the duty cycle of the PWM dimming technology to adjust the brightness of each LED street lamp. Under the premise of ensuring the safety of pedestrians walking in the park on rainy days, unnecessary lighting losses are reduced, saving energy and protecting the environment.

[0086] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, 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.

[0087] 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.

[0088] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Modifications to the technical solutions recorded in the aforementioned embodiments, or equivalent replacement of some of the technical features therein, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A lighting method based on a DC lighting system, characterized in that: The method comprises the following steps: Collect the lighting area images of each street lamp at each time on rainy days and the ambient light intensity value at each time; Identify umbrellas in the lighting area image to obtain each umbrella area in the lighting area image; extract key points of each umbrella area based on the shape features of the umbrella; analyze the distribution of corner points in the local neighborhood of each key point to determine the detail richness of each key point; Obtain each peak point in the grayscale histogram of all pixel points in the local neighborhood of each key point, and determine the pattern contrast of each key point based on the position distribution relationship and grayscale difference between all any two peak points in the local neighborhood of each key point; Based on the detail richness, pattern contrast, and gradient amplitude of each key point, the feature descriptor of each key point is determined. Combined with the lighting area images at adjacent moments, the moving direction of each umbrella is obtained to determine the pedestrian flow evaluation value of each street lamp at each moment. Analyze the grayscale difference between the pixels in the local neighborhood of each key point and all the pixels in the umbrella area, as well as the difference between the pattern contrasts of the key points, to determine the reflection intensity of each key point; obtain the optical flow direction and optical flow intensity of each pixel in the image of the illuminated area at each moment, divide the image of the illuminated area at each moment into image blocks, and determine the possibility of raindrops in each image block based on the distribution of the optical flow direction and optical flow intensity of the pixels in each image block; Threshold segmentation is performed based on the possibility of raindrops to determine the characteristic image blocks in all image blocks at each moment; the distribution of the optical flow intensity of the pixels in all the characteristic image blocks is analyzed, and the rain intensity assessment intensity at each moment is determined in combination with the reflection intensity; the pedestrian flow assessment value at each moment, the rain intensity assessment intensity, and the ambient light intensity value are comprehensively considered to determine the illuminated surface brightness coefficient at each moment, and the brightness of the street lamps is adjusted using PWM dimming technology.

2. The lighting method based on the DC lighting system according to claim 1, characterized in that: The step of extracting key points of each umbrella region based on the shape feature of the umbrella includes: Obtain the perpendicular bisectors of each boundary line of the umbrella region, determine the pixel point where the perpendicular bisectors of the umbrella region intersect the most as the umbrella tail feature point, determine the central pixel point of the line connecting the central pixel point of each boundary line and the umbrella tail feature point as the umbrella surface feature point, determine the central pixel point connecting the endpoint of each boundary line and the umbrella tail pixel point as the umbrella strut feature point, and determine the endpoint of each boundary line as the bead tail feature point; The tail feature points, umbrella surface feature points, umbrella support feature points, and bead tail feature points of the umbrella area are all recorded as key points.

3. The lighting method based on the DC lighting system according to claim 1, characterized in that: The detail richness is the ratio of the number of corner points in the local neighborhood of each key point to the maximum number of corner points in the local neighborhood of all key points in the corresponding umbrella area.

4. The lighting method based on the DC lighting system according to claim 1, characterized in that: The process of determining the pattern contrast is as follows: In the local neighborhood of each key point, for the grayscale value a corresponding to any peak point and the grayscale value b corresponding to any remaining peak point, the probability of the pixel point with grayscale value b in the neighborhood of the pixel point with grayscale value a is counted, and the average of the probabilities of all the pixel points with grayscale value a in the local neighborhood of each key point is taken as the co-occurrence probability of grayscale values ​​a and b; Calculate the difference between the grayscale value a and the grayscale value b, record it as the first difference, calculate the product of the first difference and the co-occurrence probability, calculate the sum of the product of the peak point corresponding to the grayscale value a and all the remaining peak points, normalize the accumulated result of the sum of all the peak points in the local neighborhood of each key point, and obtain the pattern contrast.

5. The lighting method based on the DC lighting system according to claim 1, characterized in that: The process of determining the pedestrian flow evaluation value is as follows: Based on the feature descriptors of each key point, a matching algorithm is used to obtain the same umbrella area in the lighting area image at adjacent moments, so as to determine the moving direction of each umbrella at each moment; the lighting area image at each moment is divided into a preset number of area blocks, and the umbrellas whose moving directions deviate from the lighting area image and are biased towards the direction of the adjacent street lamp in the area blocks close to the boundary of the lighting area image are determined as the pedestrian flow prediction values ​​of the adjacent street lamp; The sum of the number of umbrella areas in the lighting area image of each street lamp at each moment and the pedestrian flow prediction value is used as the pedestrian flow evaluation value.

6. The lighting method based on a DC lighting system according to claim 1, characterized in that: The calculation method of the reflection intensity is: In the formula, E i is the reflection intensity of the i-th key point, C max is the maximum value of the pattern contrast of all key points in the illumination area image where the i-th key point is located, C i is the pattern contrast of the i-th key point, l i is the grayscale mean of all pixels in the local neighborhood of the i-th key point, is the grayscale mean of all pixels in the umbrella area corresponding to the i-th key point.

7. The lighting method based on a DC lighting system according to claim 1, characterized in that: The rain possibility is the ratio of the normalized value of the average level of the optical flow intensity of all pixels in each image block to the normalized value of the discrete degree of the optical flow direction of all pixels.

8. The lighting method based on a DC lighting system according to claim 1, characterized in that: The determination process of the rain assessment intensity is as follows: For the illumination area image at each moment, the rain assessment intensity is the product of the average optical flow intensity of all pixels in all feature image blocks and the average reflection intensity of all key points in all umbrella areas.

9. The lighting method based on a DC lighting system according to claim 1, characterized in that: The process of determining the illumination surface brightness coefficient is as follows: For each street lamp, the multiplication result of the pedestrian flow assessment value at each moment and the rain assessment intensity is calculated, and the ratio of the multiplication result to the ambient light intensity value is determined as the illumination surface brightness coefficient at each moment.

10. The lighting method based on a DC lighting system according to claim 1, characterized in that: The method of using PWM dimming technology to adjust the brightness of street lamps includes: The PWM signal correction duty cycle at each moment is determined based on the lighting brightness coefficient, and the calculation method is: In the formula, R g，j is the duty cycle of the PWM signal correction of the g-th street lamp at the j-th moment, R g，j-1 is the corrected duty cycle of the PWM signal of the g-th street lamp at the j-1th moment, S g，j is the lighting brightness coefficient of the g-th street lamp at the j-th moment, S g，j-1 is the lighting brightness coefficient of the g-th street lamp at the j-1th moment.

Images