Multi-axial visible light vehicle detection and lighting method

Abstract

The invention discloses a multi-axial visible light vehicle detection and lighting method. On the basis of the objective fact that the light loss in the top direction is more than that in the side direction when a vehicle is parked, n visible light brightness sensors in different directions are arranged on the inner side of a face shell of the device, one visible light brightness sensor faces the sky to collect top data, and the rest visible light brightness sensors are used for collecting side data; then, after n pieces of brightness data are collected at regular time and subjected to self-adaptive operation and normalization processing, a preset algorithm and parameters are used according to different scene light rays, and a visible light vehicle detection subitem result with high confidence coefficient can be provided; according to the simplified lighting method, lighting is conducted on the left side and the right side of the berth instead of other directions, the brightness effect close to that of an empty berth is obtained, and a scientific, rapid and accurate judgment basis is provided for outputting visible light and judging whether a vehicle exists or not when necessary.

Description

Technical Field

[0001] The present invention relates to a parking vehicle detection method under the combined application of multiple sensors in the field of intelligent transportation and smart parking, and in particular to a multi-axial visible light vehicle detection and lighting method. Background Art

[0002] The existing solution of periodically collecting brightness data for before-and-after comparison through a single top-facing visible light brightness sensor has the following shortcomings in many years of project practice: 1. There is no in-depth analysis and comprehensive consideration of the huge changes in light during the day and night, and the method and means are single and rough, and the effect of using it at night is often significantly worse than that during the day; 2. In the case of non-font parking spaces where cars are closely adjacent to each other, the entry and exit of cars in the adjacent parking spaces can often cause the parking space to detect false information of visible light entering and exiting, causing unnecessary misjudgment; 3. The visible light generated by vehicles entering the parking space during the day and its scoring value can only reflect the changes in light at the moment of vehicle entering. Basically, it is invalid once used up, and it is difficult to continue to maintain it to shield the geomagnetic interference of the parking space power lines or the interference of vehicles entering and exiting the left and right parking spaces during parking. In the final analysis, only the top brightness data under the vehicle chassis can be collected during parking, and it is impossible to obtain the visible light brightness data under the actual clearance of the parking space when there is no vehicle chassis coverage. Therefore, it is difficult to provide a scientific basis for whether there is a vehicle in the current visible light and whether the scoring is maintained when needed; all these will have an adverse impact on the performance of visible light detection, and new solutions are urgently needed to overcome and solve them. Summary of the invention

[0003] The technical problem to be solved by the present invention is to avoid the shortcomings of the above-mentioned prior art and propose a multi-axial visible light vehicle detection method, which is applicable to parking type roadside equipment, multi-mode vehicle detectors or multi-mode geomagnetism, and includes: n visible light brightness sensors facing different directions are arranged on the inner side of the device shell, where n≥2, and one top sensor faces the sky to collect top visible light data, and the rest are side sensors to collect side visible light data; The device collects visible light brightness data of n sensors once every period of time; The device can adaptively calculate and obtain a normalized scale that matches the scene light according to the changes in the brightness of visible light during the day and night and in the surrounding environment, and perform normalization processing on the group of brightness data collected; The device analyzes and determines the normalized n brightness data according to the preset algorithm and parameters of the scene light corresponding to the normalized scale and obtains the visible light vehicle presence, vehicle absence or failure detection results, thereby assisting the device in multi-mode detection and outputting comprehensive vehicle detection results.

[0004] Optionally, the device face shell is a light-transmitting face shell made of a transparent material, or a face shell made of an opaque material but with a transparent lens provided through an opening to achieve a partial light-transmitting effect.

[0005] Optionally, the above-mentioned visible light brightness sensor can measure the light intensity of visible light in the wavelength range of 380nm-780nm, and convert the optical signal into an electrical signal output, including: a light intensity sensor, a photoelectric sensor, a photodiode, a phototransistor, a photoresistor, and a solar panel.

[0006] Optionally, the above n is equal to 3, that is, there are three visible light brightness sensors, namely: the above n is equal to 3, that is, there are three visible light brightness sensors, namely: the first top visible light brightness sensor whose top axis is parallel to the central axis of the device and points directly to the sky to collect top visible light data; the second lateral visible light brightness sensor whose left axis forms an angle of 70°-90° with the top axis to collect left visible light data; the third lateral visible light brightness sensor whose right axis forms an angle of 70°-90° with the top axis to collect right visible light data; when the equipment is installed and used, the left axis is set perpendicular to the left sideline of the berth, and the allowable deviation is ≤20°; the right axis is set perpendicular to the right sideline of the berth, and the allowable deviation is ≤20°.

[0007] Optionally, the above-mentioned adaptive operation includes the following steps: a. preprocessing: transform the three brightness data into the hexadecimal range of 00-FF through linear mapping; b. find the maximum value M among the three transformed brightness data; c. derive the normalization scale N according to the M value, the method is: if M≥128, then N=4; if 128>M≥64, then N=3; if 64>M≥32, then N=2; if 32>M≥16, then N=1; in other cases N=0; the normalization processing refers to shifting the three transformed hexadecimal brightness data right by N bits, and finally obtaining the hexadecimal data in the range of 0 to F to participate in the process of judging whether there is a car or not in the visible light.

[0008] Optionally, the normalized scale N (4≥N≥0) corresponds to the following five scene lights LN: L4—broad daylight, L3—dawn, dusk, cloudy or rainy daytime, L2—bright lights at dawn, dusk or night, L1—general street lighting at night, and L0—weak light at night; the preset algorithm and parameters are: for L4 and L3, if the brightness data of the side sensor with the minimum brightness is higher than that of the top sensor by a preset parameter P43, there is a car in the visible light, otherwise there is no car in the visible light; for L2, if the brightness data of the side sensor with the minimum brightness is higher than that of the top sensor by a preset parameter P2, there is a car in the visible light, otherwise there is no car in the visible light; for L1 and L0 There are three situations: A. If the brightness data of the lateral sensor with the minimum brightness is higher than that of the top sensor by the preset parameter P10-1 (when the top brightness data is non-zero) or P10-0 (when the top brightness data is zero), there is a car in the visible light, otherwise, if the three brightness data are not all zero, there is no car in the visible light; B. If the current N=0 and the three brightness data are all zero, but the three brightness data sampled in the last comprehensive no-car state are not all zero, there is a car in the visible light at this moment, otherwise there is no car in the visible light; C. At night, if according to past records, in most cases after multi-mode detection, it is comprehensively judged that there is no car or the three brightness data are all zero when there is a car, and they are still all zero at this moment, then it can be judged that the visible light fails.

[0009] At the same time, the present invention proposes a simplified lighting method for multi-axial visible light vehicle detection, which is applicable to parking-type roadside equipment, multi-mode vehicle detectors or multi-mode geomagnetism, and is characterized by comprising: Three visible light brightness sensors facing different directions are arranged on the inner side of the device cover; The first top visible light brightness sensor has a first top axis parallel to the top axis of the device and points directly to the sky to collect top visible light data; the second lateral visible light brightness sensor has a second left axis that forms an angle of 70°-90° with the first top axis to collect left visible light data; the third lateral visible light brightness sensor has a third right axis that forms an angle of 70°-90° with the first top axis to collect right visible light data; When the equipment is installed and used, the second left axis is set perpendicular to the left sideline of the berth with an allowable deviation of ≤20°; the third right axis is set perpendicular to the right sideline of the berth with an allowable deviation of ≤20°.

[0010] Optionally, the device face shell is a light-transmitting face shell made of a transparent material, or a face shell made of an opaque material but with a transparent lens provided through an opening to achieve a partial light-transmitting effect.

[0011] Compared with the prior art, the multi-axial visible light vehicle detection and lighting method of the present invention has the following technical effects: 1. According to the changing law of day and night light, a set of preprocessing and normalization methods suitable for large light ratio brightness data is proposed, and five different scene lights are defined through normalization scale, among which L4 and L3 are dominated by sunlight, L2 is the junction or mixture of sunlight and light, and L1 and L0 are dominated by light, which provides a focus for the subsequent processing of different scene lights; 2. The present invention firmly grasps the objective fact that when there is a car in the parking space, the top visible light brightness sensor loses more light and the brightness drops more sharply than the side ones. It prescribes the right remedy for the case and proposes targeted preset algorithms and parameters for different scene lights, which can provide high-confidence visible light vehicle detection sub-item results for the smooth implementation of multi-mode detection and comprehensive judgment of the presence or absence of a car for related equipment; 3. The present invention proposes a streamlined and optimized lighting method based on the length and width of the vehicle body and the characteristics of the parking space, and chooses to perform side lighting on the left and right sides of the parking space rather than the front and rear sides. The actual effect is quite close to the visible light brightness of the current parking space when it is empty, which provides a scientific, objective, fast and accurate judgment basis for outputting visible light to determine whether there is a car when necessary. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a flow chart of a multi-axial visible light vehicle detection method of the present invention; Figure 2 It is a schematic plan view of arranging three-axis photosensitivity on a multi-mode geomagnetic transparent surface shell for lighting in an embodiment of the present invention; Figure 3 This is a field installation diagram of setting a three-axis photosensor on a multi-mode geomagnetic transparent surface shell for lighting in an embodiment of the present invention; Figure 4 It is a summary analysis table of measured data of the embodiment of the present invention. DETAILED DESCRIPTION

[0013] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, it is preferred to use the implementation of visible light vehicle detection on a multi-mode geomagnetic device with a translucent surface shell as an example for explanation. It should be understood that this is only done for the purpose of easier understanding and is not used to limit the scope of protection of the present application.

[0014] like Figure 1 As shown, it is a flow chart of a multi-axial visible light vehicle detection method of the present invention; the method may include steps S100-S400.

[0015] Step S100, n visible light brightness sensors facing different directions are arranged on the inner side of the multi-mode geomagnetic device housing, where n≥2, and one top sensor faces the sky to collect top visible light data, and the rest are lateral sensors to collect lateral visible light data; Optionally, the above-mentioned multi-mode geomagnetic device face shell is a light-transmitting face shell made of transparent material, or a face shell made of opaque material but with a transparent lens set through an opening to achieve a local light-transmitting effect; as the best, this embodiment adopts the most mature and widely used light-transmitting face shell.

[0016] Optionally, the above-mentioned visible light brightness sensor can measure the light intensity of visible light in the wavelength range of 380nm-780nm, and convert the light signal into an electrical signal for output, including: a light intensity sensor, a photoelectric sensor, a photodiode, a phototransistor, a photoresistor, and a solar panel; as the best, this embodiment selects a photoresistor with the lowest power consumption, good stability and high cost performance as the visible light brightness sensor; In order to take into account the changes in light when the sun rises in the east and sets in the west, as the best, this embodiment adopts a simplified lighting method on the light-transmitting surface shell, specifically: let n=3, that is, there are three visible light brightness sensors, such as Figure 2 As shown, they are: a first top visible light brightness sensor 91, whose top axis 1 is parallel to the central axis 0 of the device and points directly to the sky to collect top visible light data; a second lateral visible light brightness sensor 92, whose left axis 2 forms an angle of 90° with the top axis 1 to collect left visible light data; a third lateral visible light brightness sensor 93, whose right axis forms an angle of 90° with the top axis 1 to collect right visible light data; Figure 3 As shown, when the equipment is installed and used, the left axis 2 is set perpendicular to the left sideline 7 of the berth; the right axis 3 is set perpendicular to the right sideline 8 of the berth.

[0017] Step S200, the multi-mode geomagnetic device collects visible light brightness data from three sensors every 20 seconds; Step S300, the multi-mode geomagnetic device can adaptively calculate and obtain a normalized scale matching the scene light according to the changes in the brightness of visible light during the day and night and in the surrounding environment, and perform normalization processing on the group of brightness data collected; The above adaptive operation includes the following steps: a. Preprocessing: transform the 3 brightness data into the hexadecimal range of 00-FF through linear mapping; b. Find the maximum value M among the three transformed brightness data; c. Calculate the normalized scale N based on the value of M. The method is: if M ≥ 128, then N = 4; if 128> M ≥ 64, then N = 3; if 64> M ≥ 32, then N = 2; if 32> M ≥ 16, then N = 1; in other cases, N = 0; The above-mentioned normalization processing refers to shifting the three transformed hexadecimal brightness data right by N bits, which is equivalent to dividing by 2 to the Nth power, and finally obtaining hexadecimal data in the range of 0 to F to participate in the process of judging whether there is a car in the visible light.

[0018] Step S400, the multi-mode geomagnetic device analyzes and determines the three normalized brightness data according to the preset algorithm and parameters of the scene light LN corresponding to the normalized scale N, and obtains the visible light vehicle presence, vehicle absence or failure detection results, thereby assisting the device in multi-mode detection and then outputting the comprehensive vehicle detection results.

[0019] The above normalized scale N (4 ≥ N ≥ 0) corresponds to the following five scene light LN: L4 - broad daylight, L3 - dusk, cloudy or rainy daytime, L2 - dawn, dusk or bright lights at night, L1 - general street lighting at night, L0 - weak light at night; The above preset algorithm and parameters are as follows: for L4 and L3, if the brightness data of the side sensor with the minimum brightness is higher than that of the top sensor by a preset parameter P43 (set to 3 in this embodiment), it means that there is a car in the visible light, otherwise there is no car in the visible light; for L2, if the brightness data of the side sensor with the minimum brightness is higher than that of the top sensor by a preset parameter P2 (set to 3 in this embodiment), it means that there is a car in the visible light, otherwise there is no car in the visible light; for L1 and L0, there are three cases: A. If the brightness data of the side sensor with the minimum brightness is higher than that of the top sensor by a preset parameter P10-1 (the top brightness data is not If N=0 (this embodiment sets it to 3) or P10-0 (when the top brightness data is zero, this embodiment sets it to 1), there is a car in the visible light, otherwise if the three brightness data are not all zero, there is no car in the visible light; B. If the current N=0 and the three brightness data are all zero, but the three brightness data sampled in the last comprehensive no-car state are not all zero, there is a car in the visible light at this moment, otherwise there is no car in the visible light; C. At night, if according to past records, in most cases after multi-mode detection, it is comprehensively determined that there is no car or the three brightness data are all zero when there is a car, and they are still all zero at this moment, then it can be determined that the visible light is invalid.

[0020] Figure 4: This is a summary analysis table of the measured data of this embodiment. Specifically, the brightness data of the second lateral visible light brightness sensor 92 corresponds to "left" in the table, the brightness data of the third lateral visible light brightness sensor 93 corresponds to "right" in the table, and the brightness data of the first top visible light brightness sensor 91 corresponds to "top" in the table. These three data are hexadecimal data obtained after the adaptive calculation and normalization processing in step S300. It can be seen from the table that the first item N=3 corresponds to P43 of 3, and left>right, right-top=4, which is greater than P43 of 3, so it is determined that there is a car in the visible light, and so on; the 7th, 13th, and 15th data meet the A situation for the processing of L1 and L0 (the top brightness data is zero, P10-0 = 1), so it is determined that there is a car in the visible light; look at the 9th and 11th data, which meet the B situation of L1 and L0 processing, that is, the three brightness data are all zero at this moment, but the three brightness data sampled in the last comprehensive no-car state are not all zero, so it is determined that there is a car in the visible light; finally, look at the 16th data, N=2 and its corresponding P2 is 3, right-top=2, which is less than P2's 3, so it is determined that there is no car in the visible light.

[0021] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention shall be included in the scope of the claims of the present invention.

Claims

1. A multi-axial visible light vehicle detection method, applicable to parking-type roadside equipment, multi-mode vehicle detectors or multi-mode geomagnetism, characterized in that: The method comprises: n visible light brightness sensors facing different directions are arranged on the inner side of the device shell, where n≥2, and one top sensor faces the sky to collect top visible light data, and the rest are side sensors to collect side visible light data; The device collects visible light brightness data of n sensors once every certain period of time; The device can adaptively calculate and obtain a normalized scale that matches the scene light according to the changes in the brightness of visible light during the day and night and in the surrounding environment, and perform normalization processing on the group of brightness data collected; The device analyzes and determines the normalized n brightness data according to the preset algorithm and parameters of the scene light corresponding to the normalized scale and obtains the visible light vehicle presence, vehicle absence or failure detection results, thereby assisting the device in multi-mode detection and outputting comprehensive vehicle detection results.

2. A multi-axial visible light vehicle detection method as claimed in claim 1, characterized in that: The device face shell is a light-transmitting face shell made of a transparent material, or a face shell made of an opaque material but with a transparent lens provided through an opening to achieve a partial light-transmitting effect.

3. A multi-axial visible light vehicle detection method as claimed in claim 1, characterized in that: The visible light brightness sensor can measure the light intensity of visible light in the wavelength range of 380nm-780nm and convert the light signal into an electrical signal for output, including: a light intensity sensor, a photoelectric sensor, a photodiode, a phototransistor, a photoresistor, and a solar panel.

4. The multi-axial visible light vehicle detection method according to any one of claims 1 to 3, characterized in that: The n is equal to 3, that is, there are three visible light brightness sensors, namely: the first top visible light brightness sensor, whose top axis is parallel to the central axis of the equipment and points directly to the sky to collect top visible light data; the second lateral visible light brightness sensor, whose left axis forms an angle of 70°-90° with the top axis to collect left visible light data; the third lateral visible light brightness sensor, whose right axis forms an angle of 70°-90° with the top axis to collect right visible light data; when the equipment is installed and used, the left axis is set perpendicular to the left sideline of the berth, and the allowable deviation is ≤20°; the right axis is set perpendicular to the right sideline of the berth, and the allowable deviation is ≤20°.

5. A multi-axial visible light vehicle detection method as claimed in any one of claims 4, characterized in that: The adaptive operation includes the following steps: a. preprocessing: transforming the three brightness data into the hexadecimal range of 00-FF through linear mapping; b. finding the maximum value M among the three transformed brightness data; c. deriving the normalization scale N according to the M value, the method is: if M≥128, then N=4; if 128>M≥64, then N=3; if 64>M≥32, then N=2; if 32>M≥16, then N=1; in other cases N=0; the normalization processing refers to shifting the three transformed hexadecimal brightness data right by N bits, and finally obtaining the hexadecimal data in the range of 0 to F for participating in the process of judging whether there is a car or not in the visible light.

6. Any multi-axial visible light vehicle detection method according to claim 5, characterized in that: The normalized scale N (4≥N≥0) corresponds to the following five scene lights LN: L4 - broad daylight, L3 - dusk, cloudy or rainy daytime, L2 - dawn, dusk or bright lights at night, L1 - general street lighting at night, L0 - weak light at night; the preset algorithm and parameters are: for L4 and L3, if the brightness data of the side sensor with the minimum brightness is higher than that of the top sensor by a preset parameter P43, there is a car in the visible light, otherwise there is no car in the visible light; for L2, if the brightness data of the side sensor with the minimum brightness is higher than that of the top sensor by a preset parameter P2, there is a car in the visible light, otherwise there is no car in the visible light; for L1 and L0, the algorithm is divided into: There are three situations: A. If the brightness data of the side sensor with the minimum brightness is higher than that of the top sensor by the preset parameter P10-1 (when the top brightness data is non-zero) or P10-0 (when the top brightness data is zero), there is a car in the visible light, otherwise, if the three brightness data are not all zero, there is no car in the visible light; B. If the current N=0 and the three brightness data are all zero, but the three brightness data sampled in the last comprehensive no-car state are not all zero, there is a car in the visible light at this moment, otherwise there is no car in the visible light; C. At night, if according to past records, in most cases after multi-mode detection, it is comprehensively judged that there is no car or the three brightness data are all zero when there is a car, and they are still all zero at this moment, then it can be judged that the visible light is invalid.

7. A simplified lighting method for multi-axial visible light vehicle detection, applicable to parking-type roadside equipment, multi-mode vehicle detectors or multi-mode geomagnetism, characterized in that: include: Three visible light brightness sensors facing different directions are arranged on the inner side of the device cover; The first top visible light brightness sensor has a first top axis parallel to the top axis of the device and points directly to the sky to collect top visible light data; the second lateral visible light brightness sensor has a second left axis that forms an angle of 70°-90° with the first top axis to collect left visible light data; the third lateral visible light brightness sensor has a third right axis that forms an angle of 70°-90° with the first top axis to collect right visible light data; When the equipment is installed and used, the second left axis is set perpendicular to the left sideline of the berth with an allowable deviation of ≤20°; the third right axis is set perpendicular to the right sideline of the berth with an allowable deviation of ≤20°.

8. The method for simplifying lighting for multi-axial visible light vehicle detection according to claim 7, characterized in that: The device face shell is a light-transmitting face shell made of a transparent material, or a face shell made of an opaque material but with a transparent lens provided through an opening to achieve a partial light-transmitting effect.

Images