Vehicle-mounted lamp and intelligent light emitting control method thereof

Abstract

The invention relates to the technical field of image processing, in particular to a vehicle-mounted lamp and an intelligent light-emitting control method thereof, and the method comprises the following steps: an image obtaining and preprocessing module and a vehicle camera capture continuous image data, carry out the formatting processing of an image, and generate formatted image data; and optimizing the formatted image data through contrast adjustment to generate optimized image data. According to the invention, the continuous image data is captured through the vehicle camera, formatting and contrast adjustment optimization are carried out on the image, the target detection accuracy is improved, and interference factors caused by ambient light changes are reduced; and target detection is executed on the basis of the optimized image, vehicles and road signs on the road are recognized in real time, and the real-time distance state of the front vehicle is evaluated, so that the perception of the front road condition is more accurate, and the intelligence and real-time performance of vehicle light regulation and control are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a vehicle-mounted lamp and an intelligent light-emitting control method thereof. Background Art

[0002] The field of image processing technology refers to the technology of using computers to analyze, enhance, reconstruct and understand images, including steps such as image acquisition, preprocessing, feature extraction, recognition and classification, image enhancement, target detection and tracking, etc. This technology can achieve accurate perception, recognition and decision-making of the environment or target through the analysis and processing of image information. It is widely used in the fields of autonomous driving, security monitoring, medical diagnosis, face recognition, industrial detection, etc. to improve the intelligence, automation and accuracy of the system.

[0003] In the existing technology, when a vehicle is driving at high speed, if the vehicle ahead suddenly brakes or the road conditions deteriorate rapidly, the existing technology is subject to information analysis lag, resulting in slow response of headlight flashing or untimely adjustment of lighting, so that the driver cannot obtain accurate and effective visual warning signals at the critical moment, increasing the risk of collision or accident. Therefore, it is necessary to improve the lighting control. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a vehicle-mounted lamp and an intelligent light-emitting control method thereof.

[0005] In order to achieve the above object, the present invention adopts the following technical solution, an intelligent light control method, comprising the following steps:

[0006] The image acquisition and preprocessing module, in which the vehicle camera captures continuous image data, formats the image, and generates formatted image data; optimizes the formatted image data by contrast adjustment, and generates optimized image data;

[0007] An image analysis and parsing module performs target detection based on the optimized image data, identifies vehicles and road signs, and obtains target detection results; analyzes the target detection results, evaluates the distance to the preceding vehicle, and generates a preceding vehicle status analysis result;

[0008] A headlight control decision module determines whether the headlights need to be flashed based on the preceding vehicle status analysis result and generates a headlight control instruction;

[0009] The emergency response and feedback module monitors the emergency brake signals of the vehicle ahead and sudden road conditions, adjusts the flashing frequency of the headlights according to the monitored emergency situations, and obtains the emergency response execution results.

[0010] Preferably, the step of acquiring the formatted image data is: the vehicle camera captures continuous image data, and adjusts the ISO sensitivity and aperture size of the camera according to lighting conditions to obtain unprocessed raw image data;

[0011] According to the unprocessed original image data, the RGB color mode is converted into a YUV color mode to obtain formatted image data;

[0012] Based on the formatted image data, a high pass filter and edge enhancement are applied to obtain formatted image data.

[0013] Preferably, the step of obtaining the optimized image data is: calculating the contrast score according to the formatted image data, and the calculation formula is:

[0014]

[0015] in, To rate the contrast, For the Line The brightness value of the column pixels, and are the maximum and minimum brightness values ​​in the image, respectively. is the overall average brightness of the image, and are the number of rows and columns of the image, respectively. and Respectively represent the brightness values ​​of the upper pixel and the left pixel adjacent to the current pixel;

[0016] Based on the contrast score, the grayscale mapping curve of the image is adjusted to make the brightness value distribution uniform, thereby obtaining optimized image data.

[0017] Preferably, the step of obtaining the target detection result is: inputting the optimized image data into a deep learning model to extract target area features, using a convolutional neural network to perform layer-by-layer convolution operations on image blocks of candidate areas, combining pooling operations to reduce redundant information, calculating the category probability distribution of each candidate area, and converting it into a vector of fixed dimension through a fully connected layer to generate target recognition result data;

[0018] Based on the target recognition result data, targets whose category confidences meet a set threshold are screened to obtain target detection results.

[0019] Preferably, the step of obtaining the state analysis result of the preceding vehicle is: based on the target detection result, extracting the position coordinate information of the preceding vehicle, combining the focal length of the camera and the pixel density of the image sensor, calculating the scale ratio of the preceding vehicle in the image coordinate system, and converting it into distance through perspective projection relationship to obtain preliminary distance data of the preceding vehicle;

[0020] According to the preliminary distance data of the preceding vehicle, the actual travel distance of the preceding vehicle is calculated using the following formula:

[0021]

[0022] in, is the actual driving distance of the front vehicle, is the focal length of the camera, is the front vehicle width, is the pixel width of the target detection box, is the single pixel size of the image sensor;

[0023] Based on the actual driving distance of the preceding vehicle, the speed change trend of the preceding vehicle is analyzed, and the acceleration change in the time series is calculated in combination with the displacement of the preceding vehicle between consecutive frames. The driving state of the preceding vehicle is evaluated according to the dynamic relationship between speed and distance, and a preceding vehicle state analysis result is generated.

[0024] Preferably, the step of acquiring the headlight control instruction is: based on the analysis result of the state of the preceding vehicle, extracting the acceleration information of the preceding vehicle and the acceleration information of the vehicle, calculating the relative acceleration change between the two, extracting the instantaneous speed data of the preceding vehicle and the instantaneous speed data of the vehicle, and obtaining the preliminary determination data of the headlight control;

[0025] According to the preliminary determination data of the headlight control, the headlight flashing trigger value is calculated, and the calculation formula is:

[0026]

[0027] in, The trigger value for the flashing of the lights. is the current acceleration of the front vehicle, is the current acceleration of the vehicle, is the instantaneous speed of the preceding vehicle, is the instantaneous speed of the vehicle, is the relative distance between the preceding vehicle and the vehicle itself, is the speed of the preceding vehicle at the previous moment, is the vehicle’s speed at the last moment;

[0028] Based on the headlight flashing trigger value, a flashing trigger threshold is set, and condition judgment is performed. If the headlight flashing trigger value exceeds the trigger threshold, the headlight flashing is triggered to obtain a headlight control instruction.

[0029] Preferably, the step of acquiring the emergency response execution result is: monitoring the emergency brake signal of the preceding vehicle and the emergency situation on the road, including the sudden appearance of obstacles or the crossing of pedestrians, to obtain emergency situation data;

[0030] According to the emergency data, the headlight flashing adjustment index is calculated using the following formula:

[0031]

[0032] in, Adjust the flashing index for the headlights. is the actual distance to the vehicle ahead or obstacle, is the safety distance threshold, is the speed difference, is the reaction time constant;

[0033] Based on the vehicle light flicker adjustment index, the flicker frequency is set and an emergency response execution result is generated.

[0034] The present invention also provides a vehicle-mounted lamp, applicable to any of the above-mentioned intelligent light control methods, comprising: a lamp housing with a heat dissipation through hole on the outer wall, a plug for conducting electricity installed at one end of the lamp housing, a first heat dissipation fan and a second heat dissipation fan correspondingly arranged inside the lamp housing, a light source substrate arranged horizontally between the first heat dissipation fan and the second heat dissipation fan, and LED lamp chips arranged on both sides of the light source substrate;

[0035] A control circuit board corresponding to the heat dissipation through hole is provided at one end of the light source substrate, a rectifier circuit board electrically connected to the plug is provided outside the control circuit board, a connection terminal electrically connected to the first heat dissipation fan is provided on the light source substrate, and the second heat dissipation fan is located between the rectifier circuit board and the control circuit board, the rectifier circuit board is connected to the control circuit board via a pin row, and the second heat dissipation fan is electrically connected to the rectifier circuit board.

[0036] Furthermore, the outside of the lamp housing is also provided with air vents corresponding to the first cooling fan and the second cooling fan one by one, and the outside of the lamp housing is also provided with openings corresponding to the light-emitting surface of the LED lamp chip, and the two ends of the guide bar hole are respectively connected to the first cooling fan and the second cooling fan.

[0037] Furthermore, the control circuit board is provided with hot air holes for hot air circulation and mounting sockets for assembling the end of the light source substrate.

[0038] Compared with the prior art, the advantages and positive effects of the present invention are:

[0039] The present invention captures continuous image data through a vehicle camera, formats and optimizes the contrast adjustment of the image, improves the accuracy of target detection, and reduces interference factors caused by changes in ambient light; performs target detection based on the optimized image, identifies vehicles and road signs on the road in real time, and evaluates the real-time distance status of the vehicle in front, so as to make the perception of the road condition in front more accurate and improve the intelligence and real-time performance of vehicle lighting control; generates headlight flashing control instructions in a timely manner according to the distance status of the vehicle in front, effectively prompts the driver of potential risks in advance and reduces the driver's reaction delay; at the same time, continuously monitors the emergency braking signal of the vehicle in front and sudden road conditions, dynamically adjusts the headlight flashing frequency, responds to sudden conditions more quickly and accurately, and improves driving safety and driving stability; realizes the deep intelligence of vehicle lighting control strategy, and effectively improves the driving vision quality and safety level at night or under complex road conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of the steps of the present invention;

[0041] Figure 2 It is a schematic diagram of the overall external structure of the vehicle-mounted lamp in the present invention;

[0042] Figure 3 It is a schematic diagram of the overall exploded structure of the vehicle-mounted lamp in the present invention;

[0043] Figure 4 It is a schematic diagram of the overall internal structure of the vehicle-mounted lamp in the present invention.

[0044] Figure 5 For the present invention Figure 3 A is an enlarged structural diagram of FIG.

[0045] Figure 6 For the present invention Figure 3 Schematic diagram of the enlarged structure at B.

[0046] Reference numerals:

[0047] 1. Lamp housing; 3. Plug; 4. Rectifier circuit board; 5. Control circuit board; 10. Heat dissipation hole; 11. Light source substrate; 12. Air outlet; 13. Opening; 21. First cooling fan; 22. Second cooling fan; 50. Hot air hole; 51. Mounting bayonet; 99. Pin header; 110. LED lamp chip; 210. Connection terminal. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0049] See also Figure 1 The present invention provides a technical solution, an intelligent light emitting control method, comprising the following steps:

[0050] Image acquisition and preprocessing module: the vehicle camera captures continuous image data, formats the image, and generates formatted image data; optimizes the formatted image data by contrast adjustment to generate optimized image data;

[0051] The image analysis and parsing module performs target detection based on optimized image data, identifies vehicles and road signs, and obtains target detection results; analyzes the target detection results, evaluates the distance to the preceding vehicle, and generates preceding vehicle status analysis results;

[0052] The headlight control decision module determines whether the headlights need to be flashed based on the status analysis results of the vehicle in front and generates headlight control instructions;

[0053] The emergency response and feedback module monitors the emergency brake signals of the vehicle ahead and sudden road conditions, adjusts the flashing frequency of the headlights according to the monitored emergency situations, and obtains the emergency response execution results.

[0054] The steps of obtaining the formatted image data are as follows: the vehicle camera captures continuous image data, and the ISO sensitivity and aperture size of the camera are adjusted according to lighting conditions to obtain unprocessed raw image data;

[0055] According to the unprocessed original image data, the RGB color mode is converted into the YUV color mode to obtain formatted image data;

[0056] Based on the formatted image data, a high pass filter and edge enhancement are applied to obtain formatted image data.

[0057] Specifically, a camera is placed at the shooting scene, and the ambient light sensor is used to obtain the light intensity and compare it with the range of 0lx to 30000lx. If the measured light intensity is lower than 500lx, a higher ISO value is selected within the range of ISO100 to ISO800 and the aperture is set to f / 2.8 to f / 4. If the light intensity is between 500lx and 2000lx, an ISO value is selected within the range of ISO100 to ISO400 and the aperture is set to f / 5.6 to f / 8. These values, such as 500lx and 2000lx, are set based on data collected in multiple shooting environments and combined with empirical statistics, and can also be visualized. In actual on-site conditions, 500lx is replaced by 300lx or 700lx, and 2000lx is replaced by 1500lx or 2500lx. After each illumination detection, the ISO and aperture adjustment process is repeated and the same operation is performed for each frame of continuous shooting. When the illumination exceeds 2000lx, the ISO can be controlled between 100 and 200 and the aperture is set to f / 8 to f / 16. In darker environments, the ISO can be appropriately increased to 800 or above and the aperture value can be tried to be reduced to below f / 2.8. During this period, the correspondence between the light measurement values ​​and the camera parameters can be recorded in chronological order to finally obtain the unprocessed original image data.

[0058] When reading the acquired unprocessed raw image data, first extract the red channel, green channel, and blue channel values ​​pixel by pixel and record them in a temporary array. Then construct a conversion matrix based on ITU-R BT.709 or other common standards. For example, the formula can be used: , in the above matrix, the required coefficients are obtained by referring to public literature or by testing and fitting typical image samples, and can be adjusted slightly in the range of 0.001 to 1.000 to adapt to specific image features. Next, perform linear operations on the RGB components of each pixel in the image, and write the calculation results into the corresponding YUV component positions. If there is an Alpha channel in the image, keep the Alpha data unchanged. The entire conversion process needs to traverse all the pixels of the image. When the number of pixels is large, consider processing row by row, that is, processing the next row after each row of data is calculated. After all rows have been processed, the converted YUV data is combined into a new image format to finally obtain formatted image data.

[0059] When reading the formatted image data obtained in the previous step, first perform high-pass filtering on its brightness component (Y). You can choose a 3×3 convolution kernel, for example, the center position is , the up, down, left and right positions are , a matrix with diagonal positions of 0. Calculated pixel by pixel: , to highlight the high-frequency components of the image. The coefficients of the convolution kernel are determined after debugging the noise and detail performance in combination with the test image. If the contrast of the local area after high-pass filtering is found to be too strong, an attenuation coefficient can be adjusted in the range of 0.1 to 1.0, and the filtering result can be multiplied by the attenuation coefficient and then added to the original Y component to form a preliminary edge enhancement effect. After that, the U and V components can be differentially operated in the same way, or only the Y component can be processed. If the strength of certain edges needs to be limited, judgment conditions can be added, such as only enhancing processing when the filtering result is greater than a preset threshold. The threshold can be determined by estimating the contrast of the noise area in the sample image, generally between 10 and 50, and can be adjusted in steps of 5. After all pixels are processed, the new YUV data is summarized and recorded as a new image output state, and finally the formatted image data is obtained.

[0060] The steps for obtaining optimized image data are: based on the formatted image data, calculate the contrast score, and the calculation formula is:

[0061]

[0062] in, To rate the contrast, For the Line The brightness value of the column pixels, and are the maximum and minimum brightness values ​​in the image, respectively. is the overall average brightness of the image, and are the number of rows and columns of the image, respectively. and Respectively represent the brightness values ​​of the upper pixel and the left pixel adjacent to the current pixel;

[0063] Based on the contrast score, the grayscale mapping curve of the image is adjusted to make the brightness value distribution uniform and obtain optimized image data.

[0064] Specifically, the formula is beneficial in that by performing multiple measurements on the difference between the pixel and the extreme value of the image brightness range, the difference between the pixel and the overall average brightness, and the difference between adjacent pixels, the image brightness distribution, local contrast, and neighborhood changes can be comprehensively reflected in a single score, thereby more accurately reflecting the overall contrast. The following is a detailed description of the steps for obtaining each parameter:

[0065] The steps to obtain the parameters are: This parameter represents the image Line To obtain this brightness value, you need to convert the RGB components of each pixel in the image into grayscale values. Or other methods. If the traditional grayscale formula is used, you can use The values ​​of the three channels are combined to obtain an integer brightness value in the range of 0 to 255. When reading the entire 1920×1080 image, the rows and columns are scanned in sequence to obtain the R, G, and B values ​​of the corresponding pixels. After calculation using the above formula, we get . Take a certain pixel as an example: if its R=100, G=150, B=200, then the corresponding After obtaining all pixel brightness, it can be directly used in the subsequent calculation of contrast score.

[0066] The steps to obtain the parameter are as follows: This parameter represents the maximum brightness value in the processed image. The acquisition method is to traverse all pixel brightness data, store it in an array, and then find the brightness with the highest value from the array as For example, among all the brightness values ​​obtained statistically, the highest brightness value is 247. In actual operation, the entire 1920×1080 image will be scanned first, the brightness of each pixel will be compared item by item, the current maximum brightness will be recorded and updated in real time, and finally the entire image will be scanned to obtain a clear maximum brightness value.

[0067] The steps to obtain the parameter are: This parameter represents the minimum brightness value in the processed image. The acquisition of is similar, it is necessary to traverse the brightness values ​​of all pixels and record the minimum brightness by comparison. If the brightness value is 12 in the complete scan and this is the minimum, then If there is a lower value in the image, such as 0, it will eventually be updated to 0. During the whole process, it is necessary to ensure that all pixels are visited. According to the actual measured results, it is assumed that the statistics are .

[0068] The steps to obtain the parameter are: This parameter represents the overall average brightness of the image. The acquisition process is to accumulate the brightness of each pixel after obtaining it, and record it as Then divide the accumulated result by the total number of image pixels When the total number of image pixels is 1920×1080, the total number of pixels is approximately 2073600. If the sum of the pixel brightness is 2.50×10^8, then we can calculate .

[0069] and The steps to obtain the parameters are: Refers to the number of rows in the image. Refers to the number of columns of the image. Both are the pixel resolution information of the image itself. If the image resolution is 1920×1080, then , This information is obtained by reading the image file header or metadata. For example, if the image width is 1920 pixels and the height is 1080 pixels, it can be determined and .

[0070] and The steps to obtain the parameters are as follows: these two parameters represent the brightness of the upper pixel adjacent to the current pixel in the vertical direction, and the brightness of the left pixel adjacent to the current pixel in the horizontal direction. The row and column data need to be read once when traversing the pixels. Taking a certain pixel coordinate (i, j) as an example, the brightness values ​​of (i-1, j) and (i, j-1) can be read and stored in the cache before accessing the current pixel, or dynamically read during the calculation process. For example, when scanning to the 3rd row and 10th column, the brightness values ​​of the 2nd row and 10th column and the 3rd row and 9th column are taken out from the memory. If the brightness of the upper pixel is 140 and the brightness of the left pixel is 138 at this time, then , .

[0071] Calculation process:

[0072] Assume the image size is 1080 rows × 1920 columns, that is , .

[0073] Statistics show , , .

[0074] Scan the pixel brightness of the entire image and calculate the brightness of each pixel in the formula The following accumulation process is performed:

[0075]

[0076] Here, the three difference items of each pixel are summed separately and then added to the total.

[0077] Let the sum of the previous step be , and then divided by the total number of pixels .

[0078] Finally, find the square root, .

[0079] In the example, if the scan statistics are It is approximately 3.32×10^8, then: ;

[0080] The result shows that the overall contrast score of the image is 12.65. The higher the value, the more significant the brightness difference. Different pictures may range from 6 to 20. The value can be used to identify which image has a larger or smaller contrast, and then provide a reference for subsequent grayscale mapping or other equalization operations.

[0081] Based on contrast rating , combined with the minimum and maximum brightness values ​​and average brightness data in the image, the previously obtained brightness distribution records are read line by line and a specific interval is selected for targeted analysis. First, the proportion of pixels in the current row concentrated in the range of 80 to 180 is identified from the previously recorded brightness distribution. If the proportion of the number of brightness pixels in the range of 80 to 180 in the image exceeds 40%, it is determined that this part of the range is regarded as a medium brightness area. Then, for the low brightness area, that is, the range of 0 to 80, if the proportion is observed to be higher than 30%, the area is taken as the focus for grayscale mapping and enhancement processing. For the high brightness area, that is, the range of 180 to 255, if the proportion is greater than 20%, the area is compressed or redistributed through the correction curve. In the process, the pixel brightness will be mapped to an adjustable curve and recorded in the real-time cache. The curve shape can be defined in segments according to the gain coefficient and threshold information. For example, when the brightness is lower than 80, the segmented gain coefficient is set to 1.2 to 1.4 To enhance the brightness of the dark area, when the brightness falls between 180 and 255, the gain coefficient is set to 0.8 to 1.0 to balance the excessive brightness distribution and set the attenuation value, for example, 0.05 to 0.1, at the curve segment connection position so that there will not be too large brightness jumps at the transition. The specific values ​​of the above thresholds and gains are determined by statistical analysis of past image processing experience. For example, 20 images with obvious dark and bright parts are counted and the brightness histogram distribution is analyzed to determine the dividing line between the dark and bright parts. After multiple tests, it is found that the attenuation value in the range of 0.05 to 0.15 is more appropriate and the gain in the range of 1.2 to 1.5 can provide a more obvious brightness gain for the dark part. Whether in the dark or bright part, segmented processing will be performed instead of uniform gain for the entire image to avoid excessive enhancement or reduction of local details. After the recording is completed, the pixel brightness distribution mapped by different segments is summarized to form a new brightness table, and then the updated grayscale value is generated row by row to finally obtain the optimized image data.

[0082] The steps for obtaining the target detection results are as follows: input the optimized image data into the deep learning model to extract the target area features, use the convolutional neural network to perform layer-by-layer convolution operations on the image blocks of the candidate area, and combine the pooling operation to reduce redundant information, calculate the category probability distribution of each candidate area, and convert it into a vector of fixed dimension through the fully connected layer to generate the target recognition result data;

[0083] Based on the target recognition result data, the targets whose category confidence meets the set threshold are screened to obtain the target detection results.

[0084] Specifically, the optimized image data is input into the deep learning model for target area feature extraction. First, the existing training samples are annotated and the bounding box coordinates and category information of the target area in each sample are recorded. Then, these annotated samples are divided into training set and validation set, and it is ensured that the training set contains no less than 10,000 images and covers a variety of scenes and target categories. The optimized image data of each image in the training set is combined with the corresponding label and sent to the convolutional neural network and the learning rate is set to 0.001, the batch size is 32, and the number of training rounds is 50. When starting layer-by-layer training, the optimized image data is read at the network input and a size of 1 is applied. The convolution kernel of is used to operate on the image block. During the convolution process, the weights of each channel are recorded and the parameters are updated in combination with the mean shift. After each convolution, a size of The pooling operation is performed and only the maximum value in the local area is retained in the pooling layer to compress redundant data. The cross entropy loss is calculated in each iteration of training and the convolution kernel and the fully connected layer parameters are back-propagated and updated. The iteration is stopped when the cross entropy loss converges to less than 0.02 for 3 consecutive rounds on the validation set. Then the network parameters are fixed and new optimized image data is received as input. For each image block in the candidate area, the edge, texture and higher-level feature information are extracted step by step in the convolution layer and the pooling layer, and the feature map is converted into a fixed-dimensional vector in the fully connected layer, so as to calculate the probability distribution of each area belonging to each target category. In the process, a number of output neurons are set according to the number of different categories and the total probability is constrained to 1. When the probability value of a certain category is the highest among all outputs, it is regarded as the recognition result of the candidate area. The category distribution and probability score of each area are recorded and integrated into the output structure to finally generate the target recognition result data.

[0085] Based on the target recognition result data, the category confidence value of each candidate area is read and compared with the best confidence threshold previously obtained through statistics of about 2000 images. The threshold is finally set to 0.75 with precision and recall as evaluation indicators in multiple experiments. The specific method is to select different confidence thresholds from 0.5 to 0.9 and increase by 0.05 in sequence for testing and record the detection accuracy and missed detection rate at each threshold. After repeatedly comparing the same batch of images, the threshold of 0.75 with the best compromise between accuracy and missed detection rate is selected. Subsequently, when analyzing the current target recognition result data, if the category confidence exceeds 0.75, the candidate area is marked as a valid target and the target category name and confidence value are attached when output. If the category confidence is lower than 0.75, the output of this candidate area is abandoned. No matter how many candidate areas there are, the same judgment will be repeated and the targets that finally meet the requirements will be classified and integrated. The number of all targets that meet the threshold is recorded and their positioning coordinates are output, and finally the target detection result is obtained.

[0086] The steps for obtaining the status analysis results of the preceding vehicle are as follows: based on the target detection results, the position coordinate information of the preceding vehicle is extracted, the scale ratio of the preceding vehicle in the image coordinate system is calculated by combining the focal length of the camera and the pixel density of the image sensor, and the scale ratio is converted into distance through the perspective projection relationship to obtain the preliminary distance data of the preceding vehicle;

[0087] According to the preliminary distance data of the preceding vehicle, the actual driving distance of the preceding vehicle is calculated using the following formula:

[0088]

[0089] in, is the actual driving distance of the front vehicle, is the focal length of the camera, is the front vehicle width, is the pixel width of the target detection box, is the single pixel size of the image sensor;

[0090] Based on the actual driving distance of the front vehicle, the speed change trend of the front vehicle is analyzed. Combined with the displacement of the front vehicle between consecutive frames, the acceleration change in the time series is calculated. According to the dynamic relationship between speed and distance, the driving state of the front vehicle is evaluated and the front vehicle state analysis results are generated.

[0091] Specifically, based on the target detection result, the pixel coordinate range of the front vehicle is read and a group of reference pixel points located in the center are selected as the basis for measurement. First, the calibration file corresponding to the focal length parameter and the pixel density of the image sensor is confirmed at the camera installation location and the physical size corresponding to the focal length value per unit pixel is recorded. Then, the rectangular area where the front vehicle is located is segmented according to the identified front vehicle position coordinates and the pixel width and height of this area are counted. The preliminary proportional information of the corresponding length can be obtained by multiplying the pixel size by the pixel density value registered in the calibration file. In order to maintain consistency in strong light or nighttime environments, multiple groups of environmental data in the calibration file will be called before each shooting for comparison and the set of contrast correction parameters closest to the current brightness range will be selected. Once the available contrast correction parameters are confirmed, they will be applied to the recognition stage of the camera output image and the position boundary changes of the front vehicle in consecutive frames will be recorded. The difference in pixel width and height of the rectangular area of ​​the front vehicle in adjacent frames is compared to determine its brightness. The zoom trend is compared with the relative position of the optical center of the lens. When the pixel width and height show an increasing trend at the same time and exceed the reference threshold specified in the calibration file, it is determined that the distance between the front vehicle and the vehicle is decreasing and the relationship between the pixel increment and the optical transformation is calculated in real time. The reference threshold is determined by testing a variety of vehicle sizes and obtaining the optimal distinction when the focal length is fixed. It is usually between 5 and 20 pixel increments. If the increment exceeds 10 pixels, it is regarded as a significant change and the perspective projection conversion is performed immediately. The pixel increment is associated with the lens principal point and the imaging plane distance to deduce the relative distance change between the front vehicle and the vehicle. Then, combined with parameters such as the camera focal length, the size ratio of the front vehicle in the image plane is calculated and matched with the previously registered projection matrix table items. The matrix table is obtained by calibration using a fixed-point target during the installation phase of the vehicle detection system. There is a unique corresponding distance information for each pixel ratio. Finally, the results of multi-frame comparison are combined to integrate a relatively stable distance average value and serve as the preliminary distance data of the front vehicle.

[0092] The benefit of the formula is that by introducing multi-dimensional factors such as pixel width, focal length, real vehicle width, and sensor single pixel size, the pixel measurement value in the image can be accurately mapped to the distance dimension in the real scene in a relatively concise form, thereby providing higher accuracy for vehicle distance measurement; the following is a description of the steps for obtaining each parameter one by one:

[0093] The steps to obtain the parameter are as follows: This parameter represents the focal length of the camera, which is usually calibrated by the lens manufacturer before leaving the factory, or can be measured through a high-precision calibration program. During the specific measurement, a target is placed at a fixed distance, the camera is aimed at the target, and the distance between the center of the lens and the imaging surface when a clear image is formed is recorded, and quantified in millimeters. For example, the focal length obtained through actual calibration is Mm.

[0094] The steps for obtaining the parameter are: This parameter refers to the width of the vehicle in front.

[0095] The steps to obtain the parameters are as follows: This parameter represents the pixel width of the front vehicle detection frame in the image, which needs to be obtained by analyzing the target detection results. The specific method is to read the left and right edge position coordinates of the front vehicle in the image plane and calculate the difference between the two, and store them in pixels. If the left edge of the image coordinate system is known to be pixels, right edge is Pixels, then During the detection, the vehicle contour area is marked with pixel coordinates by the vehicle recognition algorithm, and then the coordinate difference is counted as For example, in a 1920×1080 image, if the left edge of the front vehicle detection frame is recorded as 420 pixels and the right edge is 720 pixels, then .

[0096] The steps to obtain the parameter are as follows: This parameter is the single pixel size of the image sensor, which needs to be checked or calibrated with the specific camera model. When calibrating, you can use a micron measurement tool that can be traced back to international units to measure the total size of the sensor and calculate the single pixel size based on the number of effective pixels. For example, if the horizontal size of the sensor is 3.456 mm and the number of effective pixels is 1920, you can get .

[0097] Calculation process:

[0098] The first step is to list all parameters:

[0099]

[0100] The second step is to substitute the above values ​​into the formula:

[0101]

[0102] The third step is to calculate the denominator :

[0103]

[0104] Step 4: Calculate ;

[0105] The value can be obtained by looking up the table or calculating Radians are approximately equivalent to 89.97 degrees.

[0106] Step 5. Arrange the denominator and complete the calculation:

[0107]

[0108]

[0109]

[0110] The result shows that the actual driving distance of the front vehicle in the current image is about 3.437 meters. If it increases over time, it means that the distance to the vehicle in front is shortening. If it continues to decrease, it means that the distance is increasing. Therefore, this formula can reflect the changes in the vehicle spacing in real time when calculating multiple frames.

[0111] Based on the actual driving distance of the front vehicle, the displacement of the front vehicle in the ground coordinate system in the continuous image frames is first recorded, and the sampling time of each frame interval is fixed to 0.04 seconds. When reading the distance difference between the kth frame and the k+1th frame, if the distance difference falls between 0.01 meters and 0.3 meters, it indicates that the front vehicle is in a small range of movement. If it exceeds 0.3 meters, it is considered to be a rapid displacement. Therefore, when calculating the speed, this segment of displacement will be regarded as the focus of acceleration analysis. The instantaneous speed of the front vehicle is obtained by dividing the distance difference between adjacent frames by 0.04 seconds and then compared again between adjacent frames. If the speed increment is between 0.1 meters per second and 3 meters per second, the value is recorded and added to the acceleration sequence. If the speed increment is less than 0.1 meters per second, it means that the speed remains basically unchanged and is not included in the acceleration calculation. If the increment exceeds 3 meters per second, it is considered to have a drastic change and the next frame data will continue to be monitored. During this period, reference A speed threshold table built based on the previous 600-frame acquisition results is used to determine whether there is abnormal fluctuation. The speed threshold table is divided into a low-speed interval below 3 meters per second, a normal interval from 5 meters per second to 15 meters per second, and a fast interval above 15 meters per second according to urban road traffic conditions. It can be further subdivided for different vehicle types. Subsequently, when analyzing the acceleration sequence, each acceleration value is first compared with a reasonable interval. For example, the acceleration is compared with the range within the interval of 0 meters per square to 10 meters per square. If it exceeds 10 meters per square, it is marked as an extreme acceleration value and the subsequent speed changes of the video frame are continued to be observed. Finally, all speed and acceleration results are combined into the time series and the driving state of the leading vehicle in the following process is evaluated by calculating the average speed and median acceleration of each frame. Combined with the distance change trend, possible congestion or barrier-free passage can be distinguished, thereby generating a leading vehicle state analysis result.

[0112] The steps of obtaining the headlight control command are as follows: based on the analysis result of the state of the preceding vehicle, extracting the acceleration information of the preceding vehicle and the acceleration information of the own vehicle, calculating the relative acceleration change between the two, extracting the instantaneous speed data of the preceding vehicle and the instantaneous speed data of the own vehicle, and obtaining the preliminary determination data of the headlight control;

[0113] According to the preliminary judgment data of the headlight control, the headlight flashing trigger value is calculated. The calculation formula is:

[0114]

[0115] in, The trigger value for the flashing of the lights. is the current acceleration of the front vehicle, is the current acceleration of the vehicle, is the instantaneous speed of the preceding vehicle, is the instantaneous speed of the vehicle, is the relative distance between the preceding vehicle and the vehicle itself, is the speed of the preceding vehicle at the previous moment, is the vehicle’s speed at the last moment;

[0116] Based on the headlight flashing trigger value, a flashing trigger threshold is set, and conditional judgment is performed. If the headlight flashing trigger value exceeds the trigger threshold, the headlight flashing is triggered and a headlight control instruction is obtained.

[0117] Specifically, based on the results of the leading vehicle state analysis, first read the leading vehicle acceleration information and the own vehicle acceleration information collected in the previous stage and check whether the sources of the two sets of acceleration values ​​are complete. The leading vehicle acceleration information is calculated by the displacement difference and the corresponding time series identified in continuous multi-frame video images, and the time interval is set to 0.04 seconds. The own vehicle acceleration information is the longitudinal acceleration data output by the on-board sensor accumulated within the same 0.04 second time step and split into each specific moment value. Then, by comparing the values ​​of the leading vehicle acceleration and the own vehicle acceleration at the same timestamp, the relative acceleration change between the two is calculated, recorded as a new data sequence and recorded. When the relative acceleration change is greater than the upper limit value of 5 meters per square measured in advance in the road test, it is marked as a rapid change interval and compared at the next moment. If the relative acceleration change is less than 0 meters per square, it is marked as a deceleration interval and continues to track the acceleration in the next frame of data. Then, from the vehicle The instantaneous speed data of the preceding vehicle and the instantaneous speed data of the own vehicle are extracted from the speed data set. The instantaneous speed data of the preceding vehicle is obtained by dividing the distance increment of the preceding vehicle between consecutive frames by 0.04 seconds and stored in the speed sequence frame by frame. The instantaneous speed data of the own vehicle is taken from the on-board speed sensor and is also recorded with a sampling period of 0.04 seconds. The speed value corresponding to the previous moment is also read synchronously and recorded together according to the numerical position of the preceding vehicle instantaneous speed sequence and the own vehicle instantaneous speed sequence. To ensure the accuracy of speed matching, the timestamp difference is compared to keep the error within 0.01 seconds. When it is found that the speed difference between the preceding vehicle and the own vehicle is greater than 5 meters per second, it is registered as a high difference interval and continues to confirm whether it is still in a high difference state in the next cycle. If the speed difference continues to be high for 3 or more adjacent moments, a reminder message is output to indicate that there is a significant difference between the current vehicle speeds. When all these values ​​are associated, they are integrated in chronological order to generate preliminary judgment data for headlight control.

[0118] The benefit of the formula is that by introducing information such as the acceleration difference, speed difference, and distance between the preceding vehicle and the vehicle itself, the dynamic relationship between the two vehicles can be measured in a compact expression, and a more comprehensive reference can be provided for the determination of vehicle light flashing. The following is a description of the steps for obtaining each parameter one by one:

[0119] The steps to obtain the parameter are as follows: This parameter represents the current acceleration of the front vehicle, and the unit can be selected as meter per square. After dividing the displacement change of the front vehicle between consecutive frames by the frame interval to obtain the speed, the first-order difference of the speed is divided by the time step to obtain the acceleration. For example, when testing on urban roads, it is found that the average acceleration of the front vehicle is approximately in the range of 0 meters per square to 5 meters per square. After real-time calculation, an accurate instantaneous acceleration value is formed at each sampling moment. For example, 2.2 meters per square may be recorded at the current moment. At this time .

[0120] The steps to obtain the parameter are as follows: This parameter represents the current acceleration of the vehicle, and the unit is also meters per square. It is mainly obtained through the body acceleration sensor and the vehicle ECU output data. In the specific operation, read the value of the longitudinal acceleration sensor at this moment. If it is in the urban expressway or highway scene, the reference vehicle test results show that the value is mostly distributed between 0 meters per square and 4 meters per square. If it exceeds this range, it will be re-calibrated to confirm whether the sensor has a measurement deviation. For example, if the actual measured acceleration value of the vehicle at this time is 1.5 meters per square, then in the calculation of this formula .

[0121] The steps to obtain the parameter are as follows: This parameter is the instantaneous speed of the vehicle in front, and the unit is meter per second. Similarly, the distance change between image frames and the time interval are calculated. If the speed of the front vehicle on a city road is approximately in the range of 0 m / s to 20 m / s. For example, when the instantaneous speed of the front vehicle is 15.0 m / s at a certain moment, then .

[0122] The steps to obtain the parameter are as follows: This parameter is the instantaneous speed of the vehicle, in meters per second, obtained from the vehicle's own instrument, ECU or speed sensor and recorded in the software. According to the benchmark test, the speed of the vehicle on ordinary urban roads is mainly distributed in the range of 0 meters per second to 25 meters per second. Assuming that the current value read is 17.5 meters per second, then .

[0123] The steps to obtain the parameter are as follows: This parameter is the relative distance between the front vehicle and the vehicle itself, and the unit can be selected as meters. It is the net value obtained by subtracting the reference distance from the front of the vehicle to the camera installation point from the previously obtained front vehicle distance measurement result. If the front vehicle distance data has been obtained by camera ranging or millimeter wave radar in the previous stage, the corresponding value can be directly taken as For urban driving scenarios, L can be judged within a range of 2 meters to 100 meters. For example, if the calibration distance result shows that the center-to-center distance between the front vehicle and the vehicle is 30.0 meters, and the distance from the front of the vehicle to the camera is 1.0 meter, then .

[0124] The steps to obtain the parameter are as follows: This parameter represents the speed of the preceding vehicle at the previous moment, and the unit is also meters per second. The value of the preceding vehicle speed sequence stored in history at the time index before the current moment is taken out as . In offline or online processing, you can quickly query the speed value of the previous frame or the previous sampling period. For example, if the current speed of the front vehicle is 15.0 meters per second, and the previous record is 14.0 meters per second, then .

[0125] The steps to obtain the parameter are as follows: This parameter is the speed of the vehicle at the last moment, in meters per second, and is also obtained from the position of the previous index in the vehicle speed sequence in time. In the urban vehicle scene, the vehicle speed often fluctuates in a short period of time, so a high-frequency record is maintained in the time series. If the speed at the last moment is 17.0 meters per second, then this value is written .

[0126] Calculation process:

[0127] The first step is to select the parameters at a certain moment:

[0128]

[0129] The second step is to calculate the numerator :

[0130]

[0131] Step 3: Calculate the denominator :

[0132]

[0133]

[0134]

[0135] Step 4: Calculate the speed ratio in brackets :

[0136]

[0137]

[0138]

[0139]

[0140] Step 5: Combine the above results:

[0141]

[0142]

[0143]

[0144] The results show that when If it is necessary to determine whether to flash the lights in the subsequent steps, this value will be compared with the light flashing trigger threshold. If it exceeds the threshold, the light flashing will be started, and if it is lower than the threshold, the current light state will be maintained.

[0145] Based on the headlight flashing trigger value, first select the current value from the trigger value sequence obtained above and compare it with the headlight flashing trigger threshold. The threshold is set by referring to the vehicle following data for more than 300 hours in the past and selecting boundary points in the curve diagram of driving speed and relative acceleration to determine the critical range. For example, when the distance between the vehicle and the vehicle in front is between 5 meters and 30 meters, continue to observe the speed and acceleration difference and record the distribution of the trigger value. Finally, a range between 0.2 and 0.3 is selected as the feasible threshold range. After testing one by one under multiple working conditions, 0.25 is set as the headlight flashing trigger threshold. If the headlight flashing trigger value calculated at the current moment exceeds 0.25, the headlight flashing trigger value calculated at the current moment will be 0.25. It is determined that the flashing conditions are met and the light output switching frequency is set in the headlight control program. For example, the light flashes within the range of 0.8 to 1.2 seconds. If it does not exceed 0.25, the current headlight state is maintained, and the trigger value at the current moment is recorded together with the time index into the trigger value sequence. The sequence will be extended and monitored later. If the trigger values ​​at five consecutive sampling moments exceed the threshold, the flashing time is extended or the flashing frequency is increased. If the threshold is only exceeded for a short time, the flashing is terminated after the vehicle speed or acceleration drops. All flashing actions must be compared with the brake signal status in the vehicle's electronic system to complete the final output action, and finally settled as a headlight control instruction in the headlight control program.

[0146] The steps for obtaining the emergency response execution results are: monitoring the emergency brake signal of the preceding vehicle and the sudden conditions on the road, including the sudden appearance of obstacles or the crossing of pedestrians, to obtain emergency data;

[0147] According to the emergency data, the headlight flashing adjustment index is calculated using the following formula:

[0148]

[0149] in, Adjust the flashing index for the headlights. is the actual distance to the vehicle ahead or obstacle, is the safety distance threshold, is the speed difference, is the reaction time constant;

[0150] Based on the vehicle light flashing adjustment index, the flashing frequency is set and the emergency response execution result is generated.

[0151] Specifically, monitor the emergency brake signals of the vehicle in front and sudden situations on the road, synchronously read the measured information through the front camera and the on-board brake monitoring device, and record the vehicle speed deceleration and road image data at that moment, set the continuous monitoring time interval to 0.05 seconds, and repeatedly compare the brake light status and acceleration count value of the vehicle in front during this interval. If the brake light is high and the corresponding acceleration drops below -4 meters per square, it will be registered as an emergency deceleration. If a new obstacle or pedestrian position is detected in the picture and its coordinates are greatly displaced from the original frame to the current frame, the situation will be marked as a sudden change. When braking or obstacle detection marks appear for three consecutive samples, it is confirmed that an emergency event has occurred and multiple data of the emergency event are written into the emergency data sequence, which includes Including the brake mark of the preceding vehicle, the pixel coordinate range of the obstacle, and the detected pedestrian speed value, etc. If multiple pedestrians are detected crossing the road during the road monitoring period, their travel distance of 10 to 30 meters will be counted one by one and the number of frames in which they appear in the camera screen will be multiplied by 0.05 seconds per frame as the continuous walking time. After an emergency situation exceeding the respective thresholds is identified on the lane, the subsequent displacement changes and vehicle acceleration will continue to be tracked and updated to the data sequence. If the emergency only occurs within one sampling cycle, it will be verified in subsequent cycles whether it has been resolved. All marks will be included in this emergency data for continued use and recorded time by time, so that the subsequent processing stage can accurately locate the moment when emergency braking or obstacle appearance occurs based on the emergency data, and finally obtain the emergency data.

[0152] The benefit of the formula is that it quantitatively characterizes the actual distance and speed difference between the vehicle and the obstacle or the vehicle ahead by combining the exponential decay and exponential growth characteristics, thereby comprehensively reflecting multiple risk factors in the same expression; the following is a description of the steps for obtaining each parameter one by one:

[0153] The steps to obtain the parameters are as follows: This parameter represents the actual distance to the vehicle in front or the obstacle, which can be obtained through the vehicle ranging radar or the image perspective ranging method, and the unit is meter. The measured distance will be updated every 0.05 seconds or 0.1 seconds of each sampling interval. To ensure accuracy, it can be calibrated in combination with the camera focal length and the reference list of known obstacle sizes. When the distance between vehicles detected on the spot falls between 2 meters and 100 meters, it is regarded as the normal following vehicle or general obstacle distance range. If it is less than 2 meters, it is marked as an extreme proximity situation and noted in the ranging record. The following example: After measurement, the distance to the obstacle at this moment is recorded as meters, it is brought into subsequent operations.

[0154] The steps to obtain the parameters are as follows: This parameter is the safety distance threshold, in meters, and needs to be determined by referring to road driving regulations and vehicle brake test data before the vehicle goes on the road. Usually, it can be set between 10 meters and 30 meters in urban environments. If the driving speed is high or the road environment is special, the high end value will be set, such as 25 meters or 30 meters. In specific operations, multiple vehicles can be simulated to perform emergency braking tests at different speeds on dry roads. By measuring the braking distance and combining the requirements of laws and regulations on safe vehicle distance, the appropriate In this example, rice.

[0155] The steps to obtain the parameter are as follows: This parameter represents the speed difference, which is generally determined by the speed difference between the front vehicle and the vehicle itself or the relative speed difference between the vehicle and the obstacle. The unit is meters per second. The speed reading at that moment can be compared. If the speed of the front vehicle is 50 kilometers per hour and the speed of the vehicle itself is 40 kilometers per hour, the difference is 10 kilometers per hour, which is converted to Meters per second, denoted as .

[0156] The steps to obtain the parameters are as follows: This parameter is the reaction time constant, in seconds, which is used to characterize the vehicle's response delay to emergency situations. The average time required for the driver to detect an obstacle and effectively decelerate can be counted in multiple road tests, or the delay from the sensor reading the vehicle's posture change to the execution of the brake signal can be measured by the vehicle control system. If the most common human-machine reaction time falls between 0.8 seconds and 1.2 seconds based on the evaluation of about 500 people in an urban driving environment, the selected Seconds is a practical value.

[0157] Calculation process:

[0158] The first step is to list the parameters:

[0159]

[0160] The second step is to calculate :

[0161]

[0162]

[0163] Step 3: Calculate the denominator :

[0164]

[0165]

[0166]

[0167] Step 4: Substitute into the formula:

[0168]

[0169] The results show that when the current distance is 10 meters and the speed difference is about 2.78 meters per second, the headlight flash adjustment index is 0.676. A higher value means that the headlights can flash at a medium frequency in the current situation. If the value continues to rise to 0.9 or above, it can be determined that the proximity and relative speed difference with the vehicle in front or obstacle are further increasing, and a higher frequency light signal is required to provide a warning.

[0170] Based on the headlight flicker adjustment index, the calculated index value is first compared with the flicker frequency distribution range obtained from the three groups of survey data. The survey data is based on 300 hours of urban road and 200 hours of highway driving tests. The flicker frequency is selected in multiple grades according to the speed difference and distance. The flicker frequency is listed in the range of 1 to 4 times per second in a step of 0.5 times per second. When the headlight flicker adjustment index reaches between 0.6 and 0.7, it is corresponding to the range of 2 to 2.5 flashes per second. If the index rises to 0.8 to 0.9, it is in the range of 3 to 3.5 flashes. If the index exceeds 0.9, it is in the range of 4 flashes. Then, the changes in the instantaneous index are recorded and maintained during the operation of the vehicle. Continue to compare. If the flicker adjustment index remains around 0.8 to 0.9 at multiple consecutive sampling points, the flicker frequency will be fixed at 3 times per second and a check will be made at subsequent moments to determine whether it continues to increase. If the index is less than 0.2, it will be judged as a low-risk situation and the flickering will be terminated or the taillights will be turned on only in a constant light mode. During the entire process, the frequency settings at each sampling moment will be overwritten in real time to ensure that the subsequent results can overwrite the previous settings. When the index suddenly drops or re-enters another interval, the flicker frequency will be refreshed to the lower limit of the corresponding value range and the actual frequency value selected at each moment will be recorded. After a period of time, the system will integrate these frequency control actions to obtain the final lighting control behavior and output it as the emergency response execution result.

[0171] See also Figure 2-6 The present invention also provides a vehicle-mounted lamp, which is applicable to the intelligent light control method in the above embodiment, comprising: a lamp housing 1 having a heat dissipation through hole 10 on the outer wall, a plug 3 for conducting electricity installed at one end of the lamp housing 1, a first heat dissipation fan 21 and a second heat dissipation fan 22 are correspondingly arranged inside the lamp housing 1, a light source substrate 11 is horizontally arranged between the first heat dissipation fan 21 and the second heat dissipation fan 22, and LED lamp chips 110 are arranged on both sides of the light source substrate 11;

[0172] A control circuit board 5 corresponding to the heat dissipation through hole 10 is provided at one end of the light source substrate 11, a rectifier circuit board 4 electrically connected to the plug 3 is provided outside the control circuit board 5, a connection terminal 210 electrically connected to the first cooling fan 21 is provided on the light source substrate 11, the second cooling fan 22 is located between the rectifier circuit board 4 and the control circuit board 5, the rectifier circuit board 4 is connected to the control circuit board 5 via a pin row 99, and the second cooling fan 22 is electrically connected to the rectifier circuit board 4.

[0173] With regard to the above description, it should be noted that a groove for placing the first cooling fan 21 and the second cooling fan 22 is also provided inside the lamp housing 1. During installation, the first cooling fan 21 can be electrically connected to the light source substrate 11, and the second cooling fan 22 can be electrically connected to the rectifier circuit board 4, thereby realizing power supply to the first cooling fan 21 and the second cooling fan 22.

[0174] With regard to the above description, it should be noted that a hot air hole 50 and a mounting bayonet 51 are respectively provided on the control circuit board 5 . The hot air hole 50 is used for hot air circulation, and the mounting bayonet 51 is adapted to the end of the light source substrate 11 .

[0175] In view of the above description, it is particularly noted that during installation and use, air outlets 12 corresponding to the first cooling fan 21 and the second cooling fan 22 may be opened on the outside of the lamp housing 1, and an opening 13 corresponding to the LED lamp chip 110 may be opened on the outside of the lamp housing 1. The two ends of the opening 13 are respectively connected to the first cooling fan 21 and the second cooling fan 22, and during the heat dissipation process, the directions of the first cooling fan 21 and the second cooling fan 22 are made consistent so that the hot air flows in the same direction, and the hot air flows through the opening 13 to achieve a diversion effect.

[0176] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. An intelligent light emitting control method, characterized in that: The following steps are involved: The image acquisition and preprocessing module, in which the vehicle camera captures continuous image data, formats the image, and generates formatted image data; optimizes the formatted image data by contrast adjustment, and generates optimized image data; An image analysis and parsing module performs target detection, identifies vehicles and road signs, and obtains target detection results based on the optimized image data; Analyze the target detection result, evaluate the distance to the preceding vehicle, and generate a preceding vehicle status analysis result; A headlight control decision module determines whether the headlights need to be flashed based on the preceding vehicle status analysis result and generates a headlight control instruction; The emergency response and feedback module monitors the emergency brake signals of the vehicle ahead and sudden road conditions, adjusts the flashing frequency of the headlights according to the monitored emergency situations, and obtains the emergency response execution results.

2. The intelligent light emitting control method according to claim 1, characterized in that: The step of acquiring the formatted image data is as follows: the vehicle camera captures continuous image data, and adjusts the ISO sensitivity and aperture size of the camera according to lighting conditions to obtain unprocessed raw image data; According to the unprocessed original image data, the RGB color mode is converted into a YUV color mode to obtain formatted image data; Based on the formatted image data, a high pass filter and edge enhancement are applied to obtain formatted image data.

3. The intelligent light emitting control method according to claim 1, characterized in that: The step of obtaining the optimized image data is: calculating the contrast score according to the formatted image data, and the calculation formula is: in, To rate the contrast, For the Line The brightness value of the column pixels, and are the maximum and minimum brightness values ​​in the image, respectively. is the overall average brightness of the image, and are the number of rows and columns of the image, respectively. and Respectively represent the brightness values ​​of the upper pixel and the left pixel adjacent to the current pixel; Based on the contrast score, the grayscale mapping curve of the image is adjusted to make the brightness value distribution uniform, thereby obtaining optimized image data.

4. The intelligent light emitting control method according to claim 1, characterized in that: The target detection result is obtained by: inputting the optimized image data into a deep learning model to extract target area features, using a convolutional neural network to perform layer-by-layer convolution operations on image blocks of candidate areas, combining pooling operations to reduce redundant information, calculating the category probability distribution of each candidate area, and converting it into a vector of fixed dimension through a fully connected layer to generate target recognition result data; Based on the target recognition result data, targets whose category confidences meet a set threshold are screened to obtain target detection results.

5. The intelligent light emitting control method according to claim 1, characterized in that: The step of obtaining the front vehicle state analysis result is: based on the target detection result, extracting the position coordinate information of the front vehicle, combining the focal length of the camera and the pixel density of the image sensor, calculating the scale ratio of the front vehicle in the image coordinate system, and converting it into distance through perspective projection relationship to obtain preliminary distance data of the front vehicle; According to the preliminary distance data of the preceding vehicle, the actual travel distance of the preceding vehicle is calculated using the following formula: in, is the actual driving distance of the front vehicle, is the focal length of the camera, is the front vehicle width, is the pixel width of the target detection box, is the single pixel size of the image sensor; Based on the actual driving distance of the preceding vehicle, the speed change trend of the preceding vehicle is analyzed, and the acceleration change in the time series is calculated in combination with the displacement of the preceding vehicle between consecutive frames. The driving state of the preceding vehicle is evaluated according to the dynamic relationship between speed and distance, and a preceding vehicle state analysis result is generated.

6. The intelligent light emitting control method according to claim 1, characterized in that: The step of obtaining the headlight control instruction is: based on the analysis result of the state of the preceding vehicle, extracting the acceleration information of the preceding vehicle and the acceleration information of the vehicle, calculating the relative acceleration change between the two, extracting the instantaneous speed data of the preceding vehicle and the instantaneous speed data of the vehicle, and obtaining the preliminary determination data of the headlight control; According to the preliminary determination data of the headlight control, the headlight flashing trigger value is calculated, and the calculation formula is: in, The trigger value for the flashing of the lights. is the current acceleration of the front vehicle, is the current acceleration of the vehicle, is the instantaneous speed of the preceding vehicle, is the instantaneous speed of the vehicle, is the relative distance between the preceding vehicle and the vehicle itself, is the speed of the preceding vehicle at the previous moment, is the vehicle’s speed at the last moment; Based on the headlight flashing trigger value, a flashing trigger threshold is set, and condition judgment is performed. If the headlight flashing trigger value exceeds the trigger threshold, the headlight flashing is triggered to obtain a headlight control instruction.

7. The intelligent light emitting control method according to claim 1, characterized in that: The step of acquiring the emergency response execution result is: monitoring the emergency brake signal of the preceding vehicle and the emergency situation on the road, including the sudden appearance of obstacles or the crossing of pedestrians, to obtain emergency situation data; According to the emergency data, the headlight flashing adjustment index is calculated using the following formula: in, Adjust the flashing index for the headlights. is the actual distance to the vehicle ahead or obstacle, is the safety distance threshold, is the speed difference, is the reaction time constant; Based on the vehicle light flicker adjustment index, the flicker frequency is set and an emergency response execution result is generated.

8. A vehicle-mounted lamp, applicable to the intelligent light control method as claimed in any one of claims 1 to 7, characterized in that: include: A lamp housing (1) having a heat dissipation through hole (10) on its outer wall, a plug (3) for conducting electricity being installed at one end of the lamp housing (1), a first heat dissipation fan (21) and a second heat dissipation fan (22) being arranged inside the lamp housing (1) correspondingly, a light source substrate (11) being arranged horizontally between the first heat dissipation fan (21) and the second heat dissipation fan (22), and LED lamp chips (110) being arranged on both sides of the light source substrate (11); A control circuit board (5) corresponding to the heat dissipation through hole (10) is provided at one end of the light source substrate (11); a rectifier circuit board (4) electrically connected to the plug (3) is provided outside the control circuit board (5); a connection terminal (210) electrically connected to the first heat dissipation fan (21) is provided on the light source substrate (11); and the second heat dissipation fan (22) is located between the rectifier circuit board (4) and the control circuit board (5); the rectifier circuit board (4) is connected to the control circuit board (5) via a pin header (99); and a terminal of the second heat dissipation fan (22) is connected to the rectifier circuit board (4).

9. The vehicle-mounted lamp according to claim 8, characterized in that: The lamp housing (1) is also provided with air vents (12) on the outside corresponding to the first cooling fan (21) and the second cooling fan (22), and the lamp housing (1) is also provided with an opening (13) on the outside corresponding to the light emitting surface of the LED lamp chip (110), and two ends of the opening (13) are respectively connected to the first cooling fan (21) and the second cooling fan (22).

10. The vehicle-mounted lamp according to claim 8, characterized in that: The control circuit board (5) is provided with hot air holes (50) for hot air circulation and mounting bayonet holes (51) for assembling the end of the light source substrate (11).

Images