License plate recognition method, device, electronic equipment and system

Abstract

The invention provides a license plate recognition method, device, electronic equipment and system, and relates to the technical field of license plate recognition, and the method comprises the steps: collecting a long-exposure image under the condition of rainy and snowy weather; the long exposure image comprises a target object; determining the natural falling speed of the target object based on the long exposure image; determining a target wind speed based on the natural falling speed, and sending the target wind speed to the blowing equipment; the target wind speed is used for indicating the blowing equipment to blow the target area based on the target wind speed; the target area is a license plate snapshot area of camera equipment; and after the vehicle drives into the target area, identifying the license plate of the vehicle. When the vehicle is driven into the target area, the blowing equipment reduces rain and snow in the area of the license plate captured by the camera equipment, or blows the original disordered rain and snow which cause diffuse reflection into regular rain and snow, so that the influence of the rain and snow between the camera equipment and the position of the license plate on license plate recognition is reduced, and the accuracy of license plate recognition can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of license plate recognition, and in particular to a license plate recognition method, device, electronic equipment and system. Background Art

[0002] Currently, many residential areas, parks, and supermarket entrances have implemented entrance and exit license plate recognition systems to control vehicle access and parking fees. However, the license plate recognition effect in rainy and snowy weather has always been a problem. Rain and snow stay in the picture and block part of the license plate, or rain and snow adhere to the license plate and block the license plate or produce reflections, or rain and snow adhere to the camera lens and block part of the lens image. These problems have always affected the efficiency of vehicle access and the accuracy of license plate recognition.

[0003] In the related art, a rain cover is usually provided at the lens of the camera device to prevent rain and snow from adhering to the front of the lens, so as to avoid the rain and snow adhering to the lens from affecting the license plate recognition.

[0004] However, the above-mentioned related technologies can only prevent rain and snow from adhering to the lens, but cannot prevent rain and snow between the camera device and the license plate position. During license plate recognition, the rain and snow between the camera device and the license plate position will still affect the license plate recognition, thereby reducing the accuracy of license plate recognition. Summary of the invention

[0005] In view of the problems existing in the prior art, the embodiments of the present invention provide a license plate recognition method, device, electronic equipment and system.

[0006] The present invention provides a license plate recognition method, comprising:

[0007] In rainy and snowy weather, a long exposure image is collected; the long exposure image includes a target object, and the target object includes raindrops and / or snowflakes;

[0008] determining a natural falling speed of the target object based on the long exposure image;

[0009] Determine a target wind speed based on the natural falling speed, and send the target wind speed to the blowing device; the target wind speed is used to instruct the blowing device to blow air to a target area based on the target wind speed; the target area is the area where the camera device captures the license plate;

[0010] After the vehicle enters the target area, the license plate of the vehicle is recognized.

[0011] According to a license plate recognition method provided by the present invention, the determining of the target wind speed based on the natural falling speed comprises:

[0012] Get the direction of natural wind;

[0013] Determining the angle between the wind direction of the natural wind and the blowing direction of the blowing device;

[0014] Determining a speed at which the natural wind blows toward the target object based on the natural falling speed and the angle;

[0015] In the case where the angle is less than a preset angle, determining a speed difference between the natural falling speed and the speed at which the natural wind blows toward the target object, and determining a speed greater than the speed difference as the target wind speed;

[0016] When the angle is greater than or equal to the preset angle, a speed and value of the natural falling speed and the speed of the natural wind blowing toward the target object are determined, and a speed greater than the speed and value is determined as the target wind speed.

[0017] According to a license plate recognition method provided by the present invention, the determining the speed of the natural wind blowing toward the target object based on the natural falling speed and the angle includes:

[0018] Determining a falling angle of the target object based on the long exposure image;

[0019] Determining a vertical falling speed of the target object based on the natural falling speed and the falling angle;

[0020] The speed at which the natural wind blows toward the target object is determined based on the vertical falling speed and the angle.

[0021] According to a license plate recognition method provided by the present invention, determining the natural falling speed of the target object based on the long exposure image includes:

[0022] Acquire a preset number of target lines in the long exposure image; the target lines are used to characterize the shape of the target object in the long exposure image;

[0023] The natural falling speed is determined based on the length of each of the target lines, the shutter value of the camera device and the preset number.

[0024] According to a license plate recognition method provided by the present invention, the method further includes:

[0025] When the vehicle is detected to enter the shooting screen, predicting the license plate position of the vehicle after entering the target area based on the speed of the vehicle;

[0026] The license plate position is sent to the blowing device; the license plate position is used to instruct the blowing device to turn to the license plate position and blow air based on the target wind speed.

[0027] According to a license plate recognition method provided by the present invention, the method further includes:

[0028] When it is detected that the vehicle leaves the shooting screen, a closing instruction is sent to the blowing device; the closing instruction is used to instruct the blowing device to stop blowing.

[0029] The present invention also provides a license plate recognition device, comprising:

[0030] A collection unit, used for collecting long exposure images in rainy and snowy weather; the long exposure images include target objects, and the target objects include raindrops and / or snowflakes;

[0031] a first determining unit, configured to determine a natural falling speed of the target object based on the long exposure image;

[0032] a second determining unit, configured to determine a target wind speed based on the natural falling speed;

[0033] A first sending unit is used to send the target wind speed to the blowing device; the target wind speed is used to instruct the blowing device to blow air to a target area based on the target wind speed; the target area is an area where the camera device captures the license plate;

[0034] The recognition unit is used to recognize the license plate of the vehicle after the vehicle enters the target area.

[0035] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the above-mentioned license plate recognition methods is implemented.

[0036] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the license plate recognition method as described above is implemented.

[0037] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the license plate recognition method described above is implemented.

[0038] The present invention also provides a license plate recognition system, comprising a camera device and a blower device, wherein the camera device and the blower device are in communication connection;

[0039] The camera device is used to collect long-exposure images in rainy and snowy weather; the long-exposure images include target objects, and the target objects include raindrops and / or snowflakes;

[0040] The camera device is further used to determine the natural falling speed of the target object based on the long exposure image, determine the target wind speed based on the natural falling speed, and send the target wind speed to the blowing device;

[0041] The blowing device is used to blow air to a target area based on the target wind speed; the target area is the area where the camera device captures the license plate;

[0042] The camera device is also used to identify the license plate of the vehicle after the vehicle enters the target area.

[0043] The license plate recognition method, device, electronic device and system provided by the present invention collect long exposure images in rainy and snowy weather, determine the natural falling speed of the target object based on the long exposure image, the target object includes raindrops and / or snowflakes, determine the target wind speed based on the natural falling speed, and send the target wind speed to the blowing device, so that the blowing device blows on the target area based on the target wind speed, and the target area is the area where the camera device captures the license plate. In this way, when the vehicle enters the target area, the blowing device has reduced the rain and snow in the area where the camera device captures the license plate, or has blown the originally chaotic rain and snow that caused diffuse reflection into regular rain and snow, thereby reducing the influence of rain and snow between the camera device and the license plate position on the license plate recognition, and can improve the accuracy of license plate recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 It is a raindrop size statistical graph provided by the present invention;

[0046] Figure 2 It is a schematic diagram of a license plate recognition scenario provided by the present invention;

[0047] Figure 3 This is one of the flowcharts of the license plate recognition method provided by an embodiment of the present invention;

[0048] Figure 4 This is the second flow chart of the license plate recognition method provided by the embodiment of the present invention;

[0049] Figure 5 is a schematic diagram of a long exposure image provided by the present invention;

[0050] Figure 6 This is a third flow chart of the license plate recognition method provided by an embodiment of the present invention;

[0051] Figure 7 is a schematic diagram of a license plate recognition scenario provided by an embodiment of the present invention;

[0052] Figure 8 This is a fourth flow chart of the license plate recognition method provided by an embodiment of the present invention;

[0053] Fig. 9 It is a schematic diagram of a picture captured under light at night when it rains, provided by an embodiment of the present invention;

[0054] Fig.10 It is a schematic diagram of a picture captured under light after being blown by a blowing device when it rains at night provided by an embodiment of the present invention;

[0055] Fig.11 is a schematic structural diagram of a blowing device provided by an embodiment of the present invention;

[0056] Fig.12 is a diagram of an actual application scenario of a license plate recognition system provided by an embodiment of the present invention;

[0057] Fig.13 is a schematic diagram of the structure of a license plate recognition device provided by an embodiment of the present invention;

[0058] Fig.14 It is a schematic diagram of the physical structure of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] Regarding the speed of raindrops, that is, the speed of raindrops when they reach the ground, what we should discuss here is to estimate the speed of raindrops without using instruments to measure speed. The simplest method is to ignore the resistance of the air and treat the raindrops as free fall to estimate the speed of raindrops. Assume that the height of the raindrops is h, the mass of the raindrops is m, and the speed of the raindrops when they reach the ground is v. Because the potential energy of the raindrops at high altitudes will be converted into kinetic energy, when the raindrops reach the ground:

[0061] m*g*h=1 / 2*m*v 2

[0062] After sorting, we get:

[0063] v 2=2g*h

[0064] According to meteorologists, raindrops are formed at an altitude of about 3,000 to 4,000 meters. We take the middle value of 3,500 meters and substitute it into v 2 =2g*h, we get v 2 =2*9.8*3500=68600.

[0065] v=262m / s(meters per second)

[0066] The speed of 262 meters per second is close to the speed of sound. If we change the units, we can get:

[0067] v = 262 * 60 * 60 = 943200 m / h = 943 kilometers per hour

[0068] This is more than three times faster than a high-speed train and the cruising speed of an airplane. If a raindrop of this speed hits you, the consequences will be quite serious. Obviously, the actual speed of raindrops is not that fast. This proves that air resistance plays a great role in slowing down the raindrops as they fall, and the effect of air resistance on the speed of raindrops must be taken into account. Let's study the falling speed of raindrops from an aerodynamic perspective.

[0069] According to fluid mechanics, when an object moves in a fluid, the resistance f it encounters is calculated differently according to its speed conditions. When the speed is low, f = k*v, where v is the speed of the object and k is the viscosity coefficient of the fluid. When the speed of the object is high: f = d*C*S*v 2 / 2, where d is the density of the fluid, which is the density of air here, C is the air resistance coefficient, and S is the cross-sectional area of ​​the object in the forward direction. According to investigation, the falling raindrops should be in the case of faster speed, that is, f = d*C*S*v 2 / 2. Let the air resistance be f1, and the air buoyancy on the raindrop be f2. When the raindrop speed increases to a certain speed, the resistance increase and gravity reach a balance, and then the raindrop starts to move at a uniform speed. Then we can list the formula m*g=f1+f2, where m is the mass of the raindrop and g is the acceleration due to gravity. The following is a specific calculation of the values ​​of f1, f2 and m.

[0070] m=d1*V=4 / 3*π*r 3 *d1

[0071] f1=d*C*S*v 2 / 2=d*C*π*r 2 *v 2 / 2

[0072] f2=d*V*g=d*4 / 3*π*r 3 *g

[0073] Where d is the density of air, d1 is the density of water, r is the radius of the raindrop, and v is the speed of the raindrop. Substituting the values ​​of f1, f2, and m into m*g=f1+f2, we get 4 / 3*π*r 3 *d1*g=d*C*π*r 2 *v 2 / 2+d*4 / 3*π*r 3 *g.

[0074] Sorting out the above

[0075] v 2 =8*r*g(d1-d) / 3*d*C

[0076] Consult the relevant physical properties manual to obtain:

[0077] Density of air d = 1.65 kg / m 3 (kilograms per cubic meter), the density of water is d1 = 1000kg / m 3 , air resistance coefficient C = 0.42, then v 2 =37619*r.

[0078] Assuming that the radius of a raindrop is between 0.001m and 0.005m, the range of the raindrop velocity is:

[0079] v1=6.1(m / s)

[0080] v2=13.7(m / s)

[0081] That is to say, the speed of raindrops ranges from 6.1 to 13.7 m / s. According to the formula v 2 =37619*r. As long as the radius of the raindrop is large enough, its falling speed will become very large. If the diameter of the raindrop is 10 mm, substitute v 2 =37619*r, we can get the final speed of 19m / s, which is equivalent to 70 kilometers per hour. This kind of raindrops is dangerous if they hit people. But in fact, when the diameter of raindrops becomes larger, due to friction, raindrops will break into small raindrops, so there are actually few raindrops with a radius of more than 5mm. Figure 1 is the actual measured data of raindrop size, the horizontal axis is the diameter of the raindrop, and the vertical axis is the proportion of raindrops of different sizes to the total number of raindrops. Figure 1 It can be seen that the radius of most raindrops is between 1mm and 3mm.

[0082] Wind speed refers to the speed of air moving relative to a fixed point on the earth. The commonly used unit is m / s (meters per second), 1m / s = 3.6km / h (kilometers per hour). There is no level of wind speed, only wind force has levels. Wind speed is the basis for the classification of wind force. Generally speaking, the greater the wind speed, the higher the wind force level, and the more destructive the wind is.

[0083] Theoretically, as long as the wind speed (wind force 6 or above) is greater than the speed of raindrops, the wind can be used to blow away the raindrops. Figure 2 As shown, the scene is fixed and the capture position of the vehicle is relatively fixed, so a blower can be installed at a fixed position to blow away the rain in the designated area (the area between the license plate recognition camera and the actual license plate position). Similarly, snowflakes can also be blown away by a blower.

[0084] Based on this, the present invention proposes a license plate recognition method, which collects long exposure images in rainy and snowy weather, determines the natural falling speed of the target object based on the long exposure image, the target object includes raindrops and / or snowflakes, determines the target wind speed based on the natural falling speed, and sends the target wind speed to a blowing device, so that the blowing device blows on a target area based on the target wind speed, and the target area is the area where the camera device captures the license plate. In this way, when the vehicle enters the target area, the blowing device has reduced the rain and snow in the area where the camera device captures the license plate, or has blown the originally messy rain and snow that caused diffuse reflection into regular rain and snow, thereby reducing the influence of rain and snow between the camera device and the license plate position on the license plate recognition, and can improve the accuracy of license plate recognition.

[0085] Combine the following Figures 3 to 11 The license plate recognition method of the present invention is described. The execution subject of the method can be a camera device, or a license plate recognition device set in the camera device, and the license plate recognition device can be implemented by software, hardware, or a combination of the two. The method can be applied in scenes such as residential areas, parks, and supermarket entrances for controlling vehicle access and parking fees.

[0086] Figure 3 is one of the flow charts of the license plate recognition method provided by the embodiment of the present invention, such as Figure 3 As shown, the license plate recognition method includes the following steps:

[0087] Step 301: Under the condition of rainy and snowy weather, a long exposure image is collected; the long exposure image includes a target object, and the target object includes raindrops and / or snowflakes.

[0088] For example, the rain and snow detection sensor can be used to detect in real time whether it is raining or snowing. When rain or snow is detected, the rain and snow detection sensor sends an acquisition instruction to the camera device, so that the camera device responds to the acquisition instruction and captures a long exposure image in a long exposure manner during idle time. If the long exposure image is captured in rainy weather, the long exposure image includes raindrops; if the long exposure image is captured in snowy weather, the long exposure image includes snowflakes; if the long exposure image is captured in rainy and snowy weather, the long exposure image includes raindrops and snowflakes. The following is an explanation using the example of a long exposure image including raindrops.

[0089] It should be noted that the rain and snow detection sensor can be a separate device or integrated into a camera device, and the present invention is not limited to this.

[0090] Step 302: Determine the natural falling speed of the target object based on the long exposure image.

[0091] For example, taking the target object as a raindrop, when a long exposure image is collected, the natural falling speed of the raindrop can be determined based on the shape of each raindrop in the long exposure image and the falling angle of the raindrop.

[0092] Step 303, determine the target wind speed based on the natural falling speed, and send the target wind speed to the blowing device; the target wind speed is used to instruct the blowing device to blow air to the target area based on the target wind speed; the target area is the area where the camera device captures the license plate, that is, the area where the camera device can identify the position of the license plate when the vehicle enters or exits.

[0093] Since the license plate recognition scene is relatively fixed, the approximate blowing direction of the blowing device can be pre-adjusted based on the actual license plate recognition scene to ensure that the blowing device can blow to the area where the camera device captures the license plate.

[0094] For example, when the natural falling speed of raindrops is obtained, the target wind speed can be determined based on the current natural wind direction, the blowing direction of the blowing device and the natural falling speed of raindrops. The target wind speed can also be set to a value greater than a preset multiple of the natural falling speed. Of course, the value greater than the preset multiple of the natural falling speed needs to be determined based on the maximum power of the blowing device, and the target wind speed is sent to the blowing device that is communicatively connected to the camera device. When the blowing device receives the target wind speed, the blowing device blows based on the target wind speed and the pre-adjusted approximate blowing direction to blow away raindrops in the area where the camera device captures the license plate.

[0095] Step 304: After the vehicle enters the target area, the license plate of the vehicle is identified.

[0096] For example, after the camera device detects that a vehicle enters the target area through real-time monitoring, at this time, the blowing device has reduced the raindrops in the area where the camera device captures the license plate, or blown the originally messy raindrops that cause diffuse reflection into regular raindrops, thereby reducing the influence of raindrops between the camera device and the license plate position of the vehicle on license plate recognition in rainy weather, and the license plate can be recognized more accurately.

[0097] The license plate recognition method provided by the present invention collects long exposure images in rainy and snowy weather, determines the natural falling speed of the target object based on the long exposure image, the target object includes raindrops and / or snowflakes, determines the target wind speed based on the natural falling speed, and sends the target wind speed to the blowing device, so that the blowing device blows on the target area based on the target wind speed, and the target area is the area where the camera device captures the license plate. In this way, when the vehicle enters the target area, the blowing device has reduced the rain and snow in the area where the camera device captures the license plate, or has blown the originally chaotic rain and snow that caused diffuse reflection into regular rain and snow, thereby reducing the influence of rain and snow between the camera device and the license plate position on the license plate recognition, and can improve the accuracy of license plate recognition.

[0098] In one embodiment, Figure 4 FIG. 2 is a flow chart of the license plate recognition method provided by an embodiment of the present invention. Figure 4 As shown, the determination of the target wind speed based on the natural falling speed in the above step 303 can be specifically implemented by the following steps:

[0099] Step 3031: Obtain the direction of natural wind.

[0100] For example, the wind direction of the natural wind may be detected by a wind and rain detection sensor, and the detected wind direction of the natural wind may be sent to the camera device.

[0101] Step 3032: determine the angle between the wind direction of the natural wind and the blowing direction of the blowing device.

[0102] For example, the camera device may determine the angle β between the wind direction of the natural wind and the blowing direction of the blowing device based on the wind direction of the natural wind and the blowing direction of the set blowing device.

[0103] Step 3033: Determine the speed at which the natural wind blows toward the target object based on the natural falling speed and the angle.

[0104] Optionally, based on the long exposure image, the falling angle of the target object is determined; based on the natural falling speed and the falling angle, the vertical falling speed of the target object is determined; and based on the vertical falling speed and the angle, the speed of the natural wind blowing toward the target object is determined.

[0105] For example, Figure 5is a schematic diagram of a long exposure image provided by the present invention, such as Figure 5 As shown, affected by the natural wind, the falling angle of raindrops usually has a certain angle α with the vertical direction, that is, the falling angle of raindrops can be obtained from the long exposure image to be α. When the natural falling speed V1 of raindrops and the falling angle α of raindrops are known, the vertical falling speed of raindrops V2=V1*cosα can be obtained based on the calculation of trigonometric functions, and the angle β between the wind direction of natural wind and the blowing direction of the blowing equipment can be obtained. The speed of natural wind blowing to raindrops V3=V2 / sinβ can be obtained.

[0106] It should be noted that the weather forecast can also be obtained to know the speed and direction of the natural wind based on the weather forecast, and then the speed of the natural wind blowing toward the target object can be determined based on the speed and direction of the natural wind.

[0107] It should be noted that a pressure sensor can also be set at the air inlet of the blowing equipment to determine the speed and direction of the natural wind through the pressure signal collected by the pressure sensor, and then determine the speed of the natural wind blowing towards the target object based on the speed and direction of the natural wind.

[0108] Step 3034: When the angle is smaller than a preset angle, determine the speed difference between the natural falling speed and the speed of the natural wind blowing toward the target object, and determine the speed greater than the speed difference as the target wind speed.

[0109] The preset angle can be set based on demand, for example, the preset angle is 90 degrees.

[0110] For example, when the angle β is less than the preset angle, it means that the falling direction of the raindrops is in the forward direction relative to the natural wind direction. At this time, the speed difference V1-V3 between the natural falling speed V1 and the speed V3 of the natural wind blowing toward the target object is determined, and the target wind speed that the blowing equipment needs to provide is set to a speed greater than V1-V3.

[0111] Step 3035: When the angle is greater than or equal to the preset angle, determine the speed and value of the natural falling speed and the speed of the natural wind blowing toward the target object, and determine the speed greater than the speed and value as the target wind speed.

[0112] For example, when the angle β is greater than or equal to the preset angle, it means that the falling direction of the raindrops is opposite to the direction of the natural wind. At this time, the speed sum V1+V3 between the natural falling speed V1 and the speed V3 of the natural wind blowing toward the target object is determined, and the target wind speed that the blowing equipment needs to provide is set to a speed greater than V1+V3.

[0113] In this embodiment, the speed of natural wind blowing toward the target object is determined based on the angle between the wind direction of the natural wind and the blowing direction of the blowing equipment, and the natural falling speed of the target object, and based on the judgment of forward and reverse directions, the target wind speed corresponding to the forward direction and the target wind speed corresponding to the reverse direction are determined, thereby improving the accuracy of wind speed determination.

[0114] In one embodiment, determining the natural falling speed of the target object based on the long exposure image in step 302 may be implemented in the following manner:

[0115] A preset number of target lines in the long exposure image are obtained, and the natural falling speed is determined based on the length of each of the target lines, the shutter value of the camera device and the preset number; the target lines are used to characterize the shape of the target object in the long exposure image.

[0116] Among them, the preset number can be set based on demand, and N is used below to represent the preset number, where N is an integer greater than or equal to 2.

[0117] For example, Figure 5 As shown, N groups of target lines, i.e., raindrop lines, are taken from the collected long exposure image. The lengths of the raindrop lines are L1, L2, L3, ..., LN, respectively. When the camera device captures the picture, the shutter value of the long exposure is h. The sum of the lengths of the raindrop lines in the long exposure image is (L1+L2+L3+...+LN). Then, (L1+L2+L3+...+LN) is divided by (h*N) to obtain the natural falling speed V1 of the raindrops.

[0118] In this embodiment, the natural falling speed of the target object is determined based on the length of each target line in the long exposure image, the shutter value of the camera device and the preset number, and the calculation method is simple and accurate.

[0119] In one embodiment, Figure 6 FIG. 3 is a flow chart of the license plate recognition method provided by an embodiment of the present invention. Figure 6 As shown, before step 304, the license plate recognition method further includes the following steps:

[0120] Step 305: When it is detected that the vehicle enters the shooting screen, based on the speed of the vehicle, predict the license plate position of the vehicle after it enters the target area.

[0121] For example, when the camera device detects that a vehicle enters the shooting screen, it determines the speed of the vehicle based on the shooting time of each frame of the continuously shot preset frame images and the position of the vehicle in the frame image, and predicts the license plate position of the vehicle after entering the target area based on the vehicle speed. Specifically, the vehicle speed can be input into the position detection model to obtain the license plate position of the vehicle entering the target area output by the position detection model.

[0122] It should be noted that the position detection model can be trained based on the following method: collect speed samples of multiple vehicles in the current license plate recognition scenario, add the license plate position label of the vehicle entering the target area to the speed sample of each vehicle, input the speed samples of multiple vehicles into the initial position detection model, obtain the predicted license plate position corresponding to each vehicle output by the initial position detection model, construct a loss based on the predicted license plate position and the corresponding license plate position label, optimize the model parameters of the initial position detection model based on the constructed loss until convergence, and obtain the position detection model. The initial position detection model can be a deep neural network DNN (Deep Neural Networks, DNN) or a convolutional neural network CNN (Convolutional Neural Network, CNN), etc.

[0123] Step 306: Send the license plate position to the blowing device; the license plate position is used to instruct the blowing device to turn to the license plate position and blow air based on the target wind speed.

[0124] For example, a pan-tilt head is installed on the air blowing device. When the camera device predicts the license plate position of the vehicle, it sends the license plate position of the vehicle to the air blowing device. When the air blowing device receives the license plate position, it can control the air blowing device to turn to the license plate position through the pan-tilt head to achieve fine-tuning of the previously set target area that can be blown, and blow air in the direction of the license plate position based on the received target wind speed. Figure 7 is a schematic diagram of a license plate recognition scenario provided by an embodiment of the present invention, such as Figure 7 As shown, the two black lines 701 represent the blowing direction of the blowing device. The blowing device blows toward the license plate position to blow away the rain and snow between the camera device and the license plate position to form a rain and snow gap area. In addition, when the blowing device turns to the license plate position to blow air, if rain and snow are attached to the license plate, the rain and snow attached to the license plate can also be blown away, further improving the accuracy of license plate recognition.

[0125] In this embodiment, when a vehicle is detected entering the shooting screen, the license plate position of the vehicle after entering the target area is predicted based on the speed of the vehicle, and the license plate position is sent to the blowing device, so that the blowing device turns to the license plate position to blow air, thereby realizing the adjustment of the blowing direction of the blowing device based on the license plate position. In this way, the rain and snow between the camera device and the license plate position can be blown away by the blowing device to form a rain and snow gap area, so that when the camera device captures the license plate, the clarity of the captured license plate image can be improved, thereby improving the accuracy of license plate recognition.

[0126] In one embodiment, Figure 8 FIG. 4 is a flow chart of the license plate recognition method provided by an embodiment of the present invention. Figure 8As shown, after step 304, the license plate recognition method further includes the following steps:

[0127] Step 307: When it is detected that the vehicle leaves the shooting screen, a closing instruction is sent to the blowing device; the closing instruction is used to instruct the blowing device to stop blowing.

[0128] For example, when the camera device detects that a vehicle leaves the shooting screen, it means that the vehicle leaves after license plate recognition, and then sends a closing instruction to the blowing device. When the blowing device receives the closing instruction, it stops blowing.

[0129] In this embodiment, when it is detected that the vehicle leaves the shooting screen, a closing instruction is sent to the hair dryer. When the hair dryer receives the closing instruction, it stops blowing, thereby reducing the power consumption of the hair dryer.

[0130] Fig. 9 is a schematic diagram of a picture captured under light at night when it rains, as provided by an embodiment of the present invention. Fig. 9 As shown in the figure, raindrops produce various scattered light under the illumination of lights, which causes the license plate image captured by the camera equipment to be unable to see the specific license plate information, thereby reducing the accuracy of license plate recognition; Fig.10 1 is a schematic diagram of a picture captured under light after being blown by a blowing device at night when it rains, as provided by an embodiment of the present invention. Fig.10 As shown, after being blown by the blowing device, the captured image can clearly show the specific license plate information, thereby improving the accuracy of license plate recognition.

[0131] It should be noted that the degree of improvement in the accuracy of license plate recognition needs to be determined based on the blowing direction of the blowing equipment, the wind speed of the blowing equipment, the falling direction of rain and snow, the direction of natural wind, the vehicle's driving trajectory, the angle of the vehicle's headlights, the installation position of the blowing equipment, etc., and can be adjusted based on demand, and the present invention will not go into details here.

[0132] Fig.11 It is a structural schematic diagram of a blowing device provided in an embodiment of the present invention, where the blowing device is used to blow air toward a target area based on a target wind speed sent by a camera device.

[0133] The invention provides a vehicle license plate recognition system, comprising a camera device and a blower device, wherein the camera device and the blower device are communicatively connected.

[0134] The camera device is used to collect long-exposure images in rainy and snowy weather; the long-exposure images include target objects, and the target objects include raindrops and / or snowflakes.

[0135] The camera device is further used to determine the natural falling speed of the target object based on the long exposure image, determine the target wind speed based on the natural falling speed, and send the target wind speed to the blowing device.

[0136] The blowing device is used to blow air to a target area based on the target wind speed; the target area is the area where the camera device captures the license plate.

[0137] The camera device is also used to identify the license plate of the vehicle after the vehicle enters the target area.

[0138] Fig.12 is a diagram of an actual application scenario of the license plate recognition system provided by an embodiment of the present invention, such as Fig.12 As shown, the camera device 1201 is responsible for vehicle detection, and provides switch signals to the blowing device 1202, provides wind speed recommendations to the blowing device 1202, provides the coordinates of the parking space position to the blowing device 1202 and other functions; the blowing device 1202 is responsible for generating corresponding wind force based on the target wind speed sent by the camera device 1201 to blow away raindrops and / or snowflakes in the target area. The present invention effectively improves the accuracy of entrance and exit license plate recognition in rainy and snowy weather through the combination of wind and rain detection sensors, camera devices and blowing devices.

[0139] The license plate recognition device provided by the present invention is described below. The license plate recognition device described below and the license plate recognition method described above can be referenced to each other.

[0140] Fig.13 is a schematic diagram of the structure of a license plate recognition device provided by an embodiment of the present invention, such as Fig.13 As shown, the license plate recognition device 1300 includes a collection unit 1301, a first determination unit 1302, a second determination unit 1303, a first sending unit 1304 and a recognition unit 1305; wherein:

[0141] The acquisition unit 1301 is used to acquire a long exposure image in rainy and snowy weather; the long exposure image includes a target object, and the target object includes raindrops and / or snowflakes;

[0142] A first determining unit 1302 is configured to determine a natural falling speed of the target object based on the long exposure image;

[0143] A second determining unit 1303 is used to determine a target wind speed based on the natural falling speed;

[0144] The first sending unit 1304 is used to send the target wind speed to the blowing device; the target wind speed is used to instruct the blowing device to blow air to the target area based on the target wind speed; the target area is the area where the camera device captures the license plate;

[0145] The recognition unit 1305 is used to recognize the license plate of the vehicle after the vehicle enters the target area.

[0146] The license plate recognition device provided by the present invention collects long exposure images in rainy and snowy weather, determines the natural falling speed of the target object based on the long exposure image, the target object includes raindrops and / or snowflakes, determines the target wind speed based on the natural falling speed, and sends the target wind speed to the blowing device, so that the blowing device blows on the target area based on the target wind speed, and the target area is the area where the camera device captures the license plate. In this way, when the vehicle enters the target area, the blowing device has reduced the rain and snow in the area where the camera device captures the license plate, or has blown the originally chaotic rain and snow that caused diffuse reflection into regular rain and snow, thereby reducing the influence of rain and snow between the camera device and the license plate position on the license plate recognition, and can improve the accuracy of license plate recognition.

[0147] Based on any of the foregoing embodiments, the second determining unit 1303 is specifically configured to:

[0148] Get the direction of natural wind;

[0149] Determining the angle between the wind direction of the natural wind and the blowing direction of the blowing device;

[0150] Determining a speed at which the natural wind blows toward the target object based on the natural falling speed and the angle;

[0151] In the case where the angle is less than a preset angle, determining a speed difference between the natural falling speed and the speed at which the natural wind blows toward the target object, and determining a speed greater than the speed difference as the target wind speed;

[0152] When the angle is greater than or equal to the preset angle, a speed and value of the natural falling speed and the speed of the natural wind blowing toward the target object are determined, and a speed greater than the speed and value is determined as the target wind speed.

[0153] Based on any of the foregoing embodiments, the second determining unit 1303 is further specifically configured to:

[0154] Determining a falling angle of the target object based on the long exposure image;

[0155] Determining a vertical falling speed of the target object based on the natural falling speed and the falling angle;

[0156] The speed at which the natural wind blows toward the target object is determined based on the vertical falling speed and the angle.

[0157] Based on any of the foregoing embodiments, the first determining unit 1302 is specifically configured to:

[0158] Acquire a preset number of target lines in the long exposure image; the target lines are used to characterize the shape of the target object in the long exposure image;

[0159] The natural falling speed is determined based on the length of each of the target lines, the shutter value of the camera device and the preset number.

[0160] Based on any of the above embodiments, the license plate recognition device 1300 further includes:

[0161] A prediction unit, configured to predict the license plate position of the vehicle after it enters the target area based on the speed of the vehicle when the vehicle is detected to enter the shooting picture;

[0162] The second sending unit is used to send the license plate position to the blowing device; the license plate position is used to instruct the blowing device to turn to the license plate position and blow air based on the target wind speed.

[0163] Based on any of the above embodiments, the license plate recognition device 1300 further includes:

[0164] The third sending unit is used to send a closing instruction to the blowing device when it is detected that the vehicle leaves the shooting screen; the closing instruction is used to instruct the blowing device to stop blowing.

[0165] Fig.14 is a schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention, such as Fig.14 As shown, the electronic device may include: a processor 1410, a communication interface 1420, a memory 1430 and a communication bus 1440, wherein the processor 1410, the communication interface 1420 and the memory 1430 communicate with each other through the communication bus 1440. The processor 1410 may call the logic instructions in the memory 1430 to execute the license plate recognition method, which includes: in rainy and snowy weather, collecting a long exposure image; the long exposure image includes a target object, and the target object includes raindrops and / or snowflakes;

[0166] determining a natural falling speed of the target object based on the long exposure image;

[0167] Determine a target wind speed based on the natural falling speed, and send the target wind speed to the blowing device; the target wind speed is used to instruct the blowing device to blow air to a target area based on the target wind speed; the target area is the area where the camera device captures the license plate;

[0168] After the vehicle enters the target area, the license plate of the vehicle is recognized.

[0169] In addition, the logic instructions in the above-mentioned memory 1430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0170] On the other hand, the present invention further provides a computer program product, the computer program product includes a computer program, the computer program can be stored in a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute the license plate recognition method provided by the above methods, the method includes: in rainy and snowy weather, collecting a long exposure image; the long exposure image includes a target object, the target object includes raindrops and / or snowflakes;

[0171] determining a natural falling speed of the target object based on the long exposure image;

[0172] Determine a target wind speed based on the natural falling speed, and send the target wind speed to the blowing device; the target wind speed is used to instruct the blowing device to blow air to a target area based on the target wind speed; the target area is the area where the camera device captures the license plate;

[0173] After the vehicle enters the target area, the license plate of the vehicle is recognized.

[0174] In another aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to perform the license plate recognition method provided by the above methods, the method comprising: collecting a long exposure image in rainy and snowy weather; the long exposure image includes a target object, and the target object includes raindrops and / or snowflakes;

[0175] determining a natural falling speed of the target object based on the long exposure image;

[0176] Determine a target wind speed based on the natural falling speed, and send the target wind speed to the blowing device; the target wind speed is used to instruct the blowing device to blow air to a target area based on the target wind speed; the target area is the area where the camera device captures the license plate;

[0177] After the vehicle enters the target area, the license plate of the vehicle is recognized.

[0178] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0179] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0180] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A license plate recognition method, characterized in that: include: Collect long exposure images in rainy and snowy weather conditions; The long exposure image includes a target object, and the target object includes raindrops and / or snowflakes; determining a natural falling speed of the target object based on the long exposure image; Determine a target wind speed based on the natural falling speed, and send the target wind speed to the blowing device; the target wind speed is used to instruct the blowing device to blow air to a target area based on the target wind speed; the target area is the area where the camera device captures the license plate; After the vehicle enters the target area, the license plate of the vehicle is recognized.

2. The license plate recognition method according to claim 1, characterized in that: The determining of the target wind speed based on the natural falling speed comprises: Get the direction of natural wind; Determining the angle between the wind direction of the natural wind and the blowing direction of the blowing device; Determining a speed at which the natural wind blows toward the target object based on the natural falling speed and the angle; In the case where the angle is less than a preset angle, determining a speed difference between the natural falling speed and the speed at which the natural wind blows toward the target object, and determining a speed greater than the speed difference as the target wind speed; When the angle is greater than or equal to the preset angle, a speed and value of the natural falling speed and the speed of the natural wind blowing toward the target object are determined, and a speed greater than the speed and value is determined as the target wind speed.

3. The license plate recognition method according to claim 2, characterized in that: The determining, based on the natural falling speed and the included angle, the speed at which the natural wind blows toward the target object comprises: Determining a falling angle of the target object based on the long exposure image; Determining a vertical falling speed of the target object based on the natural falling speed and the falling angle; The speed at which the natural wind blows toward the target object is determined based on the vertical falling speed and the angle.

4. The license plate recognition method according to claim 1, characterized in that: The determining the natural falling speed of the target object based on the long exposure image includes: Acquire a preset number of target lines in the long exposure image; the target lines are used to characterize the shape of the target object in the long exposure image; The natural falling speed is determined based on the length of each of the target lines, the shutter value of the camera device and the preset number.

5. The license plate recognition method according to claim 1, characterized in that: The method further comprises: When the vehicle is detected to enter the shooting screen, predicting the license plate position of the vehicle after entering the target area based on the speed of the vehicle; The license plate position is sent to the blowing device; the license plate position is used to instruct the blowing device to turn to the license plate position and blow air based on the target wind speed.

6. The license plate recognition method according to any one of claims 1 to 5, characterized in that: The method further comprises: When it is detected that the vehicle leaves the shooting screen, a closing instruction is sent to the blowing device; the closing instruction is used to instruct the blowing device to stop blowing.

7. A license plate recognition device, characterized in that: include: A collection unit, used to collect long-exposure images in rainy and snowy weather conditions; The long exposure image includes a target object, and the target object includes raindrops and / or snowflakes; a first determining unit, configured to determine a natural falling speed of the target object based on the long exposure image; a second determining unit, configured to determine a target wind speed based on the natural falling speed; A first sending unit is used to send the target wind speed to the blowing device; the target wind speed is used to instruct the blowing device to blow air to a target area based on the target wind speed; the target area is an area where the camera device captures the license plate; The recognition unit is used to recognize the license plate of the vehicle after the vehicle enters the target area.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the license plate recognition method according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the license plate recognition method according to any one of claims 1 to 6 is implemented.

10. A license plate recognition system, characterized in that: It includes a camera device and a blower device, wherein the camera device and the blower device are communicatively connected; The camera device is used to collect long-exposure images in rainy and snowy weather; the long-exposure images include target objects, and the target objects include raindrops and / or snowflakes; The camera device is further used to determine the natural falling speed of the target object based on the long exposure image, determine the target wind speed based on the natural falling speed, and send the target wind speed to the blowing device; The blowing device is used to blow air to a target area based on the target wind speed; the target area is the area where the camera device captures the license plate; The camera device is also used to identify the license plate of the vehicle after the vehicle enters the target area.

Images