Intelligent automobile defogging method and system based on proactive vision and temperature difference dew point cooperation

Abstract

This invention relates to the field of defogging technology for new energy vehicles, and particularly to an intelligent vehicle defogging method and system based on the synergy of front-facing vision and temperature difference dew point. By collecting and preprocessing vehicle dynamic data, front-facing image data, and environmental data (including vehicle speed and wiper operation frequency), a fog detection window area is extracted from the preprocessed image of the inside of the windshield. An image segmentation algorithm is used to divide the fog detection window area into fog feature regions and interference regions. The average gray value of the fog feature regions and the standard deviation of the gray values ​​of the interference regions are calculated, and the visual signal-to-noise ratio (SNR) is calculated based on the average gray value and the standard deviation. This invention uses image segmentation and edge detection to calculate the visual SNR and fog edge blurriness, and combines this with a visually reliable fog index to effectively distinguish fog from raindrop reflections and other interference, overcoming the shortcomings of single sensors that are prone to false triggering, and significantly improving the accuracy of fog recognition.

Description

Technical Field

[0001] This invention relates to the field of new energy vehicle defogging technology, and in particular to an intelligent vehicle defogging method and system based on the synergy of forward vision and temperature difference dew point. Background Technology

[0002] Currently, automotive windshield defogger technology is mainly divided into two categories: manual and automatic. Manual defogger relies on the driver's subjective judgment, which poses safety hazards such as delayed response and distraction.

[0003] In existing technologies, automatic defogging is mostly based on a single temperature and humidity sensor, triggering defogging by detecting the difference between the glass temperature and the dew point. This has the following drawbacks: Raindrop reflections in rainy weather are easily misinterpreted as fog, leading to fog-free startup; while localized fogging caused by occupant breathing in dry environments is often missed due to the lag in the global sensor response. Responding only after fog formation cannot suppress condensation at its source, especially under high-dynamic conditions (such as high-speed driving and occupant breathing), where the fogging rate exceeds the system's response speed. Furthermore, most systems employ binary control or fixed settings, failing to dynamically adjust based on fog concentration and distribution, resulting in energy waste or incomplete defogging.

[0004] Therefore, there is an urgent need for an intelligent defogging system that can deeply integrate vision and temperature and humidity perception, and has the ability to predict and dynamically adjust, in order to overcome the above-mentioned technical defects. Summary of the Invention

[0005] The main objective of this invention is to provide an intelligent vehicle defogging method based on the synergy of forward vision and temperature difference dew point, and further to provide an intelligent vehicle defogging system based on the synergy of forward vision and temperature difference dew point that can run and implement the above method, effectively solving the problems mentioned in the background art.

[0006] The technical solution of the present invention is as follows: Firstly, a smart car defogging method based on the synergy of front-facing vision and temperature difference dew point is proposed, which includes the following steps: S1: Collect vehicle dynamic data, front camera image data and environmental data and preprocess them. The vehicle dynamic data includes vehicle speed and windshield wiper operation frequency. The front camera image data includes camera exposure time and image of the inside of the glass. The environmental data includes the temperature of the inside of the glass, the relative humidity inside the vehicle and the ambient light intensity. S2: Extract the fog detection window area from the preprocessed inner glass image. Use an image segmentation algorithm to divide the fog detection window area into fog feature areas and interference areas. Calculate the average gray value of the fog feature areas and the gray standard deviation of the interference areas. Calculate the visual signal-to-noise ratio based on the average gray value and the gray standard deviation. Perform edge detection on the fog detection window area, construct an edge point set, and calculate the fog edge blurriness based on the edge points. S3: Calculate the visual credibility haze index based on the visual signal-to-noise ratio, fog edge blurring, and pre-processed camera exposure time; calculate the thermal boundary layer fluctuation index based on the pre-processed inner surface temperature of the glass, and obtain the wet boundary layer fluctuation index based on the pre-processed relative humidity inside the vehicle; obtain the thermal-humidity coupling disturbance index based on the thermal boundary layer fluctuation index, the wet boundary layer fluctuation index, and the pre-processed vehicle speed; generate the coupling response factor based on the visual credibility haze index and the thermal-humidity coupling disturbance index. S4: Smooth the coupling response factor, output a decision threshold based on the pre-processed vehicle speed, ambient light intensity and wiper operation frequency, construct a defogging urgency index based on the smoothed coupling response factor and decision threshold, and start the defogging strategy when the defogging urgency index is greater than or equal to 1.

[0007] A further improvement of the present invention is that step S2 includes the following specific steps: S21: Extract the fog detection window area from the preprocessed inner glass image, and use an image semantic segmentation algorithm to divide the fog detection window area into fog feature areas and interference areas; S22: Calculate the average gray value of the fog feature region. and the gray standard deviation of the interference area Based on average gray value and grayscale standard deviation Computational visual signal-to-noise ratio The formula is: ; S23: Perform edge detection on the fog detection window area and construct an edge point set; S24: Calculate fog edge blur based on edge points The formula is: ; in, The total number of edge points, For edge point indexing, For the first The pixel coordinates of the edge points For image gradient operators, It is a constant.

[0008] A further improvement of the present invention is that step S3 includes the following specific steps: S31: Based on visual signal-to-noise ratio Fog edge blur and the pre-processed camera exposure time Calculate the visual credibility haze index The formula is: ; in, For ambiguity threshold, This is the steepness coefficient. For optimal exposure time, As the exposure bias penalty factor, The standard deviation of the reference noise; S32: Based on the temperature of the inner surface of the pretreated glass Calculate the thermal boundary layer fluctuation index The formula is: ; in, For the first Temperature of the inner surface of the glass at each measuring point This represents the minimum temperature of the inner surface of the glass at all measuring points. This represents the maximum value of the inner surface temperature of the glass at all measuring points. For airflow velocity gradient, For airflow velocity, Spatial coordinates; Based on the pre-treated relative humidity inside the vehicle Obtain the wet boundary layer fluctuation index The formula is: ; in, For the first The relative humidity inside the vehicle at each measuring point This represents the total number of humidity sensors. This represents the average relative humidity inside the vehicle at all measuring points. For time variables, For integration time window, the unit is , For reference respiratory pulse intensity, the unit is... , For the local humidity of the breathing area, The background humidity.

[0009] A further improvement of the present invention is that step S3 further includes the following specific steps: S33: Based on thermal boundary layer fluctuation index and wet boundary layer fluctuation index Combined with the pre-processed vehicle speed Calculate the thermal-humid coupling perturbation index The formula is: ; in, For reference speed, The influence coefficient of vehicle speed. This is the exponential decay coefficient, in units of... , For the first The dew point temperature corresponding to each measuring point, in units of ; S34: Based on Visual Credibility Haze Index and thermal-humid coupling perturbation index Generate coupling response factor The formula is: ; in, for The coupling response factor at time t, for Visual credibility haze index at any given time. for The thermal-humid coupling perturbation index at any given time. The visual saliency threshold. The physical significance threshold. This represents the visual channel steepness coefficient. This is the physical channel steepness coefficient. for The derivative of the visual credibility haze index at any given time with respect to time. for The time derivative of the thermal-humid coupling perturbation exponent.

[0010] A further improvement of the present invention is that step S4 includes the following specific steps: S41: Coupling response factor Perform a moving average process to obtain the smoothed coupling response factor. ; S42: Based on preprocessed vehicle speed Ambient light intensity and the frequency of wiper operation Generate decision threshold The formula is: ; in, for The decision threshold at any given moment. Based on the threshold, This is the illumination correction factor. For reference light intensity, for The ambient light intensity at any given time for The speed of the car at any moment This is the vehicle speed correction factor. For reference speed, This is the wiper correction factor, in units of... , for The frequency of wiper operation at any given time, measured in units of ; S43: Based on the smoothed coupling response factor and decision threshold Construct a fog removal urgency index The formula is: ;in, for The urgency index for defogging at any given moment. for The smoothed coupling response factor at any given time; S44: When the defogging urgency index is greater than or equal to 1, the defogging strategy is activated.

[0011] A further improvement of the present invention is that S44 further includes: the defogging strategy specifically refers to, when the defogging urgency index... When the condition of being greater than or equal to 1 and less than the first preset threshold is met, the warning-level defogging mode is triggered; when the defogging urgency index is... When the condition of being greater than or equal to the first preset threshold and less than the second preset threshold is met, a light-level defogging mode is triggered; when the defogging urgency index... When the condition of being greater than or equal to the second preset threshold and less than the third preset threshold is met, the medium-level defogging mode is triggered; when the defogging urgency index... When the condition of being greater than or equal to the third preset threshold is met, the heavy-level defogging mode is triggered.

[0012] Secondly, a smart car defogging system based on forward-facing vision and temperature difference / dew point synergy is proposed, including: Data acquisition module: used to collect vehicle dynamic data, front camera image data and environmental data and preprocess them. The vehicle dynamic data includes vehicle speed and windshield wiper operation frequency. The front camera image data includes camera exposure time and images of the inside of the glass. The environmental data includes the temperature of the inside of the glass, the relative humidity inside the vehicle and the ambient light intensity. Image processing module: This module extracts the fog detection window area from the preprocessed image of the inside of the glass. It uses an image segmentation algorithm to divide the fog detection window area into fog feature regions and interference regions, and calculates the average gray value of the fog feature regions and the gray standard deviation of the interference regions. Based on the average gray value and the gray standard deviation, it calculates the visual signal-to-noise ratio. It also performs edge detection on the fog detection window area, constructs an edge point set, and calculates the fog edge blurriness based on the edge points. Feature calculation module: Calculates the visually reliable haze index based on visual signal-to-noise ratio, fog edge blurring, and pre-processed camera exposure time; calculates the thermal boundary layer fluctuation index based on the pre-processed inner surface temperature of the glass; obtains the wet boundary layer fluctuation index based on the pre-processed relative humidity inside the vehicle; and obtains the thermal-humidity coupling perturbation index based on the thermal boundary layer fluctuation index, the wet boundary layer fluctuation index, and the pre-processed vehicle speed; and generates the coupling response factor based on the visually reliable haze index and the thermal-humidity coupling perturbation index. Decision module: Used to perform moving average processing on the coupled response factor. Based on the preprocessed vehicle speed, ambient light intensity and wiper operation frequency, it outputs a decision threshold. Based on the smoothed coupled response factor and the decision threshold, it constructs a defogging urgency index. When the defogging urgency index is greater than or equal to 1, the defogging strategy is activated.

[0013] The technical effects of this invention are as follows: A smart vehicle defogging method based on front-facing vision and temperature difference dew point synergy was constructed. This method calculates the visual signal-to-noise ratio and fog edge ambiguity through image segmentation and edge detection, respectively. Combined with the visual credibility fog index, it effectively distinguishes fog from interference such as raindrop reflection, overcoming the defect of single sensor being prone to false triggering, and significantly improving the fog recognition accuracy. By constructing a thermal-humidity coupling perturbation index that integrates the effects of temperature field inhomogeneity, humidity pulses, and vehicle speed, the risk of fogging can be detected in advance. Combined with the comparison of preset thresholds, warnings can be issued or pre-fogging can be initiated before fog formation, realizing a leap from passive response to active suppression. The decision threshold is dynamically generated based on real-time environmental data and compared with the smoothed coupling response factor to obtain the defogging urgency index, which then triggers a graded defogging strategy. This ensures that the defogging intensity is accurately matched with the current fog concentration and environmental conditions, avoiding energy waste or insufficient defogging caused by constant power defogging, and significantly improving energy efficiency. Attached Figure Description

[0014] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a flowchart illustrating the intelligent vehicle defogging method based on the synergy of forward vision and temperature difference dew point in Embodiment 1 of the present invention. Figure 2 This is a schematic diagram of the structure of the intelligent car defogging system based on the synergy of forward vision and temperature difference dew point in Embodiment 2 of the present invention. Detailed Implementation

[0015] Example 1

[0016] This embodiment constructs an intelligent vehicle defogging method based on front-facing vision and temperature difference dew point synergy. This method calculates the visual signal-to-noise ratio and fog edge ambiguity through image segmentation and edge detection, respectively. Combined with a visual credibility fog index, it effectively distinguishes fog from interference such as raindrop reflection, overcoming the shortcomings of single sensors being prone to false triggering, and significantly improving fog recognition accuracy. By constructing a thermal-humidity coupling perturbation index that integrates the effects of temperature field inhomogeneity, humidity pulses, and vehicle speed, it can detect fogging risks in advance. Combined with a preset threshold comparison, it can issue warnings or initiate pre-defogging before fog formation, achieving a leap from passive response to active suppression. A decision threshold is dynamically generated based on real-time environmental data and compared with a smoothed coupling response factor to obtain a defogging urgency index, thereby triggering a tiered defogging strategy. This ensures that the defogging intensity is precisely matched with the current fog concentration and environmental conditions, avoiding energy waste or insufficient defogging caused by constant power defogging, and significantly improving energy efficiency.

[0017] Intelligent vehicle defogging methods based on the synergy of front-facing vision and temperature difference dew point, such as Figure 1 As shown, the specific steps include the following: S1: Collects and preprocesses vehicle dynamic data, front camera image data, and environmental data. Vehicle dynamic data includes vehicle speed and windshield wiper operation frequency. Front camera image data includes camera exposure time and images of the inside of the glass. Environmental data includes the temperature of the inside of the glass, relative humidity inside the vehicle, and ambient light intensity.

[0018] In this embodiment, the vehicle speed is collected by a vehicle speed sensor located inside the car, the wiper operation frequency is collected by a sensor located at the wiper motor position, the camera exposure time is read directly from the camera ISP (Image Signal Processor) register, the image of the inside of the glass is collected by the camera module, the surface temperature of the inside of the glass is measured by a distributed thin-film temperature sensor array, the relative humidity inside the vehicle is measured by a multi-point MEMS temperature and humidity sensor, and the ambient light intensity is measured by a light sensor.

[0019] In this embodiment, vehicle dynamic data, front camera image data, and environmental data are preprocessed. Specifically, the preprocessing method involves unifying all sensor data to the same time reference. Using the acquisition time of the camera image frame as the synchronization reference, data such as vehicle speed, wiper operation frequency, camera exposure time, inner surface temperature of the glass, relative humidity inside the vehicle, and ambient light intensity are aligned to this time using interpolation or nearest neighbor matching methods to form a synchronized dataset. Next, outlier detection and removal are performed on each sensor data. A first-order low-pass filter is used to low-pass filter the data to suppress high-frequency noise. Finally, the data is normalized.

[0020] S2: Extract the fog detection window area from the preprocessed inner glass image. Use an image segmentation algorithm to divide the fog detection window area into fog feature areas and interference areas. Calculate the average gray value of the fog feature areas and the gray standard deviation of the interference areas. Calculate the visual signal-to-noise ratio based on the average gray value and the gray standard deviation. Perform edge detection on the fog detection window area, construct an edge point set, and calculate the fog edge blurriness based on the edge points.

[0021] In this embodiment, S2 includes the following specific steps: S21: Extract the fog detection window area from the preprocessed inner glass image, and use an image semantic segmentation algorithm to divide the fog detection window area into fog feature areas and interference areas; In this embodiment, when the vehicle rolls off the production line, a calibration plate is attached to the inside of the windshield, covering key locations such as the main viewing area and the A-pillar blind spot. An image containing the calibration plate is captured by a camera, and a corner detection algorithm is used to identify feature points on the calibration plate. The pixel coordinates of the glass area in the image are calculated. Based on the calibration results, a polygonal or rectangular area is defined as the fog detection window area.

[0022] In this embodiment, a large number of images of the inner side of the glass, including scenes with different fog concentrations, lighting conditions, and raindrop reflections, are collected. Fog areas and interference areas (such as reflective spots and raindrop marks) in each image are labeled at the pixel level to form a training dataset. A lightweight semantic segmentation network is used for training until convergence. The model input is a fixed-size parameter, and the output is a semantic probability map of the same size as the input. Each pixel corresponds to a probability value of two categories (i.e., fog feature region and interference region). Forward inference of the model is performed to obtain the probability map. For each pixel, the category with the higher probability is selected as its final label, and a binary mask is generated to identify the fog feature region and the interference region, respectively.

[0023] S22: Calculate the average gray value of the fog feature region. and the gray standard deviation of the interference area Based on average gray value and grayscale standard deviation Computational visual signal-to-noise ratio The formula is: ; S23: Perform edge detection on the fog detection window area and construct an edge point set; S24: Calculate fog edge blur based on edge points The formula is: ; in, The total number of edge points, For edge point indexing, For the first The pixel coordinates of the edge points For image gradient operators, It is a constant.

[0024] S3: Calculate the visual credibility haze index based on the visual signal-to-noise ratio, fog edge blurring, and pre-processed camera exposure time; calculate the thermal boundary layer fluctuation index based on the pre-processed inner surface temperature of the glass, and obtain the wet boundary layer fluctuation index based on the pre-processed relative humidity inside the vehicle; obtain the thermal-humidity coupling disturbance index based on the thermal boundary layer fluctuation index, the wet boundary layer fluctuation index, and the pre-processed vehicle speed; and generate the coupling response factor based on the visual credibility haze index and the thermal-humidity coupling disturbance index.

[0025] In this embodiment, S3 includes the following specific steps: S31: Based on visual signal-to-noise ratio Fog edge blur and the pre-processed camera exposure time Calculate the visual credibility haze index The formula is: ; in, For ambiguity threshold, This is the steepness coefficient. For optimal exposure time, As the exposure bias penalty factor, The standard deviation of the reference noise; In this embodiment, the steepness coefficient The α value controls the slope of the sigmoid function near the threshold. The larger the α value, the more significant the impact of ambiguity on the output. The smaller the value, the smoother the transition, and the lower the exposure deviation penalty factor. Used to control the penalty intensity when the exposure time deviates from the optimal value.

[0026] S32: Based on the temperature of the inner surface of the pretreated glass Calculate the thermal boundary layer fluctuation index The formula is: ; in, For the first Temperature of the inner surface of the glass at each measuring point This represents the minimum temperature of the inner surface of the glass at all measuring points. This represents the maximum value of the inner surface temperature of the glass at all measuring points. For airflow velocity gradient, For airflow velocity, Spatial coordinates; In this embodiment, the airflow velocity gradient is used to represent the intensity of airflow disturbance, and its value range is designed to be 0 to 0.5. The larger the value, the stronger the airflow disturbance and the more significant the amplification effect on the temperature field inhomogeneity.

[0027] Based on the pre-treated relative humidity inside the vehicle Obtain the wet boundary layer fluctuation index The formula is: ; in, For the first The relative humidity inside the vehicle at each measuring point This represents the total number of humidity sensors. This represents the average relative humidity inside the vehicle at all measuring points. For time variables, For integration time window, the unit is , For reference respiratory pulse intensity, the unit is... , For the local humidity of the breathing area, The background humidity.

[0028] S33: Based on thermal boundary layer fluctuation index and wet boundary layer fluctuation index Combined with the pre-processed vehicle speed Calculate the thermal-humid coupling perturbation index The formula is: ; in, For reference speed, The influence coefficient of vehicle speed. This is the exponential decay coefficient, in units of... , For the first The dew point temperature corresponding to each measuring point, in units of .

[0029] S34: Based on Visual Credibility Haze Index and thermal-humid coupling perturbation index Generate coupling response factor The formula is: ; in, for The coupling response factor at time t, for Visual credibility haze index at any given time. for The thermal-humid coupling perturbation index at any given time. The visual saliency threshold. The physical significance threshold. This represents the visual channel steepness coefficient. This is the physical channel steepness coefficient. for The derivative of the visual credibility haze index at any given time with respect to time. for The time derivative of the thermal-humid coupling perturbation exponent.

[0030] S4: Smooth the coupling response factor, output a decision threshold based on the pre-processed vehicle speed, ambient light intensity and wiper operation frequency, construct a defogging urgency index based on the smoothed coupling response factor and decision threshold, and start the defogging strategy when the defogging urgency index is greater than or equal to 1.

[0031] In this embodiment, S4 includes the following specific steps: S41: Coupling response factor Perform a moving average process to obtain the smoothed coupling response factor. ; S42: Based on preprocessed vehicle speed Ambient light intensity and the frequency of wiper operation Generate decision threshold The formula is: ; in, for The decision threshold at any given moment. Based on the threshold, This is the illumination correction factor. For reference light intensity, for The ambient light intensity at any given time for The speed of the car at any moment This is the vehicle speed correction factor. For reference speed, This is the wiper correction factor, in units of... , for The frequency of wiper operation at any given time, measured in units of ; S43: Based on the smoothed coupling response factor and decision threshold Construct a fog removal urgency index The formula is: ;in, for The urgency index for defogging at any given moment. for The smoothed coupling response factor at any given time; S44: When the defogging urgency index is greater than or equal to 1, the defogging strategy is activated.

[0032] The specific defogging strategy is as follows: when the defogging urgency index is... When the condition of being greater than or equal to 1 and less than the first preset threshold is met, the warning-level defogging mode is triggered; when the defogging urgency index is... When the condition of being greater than or equal to the first preset threshold and less than the second preset threshold is met, a light-level defogging mode is triggered; when the defogging urgency index... When the condition of being greater than or equal to the second preset threshold and less than the third preset threshold is met, the medium-level defogging mode is triggered; when the defogging urgency index... When the condition of being greater than or equal to the third preset threshold is met, the heavy-level defogging mode is triggered.

[0033] In this embodiment, the first preset threshold, the second preset threshold, and the third preset threshold are determined through actual vehicle calibration. Here, the corresponding selectable values ​​are the first preset threshold, etc. 1 = 1.5, second preset threshold 2 = 2.5, the third preset threshold 3 = 4.0.

[0034] Specifically, the defogging strategy is: when 1≤ < At 1 o'clock, the warning-level defogging mode is triggered, the air conditioner switches to external circulation, the compressor intermittently starts dehumidification, the air outlets are angled 45° to blow air onto the glass, the fan speed is level 2, the windshield heater is turned off or preheated at a low temperature, and the windows automatically close according to the external humidity. 1≤ < At 2 o'clock, the mild defogging mode is triggered, the air conditioning switches to internal circulation, continuously dehumidifies, the air outlet temperature is 25℃, the fan speed is level 3, the windshield heater gradually heats up at low power (30℃), and the windows are forcibly closed. 2≤ < At 3 o'clock, the medium-level defogging mode is triggered, the air conditioner operates at maximum dehumidification power, the air outlet temperature is 28℃, the fan speed is level 5, the windshield heater wires heat up rapidly at medium power (40℃), the windshield wipers intermittently wipe (once every 5 seconds), and the windows are forcibly closed. ≥ At 3 o'clock, the heavy-duty defogging mode is activated. The air conditioner operates at maximum dehumidification capacity and the highest temperature (30℃), with the fan speed at its highest setting. The windshield heater wires heat up rapidly at high power (50℃), the wipers operate intermittently at high frequency (once every 3 seconds), and the windows are forcibly closed. When the fog level is less than 1 and the driver visually confirms that the fog has been largely eliminated, exit the defogging mode.

[0035] Example 2 This embodiment proposes an intelligent automotive defogging system based on forward-facing vision and temperature difference dew point synergy, such as... Figure 2 As shown, it includes: Data acquisition module: used to collect vehicle dynamic data, front camera image data and environmental data and perform preprocessing. Vehicle dynamic data includes vehicle speed and windshield wiper operation frequency. Front camera image data includes camera exposure time and images of the inside of the glass. Environmental data includes the temperature of the inside of the glass, relative humidity inside the vehicle and ambient light intensity. Image processing module: This module extracts the fog detection window area from the preprocessed image of the inside of the glass. It uses an image segmentation algorithm to divide the fog detection window area into fog feature regions and interference regions, and calculates the average gray value of the fog feature regions and the gray standard deviation of the interference regions. Based on the average gray value and the gray standard deviation, it calculates the visual signal-to-noise ratio. It also performs edge detection on the fog detection window area, constructs an edge point set, and calculates the fog edge blurriness based on the edge points. Feature calculation module: Calculates the visually reliable haze index based on visual signal-to-noise ratio, fog edge blurring, and pre-processed camera exposure time; calculates the thermal boundary layer fluctuation index based on the pre-processed inner surface temperature of the glass; obtains the wet boundary layer fluctuation index based on the pre-processed relative humidity inside the vehicle; and obtains the thermal-humidity coupling perturbation index based on the thermal boundary layer fluctuation index, the wet boundary layer fluctuation index, and the pre-processed vehicle speed; and generates the coupling response factor based on the visually reliable haze index and the thermal-humidity coupling perturbation index. Decision module: Used to perform moving average processing on the coupled response factor. Based on the preprocessed vehicle speed, ambient light intensity and wiper operation frequency, it outputs a decision threshold. Based on the smoothed coupled response factor and the decision threshold, it constructs a defogging urgency index. When the defogging urgency index is greater than or equal to 1, the defogging strategy is activated.

[0036] In this embodiment, the image processing module includes the following specific steps: First, the fog detection window area is extracted from the preprocessed inner glass image, and an image semantic segmentation algorithm is used to divide the fog detection window area into fog feature regions and interference regions; then, the average gray value of the fog feature regions is calculated. and the gray standard deviation of the interference area Based on average gray value and grayscale standard deviation Computational visual signal-to-noise ratio The formula is: Further edge detection is performed on the fog detection window area to construct an edge point set; finally, the fog edge blurriness is calculated based on the edge points. The formula is: ; in, The total number of edge points, For edge point indexing, For the first The pixel coordinates of the edge points For image gradient operators, It is a constant.

[0037] In this embodiment, the implementation of the feature calculation module includes the following specific steps: First, based on the visual signal-to-noise ratio... Fog edge blur and the pre-processed camera exposure time Calculate the visual credibility haze index The formula is: ; in, For ambiguity threshold, This is the steepness coefficient. For optimal exposure time, As the exposure bias penalty factor, The baseline noise standard deviation is used; then, based on the temperature of the inner surface of the pretreated glass... Calculate the thermal boundary layer fluctuation index The formula is: ; in, For the first Temperature of the inner surface of the glass at each measuring point This represents the minimum temperature of the inner surface of the glass at all measuring points. This represents the maximum value of the inner surface temperature of the glass at all measuring points. For airflow velocity gradient, For airflow velocity, Using spatial coordinates; further based on the pre-processed relative humidity inside the vehicle. Obtain the wet boundary layer fluctuation index The formula is: ; in, For the first The relative humidity inside the vehicle at each measuring point This represents the total number of humidity sensors. This represents the average relative humidity inside the vehicle at all measuring points. For time variables, For integration time window, the unit is , For reference respiratory pulse intensity, the unit is... , For the local humidity of the breathing area, Background ambient humidity. Then, based on the thermal boundary layer fluctuation index... and wet boundary layer fluctuation index Combined with the pre-processed vehicle speed Calculate the thermal-humid coupling perturbation index The formula is: ; in, For reference speed, The influence coefficient of vehicle speed. This is the exponential decay coefficient, in units of... , For the first The dew point temperature corresponding to each measuring point, in units of Finally, based on the visual credibility haze index and thermal-humid coupling perturbation index Generate coupling response factor The formula is: ; in, for The coupling response factor at time t, for Visual credibility haze index at any given time. for The thermal-humid coupling perturbation index at any given time. The visual saliency threshold. The physical significance threshold. This represents the visual channel steepness coefficient. This is the physical channel steepness coefficient. for The derivative of the visual credibility haze index at any given time with respect to time. for The time derivative of the thermal-humid coupling perturbation exponent.

[0038] In this embodiment, the implementation of the decision-making module includes the following specific steps: First, the coupling response factor is... Perform a moving average process to obtain the smoothed coupling response factor. Then based on the preprocessed vehicle speed Ambient light intensity and the frequency of wiper operation Generate decision threshold The formula is: ; in, for The decision threshold at any given moment. Based on the threshold, This is the illumination correction factor. For reference light intensity, for The ambient light intensity at any given time for The speed of the car at any moment This is the vehicle speed correction factor. For reference speed, This is the wiper correction factor, in units of... , for The frequency of wiper operation at any given time, measured in units of Further based on the smoothed coupling response factor and decision threshold Construct a fog removal urgency index The formula is: ;in, for The urgency index for defogging at any given moment. for The smoothed coupling response factor at each moment; finally, when the defogging urgency index is greater than or equal to 1, the defogging strategy is activated. Specifically, the defogging strategy is as follows: when the defogging urgency index... When the condition of being greater than or equal to 1 and less than the first preset threshold is met, the warning-level defogging mode is triggered; when the defogging urgency index is... When the condition of being greater than or equal to the first preset threshold and less than the second preset threshold is met, a light-level defogging mode is triggered; when the defogging urgency index... When the condition of being greater than or equal to the second preset threshold and less than the third preset threshold is met, the medium-level defogging mode is triggered; when the defogging urgency index... When the condition of being greater than or equal to the third preset threshold is met, the heavy-level defogging mode is triggered.

[0039] The parameters and steps for implementing the corresponding functions of each unit module in the intelligent vehicle defogging system based on forward vision and temperature difference dew point coordination of the present invention can be referred to the parameters and steps in the embodiment of the intelligent vehicle defogging method based on forward vision and temperature difference dew point coordination in Embodiment 1 above.

[0040] Example 3 This embodiment provides an electronic device, including a processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes the above-described intelligent car defogging method based on the synergy of front-facing vision and temperature difference dew point by calling the computer program stored in the memory.

[0041] The electronic device can vary considerably depending on its configuration or performance. It may include one or more Central Processing Units (CPUs) and one or more memories, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the intelligent vehicle defogging method based on forward-facing vision and temperature difference dew point synergy provided in the above-described embodiment. The electronic device may also include other components for implementing its functions; for example, it may have wired or wireless network interfaces and input / output interfaces for data input and output. Details will not be elaborated upon in this embodiment.

[0042] Those skilled in the art will recognize that this invention can be implemented as a system, method, or computer program product. Therefore, this disclosure can be embodied in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, the invention can also be implemented as a computer program product contained in one or more computer-readable media, which includes computer-readable program code.

[0043] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0044] This invention is described with reference to flowchart illustrations and block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and block diagrams, as well as combinations of blocks in the flowchart illustrations and block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0045] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and boxes Figure 1 The steps of the function specified in one or more boxes.

[0046] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. An intelligent vehicle defogging method based on the synergy of forward-facing vision and temperature difference dew point, characterized in that: The specific steps include the following: S1: Collect vehicle dynamic data, front camera image data and environmental data and preprocess them. The vehicle dynamic data includes vehicle speed and windshield wiper operation frequency. The front camera image data includes camera exposure time and image of the inside of the glass. The environmental data includes the temperature of the inside of the glass, the relative humidity inside the vehicle and the ambient light intensity. S2: Extract the fog detection window area from the preprocessed inner glass image. Use an image segmentation algorithm to divide the fog detection window area into fog feature areas and interference areas. Calculate the average gray value of the fog feature areas and the gray standard deviation of the interference areas. Calculate the visual signal-to-noise ratio based on the average gray value and the gray standard deviation. Perform edge detection on the fog detection window area, construct an edge point set, and calculate the fog edge blurriness based on the edge points. S3: Calculate the visual credibility fog index based on visual signal-to-noise ratio, fog edge blurring and pre-processed camera exposure time; The thermal boundary layer fluctuation index is calculated based on the temperature of the inner surface of the pretreated glass, and the wet boundary layer fluctuation index is obtained based on the relative humidity inside the vehicle after pretreatment. The thermal-humid coupling disturbance index is obtained based on the thermal boundary layer fluctuation index, the wet boundary layer fluctuation index, and the vehicle speed after pretreatment. Based on the visual credibility haze index and the thermal-humid coupling perturbation index, a coupling response factor is generated. S4: Smooth the coupling response factor, output a decision threshold based on the pre-processed vehicle speed, ambient light intensity and wiper operation frequency, construct a defogging urgency index based on the smoothed coupling response factor and decision threshold, and start the defogging strategy when the defogging urgency index is greater than or equal to 1.

2. The intelligent vehicle defogging method based on forward-looking vision and temperature difference dew point synergy as described in claim 1, characterized in that: S2 includes the following specific steps: S21: Extract the fog detection window area from the preprocessed inner glass image, and use an image semantic segmentation algorithm to divide the fog detection window area into fog feature areas and interference areas; S22: Calculate the average gray value of the fog feature region. and the gray standard deviation of the interference area Based on average gray value and grayscale standard deviation Computational visual signal-to-noise ratio The formula is: ; S23: Perform edge detection on the fog detection window area and construct an edge point set; S24: Calculate fog edge blur based on edge points The formula is: ； in, The total number of edge points, For edge point indexing, For the first The pixel coordinates of the edge points For image gradient operators, It is a constant.

3. The intelligent vehicle defogging method based on forward-looking vision and temperature difference dew point synergy as described in claim 2, characterized in that: S3 includes the following specific steps: S31: Based on visual signal-to-noise ratio Fog edge blur and the pre-processed camera exposure time Calculate the visual credibility haze index The formula is: ； in, For ambiguity threshold, This is the steepness coefficient. For optimal exposure time, As the exposure bias penalty factor, The standard deviation of the reference noise; S32: Based on the temperature of the inner surface of the pretreated glass Calculate the thermal boundary layer fluctuation index The formula is: ； in, For the first Temperature of the inner surface of the glass at each measuring point This represents the minimum temperature of the inner surface of the glass at all measuring points. This represents the maximum value of the inner surface temperature of the glass at all measuring points. For airflow velocity gradient, For airflow velocity, Spatial coordinates; Based on the pre-treated relative humidity inside the vehicle Obtain the wet boundary layer fluctuation index The formula is: ； in, For the first The relative humidity inside the vehicle at each measuring point This represents the total number of humidity sensors. This represents the average relative humidity inside the vehicle at all measuring points. For time variables, For integration time window, the unit is , For reference respiratory pulse intensity, the unit is... , For the local humidity of the breathing area, The background humidity.

4. The intelligent vehicle defogging method based on forward-looking vision and temperature difference dew point synergy as described in claim 3, characterized in that: S3 further includes the following specific steps: S33: Based on thermal boundary layer fluctuation index and wet boundary layer fluctuation index Combined with the pre-processed vehicle speed Calculate the thermal-humid coupling perturbation index The formula is: ； in, For reference speed, The influence coefficient of vehicle speed. This is the exponential decay coefficient, in units of... , For the first The dew point temperature corresponding to each measuring point, in units of ; S34: Based on Visual Credibility Haze Index and thermal-humid coupling perturbation index Generate coupling response factor The formula is: ； in, for The coupling response factor at time t, for Visual credibility haze index at any given time. for The thermal-humid coupling perturbation index at any given time. The visual saliency threshold. The physical significance threshold. This represents the visual channel steepness coefficient. This is the physical channel steepness coefficient. for The derivative of the visual credibility haze index at any given time with respect to time. for The time derivative of the thermal-humid coupling perturbation exponent.

5. The intelligent vehicle defogging method based on forward-looking vision and temperature difference dew point synergy as described in claim 4, characterized in that: S4 includes the following specific steps: S41: Coupling response factor Perform a moving average process to obtain the smoothed coupling response factor. ; S42: Based on preprocessed vehicle speed Ambient light intensity and the frequency of wiper operation Generate decision threshold The formula is: ； in, for The decision threshold at any given moment Based on the threshold, This is the illumination correction factor. For reference light intensity, for The ambient light intensity at any given time for The speed of the car at any moment This is the vehicle speed correction factor. For reference speed, This is the wiper correction factor, in units of... , for The frequency of wiper operation at any given time, measured in units of ; S43: Based on the smoothed coupling response factor and decision threshold Construct a fog removal urgency index The formula is: ;in, for The urgency index for defogging at any given moment. for The smoothed coupling response factor at any given time; S44: When the defogging urgency index is greater than or equal to 1, the defogging strategy is activated.

6. The intelligent vehicle defogging method based on forward-facing vision and temperature difference dew point synergy as described in claim 5, characterized in that, S44 further includes: the defogging strategy specifically refers to, when the defogging urgency index... When the condition of being greater than or equal to 1 and less than the first preset threshold is met, the warning-level defogging mode is triggered; when the defogging urgency index is... When the condition of being greater than or equal to the first preset threshold and less than the second preset threshold is met, a light-level defogging mode is triggered; when the defogging urgency index... When the condition of being greater than or equal to the second preset threshold and less than the third preset threshold is met, the medium-level defogging mode is triggered; when the defogging urgency index... When the condition of being greater than or equal to the third preset threshold is met, the heavy-level defogging mode is triggered.

7. An intelligent automotive defogging system based on forward-looking vision and temperature difference dew point synergy, according to any one of claims 1 to 6, characterized in that: include: Data acquisition module: used to collect vehicle dynamic data, front camera image data and environmental data and preprocess them. The vehicle dynamic data includes vehicle speed and windshield wiper operation frequency. The front camera image data includes camera exposure time and images of the inside of the glass. The environmental data includes the temperature of the inside of the glass, the relative humidity inside the vehicle and the ambient light intensity. Image processing module: This module extracts the fog detection window area from the preprocessed image of the inside of the glass. It uses an image segmentation algorithm to divide the fog detection window area into fog feature regions and interference regions, and calculates the average gray value of the fog feature regions and the gray standard deviation of the interference regions. Based on the average gray value and the gray standard deviation, it calculates the visual signal-to-noise ratio. It also performs edge detection on the fog detection window area, constructs an edge point set, and calculates the fog edge blurriness based on the edge points. Feature calculation module: used to calculate the visual credibility haze index based on visual signal-to-noise ratio, fog edge blurring and pre-processed camera exposure time; The thermal boundary layer fluctuation index is calculated based on the temperature of the inner surface of the pretreated glass, and the wet boundary layer fluctuation index is obtained based on the relative humidity inside the vehicle after pretreatment. The thermal-humid coupling disturbance index is obtained based on the thermal boundary layer fluctuation index, the wet boundary layer fluctuation index, and the vehicle speed after pretreatment. Based on the visual credibility haze index and the thermal-humid coupling perturbation index, a coupling response factor is generated. Decision module: Used to perform moving average processing on the coupled response factor. Based on the preprocessed vehicle speed, ambient light intensity and wiper operation frequency, it outputs a decision threshold. Based on the smoothed coupled response factor and the decision threshold, it constructs a defogging urgency index. When the defogging urgency index is greater than or equal to 1, the defogging strategy is activated.

Images