CN112153304A - Exposure adjusting method and system, driver monitoring system and advanced driving assistance system - Google Patents
Exposure adjusting method and system, driver monitoring system and advanced driving assistance system Download PDFInfo
- Publication number
- CN112153304A CN112153304A CN202011044803.9A CN202011044803A CN112153304A CN 112153304 A CN112153304 A CN 112153304A CN 202011044803 A CN202011044803 A CN 202011044803A CN 112153304 A CN112153304 A CN 112153304A
- Authority
- CN
- China
- Prior art keywords
- exposure
- area
- image
- mean value
- measured
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/70—Circuitry for compensating brightness variation in the scene
- H04N23/74—Circuitry for compensating brightness variation in the scene by influencing the scene brightness using illuminating means
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/70—Circuitry for compensating brightness variation in the scene
- H04N23/71—Circuitry for evaluating the brightness variation
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Studio Devices (AREA)
- Exposure Control For Cameras (AREA)
Abstract
The invention discloses an exposure adjusting method and system, a driver monitoring system and an advanced driving auxiliary system, which are used for adjusting the exposure of an object to be measured. Firstly, acquiring an image to be detected containing the object to be detected; then, counting the brightness mean value and the histogram data of the interest area in the image to be detected, and judging the current scene based on the brightness mean value and the histogram data; and acquiring data of the object area to be measured through the current scene, and finally calculating the exposure parameter and the gain parameter of the object to be measured. By means of the scheme, accurate exposure of the object to be tested can be guaranteed under various illumination environments. And distinguishing the foreground, the background and different scenes through the statistics of brightness and histogram data, and further setting exposure parameters and gain parameters according to the current configuration and scenes.
Description
Technical Field
The invention relates to the field of image processing, in particular to an exposure adjusting method and system, a driver monitoring system and an advanced driving assistance system.
Background
The automobile becomes an indispensable part of modern society, and with the continuous update of industrial technologies, the improvement of Driving safety has gradually become a core appeal of the automobile industry, so that the born ADAS (Advanced Driving Assistance System) senses the surrounding environment at any time in the Driving process of the automobile by using various sensors installed on the automobile, collects data, identifies, detects and tracks static objects and dynamic objects, and performs systematic operation and analysis by combining map data of a navigator, thereby enabling a driver to detect possible dangers in advance, and effectively increasing the comfort and safety of automobile Driving. A DMS (Driver Monitor System) in the ADAS can detect a fatigue state, a driving behavior, and the like of a Driver in real time while the Driver is driving. After finding that the driver has fatigue, yawning, squinting and other wrong driving states, the early warning system can analyze the behaviors in time and carry out voice or light prompt. Thereby achieving the functions of warning the driver and preventing accidents.
DMS cameras typically employ a near infrared LED of 940nm wavelength as the light source. The light source with the wavelength has different light reflection rates aiming at different materials, and the human face skin has higher light reflection rate, so the brightness is higher generally. The wavelength band of 940nm exists in sunlight, and when a vehicle runs in the reverse direction and sunlight directly irradiates a human face, the exposure of the human face deviates from a target value. The DMS camera output data is usually connected with application units such as machine learning, and the data is used for face recognition and eyeball tracking calculation. Because the vehicle-mounted illumination environment changes violently, the exposure parameters and the gain parameters of the human face cannot be adjusted accurately, and errors occur in subsequent machine learning calculation.
Therefore, it is necessary to provide a scheme for automatically adjusting the exposure parameter and the gain parameter.
Disclosure of Invention
The invention aims to provide an exposure adjusting method and system, a driver monitoring system and an advanced driving auxiliary system, which are used for solving the problem that in the prior art, because the vehicle-mounted illumination environment changes violently, the exposure parameter and the gain parameter of a human face cannot be accurately adjusted, and the subsequent machine learning calculation is wrong.
In order to solve the above technical problem, the present invention provides an exposure adjustment method for adjusting an exposure of an object to be measured, the exposure adjustment method comprising the steps of:
acquiring a to-be-detected image containing the to-be-detected object;
counting the brightness mean value and histogram data of the interest region in the image to be detected, wherein the histogram data comprises a statistical distribution graph of each pixel value in the image;
judging a current scene based on the brightness mean value and the histogram data;
acquiring data of an area to be detected based on a current scene;
and calculating the exposure parameter and the gain parameter of the object to be measured based on the current scene and the data of the object to be measured.
Optionally, the region of interest comprises a core region and a background region;
the core area is the part of the interested area where the object to be detected is located, and the background area is the part of the interested area where the background is located.
Optionally, the counting the brightness mean and histogram data of the region of interest in the image to be measured includes:
setting the core area and the background area;
respectively counting the image data of the core area and the image data of the background area;
counting the brightness mean value of the core area based on the image data of the core area, and counting the brightness mean value of the background area based on the image data of the background area;
and counting the histogram data of the image to be detected and the histogram data of the core area.
Optionally, the determining a current scene based on the brightness mean and the histogram data includes:
and judging the current scene based on the brightness mean value of the background area, the brightness mean value of the core area and the histogram data of the image to be detected.
Optionally, the scene includes a fill light environment and a backlight environment;
when the brightness mean value of the background area is smaller than that of the core area, judging that the current scene is the environment of the light supplement lamp;
and when the histogram data of the image to be detected is uniformly distributed, judging that the current scene is the backlight environment.
Optionally, the backlighting environment comprises a large backlighting environment, a side-lit environment, and a top-lit environment;
when the brightness mean value of the background area is increased, judging that the current scene is the large backlight environment;
when the brightness mean value of the background area shows that one side value is large and the other side value is small, judging that the current scene is the sidelight environment;
and when the value of the upper side of the brightness mean value of the background area is large, judging that the current scene is the top light environment.
Optionally, the acquiring data of the object region includes marking a position of the object in the region of interest by adjusting the weight.
Optionally, the weights are adjusted by:
for the area exceeding a preset threshold value in the brightness mean value of the core area, distributing weight according to a reverse S curve;
distributing weight to the area which is lower than the preset threshold value in the brightness mean value of the core area according to an S curve;
the S curve is a direct proportion relation between the brightness mean value and the weight, and the reverse S curve is an inverse proportion relation between the brightness mean value and the weight.
Optionally, the calculating an exposure parameter and a gain parameter of the object based on the current scene and the data of the object region includes:
calculating the exposure parameter and the gain parameter of the object to be measured according to the following formulas:
wherein NewExp is an exposure parameter to be calculated of the object to be measured; NewGain is a gain parameter to be calculated of the object to be measured; target _ Luma is a Target brightness value; the Current _ Luma is an average brightness value of the object to be measured in the Current core area after the weight weighting calculation; k is an adjustment control coefficient; the Current _ Exp is a Current exposure parameter of the object to be measured, and the Current _ Gain is a Current Gain parameter of the object to be measured.
Based on the same inventive concept, the present invention further provides an exposure adjustment system for adjusting the exposure of an object to be measured, the exposure adjustment system comprising:
an image acquisition unit configured to acquire an image to be measured including the object to be measured;
a statistical unit configured to count a luminance mean value and histogram data of a region of interest in the image to be measured, the histogram data including a statistical distribution map of each pixel value in the image;
a scene determination unit configured to determine a current scene based on the luminance mean value and the histogram data;
a test object region acquisition unit configured to acquire data of a test object region based on a current scene;
a calculation unit configured to calculate an exposure parameter and a gain parameter of the object based on data of a current scene and the object region.
Based on the same inventive concept, the invention also provides a driver monitoring system, which utilizes the exposure adjusting method or the exposure adjusting system in any one of the above characteristic descriptions.
Based on the same inventive concept, the invention also provides an advanced driving assistance system, which comprises the driver monitoring system.
Compared with the prior art, the invention has the following beneficial effects:
1. the exposure adjusting method provided by the invention is used for adjusting the exposure of an object to be measured. Firstly, acquiring an image to be detected containing the object to be detected; then, counting the brightness mean value and the histogram data of the interest area in the image to be detected, and judging the current scene based on the brightness mean value and the histogram data; and acquiring data of the object area to be measured through the current scene, and finally calculating the exposure parameter and the gain parameter of the object to be measured. By means of the scheme, accurate exposure of the object to be tested can be guaranteed under various illumination environments. And distinguishing the foreground, the background and different scenes through the statistics of brightness and histogram data, and further setting exposure parameters and gain parameters according to the current configuration and scenes. Due to accurate exposure time calculation, exposure can be adjusted in real time along with the environment, the requirement of machine learning on the accuracy of each frame of data is met, and the influence of frame dropping or frame breaking on the calculation result of subsequent machine learning is avoided. By using the exposure adjustment method provided by the application, accurate exposure can be realized for the face in DMS application on the premise of not executing complex algorithms such as face detection and the like by analyzing a large amount of scene data, and when the exposure adjustment method is applied in the DMS, the object to be measured is the face. By using the exposure adjustment method provided by the application, calculation and storage resources can be saved, and power consumption can be reduced. In addition, the device can be embedded into a CIS (CMOS Image Sensor) for DMS, so that the cost problems such as resource occupation and power consumption are not worried about, and the device has high cost performance.
2. Under a common condition, the scene can be divided into a light supplement lamp environment and a backlight environment, the light supplement lamp environment is a reference environment for human face exposure, and ideal exposure adjustment is most easily achieved. The backlight environment means that when 940nm band near infrared light in sunlight irradiates on the object to be detected, accurate judgment of the object area to be detected is influenced. Therefore, whether the current scene is a fill light environment or a backlight environment is judged according to the brightness mean value of the background area, the brightness mean value of the core area and the histogram data of the image to be detected. Different exposure adjustment schemes can be realized through the scene discrimination, and the accurate judgment of the area to be detected is improved.
3. In practical applications, there are other situations in a backlight environment, for example, sunlight may be incident from a side, from a top, or from a front, which is likely to cause uneven exposure of the object to be measured, and thus, accurate determination of the object region to be measured is affected. In order to further improve the accurate judgment of the area to be detected, the backlight environment is further divided into a large backlight environment, a side light environment and a top light environment.
The invention also provides an exposure adjusting system, a driver monitoring system and an advanced driving auxiliary system, which belong to the same inventive concept with the exposure adjusting method, so the exposure adjusting system has the same beneficial effects.
Drawings
Fig. 1 is a schematic flowchart of an exposure adjustment method according to an embodiment of the present invention;
fig. 2 is a schematic diagram illustrating region of interest division according to an embodiment of the present invention;
FIG. 3 is a histogram diagram of a normal environment;
FIG. 4 is a histogram diagram of a large backlight environment;
FIG. 5 is a schematic diagram of a core region weight assignment S-curve and an inverse S-curve;
FIG. 6 is a histogram of an exposure area of a face region;
FIG. 7 is a schematic view showing the distribution of the lens and the fill-in light of the DMS camera module;
FIG. 8 is a schematic diagram illustrating a light supplement method of the DMS module;
fig. 9 is a schematic diagram of a control strategy of the fill-in light;
FIG. 10 is a schematic view of an exposure adjustment system according to another embodiment of the present invention;
wherein, in fig. 2 and 10: 10-an object to be measured, 20-an area of interest, 201-a core area, 202-a background area, 30-an exposure adjusting system, 301-an image acquiring unit, 302-a statistical unit, 303-a scene judging unit, 304-an object area acquiring unit and 305-a calculating unit.
Detailed Description
The following describes in more detail embodiments of the present invention with reference to the schematic drawings. The advantages and features of the present invention will become more apparent from the following description. It is to be noted that the drawings are in a very simplified form and are not to precise scale, which is merely for the purpose of facilitating and distinctly claiming the embodiments of the present invention.
In the description of the present invention, it is to be understood that the terms "center", "upper", "lower", "left", "right", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience in describing the present invention and simplifying the description, but do not indicate or imply that the device or element referred to must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Referring to fig. 1, an embodiment of the invention provides an exposure adjustment method for adjusting an exposure of an object 10 to be measured, the exposure adjustment method including the following steps:
s1: acquiring a to-be-detected image containing the to-be-detected object 10;
s2: counting the brightness mean value and histogram data of the interest region 20 in the image to be detected, wherein the histogram data comprises a statistical distribution graph of each pixel value in the image;
s3: judging a current scene based on the brightness mean value and the histogram data;
s4: acquiring data of an area to be detected based on a current scene;
s5: based on the current scene and the data of the object region, the exposure parameter and the gain parameter of the object 10 are calculated.
The difference from the prior art is that the exposure adjustment method provided by the present invention is used for adjusting the exposure of an object 10 to be measured. Firstly, acquiring an image to be detected containing the object to be detected 10; then, counting the brightness mean value and the histogram data of the interest area 20 in the image to be detected, and judging the current scene based on the brightness mean value and the histogram data; and acquiring data of the object area to be measured through the current scene, and finally calculating the exposure parameter and the gain parameter of the object 10 to be measured. By utilizing the scheme of the application, the accurate exposure of the object to be measured 10 can be ensured under various illumination environments. And distinguishing the foreground, the background and different scenes through the statistics of brightness and histogram data, and further setting exposure parameters and gain parameters according to the current configuration and scenes. Due to accurate exposure time calculation, exposure can be adjusted in real time along with the environment, the requirement of machine learning on the accuracy of each frame of data is met, and the influence of frame dropping or frame breaking on the calculation result of subsequent machine learning is avoided. By using the exposure adjustment method provided by the application, accurate exposure can be realized for a human face in DMS application on the premise of not executing complex algorithms such as human face detection and the like by analyzing a large amount of scene data, and when the exposure adjustment method is applied in the DMS, the object to be measured 10 is the human face. By using the exposure adjustment method provided by the application, calculation and storage resources can be saved, and power consumption can be reduced. In addition, the Sensor can be embedded into a CIS (CMOS Image Sensor) for DMS, so that cost such as resource occupation and power consumption is not needed to be worried about, and higher cost performance is achieved. As will be understood by those skilled in the art, the region of interest 20 (ROI) refers to a region that needs to be processed and is delineated from the processed image in the form of a box, a circle, an ellipse, or an irregular polygon in the machine vision and image processing. In addition, the object 10 to be measured may be any object that needs exposure adjustment, for convenience of understanding the technical solution of the present application, the object 10 to be measured is specifically described below by taking a human face as an example, and other objects 10 to be measured are similar to the human face.
Further, referring to fig. 2, the region of interest 20 includes a core area 201 and a background area 202, where the core area 201 is a portion of the region of interest 20 where the object 10 to be measured is located, and the background area 202 is a portion of the region of interest 20 where the background is located.
Optionally, the step of counting the luminance mean and histogram data of the region of interest 20 in the image to be measured, that is, the step S2 includes:
setting the core area 201 and the background area 202;
respectively counting the image data of the core area 201 and the image data of the background area 202;
counting the brightness mean value of the core region 201 based on the image data of the core region 201, and counting the brightness mean value of the background region 202 based on the image data of the background region 202;
and counting the histogram data of the image to be detected and the histogram data of the core area 201.
By setting the core area 201 and the background area 202, image data of an area where a human face is located and image data of an area where a background is located are respectively counted. The ROI window of the core region 201 and the background region 202 may be arbitrarily set according to the requirement. As can be seen from fig. 2, in the embodiment of the present invention, 64 sub-regions of 8 × 8 are provided in the core region 201. It can be understood that, because the DMS camera mounting position, the height, the face shape, the sitting posture of the driver, and the like may be different, the size and the position of the core area 201 and the sub-areas thereof may be adjusted arbitrarily according to the actual situation. In addition, the core area 201 only needs to cover the face area in a large range, and is not required to be completely accurate, and the subsequent algorithm can identify the face area from the core area 201. The background region 202 is typically a plurality of small ROI statistics windows distributed around the core region 201. By setting the core area 201 and the background area 202 which can freely select the range, the requirements under various different conditions can be met, and the universality of the exposure adjusting method is improved.
It is noted that each statistical window of the region of interest requires a mean value of the image within the statistical window. Meanwhile, it is necessary to count the histogram data of the entire image and separately count the histogram data of the core region 201. Fig. 2 is only an example of a statistical region dividing method, and cannot be understood as a limitation to the division of the background region 202 and the core region 201 in the present application, in other embodiments, the core region 201 and the background region 202 have many other dividing methods, as long as the main purpose is to perform statistics on the face core region 201 and the background region 202 respectively, which can be implemented, and the statistical data is used to support functions of accurately recognizing a face and judging a current scene.
Optionally, the determining a current scene based on the brightness mean and the histogram data includes: and judging the current scene based on the brightness mean value of the background area 202, the brightness mean value of the core area 201 and the histogram data of the image to be detected. The histogram data is a statistical distribution graph of each pixel value of an image, and the distribution condition of bright and dark pixels in an image can be known through the histogram data, so that the illumination environment can be conveniently distinguished by combining the histogram data and the brightness mean value. Under different illumination environments, the background region 202, the core region 201, and the histogram data and the brightness mean value of the image to be detected may present different features, and different scenes may be distinguished based on the distinguishing features.
Further, the scene comprises a fill light environment and a backlight environment;
when the brightness mean value of the background area 202 is smaller than the brightness mean value of the core area 201, determining that the current scene is the fill light environment;
and when the histogram data of the image to be detected is uniformly distributed, judging that the current scene is the backlight environment.
Under a common condition, the scene can be divided into a light supplement lamp environment and a backlight environment, the light supplement lamp environment is a reference environment for human face exposure, and ideal exposure adjustment is most easily achieved. The backlight environment means that when the 940nm band near infrared light in the sunlight irradiates the object to be measured 10, the accurate judgment of the object area to be measured is influenced. Therefore, whether the current scene is a fill light environment or a backlight environment is judged according to the brightness mean value of the background area 202, the brightness mean value of the core area 201 and the histogram data of the image to be detected. Different exposure adjustment schemes can be realized through the scene discrimination, and the accurate judgment of the area to be detected is improved.
Preferably, the backlighting environment comprises a large backlighting environment, a side-lit environment, and a top-lit environment;
when the brightness mean value of the background area 202 becomes large, judging that the current scene is the large backlight environment;
when the brightness mean value of the background area 202 shows that one side is large and the other side is small, judging that the current scene is the side light environment;
when the value of the upper side of the brightness mean value of the background area 202 is large, the current scene is determined to be the top light environment.
In practical applications, there are other situations in a backlight environment, for example, sunlight may be incident from a side, from a top, or from a front, which is likely to cause uneven exposure of the object 10, and affect accurate determination of the object region. In order to further improve the accurate judgment of the area to be detected, the backlight environment is further divided into a large backlight environment, a side light environment and a top light environment.
As can be seen from the above analysis, the basis for scene determination is the luminance average value of each region, the histogram data of the image to be measured, and the histogram data of the core region 201. The exposure environment under the condition of the 940nm fill light can be generally divided into a pure fill light environment, a large-backlight environment under the condition of direct sunlight, a sidelight environment irradiated by the side of sunlight and a top light environment irradiated by the sunlight from the top of the head (through a skylight or a convertible car). The pure light supplement environment is a reference environment for face exposure, and ideal exposure adjustment is most easily achieved. Because the sunlight contains strong 940nm waveband near-infrared light, the phenomenon of uneven facial exposure is easy to occur in a large-backlight environment, a side-light environment and a top-light environment, and the accurate judgment of a face area is influenced. Therefore, different exposure adjustment schemes need to be implemented according to different environments.
Referring to fig. 3, fig. 3 is a histogram of a pure fill-in light environment, where the abscissa in fig. 3 is a pixel value and the ordinate in fig. 3 is a pixel number, and it can be seen from fig. 3 that the histogram shows two-side bimodal distribution in the pure fill-in light environment. The general light source is a near infrared light source with 940nm, and the light source directly irradiates the face. Therefore, only the human face and the surroundings, such as clothes, seats, etc., have the light reflecting condition, and the background light is weak, and the CIS hardly senses the background light. At this time, according to the luminance mean values counted by the region of interest 20, if the luminance mean values of the background region 202 are far lower than the luminance mean value of the core region 201, it is determined that the scene at this time is a pure fill light environment.
In a backlight environment, a 940nm light source also exists in sunlight, and the light source is strong. Therefore, the background can have stronger light reflection, the shape of the histogram which is in bimodal distribution at two sides under the pure light supplement environment can be correspondingly changed, the relatively uniform distribution characteristic of fig. 4 is presented, the abscissa in fig. 4 is the pixel value, and the ordinate in fig. 4 is the pixel number. At this time, if the brightness mean value of the region of interest 20 and the brightness mean value of the background region are uniformly increased, it can be determined that the scene is in a large backlight environment. Except for the direct sunlight environment, if the sunlight is directed through the side or the top, the determination is made by the mean brightness value of the region of interest 20. In the side-light environment, the background area 202 may have a larger average luminance value on one side and a smaller average luminance value on one side. If sunlight is irradiated from a top skylight, that is, under a top light environment, a characteristic that the average value of the luminance of the upper side of the background area 202 is large occurs.
Therefore, by the characteristics of the luminance mean value of the background area 202 of the region of interest 20 and the shape of the histogram data, each specific scene can be conveniently distinguished, so as to further guide the exposure calculation and the calculation of the fill-in light intensity of the LED fill-in light. It should be noted that the scene determination is not limited to the above-described scene, and more scene details can be distinguished according to the situations of the region of interest 20 and the histogram, which are not described herein again.
Further, the acquiring data of the object area in step S4 includes marking the position of the object 10 in the region of interest 20 by adjusting the weight. When the core area 201 is set, substantially complete coverage of the possible positions of the human face is already achieved. The purpose of step S4 is to mark the precise location of the face in the region of interest 20 by means of adjusting the weight.
Further, the weights are adjusted by:
for an area exceeding a preset threshold in the brightness mean value of the core area 201, distributing weight according to a reverse S curve;
for an area lower than the preset threshold in the brightness mean value of the core area 201, assigning a weight according to an S-curve;
the S curve is a direct proportion relation between the brightness mean value and the weight, and the reverse S curve is an inverse proportion relation between the brightness mean value and the weight.
Please refer to fig. 5 for the shapes of the S-curve and the inverse S-curve, an abscissa in fig. 5 is a luminance average value of each sub-region of the core region 201, and an ordinate in fig. 5 is a weight assigned to each sub-region. In the core area 201, the human face position has higher brightness than that of a general object due to the fact that the reflection rate of the human face is higher. According to the S-shaped curve of FIG. 5, the face region can be labeled by setting the weight of the pixel values of the image according to the trend of the S-shaped curve. If the area of the face is brighter in fig. 2, the assigned weight is higher, and the rest of the area has lower weight. Under some conditions, the reflection rate of materials of clothes and the like is higher than that of the human face, at the moment, reverse labeling needs to be carried out on an S curve, namely, a region exceeding a normal human face exposure value needs to be assigned with lower weight, and therefore the influence of the materials with higher reflection rate on the labeling of the human face region can be eliminated. Fig. 6 is a statistical histogram of the core region 201 under normal conditions, where the abscissa in fig. 6 is the pixel value and the ordinate in fig. 6 is the number of pixels. If there is only a high reflection from the face, the distribution of the histogram is unimodal. When the reflection rate is higher than that of the material of the human face, the histogram has the characteristic of bimodal distribution in a highlight area. At this time, an appropriate threshold is adaptively set according to the bimodal distribution of the histogram, the part above the threshold is weighted by using the inverse sigmoid curve in fig. 5, and the part below the threshold is weighted by using the sigmoid curve.
Further, the calculating of the exposure parameter and the gain parameter of the object to be measured 10 based on the current scene and the data of the object to be measured, that is, the step S5 specifically includes:
calculating the exposure parameter and the gain parameter of the object 10 according to the following formulas:
wherein NewExp is an exposure parameter to be calculated of the object to be measured 10; NewGain is a gain parameter to be calculated of the object to be measured 10; target _ Luma is a Target brightness value; the Current _ Luma is an average brightness value of the object 10 to be measured in the Current core area 201 after the weight weighting calculation; k is an adjustment control coefficient; current _ Exp is a Current exposure parameter of the object to be measured 10, and Current _ Gain is a Current Gain parameter of the object to be measured 10.
And taking the weighted average value obtained by the face region identification as the current face exposure value. This value is compared to a target value. When the value enters a position near the target value, the current exposure and gain configuration is maintained. But the value is further from the target value, a new exposure and calculation of a new gain value are initiated.
In formula 1, K is an adjustment control coefficient for controlling the speed of exposure and gain to reach the target brightness, so as to avoid flicker. During exposure adjustment, to
As a basis for the adjustment.
When the target brightness value is greater than the average brightness value of the object 10 to be measured in the current core area 201 after the weight weighting calculation, the ratio is greater than 1, and the exposure parameter and the gain parameter need to be increased; when the target brightness value is smaller than the average brightness value obtained by weighting and calculating the brightness value of the object 10 in the current core area 201, the exposure parameter and the gain parameter are correspondingly reduced. The specific adjustment process is as follows: when the picture is adjusted from dark to bright, the exposure is prior, and when the exposure does not reach the maximum value, the exposure is adjusted firstly; when the exposure reaches the maximum value, the gain is adjusted again. When the picture is adjusted from bright to dark, the gain is prioritized, and when the gain does not reach the minimum value, the gain is adjusted firstly; when the gain reaches a minimum, the exposure is readjusted. K can be used to control the adjustment speed, and when K is 1, the adjustment can be performed in one step, but there is a possibility of screen flicker, and in practical use, there is a trade-off between convergence speed and image stability. If K is equal to 1, the image can be adjusted in place in one step, and the purpose of rapid exposure convergence is achieved.
It should be noted that, in order to obtain a better exposure effect, in addition to adjusting the exposure parameter and the gain parameter of the object to be measured 10, the fill-in light may be adjusted accordingly. Referring to fig. 7 to 9, fig. 7 is a schematic distribution diagram of a lens of a DMS camera module and a fill-in light; FIG. 8 is a schematic diagram illustrating a light supplement method of the DMS module; fig. 9 is a schematic diagram of a control strategy of the fill-in light; and performing light supplement control and face exposure calculation according to the classification of the scene and the exposure mean value of the core region 201 after the face exposure region is identified. Generally, there are 4 or more than 4 LED fill-in lamps in the DMS module, for convenience of understanding, in the embodiment of the present invention, 4 LED fill-in lamps are taken as an example for description, and the situation of other numbers of LED fill-in lamps is similar, and thus details are not described herein. As shown in fig. 8, the range of the general fill light is designed such that the fill light substantially covers the core area 201. And (4) performing light supplement control in different scenes, and if the recognition result is a pure light supplement environment, fully turning on a light supplement lamp to directly perform exposure calculation. Referring to fig. 9, if it is identified as a large backlight environment, the 940nm component in sunlight is strong, which is likely to cause overexposure of the face region and the surrounding background region 202, so the intensity of the LED fill light is first reduced, even the fill light is completely turned off, and after the fill light is reduced, the 940nm light source in sunlight is directly used for face exposure calculation. If the side light or the top light environment is identified, the brighter side light source is turned off, and the darker side light source is turned on. The shadow part of the human face is supplemented with light through a darker side light source, so that the reflecting intensity of the human face is close to the same as the sunlight, and the condition of yin and yang faces is avoided.
Based on the same inventive concept, referring to fig. 10, another embodiment of the present invention further provides an exposure adjustment system 30 for adjusting the exposure of an object 10 to be measured by using the exposure adjustment method, wherein the exposure adjustment system 30 includes:
an image acquisition unit 301 configured to acquire an image to be measured including the object to be measured 10;
a statistic unit 302 configured to count a luminance mean value and histogram data of the region of interest 20 in the image to be tested, wherein the histogram data includes a statistical distribution graph of each pixel value in the image;
a scene determination unit 303 configured to determine a current scene based on the luminance mean value and the histogram data;
an object region acquisition unit 304 configured to acquire data of an object region based on a current scene;
a calculation unit 305 configured to calculate an exposure parameter and a gain parameter of the object 10 based on the current scene and the data of the object region.
It is understood that the image obtaining unit 301, the statistic unit 302, the scene determining unit 303, the object region obtaining unit 304, and the calculating unit 305 may be combined and implemented in one device, or any one of the modules may be divided into a plurality of sub-modules, or at least part of functions of one or more of the image obtaining unit 301, the statistic unit 302, the scene determining unit 303, the object region obtaining unit 304, and the calculating unit 305 may be combined with at least part of functions of other modules and implemented in one functional module. According to the embodiment of the present invention, at least one of the image obtaining unit 301, the statistic unit 302, the scene determining unit 303, the object region obtaining unit 304, and the calculating unit 305 may be at least partially implemented as a hardware circuit, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system on a chip, a system on a substrate, a system on a package, an Application Specific Integrated Circuit (ASIC), or may be implemented in hardware or firmware in any other reasonable manner of integrating or packaging a circuit, or implemented in a suitable combination of three implementations of software, hardware, and firmware. Alternatively, at least one of the image acquisition unit 301, the statistic unit 302, the scene determination unit 303, the object region acquisition unit 304, and the calculation unit 305 may be at least partially implemented as a computer program module, and when the program is executed by a computer, the function of the corresponding module may be executed.
Based on the same inventive concept, the embodiment of the invention further provides a driver monitoring system, which utilizes the exposure adjustment method or the exposure adjustment system described in any one of the above feature descriptions.
Based on the same inventive concept, the embodiment of the invention also provides an advanced driving assistance system, which comprises the driver monitoring system.
In conclusion, the invention has the following beneficial effects:
1. the exposure adjusting method provided by the invention is used for adjusting the exposure of an object to be measured. Firstly, acquiring an image to be detected containing the object to be detected; then, counting the brightness mean value and the histogram data of the interest area in the image to be detected, and judging the current scene based on the brightness mean value and the histogram data; and acquiring data of the object area to be measured through the current scene, and finally calculating the exposure parameter and the gain parameter of the object to be measured. By means of the scheme, accurate exposure of the object to be tested can be guaranteed under various illumination environments. And distinguishing the foreground, the background and different scenes through the statistics of brightness and histogram data, and further setting exposure parameters and gain parameters according to the current configuration and scenes. Due to accurate exposure time calculation, exposure can be adjusted in real time along with the environment, the requirement of machine learning on the accuracy of each frame of data is met, and the influence of frame dropping or frame breaking on the calculation result of subsequent machine learning is avoided. By using the exposure adjustment method provided by the application, accurate exposure can be realized for the face in DMS application on the premise of not executing complex algorithms such as face detection and the like by analyzing a large amount of scene data, and when the exposure adjustment method is applied in the DMS, the object to be measured is the face. By using the exposure adjustment method provided by the application, calculation and storage resources can be saved, and power consumption can be reduced. In addition, the Sensor can be embedded into a CIS (CMOS Image Sensor) for DMS, so that cost such as resource occupation and power consumption is not needed to be worried about, and higher cost performance is achieved.
2. Under a common condition, the scene can be divided into a light supplement lamp environment and a backlight environment, the light supplement lamp environment is a reference environment for human face exposure, and ideal exposure adjustment is most easily achieved. The backlight environment means that when 940nm band near infrared light in sunlight irradiates on the object to be detected, accurate judgment of the object area to be detected is influenced. Therefore, whether the current scene is a fill light environment or a backlight environment is judged according to the brightness mean value of the background area, the brightness mean value of the core area and the histogram data of the image to be detected. Different exposure adjustment schemes can be realized through the scene discrimination, and the accurate judgment of the area to be detected is improved.
3. In practical applications, there are other situations in a backlight environment, for example, sunlight may be incident from a side, from a top, or from a front, which is likely to cause uneven exposure of the object to be measured, and thus, accurate determination of the object region to be measured is affected. In order to further improve the accurate judgment of the area to be detected, the backlight environment is further divided into a large backlight environment, a side light environment and a top light environment.
The invention also provides an exposure adjusting system, a driver monitoring system and an advanced driving auxiliary system, which belong to the same inventive concept with the exposure adjusting method, so the exposure adjusting system has the same beneficial effects.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example" or "a specific example" or the like are intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. And the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments. Furthermore, various embodiments or examples described in this specification can be combined and combined by those skilled in the art.
The above description is only a preferred embodiment of the present invention, and does not limit the present invention in any way. It will be understood by those skilled in the art that various changes, substitutions and alterations can be made herein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (12)
1. An exposure adjustment method for adjusting exposure of an object, the exposure adjustment method comprising:
acquiring a to-be-detected image containing the to-be-detected object;
counting the brightness mean value and histogram data of the interest region in the image to be detected, wherein the histogram data comprises a statistical distribution graph of each pixel value in the image;
judging a current scene based on the brightness mean value and the histogram data;
acquiring data of an area to be detected based on a current scene;
and calculating the exposure parameter and the gain parameter of the object to be measured based on the current scene and the data of the object to be measured.
2. The exposure adjustment method according to claim 1, wherein the region of interest includes a core region and a background region;
the core area is the part of the interested area where the object to be detected is located, and the background area is the part of the interested area where the background is located.
3. The exposure adjustment method of claim 2, wherein the counting the luminance mean and histogram data of the region of interest in the image to be tested comprises:
setting the core area and the background area;
respectively counting the image data of the core area and the image data of the background area;
counting the brightness mean value of the core area based on the image data of the core area, and counting the brightness mean value of the background area based on the image data of the background area;
and counting the histogram data of the image to be detected and the histogram data of the core area.
4. The exposure adjustment method of claim 3, wherein the determining a current scene based on the luminance mean and the histogram data comprises:
and judging the current scene based on the brightness mean value of the background area, the brightness mean value of the core area and the histogram data of the image to be detected.
5. The exposure adjustment method according to claim 4, wherein the scene includes a fill light environment and a backlight environment;
when the brightness mean value of the background area is smaller than that of the core area, judging that the current scene is the environment of the light supplement lamp;
and when the histogram data of the image to be detected is uniformly distributed, judging that the current scene is the backlight environment.
6. The exposure adjustment method according to claim 5, wherein the backlight environment includes a large backlight environment, a side-light environment, and a top-light environment;
when the brightness mean value of the background area is increased, judging that the current scene is the large backlight environment;
when the brightness mean value of the background area shows that one side value is large and the other side value is small, judging that the current scene is the sidelight environment;
and when the value of the upper side of the brightness mean value of the background area is large, judging that the current scene is the top light environment.
7. The exposure adjustment method according to claim 2, wherein the acquiring data of the object region includes labeling a position of the object in the object region by an adjustment weight.
8. The exposure adjustment method according to claim 7, wherein the weight is adjusted by:
for the area exceeding a preset threshold value in the brightness mean value of the core area, distributing weight according to a reverse S curve;
distributing weight to the area which is lower than the preset threshold value in the brightness mean value of the core area according to an S curve;
the S curve is a direct proportion relation between the brightness mean value and the weight, and the reverse S curve is an inverse proportion relation between the brightness mean value and the weight.
9. The exposure adjustment method according to claim 8, wherein the calculating of the exposure parameter and the gain parameter of the object based on the current scene and the data of the object area includes:
calculating the exposure parameter and the gain parameter of the object to be measured according to the following formulas:
wherein NewExp is an exposure parameter to be calculated of the object to be measured; NewGain is a gain parameter to be calculated of the object to be measured; target _ Luma is a Target brightness value; the Current _ Luma is an average brightness value of the object to be measured in the Current core area after the weight weighting calculation; k is an adjustment control coefficient; the Current _ Exp is a Current exposure parameter of the object to be measured, and the Current _ Gain is a Current Gain parameter of the object to be measured.
10. An exposure adjustment system for adjusting exposure of an object, the exposure adjustment system comprising:
an image acquisition unit configured to acquire an image to be measured including the object to be measured;
a statistical unit configured to count a luminance mean value and histogram data of a region of interest in the image to be measured, the histogram data including a statistical distribution map of each pixel value in the image;
a scene determination unit configured to determine a current scene based on the luminance mean value and the histogram data;
a test object region acquisition unit configured to acquire data of a test object region based on a current scene;
a calculation unit configured to calculate an exposure parameter and a gain parameter of the object based on data of a current scene and the object region.
11. A driver monitoring system, characterized in that the exposure adjustment method according to any one of claims 1-9 or the exposure adjustment system according to claim 10 is used.
12. An advanced driving assistance system, characterized by comprising a driver monitoring system according to claim 11.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011044803.9A CN112153304B (en) | 2020-09-28 | 2020-09-28 | Exposure adjusting method and system, driver monitoring system and advanced driving assistance system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011044803.9A CN112153304B (en) | 2020-09-28 | 2020-09-28 | Exposure adjusting method and system, driver monitoring system and advanced driving assistance system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112153304A true CN112153304A (en) | 2020-12-29 |
| CN112153304B CN112153304B (en) | 2021-11-05 |
Family
ID=73894978
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011044803.9A Active CN112153304B (en) | 2020-09-28 | 2020-09-28 | Exposure adjusting method and system, driver monitoring system and advanced driving assistance system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112153304B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113612991A (en) * | 2021-09-15 | 2021-11-05 | 深圳市丘钛微电子科技有限公司 | Brightness uniformity detection method and device of light supplement type camera |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070177050A1 (en) * | 2006-01-30 | 2007-08-02 | Sony Corporation | Exposure control apparatus and image pickup apparatus |
| US20100309332A1 (en) * | 2009-06-03 | 2010-12-09 | Sony Corporation | Imaging apparatus and imaging control method |
| CN102196182A (en) * | 2010-03-09 | 2011-09-21 | 株式会社理光 | Backlight detection equipment and method |
| US20140176789A1 (en) * | 2012-12-21 | 2014-06-26 | Canon Kabushiki Kaisha | Image capturing apparatus and control method thereof |
| CN105100632A (en) * | 2014-05-13 | 2015-11-25 | 北京展讯高科通信技术有限公司 | Adjusting method and apparatus for automatic exposure of imaging device, and imaging device |
| CN105791709A (en) * | 2015-12-29 | 2016-07-20 | 福建星网锐捷通讯股份有限公司 | Automatic exposure processing method and apparatus with back-light compensation |
| CN106713778A (en) * | 2016-12-28 | 2017-05-24 | 上海兴芯微电子科技有限公司 | Exposure control method and device |
| CN108769544A (en) * | 2018-06-07 | 2018-11-06 | 深圳市亿联智能有限公司 | A kind of video monitoring equipment adpative exposure adjustment method |
| CN109918993A (en) * | 2019-01-09 | 2019-06-21 | 杭州中威电子股份有限公司 | A kind of control method based on human face region exposure |
-
2020
- 2020-09-28 CN CN202011044803.9A patent/CN112153304B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070177050A1 (en) * | 2006-01-30 | 2007-08-02 | Sony Corporation | Exposure control apparatus and image pickup apparatus |
| US20100309332A1 (en) * | 2009-06-03 | 2010-12-09 | Sony Corporation | Imaging apparatus and imaging control method |
| CN102196182A (en) * | 2010-03-09 | 2011-09-21 | 株式会社理光 | Backlight detection equipment and method |
| US20140176789A1 (en) * | 2012-12-21 | 2014-06-26 | Canon Kabushiki Kaisha | Image capturing apparatus and control method thereof |
| CN105100632A (en) * | 2014-05-13 | 2015-11-25 | 北京展讯高科通信技术有限公司 | Adjusting method and apparatus for automatic exposure of imaging device, and imaging device |
| CN105791709A (en) * | 2015-12-29 | 2016-07-20 | 福建星网锐捷通讯股份有限公司 | Automatic exposure processing method and apparatus with back-light compensation |
| CN106713778A (en) * | 2016-12-28 | 2017-05-24 | 上海兴芯微电子科技有限公司 | Exposure control method and device |
| CN108769544A (en) * | 2018-06-07 | 2018-11-06 | 深圳市亿联智能有限公司 | A kind of video monitoring equipment adpative exposure adjustment method |
| CN109918993A (en) * | 2019-01-09 | 2019-06-21 | 杭州中威电子股份有限公司 | A kind of control method based on human face region exposure |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113612991A (en) * | 2021-09-15 | 2021-11-05 | 深圳市丘钛微电子科技有限公司 | Brightness uniformity detection method and device of light supplement type camera |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112153304B (en) | 2021-11-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106791475B (en) | Exposure adjustment method and the vehicle mounted imaging apparatus being applicable in | |
| CN108206918B (en) | Optical compensation method and device | |
| EP2120183B1 (en) | Method and system for cancellation of ambient light using light frequency | |
| US20090015683A1 (en) | Image processing apparatus, method and program, and recording medium | |
| JP5759950B2 (en) | In-vehicle camera device | |
| US20110035099A1 (en) | Display control device, display control method and computer program product for the same | |
| US10129458B2 (en) | Method and system for dynamically adjusting parameters of camera settings for image enhancement | |
| CN102104724B (en) | Image processing apparatus and image processing method | |
| US20100033619A1 (en) | Exposure determining device and image processing apparatus | |
| US20180262667A1 (en) | Image capturing device and brightness adjusting method | |
| CN112153304B (en) | Exposure adjusting method and system, driver monitoring system and advanced driving assistance system | |
| EP1762975A2 (en) | Histogram equalization method for a vision-based occupant sensing system | |
| US20130242073A1 (en) | Image-capturing control apparatus and control method of capturing image | |
| US20140015962A1 (en) | Headlamp aiming using high-dynamic range camera | |
| CN113747008A (en) | Camera and light supplementing method | |
| US10057500B2 (en) | Image capturing method and monitoring apparatus with supplemental lighting modulation | |
| CN114286014A (en) | Image database acquisition method | |
| US8531589B2 (en) | Image pickup device and image pickup method | |
| CN111127537A (en) | Method and apparatus for detecting shadows in a head mounted device | |
| CN116524877A (en) | Vehicle-mounted screen brightness adjustment method and device, electronic equipment and storage medium | |
| TWI630818B (en) | Dynamic image feature enhancement method and system | |
| KR20140147211A (en) | Method for Detecting Fog for Vehicle and Apparatus therefor | |
| US7791670B2 (en) | Method of operation for a vision-based occupant sensing system | |
| WO2022224423A1 (en) | In-vehicle exposure control device and exposure control method | |
| CN109785251A (en) | A kind of picture quality method for improving, device and automated driving system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |