JP2006080752A - Exposure controller for camera and exposure control method for camera - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To accurately detect an object in the vicinity of a vehicle by performing proper exposure control regardless of a relative increase in an occupation rate of a road or the like, as regards an exposure controller for a camera and a control method for the camera which control the exposure of the camera on the basis of an exposure control value set in accordance with the brightness or the like of image data when processing image data from the camera to detect the object in the vicinity of the vehicle. <P>SOLUTION: An exposure control value calculation part 4 for calculating an exposure control value in accordance with the state of image data is provided with; an area dividing means 41 for dividing the area of image data into a plurality of areas; a road density value determination means 42 for calculating a density histogram showing appearance frequencies of density values of image data in each area to determine the road density value of each area by a calculation result; an area exposure control value calculation means 43 for calculating the exposure control value of each area from density values other than the road density value; and an entire exposure control value determination means 44 for determining the exposure control value of the entire area of image data from the exposure control value of each area. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  In the present invention, when image data obtained from a camera is processed to detect an object existing in the vicinity (surrounding) of a vehicle (moving body), it is set according to a state such as brightness of the image data. The present invention relates to a camera exposure control apparatus and control method for appropriately controlling exposure of a camera based on an exposure control value.

  In general, in a traveling vehicle, a camera such as an in-vehicle camera is installed in the vehicle in order to accurately detect an object (for example, another vehicle existing in front of the vehicle) existing in the vicinity (surrounding) of the vehicle. It is provided in front of the rearview mirror. An object existing in front of or behind the vehicle can be recognized by performing processing such as image data edge extraction on the image data captured by the camera by an image processing device or the like.

  Here, the “edge” refers to a portion with large shading in the image data, that is, contour lines of a plurality of objects in the image data. This edge has the advantage that there is little influence even if the shading changes due to shadows or dirt, and can be used when recognizing an object in the vicinity (surrounding) of the traveling vehicle.

  In order to process image data obtained from the camera and accurately recognize an object in the vicinity (surrounding) of the vehicle, the exposure condition of the camera is appropriately controlled in accordance with the state of the image data such as brightness. is important.

  In the conventional camera exposure control method, the average value of the density value (brightness) of the image data is calculated based on all the information of the image data input to the image processing apparatus or the like, and the average value of the density values is calculated. Based on this, the camera exposure control values (aperture and shutter speed (or exposure time)) were determined. Finally, the exposure control of the camera is performed according to the exposure control value determined in this way.

  However, in the camera exposure control method as described above, the environment in which the traveling vehicle is placed appears in the input image data. Therefore, information that is not a recognition target such as a road is included in the image data. Will be included. Therefore, when the proportion of roads, etc. occupies objects such as other vehicles increases, the exposure control value of the camera is determined according to the brightness of the roads, etc., which are originally unnecessary when recognizing objects such as other vehicles. May be determined. As a result, improper exposure control is performed especially when the ratio of roads and the like to objects such as other vehicles increases, and objects in the vicinity (around) of the traveling vehicle are accurately recognized. The problem has become difficult.

  Here, for reference, four prior art documents (Patent Documents 1 to 4) describing technical contents related to the conventional camera exposure control method are presented.

  In Patent Document 1 (Japanese Patent Laid-Open No. 7-81459), the iris value is obtained by averaging the image data representing the brightness of the pixels in the square area surrounding the preceding vehicle and obtaining the difference from the appropriate exposure value. A travel control device for an automobile that controls an iris based on an iris value is disclosed.

  In Patent Document 2 (Japanese Patent Laid-Open No. 9-181962), an imaging device for a moving body that adjusts the exposure amount by calculating the optimum exposure amount at a predetermined position in front of the vehicle from the brightness of the corresponding position on the image. An exposure amount control apparatus is disclosed.

  In Patent Document 3 (Japanese Patent Laid-Open No. 7-318644), an image is divided into a plurality of areas (areas) such as long distance / short distance, own lane / other lane / white line, and the luminance gradation of a specific area is determined. An obstacle detection apparatus that performs exposure control according to an average value is disclosed.

  In Patent Document 4 (Japanese Patent Laid-Open No. 2000-209488), a high-contrast imaging device that adjusts the exposure amount based on the maximum value and the minimum value of the average pixel value in a plurality of regions in two images having different exposure amounts. Is disclosed.

  However, in any of Patent Documents 1 to 4, there still remains a problem that it becomes difficult to accurately recognize an object in the vicinity (surrounding) of the vehicle by performing inappropriate exposure control.

JP 7-81459 A Japanese Patent Laid-Open No. 9-181962 Japanese Patent Laid-Open No. 7-318644 JP 2000-209488 A

  The present invention has been made in view of the above problems, and even when a ratio of roads or the like to an object around the traveling vehicle increases, the vehicle travels by appropriately controlling the exposure of the camera. It is an object of the present invention to provide a camera exposure control apparatus and control method that can accurately detect objects around the camera.

  In order to solve the above problems, the present invention is directed to an image recognition processing unit that processes image data obtained from a camera to detect an object around the vehicle, and performs exposure of the camera according to the state of the image data. An exposure control apparatus for a camera having an exposure control value calculation unit that calculates an exposure control value for controlling, wherein the exposure control value calculation unit divides the area of the image data into a plurality of areas In each area, a density histogram indicating the relationship between the density value of the image data and the frequency of appearance of the density value is calculated, and the roads included in each area are calculated based on the calculation result of the density histogram. Road density value determining means for determining a road density value; each area exposure control value calculating means for calculating an exposure control value for each of the areas based on a density value other than the road density value; To provide a camera for exposure control apparatus comprising a whole exposure control value determining means for determining an exposure control value in the entire area of the image data on the basis of the exposure control value of the area.

  Preferably, in the camera exposure control apparatus according to the present invention, each area exposure control value calculation means calculates an average value of the density values of the areas based on the density value of the image data excluding the road density value. The exposure control value for each area is calculated based on the calculated average value of the density values, and the overall exposure control value determination means is obtained by averaging the exposure control values for each area. The average value of the exposure control values is finally selected as the exposure control value for the entire area of the image data.

  On the other hand, according to the present invention, when processing image data obtained from a camera to detect an object around the vehicle, an exposure control value for controlling the exposure of the camera according to the state of the image data. A camera exposure control method for performing exposure control of the camera using the exposure control value, the step of dividing the area of the image data into a plurality of areas, and the image data in each of the areas Calculating a density histogram indicating a relationship between the density value of the image and the appearance frequency of the density value, determining a road density value of a road included in each of the areas based on the calculation result of the density histogram, and Calculating an exposure control value for each of the areas based on a density value other than a road density value; and the entire area of the image data based on the exposure control value for each of the areas To provide a camera exposure control method comprising the steps of determining the exposure control value.

  Preferably, in the camera exposure control method of the present invention, the density of the image data excluding the road density value in the step of calculating the exposure control value of each area based on a density value other than the road density value. An average value of density values of each area is calculated based on the value, an exposure control value of each area is calculated based on the calculated average value of the density values, and an exposure control value of each area is calculated. In the step of determining the exposure control value for the entire area of the image data based on the above, the average value of the exposure control values obtained by averaging the exposure control values for the respective areas is used as the exposure control for the entire area of the image data. The value is finally selected.

  Further preferably, in the camera exposure control method of the present invention, in the step of dividing the area of the image data into a plurality of areas, the own lane of the road on which the vehicle travels and the road adjacent to the own lane The adjacent lanes are divided by weighting.

  Further preferably, in the camera exposure control method of the present invention, in the step of dividing the area of the image data into a plurality of areas, the area having a short distance relationship with the vehicle and the long distance relationship with the vehicle. The area is divided by weighting.

  Further preferably, in the camera exposure control method of the present invention, in the step of dividing the area of the image data into a plurality of areas, the area ratio of each area is divided so as to be equal. Yes.

  Further preferably, in the camera exposure control method of the present invention, in the step of dividing the area of the image data into a plurality of areas, an area of the hood of the vehicle is set in advance, and the density value other than the road density value is set. Based on the step of calculating the exposure control value for each of the above areas, the hood area of the vehicle is excluded.

  Further preferably, in the exposure control method for a camera according to the present invention, in the step of dividing the area of the image data into a plurality of areas, an empty area is set in advance, and each is based on a density value other than the road density value. In the step of calculating the exposure control value of the area, the empty area is excluded.

  Further preferably, in the exposure control method for a camera according to the present invention, in the step of determining the road density value of the road included in each of the areas based on the calculation result of the density histogram, A density value existing within the density value range is selected as the road density value.

  Further preferably, in the camera exposure control method of the present invention, in the step of determining the road density value of the road included in each of the areas based on the calculation result of the density histogram, in the range of the specific density value. Among the density value distributions, a density value corresponding to a density value distribution near the peak value of the density value appearance frequency is selected as the road density value.

  Further preferably, in the camera exposure control method of the present invention, in the step of determining the road density value of the road included in each of the areas based on the calculation result of the density histogram, the vicinity of the white line detected in advance The road concentration value is determined based on the average value of the concentration values.

  Further preferably, in the camera exposure control method of the present invention, in the step of determining the road density value of the road included in each of the areas based on the calculation result of the density histogram, the back provided in the vehicle is provided. The road density value is determined based on image data obtained from the monitor camera.

  Further preferably, in the camera exposure control method of the present invention, in the step of determining the road density value of the road included in each area based on the calculation result of the density histogram, the radar provided in the vehicle is provided. To measure the distance to the object ahead of the vehicle and determine the road concentration value based on the road surface density value reflected from the vehicle hood to the object ahead of the vehicle. It has become.

  Still preferably, in a camera exposure control method according to the present invention, in the step of calculating an exposure control value for each of the areas based on a density value other than the road density value, based on the density value other than the road density value. In the step of calculating the exposure control value for each of the areas, the maximum value and the minimum value of density values other than the road density value are obtained, and the exposure control value is determined based on the maximum value and the minimum value of the density value. It is supposed to be.

  Further preferably, in the camera exposure control method of the present invention, in the step of calculating the exposure control value of each of the areas based on a density value other than the road density value, a first direction of the density value (for example, , The density projection value in the vertical direction) is calculated, and the exposure control value is determined according to the number of lines in which the density projection value within a specific range exists.

  Further preferably, in the camera exposure control method of the present invention, in the step of calculating the exposure control value of each of the areas based on a density value other than the road density value, a second direction of the density value (for example, , The density projection value in the horizontal direction) is calculated, and the exposure control value is determined according to the number of lines in which the density projection value within a specific range exists.

  Further preferably, in the exposure control method for a camera according to the present invention, in the step of calculating the exposure control value of each area based on a density value other than the road density value, the exposure control value exists within a specific range of the density value. The exposure control value is determined according to the number of density values to be performed.

  Further preferably, in the camera exposure control method of the present invention, in the step of determining the exposure control value of the entire area of the image data based on the exposure control value of each of the above areas, each of the above areas for each of the above areas. A constant weighting (fixed weighting) is performed on the exposure control value.

  Still preferably, in a camera exposure control method according to the present invention, in the step of determining an exposure control value for the entire area of the image data based on an exposure control value for each area, the road density for each area is determined. In accordance with the value, weighting (variable weighting) is performed on the exposure control value of each area for each area.

  Further preferably, in the camera exposure control method of the present invention, in the step of determining the exposure control value of the entire area of the image data based on the exposure control value of each of the areas, the central portion of the area of the image data is determined. Only the exposure control values in the nearby area are selected.

  Further preferably, in the camera exposure control method of the present invention, in the step of determining the exposure control value of the entire area of the image data based on the exposure control value of each of the areas, a radar provided in the vehicle is used. Based on the result of detecting the object in front of the vehicle, the weight of the exposure control value in the area where the object in front of the vehicle is present is increased.

  In summary, in the present invention, first, an area of image data obtained from a camera is divided into a plurality of areas, and exposure control of each area is performed based on density values other than road density values of roads included in each area. A value is calculated, and an exposure control value for the entire area of the image data is finally determined based on the exposure control value of each area. Therefore, even when the ratio of roads to the surrounding objects of the traveling vehicle increases, it is possible to appropriately control the exposure of the vehicle using image data excluding the road density value of the road. It becomes possible to accurately detect surrounding objects.

  Further, according to the present invention, secondly, by dividing the image data area into a plurality of areas according to appropriate division conditions, the exposure control value of each area can be calculated once, and various environmental conditions can be obtained. It is possible to appropriately set an exposure control value considering the above.

  Furthermore, in the present invention, thirdly, the road density value of the road included in each area of the image data is excluded, and the bonnet portion and the empty portion of the relatively bright vehicle are excluded, thereby more appropriately. It becomes possible to set an exposure control value for each area.

  Furthermore, in the present invention, fourthly, exposure control of each area is more appropriately performed in consideration of road information regarding a road in the vicinity of the vehicle detected using a back monitor camera or a radar provided separately in the vehicle. The value can be set.

  Furthermore, in the present invention, fifth, the exposure control value for the entire area of the image data is finally determined more appropriately by analyzing the density value of the image data excluding the road density value of the road in more detail. It becomes possible.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings (FIGS. 1 to 8).

  FIG. 1 is a block diagram showing the configuration of an embodiment according to the camera exposure control apparatus of the present invention. However, here, the configuration of the embodiment in the case where the camera exposure control apparatus 10 of the present invention is applied to the camera 1 such as an in-vehicle camera is shown in a simplified manner.

  As shown in FIG. 1, a camera 1 such as an in-vehicle camera is mounted on the front part of a traveling vehicle (for example, in front of a room mirror in the vehicle). In the camera exposure control apparatus 10 of the present invention, an image processing control unit 2 that performs analog / digital conversion processing and sampling processing of the image data Sia captured by the camera 1 and outputs the image data Sid at a predetermined timing; The image data Sid supplied from the image processing control unit 2 is processed by edge extraction or the like to recognize or detect an object in the vicinity (surrounding) of the vehicle (for example, another vehicle existing in the vicinity of the traveling vehicle). The image recognition processing unit 3 and the exposure for calculating the exposure control values (aperture and shutter speed) for adjusting the exposure of the camera 1 in accordance with the brightness of the image data Sid processed by the image recognition processing unit 3 A control value calculation unit 4 is provided. Preferably, the image recognition processing unit 3 is configured by a CPU (Central Processing Unit) of a microcomputer.

  Further, in the camera exposure control apparatus 10 of FIG. 1, the camera 1 captures a memory (memory) 5 for storing various data including a program (software) for image data processing and a program for camera exposure control. A display unit 30 such as a display for displaying the received image data and an object in the vicinity of the vehicle detected by the image recognition processing unit 3 is provided.

  The image processing control unit 2, the image recognition processing unit 3, the exposure control value calculation unit 4, the storage unit 5, and the display unit 30 are connected by a bus 6 so that data can be exchanged with each other. Yes.

  More specifically, the exposure control value calculation unit 4 in FIG. 1 includes an area dividing unit 41 that divides the area of the image data Sid supplied from the image processing control unit 2 into a plurality of areas, and roads included in each area. Road density value determining means 42 for determining the road density value. The road density value determining unit 42 generates a density histogram indicating the relationship between the density value (brightness) of the image data and the appearance frequency of the density value for each of the plurality of areas divided by the area dividing unit 41. It has a function of calculating and determining the road density value Rd of the road included in each area based on the calculation result of the density histogram.

  Preferably, the area dividing means 41 divides the area of the image data Sid into a plurality of areas by weighting the own lane of the road on which the vehicle is traveling and the adjacent lane of the road adjacent to the own lane. I am doing so. Alternatively, the area of the image data is divided into a plurality of areas by weighting an area in a short distance relationship with the traveling vehicle and an area in a long distance relationship with the vehicle. Alternatively, the area is divided so that the area ratio of each area becomes equal.

  1 calculates the exposure control value Sca of each area based on the density value of the image data excluding the road density value Rd determined by the road density value determining means 42. An area exposure control value calculating means 43 and an overall exposure control value determining means 44 for determining an exposure control value Sce for the entire area of the image data based on the exposure control value for each area are provided.

  Preferably, each area exposure control value calculation means 43 calculates the average value of the density values of each area based on the density value of the image data excluding the road density value, and sets the average value of the calculated density values. Based on this, the exposure control value Sca for each area is calculated.

  On the other hand, the overall exposure control value determining means 44 finally selects the average value of the exposure control values obtained by averaging the exposure control values of the respective areas as the exposure control value Sce of the entire area of the image data. , Feedback to the camera 1.

  Preferably, a program for executing the functions of the area dividing means 41, the road density value determining means 42, each area exposure control value calculating means 43, and the overall exposure control value determining means 44 as described above is stored in the storage unit 5. Stored in advance. The area dividing unit 41, the road density value determining unit 42, each area exposure control value calculating unit 43, and the overall exposure control value determining unit 44 are stored in a program (software) in the storage unit 5 by the CPU (image recognition processing unit 3). ) Is read and operated.

  According to the embodiment of FIG. 1, the area of the image data obtained from the camera is divided into a plurality of areas, and exposure of each area is performed based on the density value of the image data excluding the road density value of the road included in each area. A control value is calculated, and finally an exposure control value (aperture and shutter speed) for the entire area of the image data is determined based on the exposure control value of each area. Therefore, even if the image data contains a lot of information that is not subject to recognition, such as roads, it is possible to run by appropriately controlling camera exposure using image data excluding road density values of roads. It becomes possible to accurately detect an object in the vicinity of the vehicle to be operated.

  Furthermore, according to the embodiment of FIG. 1, by dividing the image data area into a plurality of areas according to an appropriate division condition considering the own lane, adjacent lane, etc., the exposure control value for each area is calculated once. Thus, it becomes possible to appropriately set the exposure control value in consideration of various environmental conditions.

  FIG. 2 is a block diagram showing a specific configuration of the image processing control unit in FIG. Hereinafter, the same components as those described above are denoted by the same reference numerals.

  As shown in FIG. 2, the image processing control unit 2 is connected to an analog / digital converter 21 (abbreviated as A / D in FIG. 2) connected to the camera 1 and the analog / digital converter 21. A first-in first-out (FIFO) field memory 22.

  Here, the analog / digital converter 21 converts the image signal for each pixel of the pixel data Sia captured by the camera 1 from the analog format to the digital format at the timing of the sampling signal Ss, and has an 8-bit density value, for example. Pixel data Sid is output.

  On the other hand, in the FIFO field memory 22, the image data Sid is written at the timing output from the analog / digital converter 21, and is written first at the timing of the capture signal Sr from the CPU (image recognition processing unit 3). Image data Sid is read first. The output of the capture signal Sr starts at the timing of the vertical synchronization signal Svs of the camera 1 and ends with the end of reading.

  Further, the image processing control unit 2 in FIG. 2 generates a sampling signal Ss and supplies it to the analog / digital converter 21, a synchronization signal detection unit 24 for detecting the vertical synchronization signal Svs, and write control. And a write control unit 25 for supplying the signal Sw to the FIFO field memory 22.

  Here, the synchronization signal detection unit 24 detects the vertical synchronization signal Svs by branching and capturing the image signal output from the camera 1 and extracting the vertical synchronization signal Svs from the image signal. The vertical synchronization signal Svs from the synchronization signal detector 24 is sent to the CPU and used for the output timing of the capture signal Sr.

  FIG. 3 is a flowchart for explaining the camera exposure control method of the present invention, FIG. 4 is an image data diagram showing an example of area division of FIG. 3, and FIG. 5 is a density in one area of FIG. It is a histogram which shows an example of value distribution. Here, the procedure for operating the CPU to execute the camera exposure control method of the present invention will be described with reference to the flowchart of FIG.

  When the camera exposure control value is calculated according to the brightness (density value) of the image data captured by the camera and the camera exposure control is performed using the exposure control value, the road density value of the road on which the vehicle travels Therefore, first, (start), as shown in step 100 of FIG. 3, the area of the image data is divided into a plurality of areas.

  When the image data area is divided into a plurality of areas, for example, as shown in FIG. 4, the area is divided into three areas A, B, and C with the white line W of the road R as a boundary. Here, three areas are set by weighting the area B including the own lane of the road on which the vehicle 7 travels and the areas A and C including the adjacent lane of the road adjacent to the own lane. In this case, the input image data (Sid) is composed of 640 pixels × 480 pixels, and each pixel has, for example, an 8-bit density value (brightness). Since there is a difference in density value between an object such as a vehicle and an object that is not a recognition object such as a road, the object such as a vehicle in each area is extracted as an edge image by edge extraction or the like. Based on the edge distribution of the edge image, an object in the vicinity of the traveling vehicle (for example, another vehicle existing ahead of the traveling vehicle) can be recognized.

  As an alternative method for dividing the area of the image data into a plurality of areas, by weighting the area in a short distance relationship with the traveling vehicle and the area in a long distance relationship with the vehicle, A method of dividing into two areas is conceivable. Furthermore, the area including the own lane of the road on which the vehicle travels, the area including the adjacent lane of the road adjacent to the own lane, the area in a short distance relationship with the traveling vehicle, and the relationship between the vehicle and the long distance A method of dividing into more areas by weighting a certain area is also conceivable. On the other hand, a method of dividing the area so that the area ratio of each area is equal is also conceivable.

  Next, as shown in step 115 of FIG. 3, in each area (A, B, and C), a density histogram indicating the relationship between the density value of the image data and the appearance frequency of the density value is calculated. Based on the calculation result of the histogram, the road density value of the road included in each area is determined. The procedure for determining the road density value of such a road is sequentially executed for each area as shown in steps 105 and 110, and is continued until the exposure control values for all the areas are determined.

  FIG. 5 shows an example of density distribution of image data when a density histogram of image data is created in a certain area. However, in the example of FIG. 5, it is assumed that each pixel constituting the image data has a density value of 8 bits (that is, 256 gradation levels from 0 bit to 255 bits).

  In FIG. 5, in one area, the density histogram is calculated with the density value (value expressed in bits) of each pixel as the horizontal axis and the number of the density values (appearance frequency) as the vertical axis. In this density histogram, the presence of a black object is detected by the peak of the frequency of appearance of density values in the range # 1 where density values are very low, and the appearance of density values in the range # 3 where density values are very high. The presence of a white object is detected by the frequency peak. By the way, it is known that the density value indicating the brightness of the road falls within a certain density value range (that is, a specific intermediate gradation level (gray level range). The peak of the frequency of appearance of the density value in the value range # 2 is considered to be caused by the road density value of the road, and image data having the density value in the range # 2 is excluded.

  Here, the road density value of the road is made variable, and among the density value distribution in a specific density value range (for example, range # 2), the density value distribution near the peak value of the frequency of appearance of the density value is changed. It is also possible to select the corresponding concentration value as the road concentration value. In this case, the range to be excluded as the road density value does not have to be unnecessarily large, so that the exposure control value of an object such as a vehicle can be determined with higher accuracy.

  Preferably, the road density value of the road is calculated based on the average value of the density values in the vicinity of the white line W by regarding the vicinity of the white line W (see FIG. 4) detected in advance in the image data as a road. It is also possible to decide.

  Further, preferably, an area of a vehicle hood having a relatively high density value in the image data area (for example, see FIG. 7 described later) is set in advance, and the area of the vehicle hood is set together with the road density value of the road. It is also possible to exclude.

  Further, preferably, an empty area having a relatively high density value in the image data area (for example, see FIG. 8 described later) is set in advance, and the empty area is excluded together with the road density value of the road. Is possible.

  Further, as shown in step 120 of FIG. 3, the exposure control value of each area is determined based on the density value of the image data excluding the road density value of the road (for example, the density value of range # 2 of FIG. 5). calculate.

  More specifically, the density value of each area is calculated using the density value of the image data excluding the road density value of the road. Furthermore, an exposure control value for each area is calculated based on the average value of the density values calculated for each area.

Further, as shown in step 125 of FIG. 3, based on the exposure control value for each area calculated in step 120 described above, the exposure control value for the entire area of the image data is determined (END).
More specifically, the average value of the exposure control values obtained by averaging the exposure control values in each area is finally selected as the exposure control value for the entire area of the image data, and fed back to the camera 1.

  According to the exposure control method for a camera of the present invention shown in FIG. 3, the area of the image data obtained from the camera is divided into a plurality of areas, and the density value of the image data is obtained by excluding the road density value of the road included in each area. Based on the exposure control value for each area, the exposure control value for the entire area of the image data is finally determined based on the exposure control value for each area. Therefore, even if the image data contains a lot of information that is not subject to recognition, such as roads, it is possible to run by appropriately controlling camera exposure using image data excluding road density values of roads. It becomes possible to accurately detect an object in the vicinity of the vehicle to be operated.

  Further, according to the exposure control method for a camera of the present invention shown in FIG. 3, the image data area is divided into a plurality of areas according to an appropriate division condition in consideration of the own lane, the adjacent lane, etc. The value can be once calculated, and the exposure control value can be set appropriately in consideration of various environmental conditions.

  Furthermore, according to the exposure control method for a camera of the present invention shown in FIG. 3, the road density value of the road included in each area of the image data is excluded, and the bonnet portion of the vehicle having a relatively high density value or the sky By excluding portions and the like, exposure control values for each area can be set more appropriately.

  FIG. 6 is a top view of a vehicle showing the arrangement relationship of components used in the camera exposure control apparatus of the present invention, and FIG. 7 is a front view thereof. Here, referring to FIG. 6 and FIG. 7, based on the information about the road in the vicinity of the vehicle detected by using the back monitor camera 8 and the radar 9 separately provided in the vehicle 7, the steps of FIG. A case where 115 (that is, a step of determining the road density value of the road in each area based on the calculation result of the density histogram) is executed will be described.

  As shown in FIGS. 6 and 7, in a traveling vehicle 7, a camera 1 such as an in-vehicle camera is usually provided in front of a room mirror 71 in the vehicle. Further, in order to recognize an object behind the vehicle 7, a back monitor camera 8 is installed in the vicinity of the tail lamp 73 (or in the vicinity of the ceiling 72 at the rear of the vehicle 7). Based on the image data obtained from the back monitor camera 8, the road density value of a specific area can be determined. Determining the road density value using the back monitor camera 8 is particularly effective when it is desired to accurately grasp the situation of the parking lot when the vehicle is backed.

  On the other hand, in the traveling vehicle 7, a radar 9 is provided in the vicinity of the hood 75 and the headlamp 74 of the vehicle in order to measure the distance from the object in front of the vehicle using electromagnetic waves. The radar 8 is used to measure the distance to the object ahead of the vehicle 7 and determine the road concentration value of a specific area based on the road surface density value reflected from the bonnet 75 to the object ahead of the vehicle. can do. The determination of the road density value using the radar 8 is particularly effective when it is desired to accurately grasp how many meters ahead the object is present.

  Furthermore, a modified example in the case of executing the above-described step 120 of FIG. 3 (that is, a step of calculating the exposure control value of each area based on the density value of the image data excluding the road density value) will be described. Here, in each area, the maximum value and the minimum value of the density value excluding the road density value are obtained from the density histogram, and the exposure control value of each area is determined based on the maximum value and the minimum value of the density value. It is like that. For example, in the density histogram of FIG. 5, in the density value range # 1, the black object is accurately grasped by obtaining the maximum value and the minimum value of the density value, and in the density value range # 3, the maximum density value is obtained. The white object can be accurately grasped by obtaining the value and the minimum value.

  FIG. 8 is a histogram showing an example of the distribution of density projection values used in calculating the exposure control value for each area. Here, another modified example in the case of executing the above-described step 120 of FIG. 3 (that is, the step of calculating the exposure control value of each area based on the density value of the image data excluding the road density value) will be described. To do.

  By merely referring to the density histogram as shown in FIG. 5, it may not be possible to accurately grasp how the density value changes depending on the location in the image data area. For this reason, as shown in FIG. 8, the density projection value in the horizontal direction of the density value of the image data excluding the road density value is calculated, and within a specific range (for example, the threshold value Vth1 of the density projection value in the horizontal direction). The exposure control value of each area is determined according to the number of lines in which density projection values within the following range exist. On the other hand, the density projection value in the vertical direction of the density value of the image data excluding the road density value is calculated, and within a specific range (for example, within the range of the threshold value Vth2 of the density projection value in the vertical direction). The exposure control value for each area is determined in accordance with the number of lines where the density projection value exists. In this way, it becomes possible to accurately grasp the portion of the sky S having a relatively high density value based on the density projection value in the vertical direction.

  Furthermore, another modified example in the case of executing the above-described step 120 of FIG. 3 (that is, the step of calculating the exposure control value of each area based on the density value of the image data excluding the road density value) will be described. . Here, among the density values of the image data excluding the road density value, each exposure control value is determined in accordance with the number of density values existing within a specific density value range. In this way, the exposure control value for the entire area of the image data can be finally determined more appropriately by analyzing the density value of the image data excluding the road density value of the road in more detail.

  Furthermore, a modified example in the case where the above-described step 125 of FIG. 3 (that is, a step of determining the exposure control value of the entire area of the image data based on the exposure control value of each area) will be described. Here, constant weighting is performed on the exposure control value of each area for each area. For example, a weighting factor of 3: 1: 1 is set for the density value of one area including the own lane and the density value of two areas including the adjacent lane, and the weight is assigned to the exposure control value of each area. An exposure control value obtained by averaging the values multiplied by the coefficients is selected as the exposure control value for the entire area of the final image data.

  Furthermore, another modified example in the case where step 125 of FIG. 3 described above is executed will be described. Here, variable weighting is performed on the exposure control value of each area for each area in accordance with the road density value of each area. In this case, taking into account the road density value of each area (for example, depending on how many vehicles are on the road in each area), the density value of one area including the own lane and the adjacent lane are included. The weighting coefficient set for the density values of the two areas can be changed.

  Furthermore, another modified example in the case where step 125 of FIG. 3 described above is executed will be described. Here, only the exposure control value for the area near the center of the image data area is selected. For example, if traffic congestion occurs in the area of the own lane, it is not necessary to substantially consider the exposure control value of the adjacent lane area, so only the exposure control value of the area near the center of the image data area is selected. To be effective.

  Furthermore, another modified example in the case where step 125 of FIG. 3 described above is executed will be described. Here, based on the result of detecting an object in front of the vehicle by a radar (for example, see FIG. 7) provided in the vehicle, the exposure control value in the area where the object in front of the vehicle exists is increased in weight. Like to do. In this way, the exposure control value for the entire area of the image data can be finally determined more appropriately by positively using the detection result by the radar.

  The camera exposure control apparatus and control method of the present invention can be applied not only to an in-vehicle camera provided in a vehicle but also to a general camera for recognizing and detecting an object.

It is a block diagram which shows the structure of one Example which concerns on the exposure control apparatus for cameras of this invention. It is a block diagram which shows the specific structure of the image processing control part of FIG. It is a flowchart for demonstrating the exposure control method for cameras of this invention. FIG. 4 is an image data diagram illustrating an example of area division in FIG. 3. 5 is a histogram showing an example of density value distribution in one area of FIG. 4. It is a top view of the vehicle which shows the arrangement | positioning relationship of the component used for the exposure control apparatus for cameras of this invention. It is a front view of the vehicle which shows the arrangement | positioning relationship of the component used for the exposure control apparatus for cameras of this invention. It is a histogram which shows an example of distribution of the density projection value used when calculating the exposure control value of each area.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Camera 2 Image processing control part 3 Image recognition processing part 4 Exposure control value calculating part 5 Memory | storage part 6 Bus 7 Vehicle 8 Back monitor camera 9 Radar 10 Camera exposure control apparatus 21 Analog / digital (A / D) converter 22 FIFO Field memory 23 Timing generation unit 24 Sync signal detection unit 25 Write control unit 30 Display unit 41 Area division unit 42 Road density value determination unit 43 Each area exposure control value calculation unit 44 Overall exposure control value determination unit 71 Room mirror 72 Ceiling 75 Bonnet

Claims (22)

An image recognition processing unit that processes image data obtained from a camera to detect an object around the vehicle, and an exposure control value that calculates an exposure control value for controlling the exposure of the camera according to the state of the image data An exposure control apparatus for a camera having a calculation unit, wherein the exposure control value calculation unit is
Area dividing means for dividing the area of the image data into a plurality of areas;
A density histogram indicating the relationship between the density value of the image data and the appearance frequency of the density value in each area is calculated, and the road density of the road included in each area is calculated based on the calculation result of the density histogram. Road concentration value determining means for determining the value;
Each area exposure control value calculation means for calculating an exposure control value of each of the areas based on a density value other than the road density value;
An exposure control apparatus for a camera, comprising: an overall exposure control value determining unit that determines an exposure control value for the entire area of the image data based on an exposure control value for each of the areas. Each area exposure control value calculation means calculates an average value of the density values of each area based on the density value of the image data excluding the road density value, and calculates the average value of the calculated density values. Based on the exposure control value of each said area based on,
The overall exposure control value determining means finally selects an average value of exposure control values obtained by averaging the exposure control values of the areas as the exposure control value of the entire area of the image data. The camera exposure control apparatus according to claim 1. When processing image data obtained from the camera to detect an object around the vehicle, an exposure control value for controlling the exposure of the camera is calculated according to the state of the image data, and the exposure control value A camera exposure control method for performing exposure control of the camera using
Dividing the area of the image data into a plurality of areas;
A density histogram indicating the relationship between the density value of the image data and the appearance frequency of the density value in each area is calculated, and the road density of the road included in each area is calculated based on the calculation result of the density histogram. Determining a value;
Calculating an exposure control value for each of the areas based on a density value other than the road density value;
And a step of determining an exposure control value for the entire area of the image data based on an exposure control value for each of the areas. In the step of calculating an exposure control value for each area based on a density value other than the road density value, an average of the density values of each area based on the density value of the image data excluding the road density value A value is calculated, an exposure control value for each of the areas is calculated based on the calculated average value of the density values,
In the step of determining the exposure control value of the entire area of the image data based on the exposure control value of each of the areas, the average value of the exposure control values obtained by averaging the exposure control values of the respective areas, 4. The camera exposure control method according to claim 3, wherein the exposure control value for the entire area of the image data is finally selected.   The step of dividing the region of the image data into a plurality of areas is performed by weighting the own lane of the road on which the vehicle travels and the adjacent lane of the road adjacent to the own lane. Item 4. A camera exposure control method according to Item 3.   In the step of dividing the area of the image data into a plurality of areas, weighting is performed on an area in a short distance relationship with the vehicle and an area in a long distance relationship with the vehicle. The exposure control method for a camera according to claim 3.   4. The exposure control method for a camera according to claim 3, wherein in the step of dividing the area of the image data into a plurality of areas, the area of each of the areas is divided so as to be equal. In the step of dividing the area of the image data into a plurality of areas, a bonnet area of the vehicle is set in advance,
4. The exposure control method for a camera according to claim 3, wherein in the step of calculating the exposure control value for each of the areas based on a density value other than the road density value, the bonnet area of the vehicle is excluded. In the step of dividing the area of the image data into a plurality of areas, an empty area is set in advance,
4. The exposure control method for a camera according to claim 3, wherein in the step of calculating an exposure control value for each of the areas based on a density value other than the road density value, the empty area is excluded.   In the step of determining the road density value of the road included in each of the areas based on the calculation result of the density histogram, the density value existing within a certain density value range in the density histogram is determined as the road density value. 4. The exposure control method for a camera according to claim 3, wherein:   In the step of determining the road density value of the road included in each area based on the calculation result of the density histogram, the frequency of occurrence of the density value in the density value distribution in the specific density value range is determined. 4. The exposure control method for a camera according to claim 3, wherein a density value corresponding to a density value distribution near a peak value is selected as the road density value.   In the step of determining the road density value of the road included in each area based on the calculation result of the density histogram, the road density value is calculated based on an average value of density values in the vicinity of the white line detected in advance. 4. The exposure control method for a camera according to claim 3, wherein the exposure control method is determined.   In the step of determining the road density value of the road included in each area based on the calculation result of the density histogram, the road density value is calculated based on image data obtained from a back monitor camera provided in the vehicle. 4. The exposure control method for a camera according to claim 3, wherein the exposure control method is determined.   In the step of determining a road density value of a road included in each area based on the calculation result of the density histogram, a distance from an object ahead of the vehicle is measured using a radar provided in the vehicle. 4. The camera exposure control method according to claim 3, wherein the road density value is determined based on a density value of a road surface of a road reflected from a hood of the vehicle to an object ahead of the vehicle.   In the step of calculating an exposure control value for each of the areas based on a density value other than the road density value, a maximum value and a minimum value of density values other than the road density value are obtained, and the maximum value and minimum value of the density value are determined. 4. The camera exposure control method according to claim 3, wherein the exposure control value is determined based on the value.   In the step of calculating an exposure control value for each of the areas based on a density value other than the road density value, a density projection value in the first direction of the density value is calculated, and the density projection within a specific range is calculated. 4. The exposure control method for a camera according to claim 3, wherein the exposure control value is determined in accordance with the number of lines having a value.   In the step of calculating an exposure control value for each of the areas based on a density value other than the road density value, a density projection value in the second direction of the density value is calculated, and the density projection within a specific range is calculated. 4. The exposure control method for a camera according to claim 3, wherein the exposure control value is determined in accordance with the number of lines having a value.   In the step of calculating an exposure control value for each of the areas based on a density value other than the road density value, the exposure control value is determined according to the number of density values existing within a certain density value range. 4. The exposure control method for a camera according to claim 3, wherein:   In the step of determining the exposure control value of the entire area of the image data based on the exposure control value of each of the areas, a constant weighting is performed on the exposure control value of each of the areas for each area. 4. The exposure control method for a camera according to claim 3, wherein:   In the step of determining an exposure control value of the entire area of the image data based on an exposure control value of each of the areas, the area of each of the areas according to the road density value of each of the areas 4. The camera exposure control method according to claim 3, wherein the exposure control value is weighted.   In the step of determining the exposure control value of the entire area of the image data based on the exposure control value of each area, only the exposure control value of the area near the center of the area of the image data is selected. The camera exposure control method according to claim 3.   In the step of determining the exposure control value of the entire area of the image data based on the exposure control value of each of the areas, based on the result of detecting an object ahead of the vehicle by a radar provided in the vehicle 4. The exposure control method for a camera according to claim 3, wherein a weighting of an exposure control value in an area where an object in front of the vehicle is present is increased.

Images