JP2009071706A5 - - Google Patents

Description

Imaging device

  The present invention relates to an imaging apparatus, and more particularly to an imaging apparatus capable of imaging a subject at two different shutter speeds.

Conventionally, when an image pickup device using a CCD or CMOS sensor is used to pick up an image from a bright subject to a dark subject, the dynamic range of the image pickup device is narrow, so that a bright part is saturated or an S / N of a dark part. However, in order to eliminate this defect, two images of a high-speed shutter image and a low-speed shutter image are taken out from the same imaging device and synthesized. Thus, a method has been proposed and put into practical use that can capture a clear image from a bright part to a dark part by expanding the dynamic range (for example, Patent Document 1 and Patent Document 2).

There is also a technology for detecting the occurrence of various image states (phenomena) such as the occurrence of backlight in an imaging device such as a camera mounted on a vehicle such as an automobile, and performing a process such as notification according to the detection result. Proposed (for example, Patent Document 3)
Japanese Unexamined Patent Publication No. 2007-129631 (FIG. 1, paragraphs [0042] to [0043]) Japanese Unexamined Patent Publication No. 2003-250094 (FIG. 1, paragraphs [0069] to [0071]) Japanese Unexamined Patent Publication No. 2000-207563

  When the above-described imaging device according to the prior art is mounted on a moving body such as an automobile, the environment in which imaging is dynamically performed changes every moment, such as in a tunnel, backlight, and between dark forests. In these environments, the dynamic range suitable for imaging is different, but in conventional imaging devices, the imaging conditions of the camera were not variable, so images were captured with an appropriate dynamic range according to changes in the environment, and good output was achieved. It was difficult to provide images.

More specifically, with a conventional in-vehicle WDR camera, a high-intensity image and a low-intensity image, which could not be captured simultaneously with a general in-vehicle camera, were made possible by using a technique called 2-screen addition. However, since the condition setting for adding two screens was not variable, the following problems occurred.
(1) A subject moving in the daytime becomes a multiple image and is unnatural.
(2) The subject in the daytime cannot be seen well (low contrast).
(3) Immediately before the tunnel population, the inside of the tunnel cannot be imaged brightly.
(4) Near the tunnel exit, the images inside and outside the tunnel are unnatural (low contrast, poor hue).
(5) The subject cannot be seen well in low light (low contrast).
(6) The night subject cannot be seen well (low contrast, poor hue).

An object of the present invention is to solve the above-described problems. More specifically, the environment is divided into a plurality of environments, and the plurality of environments are detected using luminance information obtained from the image sensor, and the detection result is used. An object of the present invention is to change the camera setting parameters of the imaging apparatus and perform imaging using camera setting parameters suitable for each environment.

As means for solving the above problems, the present invention has the following features.
A first aspect of the present invention is proposed as an imaging apparatus.
This imaging apparatus is an imaging apparatus having a function of capturing a subject at two different shutter speeds and expanding a dynamic range. The imaging apparatus alternately repeats imaging of a subject at two different shutter speeds and corresponds to each shutter speed. Imaging means for outputting a first analog image signal and a second analog image signal, gains of the first analog image signal and the second analog image signal are varied, and the first analog image signal and the second analog image signal are changed. Analog signal processing means for converting the image signal into a first digital image signal and a second digital image signal ;
The first digital image signal and the second digital image signal are converted into a first continuous image signal and a second continuous image signal, which are continuous image signals, respectively, and the first continuous image signal and the second continuous image are converted. Digital signal processing means for performing characteristic conversion using a specified characteristic conversion parameter value and outputting the first continuous image signal and the second continuous image signal subjected to the characteristic conversion;
The first continuous digital image signal and the second digital continuous image signal, to generate a third digital image signals are added by the given addition rate, the third digital image signals analog image analog image signal a digital signal processing means for converting the signal, for at least one of the first digital image signal and the second digital image signal, by dividing the digital image signal of one screen into a plurality of unit areas, the unit areas The average brightness is calculated, the environmental condition is determined based on the calculated average brightness, the camera setting data corresponding to the determined environmental condition is selected, and the electronic shutter signal is sent to the analog signal processing means based on the selected camera setting data. outputs, and characterized by a control means for outputting a value that specifies the characteristic conversion parameter value and the addition ratio to the digital signal processing means It is.

According to the present invention, dividing the environment into a plurality of detecting a plurality of environments with a luminance information obtained from the imaging device, and change the camera setting parameters of the imaging device by the detected result, camera suitable for each environment Imaging can be performed by the setting parameter.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
The embodiment of the present invention is a function (scene change function) that enables each setting to be dynamically changed in response to various conditions in a wide dynamic range camera (hereinafter referred to as a WDR camera) using a two-screen addition method. ) Is proposed as an imaging device (for example, a vehicle camera for front monitoring).

[1. Configuration of imaging device]
FIG. 1 is a functional block diagram of an imaging apparatus according to an embodiment of the present invention. Hereinafter, a configuration example of the imaging apparatus according to the present embodiment will be described with reference to FIG.
As shown in FIG. 1, the imaging apparatus 100 mainly includes an imaging unit 110 corresponding to the imaging unit of the present invention and an analog signal corresponding to the analog signal processing unit of the present invention that performs processing from analog signal to digital conversion. The processing unit 120, the digital signal obtained from the analog signal processing unit 120, the digital signal processing unit 130 corresponding to the digital signal processing means of the present invention, and the digital signal processed by the digital signal processing unit 130 The control unit 140 is configured to execute control of the entire image pickup apparatus 100 based on the control unit 140 according to the present invention.

  The output from the digital signal processing unit 130 is sent to a monitor device 200 such as an in-vehicle liquid crystal display device. The monitor device 200 displays an image according to the image signal output from the imaging device 100.

  The imaging apparatus 100 captures a subject image with the imaging unit 110 and generates an image signal (analog signal). The generated analog image signal is converted from an analog signal to a digital signal (hereinafter referred to as A / D conversion) in the analog signal processing unit 120 and then output as a digital image signal. This digital image signal is branched into two, one of which is input to the digital signal processing unit 130 for signal processing. The digital image signal after the signal processing by the digital signal processing unit 130 is converted from a digital signal to an analog signal (hereinafter referred to as D / A conversion) and output from an image signal output terminal (not shown).

The other of the image signals output from the analog signal processing unit 120 is input to the control unit 140. The control unit 140 generates a control signal and controls the imaging unit 110, the analog signal processing unit 120, and the digital signal processing unit 130. With these controls of the control unit 140, the imaging apparatus 100 can perform imaging capable of expressing a wide luminance region (wide dynamic range) even for a subject having a wide luminance difference.
Hereinafter, specific configuration examples of the above-described units will be described.

[1.1. Imaging unit]
The imaging unit 110 includes an imaging lens 111 that collects light from a subject, and an imaging device 112 (eg, an imaging device camera, a CMOS sensor, etc.). In the imaging unit 110, light from the subject is collected by the imaging lens 111 and an image of the subject is formed on the light receiving surface of the imaging element 112. The exposure condition of the image sensor 112 is determined by adjusting an electronic shutter speed controlled by an electronic shutter control unit 143 described later. The image sensor 112 performs photoelectric conversion and accumulates an amount of charge corresponding to the amount of image light. An image signal corresponding to the accumulated charge amount is output from the image sensor 112.

  When imaging a subject, the image sensor 112 of the imaging unit 110 alternately repeats imaging at two different shutter speeds. That is, imaging is performed at a low shutter speed and a high shutter speed. Image signals that are alternately output from the imaging element 112 are output from the imaging unit 110 and input to the analog signal processing unit 120.

[1.2. Analog signal processor]
The analog signal processing unit 120 sets an AGC circuit 121 that sets the gain of an analog image signal picked up by alternately inputted low-speed shutter speed and high-speed shutter speed, and A performs conversion from an analog signal to a digital signal. / D conversion circuit 122. Analog image signals input from the image sensor 112 to the analog signal processing unit 120 are sequentially transmitted to the AGC circuit 121 and the A / D conversion circuit 122 for signal processing. With the control signal from the control unit 140, the AGC circuit 121 can change the gain for each analog image signal.

  The A / D conversion circuit 122 converts the analog image signal into a digital image signal. The converted image signal, that is, the digital image signal imaged at the low shutter speed and the high shutter speed is alternately output from the analog signal processing unit 120 and is input to the digital signal processing unit 130 alternately.

[1.3. Digital signal processor]
The digital signal processing unit 130 includes a WDR signal processing circuit 131 and a camera signal processing circuit 132.

[1.3.1. WDR signal processing circuit]
FIG. 2 is a block diagram showing a specific configuration example of a WDR (Wide Dynamic Range) signal processing circuit 130.

  The image signal processing execution unit 131 is a low-speed shutter signal processing unit 131a that processes a digital image signal captured by a low-speed shutter, and a high-speed shutter signal processing unit 131b that processes a digital image signal captured at a high-speed shutter speed. With. Each of the low-speed shutter signal processing unit 131a and the high-speed shutter signal processing unit 131b includes a memory circuit, a switching circuit, and a characteristic conversion circuit. That is, the low-speed shutter signal processing unit 131a includes a low-speed shutter memory circuit 131a1, a low-speed shutter switching circuit 131a2, and a low-speed shutter characteristic conversion circuit 131a3, while the high-speed shutter signal processing unit 131b includes A high-speed shutter memory circuit 131b1, a high-speed shutter switching circuit 131b2, and a high-speed shutter characteristic conversion circuit 131b3 are provided.

Since the electronic shutter control unit 143 alternately supplies the low-speed electronic shutter signal and the high-speed electronic shutter electronic shutter signal having different shutter speeds for each field to the image sensor 112, the low-speed shutter image signal and the high-speed shutter image for each field. Signals are obtained alternately. These image signals are respectively input to a low-speed shutter memory circuit 131a1 and a high-speed shutter memory circuit 131b1 which are memories of one vertical period, and are separated into a low-speed shutter image signal and a high-speed shutter image signal.

  Signals on the input side and output side of the low-speed shutter memory circuit 131a1 are input to the low-speed shutter switching circuit 131a2. Similarly, signals on the input side and output side of the high-speed shutter memory circuit 131b1 are input to the high-speed shutter switching circuit 131b2. The low-speed shutter switching circuit 131a2 and the high-speed shutter switching circuit 131b2 each output the low-speed shutter image signal and the high-speed shutter image signal, which are intermittent signals, as continuous signals. That is, the low-speed shutter switching circuit 131a2 obtains a low-speed shutter image signal that is a continuous signal, and the high-speed shutter switching circuit 131b2 obtains a high-speed shutter image signal that is a continuous signal. These signals are input to the low-speed shutter characteristic conversion circuit 131a3 and the high-speed shutter characteristic conversion circuit 131b3. The low-speed shutter characteristic conversion circuit 131a3 and the high-speed shutter characteristic conversion circuit 131b3 perform characteristic conversion on the low-speed shutter image signal that is the continuous signal and the high-speed shutter image signal that is the continuous signal, for example, non-linear characteristics. Performs characteristic conversion to obtain gamma characteristics.

  The low-speed shutter characteristic conversion circuit 131a3 and the high-speed shutter characteristic conversion circuit 131b3 perform characteristic conversion according to the control unit 140, more specifically, the low-speed characteristic conversion control signal and the high-speed characteristic conversion control signal generated by the camera setting change unit 142. . The low-speed characteristic conversion control signal and the high-speed characteristic conversion control signal are used to construct a so-called dynamic range expansion image in which low-brightness to high-brightness is expressed by combining a low-speed shutter image and a high-speed shutter image. The control signal for optimizing the composite image.

[1.3.2. Camera signal processing circuit]
The camera signal processing circuit 132 has a function of converting the output of the WDR signal processing circuit 131 into a signal that can be displayed by the monitor device 200 and outputting the signal. FIG. 3 is a block diagram showing a specific configuration example of the camera signal processing circuit 132. In the example shown in FIG. 3, the camera signal processing circuit 132 includes an addition circuit 132a and a D / A conversion circuit 132b connected to the output of the addition circuit 132a.

  The low-speed shutter image signal subjected to characteristic conversion and the high-speed shutter image signal subjected to characteristic conversion, which are outputs of the low-speed shutter characteristic conversion circuit 131a3 and the high-speed shutter characteristic conversion circuit 131b3, are supplied to the camera signal processing circuit 132, more specifically, the addition circuit 132a. Sent to. The addition circuit 132a adds the low-speed shutter image signal whose characteristics have been converted and the high-speed shutter image signal whose characteristics have been converted at an addition ratio designated by the control unit 140. The addition circuit 132a instructs the image of the characteristic-converted low-speed shutter image signal and the characteristic-converted high-speed shutter image signal by the addition ratio control signal output by the control unit 140, more specifically, the camera setting management unit 142. Add according to the added ratio.

A digital image signal synthesized by addition, which is an output signal of the adder circuit 132a, is input to a digital / analog (D / A) conversion circuit 132b, converted from a digital image signal to an analog image signal, and externally output via an output terminal 132c. Is output.

[1.4. Control unit]
Next, returning to FIG. 1, a configuration example of the control unit 140 that describes the control unit 140 will be described.

  The control unit 140 calculates a luminance from the image signal output from the A / D conversion circuit 122, and detects and determines the environmental state based on the output of the photometry unit 141, and a plurality of pre-prepared A camera setting management unit 142 that selects camera setting data suitable for the current environmental state from the camera setting information, and an electronic shutter control unit 143 that controls and drives the shutter speed of the image sensor 112.

  The control unit 140 acquires image signal information from the low-speed shutter image signal and the high-speed shutter image signal by the photometry unit 141, and based on the acquired image signal information, the image sensor 112, the WDR signal circuit 131, the camera signal processing circuit 132, and the like. A control signal (low speed / high speed shutter speed control signal, nonlinear control signal, addition ratio control signal, etc.) to be controlled is generated by the camera setting management unit 142. Of the control signals generated by the camera setting management unit 142, a control signal for controlling the image sensor 112 is input to the electronic shutter controller 143, and a control signal for controlling the electronic shutter of the image sensor 112 provided in the image sensor 110 is used. Output.

[1.4.1. Metering unit]
First, the photometry unit 141 divides a captured image signal for one screen in order to acquire image signal information. In this embodiment, the photometry unit 141 divides a captured image signal for one screen (referred to as a frame) into a plurality of areas, for example, 100 areas each having 10 rows and 10 rows and an average of each area. Calculate the brightness. Here, the “average luminance” is a value obtained by normalizing the average luminance of all pixels in each area by 8 bits (0 to 255). The photometry unit 141 outputs the average luminance of each area to the camera setting management unit 142.

FIG. 4 is an explanatory diagram showing image signals for one screen ( one frame) divided into 100 areas. In the example shown in the figure, the image signal 400 for one screen is divided into 100 areas each consisting of 10 columns and 10 rows, and the average luminance in each area is normalized by 8 bits (0 to 255) ( (Hereinafter referred to as “average luminance value”). For example, the average luminance value of the area 401 in FIG. 4 is “43”.

The average luminance value is not limited to the average luminance in one area (hereinafter referred to as “unit area”), but the average of the average luminance in a plurality of areas (hereinafter referred to as “collection area”). Used to represent. For example, the collective area 402 in FIG. 4 is composed of two unit areas. The average luminance values of the two unit areas are “38” and “33”, respectively. In this case, the average luminance of the gathering area 402 is calculated by the following formula.
(38 + 33) /2=35.5
The average brightness value of the collective area is calculated not by the photometric unit 141 but by the camera setting management unit 142 described later.

[1.4.2. Camera Settings Manager]
The camera setting management unit 142 selects one of a plurality of camera setting data (a set of parameter values) prepared in advance based on the average luminance value of each unit area output from the photometry unit 142, and selects the selected camera Based on the setting data, a control signal for controlling the image sensor 112, the WDR signal circuit 131, the camera signal processing circuit 132, and the like is generated and transmitted. The camera setting management unit 142 is, for example, a microcomputer (microcomputer circuit) including a CPU, a ROM, and a RAM.

  FIG. 5 is a functional block diagram illustrating a configuration example of the camera setting management unit 142. In the configuration example illustrated in FIG. 5, the camera setting management unit 142 includes a main determination unit 501, a day / night determination unit 502 connected to the main determination unit 501 and the photometry unit 141, a tunnel determination unit 503, a tunnel entrance detection unit 504, The backlight detection unit 505 includes a control signal generation unit 507 connected to the main determination unit 501, and a camera setting storage unit 506 connected to the control signal generation unit 507.

  Note that the term “connected” here is used not only to be physically or electrically connected, but also to include that data is exchanged between programs and modules.

[1.4.2.1. Main judgment unit]
The main determination unit 501 receives the determination result or detection result from the day / night determination unit 502, the tunnel determination unit 503, the tunnel entrance detection unit 504, and the backlight detection unit 505, and is stored in the camera setting storage unit 506 accordingly. It has a function of determining which camera setting data is to be used among the plurality of camera setting data, and causing the control signal generation unit 507 to generate a control signal based on the camera setting data to be used. Here, “camera setting data” is a combination of various parameters that determine the operating characteristics of the imaging apparatus 100. Each camera setting data is information in which setting values of various parameters suitable for the corresponding environmental state are determined. In the present embodiment, four types of environmental states are defined, and camera settings corresponding to each environmental state are prepared. Note that the situation around the subject imaged by the imaging apparatus 100 is referred to as a current environment state, and the environment state detected by the control unit 140, more specifically, the camera setting management unit 142, is referred to as a recognition environment state.

  There are four types of recognition environment states that the imaging apparatus 100, more specifically the control unit 140, can take: “day state”, “night state”, “tunnel state”, and “backlight state”. FIG. 6 is a transition diagram showing an aspect of transition from one recognition environment state to another recognition environment state in the present embodiment.

In the imaging apparatus 100 according to the present embodiment, a transition is made from a camera activation state 601 that is a power-on state to a daytime state 602 that is a default recognition environment state. From the daytime state 602 , it is possible to transition to a night state 603, a tunnel state 604, and a backlight state 605. In addition, transition from the night state 603, the tunnel state 604, and the backlight state 605 to the day state 602 is possible. In addition, although the transition from the tunnel state 604 to the night state 603 is possible, the transition is not performed in reverse. This is because imaging in the night tunnel may be performed with camera settings corresponding to the night state 603.

Further, although it is possible to transit from the backlight state 605 to the tunnel state 604, the transition is not performed in reverse. This is because the transition from the tunnel state 604 to the backlight state 605 only needs to transit through the daytime state 602 once. Details of the determination operation performed by the main determination unit 501 will be described later.

[1.4.2.2. Day / night judgment part]
The day / night determination unit 502 has a function of determining whether the current current environmental state is the day state 602 or the night state 603 and outputting the determination result.

  The above determination is made based on the pseudo absolute luminance y of the collective area composed of 80 unit areas excluding the upper two lines in the image information for one screen divided into 10 rows and 10 columns. FIG. 7 is a diagram illustrating an example of a collection area for the day / night determination unit 502 to calculate the pseudo absolute luminance y. Of the image information 700 for one screen divided into 10 rows and 10 columns, a set area 701 composed of 80 unit areas excluding the upper two lines is used to calculate the pseudo absolute luminance y.

The pseudo absolute luminance y is obtained by the following equation.

In the above formula (1), the average brightness is an average value of the average brightness of each unit area in the collective area 701, and the exposure time is the exposure time in the image sensor 112 controlled by the electronic shutter control unit 143. . The AGC gain magnification is a gain magnification in the AGC circuit 121.

  The day / night determination unit 502 stores two threshold values to be compared with the pseudo absolute luminance y. FIG. 8 is a diagram illustrating two threshold values used by the day / night determination unit 502. The two threshold values are a threshold value Y1 for determining the establishment of the night state and a threshold value Y2 for determining the establishment of the day state. In the present embodiment, it is determined that the relationship of Y2 = Y1 + Δ is established.

  The day / night determination unit 502 determines that the current current environmental state is the night state if the pseudo absolute luminance y is less than the threshold value Y1, while the current current environmental state is determined if the pseudo absolute luminance y exceeds the threshold value Y2. Determined to be in the daytime state. When the pseudo absolute luminance y satisfies Y2 ≧ y ≧ Y1, the day / night determination unit 502 does not perform state transition. That is, when the current current environment state is the night state, the recognition environment state is left at the night state, and when the recognition environment state is the day state, the day state is kept. In any case, the day / night determination unit 502 determines that the day state is established or the night state is established when the condition is satisfied in the f times (frames) of determination. The value of f is 900, for example.

[1.4.2.3. Tunnel determination unit]
The in-tunnel determination unit 503 has a function of determining whether or not the current environmental state is a tunnel state. Similar to the case of the day / night determination unit 502, the above-described determination is a collective area 701 composed of 80 unit areas excluding the upper two lines in the image information for one screen divided into 10 rows and 10 columns (see FIG. 7)) based on the pseudo absolute luminance y. The calculation method of the pseudo absolute luminance y is the same as that in the day / night determination unit 502.

More specifically, the in-tunnel determination unit 503 performs the following three types of determination.
(1) When the imaging apparatus 100 is operating with the environmental state being the “tunnel state”, the recognition environmental state is transitioned from the “tunnel state” to the “daytime state” when the average luminance is higher than a predetermined threshold. When the detection result by the tunnel entrance detection unit 504, which will be described later, is that the tunnel has not been detected, when the imaging apparatus 100 is operating with the recognition environment state being “daytime”, when the average luminance is lower than a predetermined threshold value, “daytime” (3) When the detection result by the tunnel entrance detection unit 504 described later is the tunnel entrance detection, the average luminance is more than a predetermined threshold after a certain time from the detection. If not lowered, the recognition environment state is shifted from the “tunnel state” to the “daytime state” because the tunnel entrance detection is erroneous.

The in-tunnel determination unit 503 stores a plurality of threshold values to be compared with the pseudo absolute luminance y. FIG. 9 is a diagram illustrating the two threshold values used by the in-tunnel determination unit 503 in the determination of (1) above. The two threshold values are a threshold value Y3 for determining whether the current environmental state is a tunnel state in the daytime and a threshold value for determining whether the current environmental state is the tunnel state in the evening. Y4 is used. This is because the illuminance near the tunnel exit is higher in the daytime than in the evening, so if you decide whether you are inside the tunnel using the same threshold for both daytime and evening, This is because the recognition environment state transitions to the daytime state.

  That is, the determination unit 503 in tunnel compares the pseudo absolute luminance y with the threshold value Y3 in the daytime, and determines that the current environmental state is the “tunnel state” if y <Y3, and conversely y If it is ≧ Y3, it is determined that it is not in the tunnel, that is, not in the “tunnel state”. In the evening, the tunnel determination unit 503 compares the pseudo absolute luminance y with the threshold value Y4, and determines that the current environmental state is the “tunnel state” if y <Y4. If it is ≧ Y4, it is determined that it is not in the tunnel. Whether it is daytime or evening is determined using the value of the pseudo absolute luminance y before the in-tunnel determination unit 503 determines the tunnel state.

Further, in any case, the in-tunnel determination unit 503 determines that the daytime state is established when the condition is satisfied in the determination of f times (frames) in succession. The value of f is 30, for example.

In the case of the determinations (2) and (3) above, the in-tunnel determination unit 503 similarly makes a determination by comparing the average luminance y with a predetermined threshold value, but the predetermined threshold values to be used are different from each other. Thresholds may be used.

[1.4.2.4. Tunnel entrance detector]
The tunnel entrance detection unit 504 has a function of detecting the presence of the tunnel entrance from the image information. The tunnel entrance detection unit 504 determines the presence / absence of a tunnel entrance from the luminance characteristics of image information obtained by imaging the vicinity of the tunnel entrance. FIG. 10 is an image near the tunnel entrance and shows a continuous image that changes over time. The time elapses in the order of FIG. 10A, FIG. 10B, and FIG. 10C, and the imaging device 100 approaches the tunnel entrance. Each of FIG. 10A, FIG. 10B, and FIG. 10C includes a road surface portion 1000 and a semicircular tunnel entrance portion 1001 on the road surface portion 1000.

As shown in FIGS. 10A, 10B, and 10C, the image near the tunnel entrance has the following characteristics.
(1) The tunnel entrance portion 1001 exists as a dark ( low brightness) portion, and the dark portion that is the tunnel entrance portion 1001 spreads over time.
(2) The brightness (brightness) of the road surface portion 1000 is brighter than that of the tunnel entrance portion 1001, and the brightness of the road surface portion 1000 is maintained even if time elapses.

In order to quantify the luminance characteristics of (1) and (2) above, the tunnel entrance detection unit 504 obtains the average brightness of the tunnel entrance portion 1001 and the road surface portion 1000. FIG. 11 is a diagram (schematically) displaying an image signal for one screen divided into 100 unit areas and captured by the imaging apparatus 100 at the tunnel entrance.

As shown in FIG. 11 , six aggregate rears are set in order to quantify the tunnel entrance luminance characteristics. That is, the collective area A for the entire tunnel entrance part (2 rows and 2 columns), the collective area B for the tunnel entrance part left side (2 rows and 2 columns), the collective area C for the tunnel entrance part right side (2 rows and 2 columns), The gathering area E (2 rows and 1 column) for the upper half of the tunnel entrance portion, the gathering area H (2 rows and 1 column) for the lower half of the tunnel entrance portion, and the gathering area D for the road surface portion.

The average brightness of each set area is expressed as follows.
Average brightness of gathering area A = L (A)
Average brightness of gathering area B = L (B)
Average brightness of gathering area C = L (C)
Average brightness of gathering area D = L (D)
Average brightness of gathering area E = L (E)
Average brightness of gathering area H = L (H)
The tunnel entrance detection unit 504 calculates the following variables x ₆ and x ₈ using the above average luminance. These variables are used as variables of a correlation function by regression analysis described later.


The variable x ₆ represents the luminance ratio in road surface portion and the tunnel, the variable x ₈ shows a luminance ratio of tunnel and the tunnel top.

Collect images of multiple “tunnel entrances” and “non-tunnel entrances” as sample data. For each sample data, x ₆ and x ₈ are calculated by the above formulas (2) and (3), respectively, and plotted on a scatter plot with x ₆ on the vertical axis and x ₈ on the horizontal axis. Check whether the area can be broadly divided into “entrance” and “non-tunnel entrance”. FIG. 12 shows a scatter diagram in which data of sample images obtained by photographing “tunnel entrance” and “non-tunnel entrance” are plotted. In the figure, a value of x ₈ on the horizontal axis, the vertical axis represents the value of x _6, plotted for each screen information. In FIG. 12 , points indicated by square markers correspond to image information of “tunnel entrance”, and points indicated by asterisk markers correspond to image information of “non-tunnel entrance”.

When it is confirmed that the sample data is divided into two regions by the scatter diagram, the following function z corresponding to the boundary line (straight line 1300 in FIG. 12 ) of the two regions is defined.

In the above formula (4), alpha ₁ is a fixed _parameter, alpha _2, alpha _{3 (both} multiple regression coefficients) obtained beforehand by multiple regression analysis and stored.

The tunnel entrance detector 504 calculates the values of the variables x ₆ and x ₈ for each screen , and then applies these values and the fixed parameters α ₁ , α ₂ , and α ₃ to the above equation (4) to obtain the function z. Is calculated.

Tunnel entrance detection unit 504, based on the value of the calculated function z, determines whether it has detected the tunnel entrance. More specifically, when the value of the calculated function z is z> 0, the tunnel entrance detection unit 504 determines that the tunnel detection is successful and outputs an output indicating that the tunnel detection is successful to the main determination unit 501. hand over. On the other hand, when the value of the calculated function z is z ≦ 0, it is determined that the tunnel detection has failed (the tunnel entrance does not exist), and an output indicating that the tunnel detection is not successful is passed to the main determination unit 501.

After determining fixed parameters α ₁ , α ₂ , α ₃ by multiple regression analysis using a scatter diagram, a simulation was performed to evaluate whether tunnel detection can be performed using these parameters and variables x ₆ , x ₈ It was. FIG. 13 is a diagram in which the value (discriminant value) of the function z is calculated for each successive frame from the function z specifically obtained by the multiple regression analysis, and the time change is displayed. The horizontal axis of the graph shown in FIG. 13 represents consecutive frame numbers, and the vertical axis represents the value of the function z of each frame . Furthermore, the tunnel position and the frame positions 40m, 100m, 150m and 200m in front of the tunnel are shown in bold lines on the same graph.

There are three tunnels in the graph of FIG . The value of the function z changes from negative (z <0) to positive (z> 0) immediately before each tunnel entrance, and it can be confirmed that tunnel detection by the tunnel detection unit 504 is properly performed.

[1.4.2.5. Backlight detection unit]
Returning to FIG. 5, the backlight detection unit 505, which is one of the components of the camera setting management unit 142, will be described. The backlight detection unit 505 has a function of detecting that the current environment state is a backlight state from the luminance characteristics of the image information.

A scene where it is better to use the camera setting data corresponding to the “backlight state” means that the electronic shutter speed is increased by being pulled by the sun on the screen, and therefore the screen displayed on the monitor device 200 is not night. In this scene, the road surface is extremely dark and the sky is bright. Similarly sky is a bright scene is not a backlit sometimes like to travel to the location where wide open the surrounding like a sea of road but, instead of the average brightness, when compared with the pseudo-absolute luminance, backlit scene Is generally larger (brighter) . Therefore, paying attention to the following two points, the sky and the road surface are defined, the pseudo absolute brightness is obtained from the average brightness of each of the sky and the road surface, and the multiple regression analysis is performed on the two elements. Detect state.
(1) The road surface is dark (2) Even if the brightness of the sky is saturated, the brightness of the sun and normal sky are different

FIG. 14 is a diagram illustrating an example of a region (aggregate area) in which the backlight detection unit 505 obtains the pseudo absolute luminance of the sky portion and the road surface portion. In the image information for one screen divided into 100 unit areas, the collection area A (2 rows and 6 columns), the collection area B (2 rows and 6 columns), and the collection area C (2 rows and 6 columns) are empty areas. Is set. The lower row of the collective area A and the upper row of the collective area B overlap, and the lower row of the collective area B and the upper row of the collective area C overlap. In addition, a collective area D (2 rows and 6 columns) for calculating the pseudo average luminance of the road surface portion is set in the lower center of the screen. The upper two rows of the screen are mask areas.

The average brightness of each set area is expressed as follows.
Average brightness of gathering area A = L (A)
Average brightness of gathering area B = L (B)
Average brightness of gathering area C = L (C)
Average brightness of gathering area D = L (D)
The backlight detection unit 505 calculates the following variables x ₄ and x ₅ using the above average luminance. These variables are used as variables of a correlation function by regression analysis described later.

The variable x ₄ indicates the pseudo absolute luminance of the road surface portion, and the variable x ₅ indicates the maximum pseudo absolute luminance of the collective areas A, B, and C corresponding to the sky portion.

Images with a camera setting of “daytime” that captures the image information of the “backlight” scene, which is the scene corresponding to the backlight, the image information of the “flare” scene, and the “daytime” scene, which corresponds to the scene that is not backlit. Collect information as sample data. For each sample data, x ₄ and x ₅ are calculated by the above formulas (5) and (6), respectively, and plotted on a scatter diagram with x ₄ on the vertical axis and x ₅ on the horizontal axis. Check if it can be roughly divided into two areas: + flare and daytime. FIG. 15 is a scatter diagram in which data of sample images obtained by photographing “backlight”, “flare”, and “daytime” are plotted. In the figure, a value of x ₅ in the horizontal axis, the vertical axis represents the value of x _4, and plotted for each screen information. In FIG. 15 , points indicated by black rhombus markers are “backlight” image information, points indicated by white square markers are points indicated by “flare” image information, and white triangle markers. Corresponds to image information of “daytime”, which is a scene that is not backlit.

When it is confirmed that the sample data is divided into two regions by the scatter diagram, the following function w corresponding to the boundary line between the two regions (straight line 1600 in FIG. 15 ) is defined.
w = α ₄ · x ₄ + α ₅ · x ₅ + α ₆ (7)
In the above equation (7), α ₄ , α ₅ , and α ₆ (all of which are multiple regression coefficients) that are fixed parameters are obtained in advance by multiple regression analysis and stored.

After calculating the values of the variables x ₄ and x ₅ for each screen , the backlight detection unit 505 applies the values of these variables and the fixed parameters α ₄ , α ₅ , and α ₆ to the above equation (7) to obtain the function w Is calculated.

  The backlight detection unit 505 determines whether a backlight state is detected based on the calculated value of the function w. More specifically, when the value of the calculated function w is w> 0, the backlight detection unit 505 determines that the current environment state is the backlight state, and outputs an output indicating the backlight state as the main determination unit. Pass to 501. On the other hand, when the value of the calculated function w is w ≦ 0, it is determined that the backlight detection has failed (a state that is not the backlight state), and an output indicating that the backlight detection is not successful is passed to the main determination unit 501.

[1.4.2.6. Camera setting storage section]
Returning to FIG. 5, the description of the components of the camera setting management unit 142 will be continued.
The camera setting storage unit 506 stores data (camera setting data and camera setting data) corresponding to each of four types of current environmental conditions, “daytime state”, “night state”, “tunnel state”, and “backlight state”. Has a function of storing (calling).

19 from FIG. 16, "afternoon state", "Night state", "tunnel state" shows a data configuration example of a camera setting data corresponding to each of the "backlit." These camera setting data 1700 corresponding to the “daytime state”, “night state”, “tunnel state”, and “backlight state” are “convergence value”, “mask area”, In addition to describing the values of “number of areas”, “addition ratio”, and “non-linear”, it also describes whether or not to turn on the WDR mode.

[1.4.2.7. Control signal generator]
Returning to FIG. 5, the description of the components of the camera setting management unit 142 will be continued.
The control signal generation unit 507 reads the camera setting data corresponding to the specified environmental situation from the camera setting storage unit 506 in accordance with the signal specifying the environmental situation from the main determination unit 501, and converts the camera setting data into the read camera setting data. Based on this, the value of the non-linear characteristic is transmitted to the low-speed shutter characteristic conversion circuit 131a3 and the high-speed shutter characteristic conversion circuit 131b3 , and the characteristic-converted high-speed shutter image signal and the characteristic-converted high-speed shutter image signal generates a synthesis ratio control signal for instructing the combining ratio is a ratio to synthesize a shutter image signal performs like to send. With the various control signals generated by the control signal generation unit 507, the imaging apparatus 100 can perform the imaging process according to the camera settings corresponding to the environmental state determined by the main determination unit 501 .

[1.4.3. Electronic shutter control unit ]
Returning to FIG. 1, the description of the components of the control unit 140 is resumed.
The electronic shutter control unit 143 that is a component of the control unit 140 includes two electronic shutters having different low-speed shutter speeds and high-speed shutter speeds according to control signals from the camera setting management unit 142, more specifically, the control signal generation unit 507. The image sensor 112 is controlled at a speed.
Above, description of the structural example of the imaging device 100 is complete | finished.

[2. Example of operation when changing camera settings of imaging device]
Next, the imaging apparatus 100 having the above structure, more specifically the operation of the camera setting management unit 142 will be described with reference to FIG. 23 from FIG. 20. 20 to 23 are flowcharts illustrating an example of the operation of the imaging apparatus 100, more specifically, the camera setting management unit 142.

  The imaging apparatus 100, more specifically the camera setting management unit 142, uses the luminance characteristics of the image information obtained from the imaging element 112, which are four environmental states “day state”, “night state”, “tunnel state”, and By detecting the “backlight state” and using the camera setting data corresponding to the detected current environment state based on the detection result, a dynamic range necessary for the environment is secured and imaging is performed.

  When the power of the imaging apparatus 100 is turned on, the camera setting management unit 142, more specifically, the main determination unit 501 displays the variable “wdrState” indicating the recognition environment state, the variable “tunnelState” indicating the tunnel detection result, and the activation state. The variable "start flag" shown is set to the initial value. The camera setting management unit 142, more specifically, the main determination unit 501 stores these variables in a rewritable manner.

  Possible values of the variable “wdrState” are “daytime” indicating “daytime state”, “night” indicating “night state”, “tunnel” indicating “tunnel state”, and “backlighting” indicating “backlight state”. Is one of four types of values. Further, the values that can be taken by the variable “tunnelState” are “tunnel detection OK” indicating that the tunnel entrance detection unit 504 has detected the tunnel entrance, and “no tunnel entrance detection unit 504 has detected the tunnel entrance”. One of two values of “tunnel detection NG”. The variable “activation flag” can take two values, “true” indicating that a predetermined time t1 has elapsed after activation and “false” indicating that the predetermined time t1 has not elapsed since activation. One of them.

  Now, the camera setting management unit 142, more specifically, the main determination unit 501 sets the above three variables to default values “wdrState” = tunnel, “tunnelState” = tunnel detection OK, “start flag” = “false” after power-on. (S10).

Next, the camera setting management unit 142, more specifically, the main determination unit 501 determines whether or not the value of the stored variable “wdrState” is “daytime” (S20). When the value of the variable “wdrState” is “daytime” (S20, Yes), the camera setting management unit 142, more specifically, the main determination unit 501, determines whether or not the value of the variable “start flag” is “false”. Is determined (S30). On the other hand, if the value of the variable “wdrState” is not “daytime” in S20 (S20, No), the process proceeds to step S210 (see FIG. 22 ) described later.

When it is determined in step S30 that the value of the variable “activation flag” is “false” (S30, Yes), the main determination unit 501 refers to a timer or counter that has started counting after activation, and is predetermined after activation. It is determined whether the time t1 has elapsed (S40). On the other hand, when it is determined that the value of the variable “activation flag” is not “false” (S30, Yes), the main determination unit 501 proceeds to step S60 ( FIG. 20 ) described later.

  If it is determined in step S40 that the predetermined time t1 has elapsed after activation (S40, Yes), the main determination unit 501 rewrites the value of the variable “activation flag” to “true” and stores it (S50). On the other hand, when it is determined that the predetermined time t1 has not elapsed after activation (S40, No), the main determination unit 501 shifts the control to step S60 described later.

When the determination result is “No” in both step S30 and step S40 described above, the main determination unit 501 causes the day / night determination unit 502 to determine whether or not the current environmental state is the “night state” (S60). . The day / night determination unit 502 compares the pseudo absolute luminance y with the threshold value Y1 for determining the night state, determines whether or not it is the “night state”, and outputs the determination result to the main determination unit 501. When the determination result of the day / night determination unit 502 is “night state” (S60, Yes), the main determination unit 501 rewrites the value of the variable “wdrState” to “night” and stores this (S70). After execution of step S70, the main determination unit 501 returns control to step S20 described above. On the other hand, in the determination of step S60, when the determination result of the day / night determination unit 502 is not “night state” (S60, No, the main determination unit 501 proceeds to step S110 (see FIG. 21 ) described later.

Subsequent to step S70 and the return from the external page references 3, 4, and 6 in FIG. 20 , the main control unit 501 refers to the stored variable “wdrState” and stores the camera setting data corresponding to this variable. The control signal generator 507 is controlled so as to generate various control signals by using the control signal (S80). Thereby, the camera setting data is selected according to the value of the variable “wdrState” determined according to the detection results of the day / night determination unit 502, the tunnel determination unit 503, the tunnel entrance determination unit 504, and the backlight detection unit 505. Each part is controlled according to the camera setting data suitable for the state, and a dynamic range suitable for the current environment state is selected.

Now, with reference to FIG. 21 , the processing after step S110, which shifts from step 60, will be described.
When control is transferred to step S110, the main determination unit 501 causes the tunnel entrance detection unit 504 to execute tunnel entrance detection processing, and determines whether there is a tunnel entrance ahead (S110). The tunnel entrance detection unit 504 calculates the variables x ₆ and x ₈ based on the average luminance of the collection areas A to H shown in FIG. 11, and applies the calculated variables x ₆ and x ₈ to the above-described equation (4). The value of z is determined to determine the presence of the tunnel entrance, and the determination result is output to the main determination unit 501.

  If the tunnel entrance detection unit 504 determines that the tunnel entrance exists (S110, Yes), the main determination unit 501 rewrites and stores the value of the variable “wdrState” as “tunnel” (S120). Accordingly, various parameters of the imaging apparatus 100 are determined based on the camera setting data corresponding to the environmental state “tunnel”, and the imaging apparatus 100 operates in a state suitable for imaging in the tunnel.

After rewriting the variable “wdrState”, the main determination unit 501 returns the control to the above-described step S20 (see FIG. 20 ). On the other hand, when it is determined from the determination result of the tunnel entrance detection unit 504 that there is no tunnel entrance (No in S110), the main determination unit 501 determines whether or not the current value of the variable “activation flag” is “true”. Is determined (S130). If it is determined in step S130 that the current value of the variable “activation flag” is “true” (S130, Yes), the main determination unit 501 uses the mid-tunnel determination unit 503 to determine whether the tunnel is in progress. In-tunnel determination processing is executed to determine whether or not (S140). In this in-tunnel determination process, the in-tunnel determination unit 503 compares the pseudo absolute luminance y of the set area 701 (see FIG. 7) composed of 80 unit areas excluding the upper two lines with a predetermined threshold value, and determines the pseudo absolute luminance. If y is lower than a predetermined threshold, it is determined that the tunnel is inside, and the determination result is output to the main determination unit 501.

When the determination result of the in-tunnel determination unit 503 is in the tunnel (S140, Yes), the main determination unit 501 rewrites and stores the value of the variable “wdrState” as “tunnel” and sets the value of the variable “tunnnelState” to “ It is rewritten and stored in “tunnel detection NG” (S150). After rewriting these variables, the main determination unit 501 returns the control to the above-described step S80 (see FIG. 20 ).

When it is determined in step S130 that the current value of the variable “activation flag” is not “true” (S130, No), and when the determination result of the in-tunnel determination unit 503 is not within the tunnel in step S140 (S140) , No), the main determination unit 501 causes the backlight detection unit 505 to perform backlight detection and determines whether or not it is in the backlight state (S160). The backlight detection unit 505 calculates the variables x ₄ and x ₅ from the average luminance of each of the collective areas A to D shown in FIG. 14 , applies the calculated variables x ₄ and x ₅ to the above equation (7), and sets the function The value of w is determined to determine the presence of the tunnel entrance, and the determination result is output to the main determination unit 501.

In step S160, when the determination result of the backlight detection unit 505 is the backlight state (S160, Yes), the main determination unit 501 rewrites and stores the value of the variable “wdrState” as “backlight” (S170). After rewriting the variable “wdrState”, the main determination unit 501 returns the control to the above-described step S80 (see FIG. 20 ). On the other hand, when the determination result of the backlight detection unit 505 is not the backlight state (S160, No), the main determination unit 501 shifts the control to the above-described step S80 (see FIG. 20 ) without performing any processing.

Next, with reference to FIG. 22 , the processing when it is determined in step S20 that the value of the variable “wdrState” is not “daytime” ( FIG. 20 , S20, No) will be described.
In this case, the main determination unit 501 determines whether or not the value of the stored variable “wdrState” is “tunnel” (S210). If the value of the variable “wdrState” is not “tunnel” (S210, No), the process proceeds to step S410 (see FIG. 23 ) described later. On the other hand, when the value of the variable “wdrState” is “tunnel” (S210, Yes), the main determination unit 501 determines whether or not the value of the stored variable “tunnelState” is “tunnel detection NG”. (S220). When the value of the variable “tunnelState” is not “tunnel detection NG” (S220, No), the process proceeds to step S270 described later.

  On the other hand, when the value of the variable “tunnelState” is “tunnel detection NG” (S220, Yes), the main determination unit 501 determines whether or not to transition the environmental state from the “tunnel state” to the “daytime state”. Therefore, the in-tunnel determining unit 503 is controlled to detect the tunnel exit (S230). This is a process for transitioning the environmental state from the “tunnel state” to the “daytime state” when going out of the tunnel from inside the tunnel.

  The in-tunnel determination unit 503 compares the pseudo absolute luminance y of the set area 701 (see FIG. 7) composed of 80 unit areas excluding the upper two lines with the predetermined threshold, and the pseudo absolute luminance y is higher than the predetermined threshold. In the case, it is determined that the vicinity of the tunnel exit is approached, and the determination result is output to the main determination unit 501.

When the determination result of the in-tunnel determination unit 503 is close to the tunnel exit (S230, Yes), the main determination unit 501 rewrites the value of the variable “wdrState” to “daytime” and stores it (S240). After the rewriting of the variable is completed, the main determination unit 501 shifts the control to the above-described step S80 (see FIG. 20 ).

On the other hand, in step S230, when the determination result of the in-tunnel determination unit 503 is not close to the vicinity of the tunnel exit (No in S230), the main determination unit 501 determines whether or not the predetermined time t2 has elapsed ( S250). When it is determined that the predetermined time t2 has elapsed (S250, Yes), the main determination unit 501 rewrites the value of the variable “wdrState” to “night” and stores it (S260). After completing the rewriting of the variable “wdrState”, the main determination unit 501 returns the control to the above-described step S20 (see FIG. 20 ). This is a process for transitioning to the “night state” if the “tunnel state” does not end even after a certain time (t2) has elapsed. On the other hand, when it is determined in step S250 that the predetermined time t2 has not elapsed (S250, No), the main determination unit 501 moves the control to the above-described step S80 (see FIG. 20 ) without performing any processing.

  Here, the description returns to step S220. If the value of the variable “tunnelState” is not “tunnel detection NG” in step S220 (S220, No), the main determination unit 501 uses the in-tunnel determination unit 503 to determine whether the current environment state is “tunnel state”. It is determined whether or not (S270). The in-tunnel determination unit 503 compares the pseudo absolute luminance y of the aggregate area 701 (see FIG. 7) composed of 80 unit areas excluding the upper two lines, and the pseudo absolute luminance y is lower than the predetermined threshold. If not, it is determined that it is inside the tunnel, and if not, it is determined that it is not inside the tunnel, and the determination result is output to the main determination unit 501. In the determination in step S270, as shown in FIG. 9, it is determined whether or not the vehicle is inside the tunnel using threshold values Y3 and Y4 that are different between daytime and evening.

  When the determination result of the in-tunnel determination unit 503 is not in the tunnel (No in S270), the main determination unit 501 shifts control to step S300 described later. On the other hand, when the determination result of the in-tunnel determination unit 503 is in the tunnel (S270, Yes), the main determination unit 501 determines whether or not the tunnel determination unit 503 changes the environmental state from “tunnel state” to “daytime state”. The in-tunnel determining unit 503 is controlled to detect the tunnel exit (S280). Since the processing content in step S280 is the same as that in step S230 described above, detailed description thereof is omitted.

When the determination result of the in-tunnel determination unit 503 approaches the vicinity of the tunnel exit (S280, Yes), the main determination unit 501 rewrites the value of the variable “wdrState” to “daytime” and stores it (S290). After completion rewritten for this variable, the main determining section 501 returns control to the aforementioned step S20 (see FIG. 20).

  On the other hand, in step S280, when the determination result of the in-tunnel determination unit 503 is not close to the vicinity of the tunnel exit (No in S280), the main determination unit 501 determines whether or not the predetermined time t2 has passed ( S250). Since the subsequent processing is the same as that described above for step S250, the description thereof is omitted here.

Here, the description returns to step S270 described above. When the determination result of the in-tunnel determining unit 503 is not within the tunnel in step S270 (S270, No), the main determining unit 501 determines whether or not the predetermined time t3 has elapsed (S300). When it is determined that the predetermined time t3 has elapsed (S300, Yes), the main determination unit 501 rewrites the value of the variable “wdrState” to “daytime” and stores it (S310). After the rewriting of the variable “wdrState” is finished, the main determination unit 501 returns the control to the above-described step S80 (see FIG. 20 ).

Next, with reference to FIG. 23 , processing after the case where the value of the variable “wdrState” is determined not to be “tunnel” in the above-described step S210 (see FIG. 22 ) (No in FIG. 22 , S210) will be described.

When it is determined in step S210 that the value of the variable “wdrState” is not “tunnel”, the main determination unit 501 determines whether the value of the stored variable “wdrState” is “backlight” (S410). ). If the value of the variable "wdrState" is not "backlit" in step S410 (S410, No), it returns control to the aforementioned step S80 (see FIG. 20) without performing the main determination unit 501 any processing. On the other hand, when the value of the variable “wdrState” is “backlight” in step S410 (S410, Yes), the main determination unit 501 causes the tunnel entrance detection unit 504 to execute the tunnel entrance detection process, and there is a tunnel entrance ahead. It is determined whether or not to perform (S420). Since the contents of the tunnel entrance process detection are the same as the contents of step S110 described above, a detailed description thereof will be omitted. When the main determination unit 501 determines from the determination result of the tunnel entrance detection unit 504 that a tunnel entrance exists (S420, Yes), the main determination unit 501 rewrites and stores the value of the variable “wdrState” as “tunnel” (S430). ). After rewriting the variable “wdrState”, the main determination unit 501 shifts the control to the above-described step S80 (see FIG. 20 ).

On the other hand, if it is determined in step S420 that the tunnel entrance does not exist from the determination result of the tunnel entrance detection unit 504 (S420, No), the main determination unit 501 uses the mid-tunnel determination unit 503 to determine whether or not the tunnel is in the tunnel. In-tunnel determination processing is executed to determine whether or not (S440). Note that the content of the in-tunnel determination process is the same as the content performed in step S140 (see FIG. 21 ) described above, and a detailed description thereof will be omitted.

When the determination result of the in-tunnel determination unit 503 is in the tunnel (S440, Yes), the main determination unit 501 rewrites and stores the value of the variable “wdrState” as “tunnel” and sets the value of the variable “tunnnelState” to “ It is rewritten and stored in “tunnel detection NG” (S450). After rewriting these variables, the main determination unit 501 returns the control to the above-described step S80 (see FIG. 20 ).

On the other hand, when the determination result of the in-tunnel determining unit 503 is not in the tunnel in step S440 (S440, No), the main determining unit 501 determines whether or not the predetermined time t4 has passed (S460). When it is determined that the predetermined time t4 has elapsed (S460, Yes), the main determination unit 501 rewrites the value of the variable “wdrState” to “daytime” and stores it (S470). After the rewriting of the variable “wdrState” is completed, the main determination unit 501 shifts the control to the above-described step S80 (see FIG. 20 ). This is a process for transitioning to the “daytime state” when the “backlight state” does not end even after a lapse of a certain time (t3).

On the other hand, in step S460, back to when it is determined that the predetermined time t4 has not elapsed (S460, No), (see Fig. 20) step S80 of the aforementioned control without the main judgment unit 501 any processing.

  By executing the above processing, the environment information such as “daytime”, “nighttime”, “tunnel”, and “backlight” is detected using the luminance information obtained from the image sensor, and the camera settings of the imaging device are set based on the detection result. It is possible to change the parameters and perform imaging with camera setting parameters suitable for each environmental state.

Functional block diagram of an imaging apparatus according to an embodiment of the present invention Block diagram showing a specific configuration example of a WDR signal processing circuit Block diagram showing a specific configuration example of a camera signal processing circuit Explanatory diagram showing image signals for one screen ( one frame) divided into 100 unit areas Functional block diagram showing a configuration example of the camera setting management unit Transition diagram showing an aspect of transition from one recognition environment state to another recognition environment state in the present embodiment The figure which showed the example of the collection area for the day / night determination part to calculate the pseudo absolute luminance y The figure explaining two threshold values which the day / night determination part 502 uses The figure explaining the two threshold values used by the mid-tunnel determination unit 502 for determination (A) is an image near the tunnel entrance, showing a continuous image that changes over time, (b) is a diagram of an image following (a), and (c) is an image following (b). Figure of the The figure which displays the image signal which the imaging device photographed the tunnel entrance as a figure Scatter plot with sample data plotted for image information of "tunnel entrance" and "non-tunnel entrance" Figure showing the time change of function z value (discriminant value) calculated for each frame of continuous data from function z obtained by multiple regression analysis The figure which shows the example of the area | region (gathering area) from which a backlight detection part calculates | requires the pseudo absolute brightness | luminance of a sky part and a road surface part Scatter plot with sample data plotted for image information of "Backlight", "Flare" and "Daytime" An example of the data structure of camera setting data corresponding to “daytime” An example of the data structure of the camera setting data corresponding to “Night state” An example of the data structure of the camera setting data corresponding to the “tunnel state” is shown. An example of the data structure of the camera setting data corresponding to the “backlight state” Flow chart showing an operation example of an imaging apparatus, more specifically, a camera setting management unit Flow chart showing an operation example of an imaging apparatus, more specifically, a camera setting management unit Flow chart showing an operation example of an imaging apparatus, more specifically, a camera setting management unit Flow chart showing an operation example of an imaging apparatus, more specifically, a camera setting management unit

DESCRIPTION OF SYMBOLS 1 ... Imaging device 110 ... Imaging part 120 ... Analog signal processing part 130 ... Digital signal processing part 140 ... Control part 501 ... Main judgment part 502 ... Day / night judgment part 503 ... In-tunnel judgment part 504 ... Tunnel entrance detection part 505 ... Backlight detection Unit 506 ... Camera setting storage unit 507 ... Control signal generation unit

Claims (4)

An imaging apparatus having a function of imaging a subject at two different shutter speeds and expanding a dynamic range,
Imaging means for alternately repeating imaging of a subject at two different shutter speeds, and outputting a first analog image signal and a second analog image signal corresponding to each shutter speed;
The gains of the first analog image signal and the second analog image signal are varied, and the first analog image signal and the second analog image signal are converted into the first digital image signal and the second digital image signal. Analog signal processing means;
The first digital image signal and the second digital signal are converted into a first continuous image signal and a second continuous image signal, which are continuous image signals, respectively, and the first continuous image signal and the second continuous image are converted. Digital signal processing means for performing characteristic conversion using a specified characteristic conversion parameter value and outputting the first continuous image signal and the second continuous image signal subjected to characteristic conversion;
The first digital image signal and the second digital image signal are added at a specified addition ratio to generate a third digital image signal, and the third digital image signal is converted into an analog image signal. Digital signal processing means for outputting;
For at least one of the first digital image signal and the second digital image signal, by dividing the digital image signal of one screen into a plurality of unit areas, and calculates an average brightness of the unit areas, to the calculated average brightness An environmental state is determined based on the selected camera setting data according to the determined environmental state, an electronic shutter signal is output to the analog signal processing unit based on the selected camera setting data, and the digital signal processing unit imaging apparatus characterized by a control means for outputting a value that specifies the characteristic conversion parameter value and the addition ratio. The control means includes an average luminance of a first collective area consisting of a plurality of unit areas corresponding to a tunnel entrance portion, an average brightness of a second collective area consisting of a plurality of unit areas corresponding to a road surface portion, and a tunnel entrance based on image information and the image information of the non-tunnel inlet of the function using the obtained constants by regression analysis, and the tunnel inlet detection means for detecting a tunnel inlet portion,
Camera setting storage means for storing camera setting data corresponding to the tunnel ;
When the tunnel entrance detection means detects a tunnel entrance, a value specifying the control signal and the addition ratio indicating the characteristic conversion parameter value according to the camera setting data corresponding to the tunnel stored in the camera setting storage means The image pickup apparatus according to claim 1, further comprising: a control signal generation unit that generates a control signal that instructs The camera setting management means includes a pseudo absolute luminance of a third collective area consisting of a plurality of unit areas corresponding to a sky portion, a pseudo absolute luminance of a fourth collective area consisting of a plurality of unit areas corresponding to a road surface portion, and A backlight detection means for detecting a backlight state based on a function using image information corresponding to a backlight scene and a constant obtained by regression analysis of image information of a scene that is not backlight;
Camera setting storage means for storing camera setting data corresponding to the backlight state;
When the backlight detection unit detects a backlight state, a value specifying a control signal and an addition ratio indicating the characteristic conversion parameter value according to camera setting data corresponding to the backlight state stored in the camera setting storage unit The imaging apparatus according to claim 1, further comprising: a control signal generation unit that generates a control signal that instructs   The camera setting data stores camera setting data corresponding to each of the four environmental states of daytime, nighttime, tunneling, and backlighting. Each camera setting data includes the characteristic conversion parameter value and the addition. The information specifying at least one of execution or cancellation of the function for expanding the dynamic range, brightness convergence value, photometric area, and mask area is stored in addition to the ratio. Item 4. The imaging device according to any one of Items 3.