JP7277205B2 - Traffic light detection method and traffic light detection device - Google Patents

Description

The present invention relates to a traffic signal detection method and a traffic signal detection device.

As a technology for recognizing a traffic signal even when the overall shape of the traffic signal cannot be seen due to nighttime or conditions where the traffic signal is partially blocked, we focused on the fact that the brightness of the lamp of the traffic signal changes due to the frequency of the input AC power. However, a traffic light detection device and a traffic light detection method using a synchronous image generation unit that calculates frequency components by extracting a region where luminance changes at the same frequency as the lamp of the traffic light by frequency analysis has been proposed (see Patent Document 1). .

WO2015/136594

However, according to the technique described in Patent Document 1, since it is configured to extract a region having the same frequency component as the lamp of the traffic light, when a high-brightness object moves on the pixel, an object other than the traffic light may be misidentified as a traffic light.

For example, when a high-brightness object such as a tail lamp passes over a specific pixel, the brightness value of the pixel gradually increases as the high-brightness object moves, and after reaching the maximum brightness value , the luminance value of the pixel gradually decreases, so that the luminance change in the pixel contains frequency components. For this reason, luminance changes accompanying the movement of a high-luminance object may also include the same frequency components as the lamp of a traffic light, and there is a risk of erroneously detecting an object that is not a traffic light as a traffic light.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and its object is to erroneously detect a high-intensity object such as the tail lamp of another vehicle that moves relatively to the own vehicle as a traffic signal. To provide a traffic signal detection method and a traffic signal detection device capable of preventing a situation in which a traffic signal is lost and improving signal recognition performance.

In order to solve the above-described problems, a traffic light detection method and a traffic light detection device according to an aspect of the present invention are based on a first image acquired by an imaging unit mounted on the vehicle, and the vehicle moves relative to the vehicle. A second image is created in which temporal change in luminance value due to an object is suppressed, and among the pixels of the second image, a pixel whose magnitude of a predetermined frequency component included in the temporal change in luminance value exceeds a predetermined threshold value is detected as a traffic signal. A traffic light is detected by extracting it as a pictured traffic light pixel.

According to the present invention, when one's own vehicle approaches an intersection, it is possible to prevent a situation in which a high-intensity object such as the tail lamp of another vehicle that moves relatively to one's own vehicle is erroneously detected as a traffic signal. Recognition performance can be improved.

FIG. 1 is a block diagram showing the configuration of a traffic light detection device according to one embodiment of the present invention. FIG. 2 is a block diagram showing the configuration of the frequency component calculator in the traffic light detection device according to the embodiment of the present invention. FIG. 3 is a flow chart showing the processing procedure of the traffic light detection device according to one embodiment of the present invention. FIG. 4 is a diagram showing how the luminance value of the lamp of the traffic signal changes over time. FIG. 5 is a diagram showing how the luminance value changes in a pixel through which a high-luminance object passes.

Embodiments of the present invention will now be described in detail with reference to the drawings. In the explanation, the same reference numerals are given to the same parts, and redundant explanations are omitted.

[Configuration of traffic light detection device]
FIG. 1 is a block diagram showing the configuration of a traffic light detection device according to this embodiment. As shown in FIG. 1 , the traffic light detection device according to this embodiment includes an imaging unit 20 and a controller 100 . The traffic light detection device is mounted on a vehicle (not shown).

The imaging unit 20 is a digital camera having a solid-state imaging device such as a CCD or CMOS, and acquires an image that can be processed as a first image. The imaging unit 20 has a wide-angle lens with a wide angle of view, and the imaging range (angle of view) of the camera includes the vehicle's traveling direction to the vehicle's left and right directions, and also includes road shoulders around the vehicle. . The imaging unit 20 repeatedly images the surroundings of the vehicle at predetermined time intervals to obtain a plurality of continuous first images (frames).

The sampling frequency (the reciprocal of the predetermined time interval at which imaging is performed) when the imaging unit 20 performs imaging is desirably higher than the frequency of the AC power supplied to the lamp of the traffic signal to be detected. More preferably, the sampling frequency is at least twice the frequency of the AC power supply.

The controller 100 (an example of a control unit or processing unit) is a general-purpose microcomputer including a CPU (Central Processing Unit), memory, and an input/output unit. The controller 100 is installed with a computer program (traffic light detection device program) for functioning as a traffic light detection device. By executing the computer program, the controller 100 functions as a plurality of information processing circuits (51, 53, 55, 57, 61, 63) included in the traffic light detection device.

Here, an example of realizing a plurality of information processing circuits (51, 53, 55, 57, 61, 63) included in the traffic light detection device by software is shown. However, it is also possible to configure the information processing circuits (51, 53, 55, 57, 61, 63) by preparing dedicated hardware for executing each information processing described below. Also, the plurality of information processing circuits (51, 53, 55, 57, 61, 63) may be configured by individual hardware. Furthermore, the information processing circuits (51, 53, 55, 57, 61, 63) may also be used as an electronic control unit (ECU) used for other controls related to the vehicle.

The controller 100 includes, as a plurality of information processing circuits (51, 53, 55, 57, 61, 63), a variation suppression unit 51, an area determination unit 53, a frequency component calculation unit 55, an area extraction unit 57, a color determination unit 61, An output unit 63 is provided. Furthermore, as shown in FIG. 2 , the frequency component calculator 55 includes a multiplier 81 , a low-pass filter 83 (LPF), a reference signal generator 85 and a phase determiner 87 .

The region determination unit 53 extracts pixels (target pixels) whose luminance value changes over time are to be suppressed based on the time-series luminance value changes of the pixels forming the first image. For example, the region determination unit 53 compares the luminance values between temporally successive frames, and determines pixels whose luminance values continuously increase or whose luminance values decrease continuously for a predetermined period of time as target pixels. Extract as Also, the area determination unit 53 may extract the target pixel based on pattern matching.

In addition, the region determination unit 53 compares the luminance values between the temporally preceding and succeeding frames, and for pixels whose luminance value continuously increases or whose luminance value continuously decreases for a predetermined time, In addition, a pixel having a difference in luminance with an adjacent pixel equal to or greater than a predetermined value may be extracted as a target pixel.

Note that the predetermined time used when extracting pixels whose luminance values continuously increase or whose luminance values decrease continuously as target pixels is a value equal to or greater than the reciprocal of the frequency of the AC power supplied to the traffic light. It is desirable to have

The variation suppressing unit 51 targets the target pixels extracted by the area determining unit 53 based on changes in the time-series luminance values of the pixels forming the first image, and detects an object moving relative to the vehicle. A process for suppressing the time change of the luminance value due to is performed. For example, the variation suppressing unit 51 obtains the difference in the luminance values of the target pixel between the temporally preceding and succeeding frames, and creates the second image using the absolute value of the difference as the new luminance value. Obtaining the difference in brightness value between frames corresponds to first-order temporal differentiation of the brightness value in the first image, that is, processing with a first-order differential filter. The variation suppressing unit 51 may create the second image by setting the luminance value of the target pixel to be a constant.

The variation suppressing unit 51 performs a process of suppressing a temporal change in the luminance value of the target pixel extracted by the area determining unit 53, and creates a second image. The luminance values of the pixels other than the target pixel in the second image are the same as the luminance values of the pixels in the first image. The variation suppression unit 51 sequentially processes the first images repeatedly acquired by the imaging unit 20 to sequentially create second images.

The frequency component calculation unit 55 calculates a predetermined frequency component included in the temporal change in luminance value based on the time-series change in the luminance value of each pixel forming the second image created by the variation suppressing unit 51 . .

For the calculation, the frequency component calculator 55 uses the reference signal generated by the reference signal generator 85 . The reference signal is a sine wave. The reference signal generation unit 85 generates a normal signal having the same frequency as the AC power supplied to the traffic light, a delayed signal having a predetermined delay phase with respect to the normal signal, a leading signal having a predetermined lead phase with respect to the normal signal, to generate three types of reference signals.

The multiplier 81 multiplies each of the three types of reference signals by a time-series luminance value to calculate three types of detection signals corresponding to each of the three types of reference signals. A low-pass filter 83 removes harmonic components from the detection signal and outputs an amplitude detection signal. The frequency component calculator 55 outputs the amplitude detection signal having the largest amplitude among the three types of amplitude detection signals.

The reason for using three types of reference signals is that the phase difference between the AC power actually supplied to the traffic signal and the reference signal is generally unknown, and the amplitude of the amplitude detection signal changes due to the phase difference. This is to deal with what may happen. The phase determination section 87 controls the reference signal generation section 85 so that the reference signal generation section 85 generates a normal signal in synchronization with the phase of the amplitude detection signal having the largest amplitude among the three types of amplitude detection signals. do.

In addition to the above, the frequency component calculator 55 may calculate a predetermined frequency component by performing a discrete time Fourier transform based on frame data of a finite time length. In this case, no reference signal is required.

Based on the output of the frequency component calculation unit 55, the area extraction unit 57 extracts pixels in which the magnitude (amplitude) of the component of a predetermined frequency exceeds a predetermined threshold as traffic signal pixels representing the traffic signal. In addition, the area extraction unit 57 may extract traffic light pixels and group adjacent traffic light pixels. A set of pixels obtained by grouping the number of which is equal to or less than a predetermined number of pixels may be extracted as an isolated point region. Alternatively, an isolated point area may be determined as a traffic signal candidate area.

The color determination unit 61 refers to the color information (for example, RGB values) of the pixels on the first image corresponding to the traffic light pixels to determine the signal color of the traffic light.

The output unit 63 notifies the driver of the signal color of the traffic signal determined by the color determination unit 61 . For example, the output unit 63 may notify the driver of the signal color of the traffic light as visual information via an image display unit (not shown). Further, the output unit 63 may notify the driver of the signal color of the traffic light as auditory information through a speaker (not shown).

[Processing procedure of traffic light detection device]
Next, a processing procedure for traffic signal detection by the traffic signal detection apparatus according to the present embodiment will be described with reference to the flowchart of FIG. The signal detection process shown in FIG. 3 is started when the ignition of the vehicle is turned on, and is repeatedly executed while the ignition is on.

In step S101, the imaging unit 20 repeatedly images the surroundings of the vehicle to obtain a plurality of consecutive first images (frames). The obtained first image is recorded in a memory (not shown) or the like. The imaging unit 20 performs multiple imagings during one cycle of the frequency of the AC power supplied to the lamp of the traffic light.

In step S<b>103 , the area determination unit 53 refers to the plurality of recorded first images to calculate the change in luminance value between frames that temporally precede and follow each pixel. Then, in step S105, in the pixel for which the change in luminance value is calculated, if the luminance value continuously increases or decreases continuously for a predetermined period of time (if YES in step S105), , the region determination unit 53 extracts the pixel as a target pixel, and in step S107, performs a process of suppressing a temporal change in the luminance value of the extracted target pixel. If NO in step S105, the process of suppressing the time change of the luminance value is not performed. Information as to whether or not pixels forming the first image have been extracted as target pixels is recorded in a memory or the like (not shown).

In step S109, the reference signal generation unit 85 included in the frequency component calculation unit 55 generates a normal signal having the same frequency as the AC power supplied to the traffic signal, a delay signal having a predetermined delay phase with respect to the normal signal, a normal signal and a leading signal having a predetermined leading phase with respect to .

In step S111, the multiplication unit 81 multiplies each of the three types of reference signals calculated in step S109 by the time-series luminance value, and obtains three types of detection signals corresponding to each of the three types of reference signals. Calculate Then, the low-pass filter 83 removes harmonic components from the detection signal and outputs an amplitude detection signal.

In step S113, the frequency component calculator 55 detects the amplitude detection signal having the largest amplitude among the three types of amplitude detection signals calculated in step S111. The frequency component calculator 55 outputs the detected amplitude detection signal.

In step S<b>115 , based on the output of the frequency component calculator 55 , the region extractor 57 extracts pixels whose magnitude of the component of the predetermined frequency exceeds a predetermined threshold value as traffic signal pixels representing the traffic signal. The region extracting unit 57 extracts traffic light pixels, groups adjacent traffic light pixels, and extracts a set of pixels obtained by grouping the number of which is equal to or less than a predetermined number of pixels as an isolated point region.

In step S117, the region extraction unit 57 performs image binarization based on the first image. Specifically, in the first image, pixels exhibiting a predetermined luminance value or more are extracted as high-luminance pixels. In step S119, the region extracting unit 57 extracts pixels included in the isolated point region, which are pixels extracted as high-brightness pixels, as traffic light pixels in which the traffic light is captured. As a result, the isolated point area can be determined as a signal candidate area.

In step S121, the region extracting unit 57 performs tracking processing for the isolated point region determined as the traffic signal candidate region. There are various methods of tracking processing, and for example, tracking processing is performed based on the normalized correlation method.

In step S123, the area extracting unit 57 determines whether or not the number of times tracked in step S121 is equal to or greater than a predetermined threshold for each signal candidate area. case), in step S131, color determination is performed for the traffic signal candidate area. If NO in step S123, color determination is not performed.

In step S131, the color determination unit 61 refers to the color information of the pixels on the first image corresponding to the traffic signal pixels to determine the signal color of the traffic signal. More specifically, the signal color of the traffic signal is determined for the traffic signal candidate area for which the number of times of tracking is determined to be equal to or greater than the predetermined threshold in step S123. For example, the hue is determined by referring to the RGB values as the color information of the pixel, and if it is determined as "red" or "blue" ("green"), it is regarded as a signal, and the color is recorded as the signal color.

In step S<b>133 , the output unit 63 notifies the driver of the signal color of the traffic signal determined by the color determination unit 61 . For example, the output unit 63 notifies the driver of the signal color of the traffic light as visual information via an image display unit (not shown). Alternatively, the output unit 63 notifies the driver of the signal color of the traffic light as auditory information through a speaker (not shown).

[Effects of Embodiment]
As described in detail above, the traffic light detection method and the traffic light detection device according to the present embodiment are based on the first image acquired by the imaging unit mounted on the vehicle. A second image is created in which temporal change in luminance value is suppressed, a component of a predetermined frequency included in the temporal change in luminance value is calculated for each pixel of the second image, and the magnitude (amplitude) of the component exceeds a predetermined threshold. A pixel is extracted as a traffic light pixel in which the traffic light appears.

As a result, it becomes difficult to observe from the second image the same frequency component as the temporal change in the luminance value of the lamp of the traffic light, which can occur in pixels in which a moving high-brightness object is captured. Therefore, it is possible to reduce the possibility of erroneously detecting a pixel in which a moving high-brightness object is captured as an area in which a traffic light exists. As a result, only the lamp portion of the traffic signal can be detected with high accuracy. In addition, the detected parts can be grouped and extracted, and the accuracy of the detection of the traffic signal and the recognition of the lighting state of the traffic signal can be improved.

In addition, in the traffic light detection method and the traffic light detection device according to the present embodiment, the first-order time differentiation of the luminance value of the pixel of the first image is used as a new luminance value for each pixel of the first image, so that the second image is It may be created. As a result, the same frequency component as the temporal change in the luminance value of the lamp of the traffic signal is less likely to be observed from the second image, and the possibility of erroneously detecting the area where the traffic signal exists can be reduced.

The reason why the first-order temporal differentiation of the luminance value makes it difficult to observe frequency components that can occur in pixels in which a moving high-luminance object appears will be described with reference to FIGS. 4 and 5. FIG.

FIG. 4 is a diagram showing how the luminance value of the lamp of the traffic signal changes over time. As shown in FIG. 4, in general, the lamp of the traffic signal changes like curves F1 and F2 depending on the voltage change of the AC power supply supplied to the lamp of the traffic signal. In FIG. 4, the curve F1 shows the time change of the luminance value of a lamp with high responsiveness such as an LED, and the curve F2 shows the time change of the luminance value of a lamp with low responsiveness such as an incandescent lamp. ing. Since the first-order time differentiation of the curves F1 and F2 also shows the same change as in FIG. 4, the first-order time differentiation does not suppress the temporal change in the luminance value of the lamp of the traffic signal.

On the other hand, FIG. 5 is a diagram showing how the luminance value changes in a pixel through which a high-luminance object passes. FIG. 5 shows a case where a high-brightness object BR passes through a region R1 that is reflected in a certain pixel. A state (state B) in which the bright object BR is completely captured in the pixel, and a state (state C) in which the state in which the high brightness object BR is captured in the pixel ends at time t3 are shown. The time change of the brightness value due to the movement of the high-brightness object does not continue steadily as shown in FIG. (from time t2 to time t3) and does not continue. Therefore, the first-order time differentiation suppresses frequency components that may occur in pixels in which a moving high-brightness object appears.

Further, in the traffic signal detection method and the traffic signal detection apparatus according to the present embodiment, a pixel whose luminance value continuously increases or whose luminance value continuously decreases for a predetermined time is extracted as a target pixel, and the target pixel is The second image may be created by suppressing the time change of the luminance value at .

As shown in FIG. 5, it is possible to extract pixels whose luminance values continuously increase or whose luminance values decrease continuously as target pixels. Therefore, it is possible to suppress temporal change in the luminance value of the pixel. As a result, the same frequency component as the time change of the luminance value of the lamp of the traffic signal is less likely to be observed from the second image, and the possibility of erroneously detecting the area where the traffic signal exists can be reduced.

Further, in the traffic signal detection method and the traffic signal detection apparatus according to the present embodiment, pixels whose luminance values continuously increase or whose luminance values continuously decrease for a predetermined period of time, and the luminance of adjacent pixels A pixel having a difference equal to or greater than a predetermined value may be extracted as a target pixel, and the second image may be generated by suppressing temporal change in luminance value of the target pixel.

In order to extract, as target pixels, pixels whose luminance difference from adjacent pixels is equal to or greater than a predetermined value, target pixels are extracted using information about edges that may exist around pixels in which a high-luminance object appears. Become. Therefore, it is possible to more accurately detect a pixel through which a high-brightness object passes, and to suppress temporal change in the brightness value of the pixel. As a result, the same frequency component as the time change of the luminance value of the lamp of the traffic signal is less likely to be observed from the second image, and the possibility of erroneously detecting the area where the traffic signal exists can be reduced.

Furthermore, in the traffic light detection method and the traffic light detection device according to the present embodiment, the second image may be created by setting the new luminance value of the pixel extracted as the target pixel to be a constant. Pixels through which a high-brightness object passes can be captured more accurately, and temporal changes in brightness values at the pixels can be removed.

Further, in the traffic signal detection method and the traffic signal detection device according to the present embodiment, the predetermined time may be longer than the reciprocal of the predetermined frequency. A pixel whose luminance value continuously increases or whose luminance value decreases continuously over a period of time longer than one cycle of the temporal change in luminance value occurring in the lamp of a traffic light is extracted as a target pixel. Therefore, it is possible to more accurately detect a pixel through which a high-brightness object passes, and to suppress temporal change in the brightness value of the pixel. As a result, the same frequency component as the time change of the luminance value of the lamp of the traffic signal is less likely to be observed from the second image, and the possibility of erroneously detecting the area where the traffic signal exists can be reduced.

Furthermore, in the traffic signal detection method and the traffic signal detection device according to the present embodiment, the predetermined frequency may be twice the frequency of the AC power supplied to the traffic signal. Since the frequency of the time change of the luminance value occurring in the traffic signal lamp is often twice the frequency of the AC power supply, the time change of the luminance value occurring in the traffic light lamp can be captured more accurately.

Further, in the traffic signal detection method and the traffic signal detection device according to the present embodiment, the signal color of the traffic signal may be determined based on the color information of the traffic signal pixels, and the signal color may be notified to the driver of the vehicle. Even in an environment where it is difficult for the driver to recognize the traffic signal, it is possible to notify the driver of the signal color of the traffic signal detected based on the time change in the brightness value of the lamp of the traffic signal. The lighting state can be recognized.

Each function illustrated in the above embodiments may be implemented by one or more processing circuits. Processing circuitry includes programmed processors, electrical circuits, etc., as well as devices such as application specific integrated circuits (ASICs) and circuit components arranged to perform the described functions. etc. are also included.

Although the contents of the present invention have been described above according to the embodiments, it is obvious to those skilled in the art that the present invention is not limited to these descriptions and that various modifications and improvements are possible. The discussion and drawings forming part of this disclosure should not be understood as limiting the invention. Various alternative embodiments, implementations and operational techniques will become apparent to those skilled in the art from this disclosure.

Of course, the present invention includes various embodiments and the like that are not described here. Therefore, the technical scope of the present invention is defined only by the matters specifying the invention according to the valid scope of claims based on the above description.

20 Imaging unit 51 Fluctuation suppression unit 53 Area determination unit 55 Frequency component calculation unit 57 Area extraction unit 61 Color determination unit 63 Output unit 81 Multiplication unit 83 Low-pass filter (LPF)
85 reference signal generation unit 87 phase determination unit 100 controller

Claims (10)

A traffic light detection method for detecting a traffic light based on a first image acquired by an imaging unit mounted on a vehicle,
Based on the first image, for each pixel of the first image, the first time derivative of the luminance value of the pixel of the first image is set as a new luminance value, thereby moving relative to the vehicle. creating a second image that suppresses temporal changes in luminance values due to objects ;
calculating a predetermined frequency component included in the time change of the luminance value for each pixel of the second image;
Extracting pixels in which the size of the component exceeds a predetermined threshold as traffic light pixels representing the traffic light.
A traffic light detection method characterized by: A traffic light detection method for detecting a traffic light based on a first image acquired by an imaging unit mounted on a vehicle,
Based on the first image, a pixel whose luminance value continuously increases or whose luminance value continuously decreases for a predetermined time is extracted as a target pixel,
creating a second image in which temporal change in luminance value of the target pixel is suppressed to suppress temporal change in luminance value of an object moving relative to the vehicle ;
calculating a predetermined frequency component included in the time change of the luminance value for each pixel of the second image;
Extracting pixels in which the size of the component exceeds a predetermined threshold as traffic light pixels representing the traffic light.
A traffic light detection method characterized by: A traffic light detection method for detecting a traffic light based on a first image acquired by an imaging unit mounted on a vehicle,
A pixel whose luminance value continuously increases or whose luminance value continuously decreases for a predetermined period of time based on the first image , and whose luminance difference from an adjacent pixel is equal to or greater than a predetermined value. is extracted as the target pixel,
creating a second image in which temporal change in luminance value of the target pixel is suppressed to suppress temporal change in luminance value of an object moving relative to the vehicle ;
calculating a predetermined frequency component included in the time change of the luminance value for each pixel of the second image;
Extracting pixels in which the size of the component exceeds a predetermined threshold as traffic light pixels representing the traffic light.
A traffic light detection method characterized by: The traffic signal detection method according to claim 2 or 3 ,
A traffic light detection method, wherein the second image is created by setting new luminance values of the pixels extracted as the target pixels as constants. The traffic signal detection method according to any one of claims 2 to 4 ,
The traffic signal detection method, wherein the predetermined time is equal to or longer than the reciprocal of the predetermined frequency. The traffic light detection method according to any one of claims 1 to 5 ,
The traffic signal detection method, wherein the predetermined frequency is twice the frequency of an AC power supply supplied to the traffic signal. The traffic light detection method according to any one of claims 1 to 6 ,
determining the signal color of the traffic light based on the color information of the traffic light pixels;
A traffic light detection method, wherein the signal color is notified to a driver of the vehicle. an imaging unit mounted on a vehicle;
a controller that processes the first image acquired by the imaging unit;
A traffic light detection device comprising:
The controller is
Based on the first image, for each pixel of the first image, the first time derivative of the luminance value of the pixel of the first image is set as a new luminance value, thereby moving relative to the vehicle. creating a second image that suppresses temporal changes in luminance values due to objects;
calculating a predetermined frequency component included in the time change of the luminance value for each pixel of the second image;
extracting a pixel whose magnitude of the component exceeds a predetermined threshold as a traffic light pixel representing a traffic light;
A traffic light detection device characterized by: an imaging unit mounted on a vehicle;
a controller that processes the first image acquired by the imaging unit;
A traffic light detection device comprising:
The controller is
Based on the first image, a pixel whose luminance value continuously increases or whose luminance value continuously decreases for a predetermined time is extracted as a target pixel,
creating a second image in which temporal change in luminance value of the target pixel is suppressed to suppress temporal change in luminance value of an object moving relative to the vehicle;
calculating a predetermined frequency component included in the time change of the luminance value for each pixel of the second image;
extracting a pixel whose magnitude of the component exceeds a predetermined threshold as a traffic light pixel representing a traffic light;
A traffic light detection device characterized by: an imaging unit mounted on a vehicle;
a controller that processes the first image acquired by the imaging unit;
A traffic light detection device comprising:
The controller is
A pixel whose luminance value continuously increases or whose luminance value continuously decreases for a predetermined period of time based on the first image, and whose luminance difference from an adjacent pixel is equal to or greater than a predetermined value. is extracted as the target pixel,
creating a second image in which temporal change in luminance value of the target pixel is suppressed to suppress temporal change in luminance value of an object moving relative to the vehicle;
calculating a predetermined frequency component included in the time change of the luminance value for each pixel of the second image;
extracting a pixel whose magnitude of the component exceeds a predetermined threshold as a traffic light pixel representing a traffic light;
A traffic light detection device characterized by:

Images