JP2008236062A - Imaging device, and obstacle detection apparatus and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To surely detect an obstacle by obtaining a high contrast images. <P>SOLUTION: An image processing part 30 detects a pedestrian on the basis of a thermal image generated by a far infrared light image sensor 20 in the state of a substrate temperature Ta. The image processing part 30 then detects whether or not the pedestrian is detected, ends pedestrian detection since there is no decline in contrast in the affirmative judgement, and detects the pedestrian on the basis of the thermal image of the substrate temperature Tb (Tb<Ta, for instance) in the negative judgement. The image processing part 30 further detects whether or not the pedestrian is detected, detects the pedestrian since there is no decline in the contrast in the affirmative judgement, and judges that the pedestrian is not within the visual field of a driver and ends the pedestrian detection in the negative judgement. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an image sensor, an obstacle detection apparatus, and a method.

  Of the annual traffic fatalities in Japan, about 20% of nighttime pedestrian fatalities account for. Many of these accidents are thought to have occurred because the driver was delayed in noticing the presence of a forward pedestrian. In view of this, a pedestrian recognition support system has been developed in which a far-infrared camera is used to detect a front crossing pedestrian or a pedestrian on the runway and notify the driver. As a pedestrian recognition support system, techniques for extracting obstacles such as pedestrians and vehicles using far-infrared images are disclosed (see Patent Documents 1 and 2).

  The pedestrian detection device of Patent Document 1 cuts out an area estimated to be a human from an image obtained by an imaging unit, calculates the area of the cut out human estimated area as a feature amount, and calculates the variation from time-series data of the feature amount. The calculated statistic is calculated, and when the calculated statistic is larger than the determination threshold, it is determined that the image corresponding to the human estimation region is a pedestrian.

The vehicle periphery monitoring device of Patent Literature 2 has an image processing unit including a CPU. As shown in FIG. 1 of Patent Document 1, the image processing unit is connected to two far-infrared cameras capable of detecting far-infrared rays, a yaw rate sensor, a vehicle speed sensor, and a brake sensor. A moving object such as a pedestrian or animal in front of the vehicle is detected from a signal indicating the running state of the vehicle, and an alarm is issued when it is determined that the possibility of a collision is high.
JP 2001-28050 A JP 2005-354597 A

  By the way, for example, when the temperature rises in summer and the temperature difference between the pedestrian and the background of the road surface or outer wall becomes small, or when the temperature difference between the background and the pedestrian becomes small due to rain, the emissivity and temperature are different. However, the contrast between the pedestrian and the background may be reduced or assimilated. Under such conditions, there is a problem that obstacles such as pedestrians cannot be detected even if far-infrared images are used as described in Patent Documents 1 and 2.

  The present invention has been proposed to solve the above-described problems, and an object thereof is to provide an imaging device, an obstacle detection device, and a method for reliably detecting an obstacle by obtaining a high-contrast image. .

  The image pickup device of the present invention includes an image pickup device substrate that generates an image corresponding to far infrared rays from a subject, and a temperature control unit that controls the image pickup device substrate to different substrate temperatures in terms of time. The image pickup device of the present invention includes an image pickup device substrate on which first and second light receiving element arrays that generate images corresponding to far infrared rays from a subject are mounted, and first and second image pickup devices on the image pickup device substrate. Temperature control means for controlling each region of the light receiving element array to a different substrate temperature.

  The obstacle detection apparatus according to the present invention provides an obstacle based on any one of the above-described imaging elements and images generated by the imaging element substrate in each state controlled to different substrate temperatures by the temperature control means. And an obstacle detection means for detecting.

  The image sensor substrate generates an image according to far infrared rays from the subject. At this time, even if the emissivity and the temperature of the subject and the background are different, the luminance of the image is almost the same, and they may be assimilated. Therefore, the temperature control device controls the substrate temperature of the image sensor substrate in two or more stages. Thereby, the brightness | luminance of each image can be changed and assimilation can be avoided. The obstacle detecting means detects the obstacle based on each image generated by the image pickup device substrate in each state controlled to the substrate temperature of two or more stages.

  Therefore, according to the above invention, even if the emissivity and temperature are different, even if the luminance of the image is almost the same, the luminance of each image can be changed by changing the substrate temperature. An object can be reliably detected.

  The present invention reliably detects an obstacle by obtaining a high-contrast image.

  Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings.

[Principle of the Invention]
All objects of absolute zero or higher emit far infrared rays according to Stefan Boltzmann's law. A black body with an emissivity of 1.0 emits radiant energy of the following equation (1), and other objects emit radiant energy multiplied by the emissivity of the object surface as shown in equation (2).
E = σT ⁴ [W / m ² ] (1)
E = εσT ⁴ [W / m ² ] (2)
In addition,
σ: Stefan Boltzmann constant 5.67e-8 [W / m ² K]
ε: Emissivity (0 to 1.0)
T: Temperature [K]
It is.

  The thermal far-infrared camera converts a minute temperature change due to absorption of radiant energy from the object into an electrical signal and visualizes it. Here, each object has a different emissivity depending on the material, color, surface condition, and the like.

  FIG. 1 is a diagram illustrating an example of a radiation rate. For example, the skin has a value of 0.98, asphalt 0.95, fiber 0.9, brick 0.8 and so on. Moreover, even if it is the same material, it becomes a different value by a color, the unevenness | corrugation of a surface, flatness, and an absorptance.

  Conventional far-infrared cameras, such as when the temperature of the road surface and outer wall in summer rises and the temperature difference with pedestrians becomes smaller, or when the road surface, outer wall, and pedestrians are all cooled by rain, the temperature difference becomes smaller, etc. There is a problem that the contrast between the pedestrian and the background is lowered, and the pedestrian cannot be recognized.

  In general, a far-infrared camera capable of measuring a non-contact temperature is controlled at a constant temperature of about 22 ° C. (295 K) by using a Peltier element as an image sensor. This is because the temperature change of the sensor light receiving part due to the far infrared rays from the object is very small, about 1 mK, and if the temperature of the substrate that becomes the escape route of the heat generated by the far infrared ray absorption is not stable, stable measurement, temperature information Is not output.

FIG. 2 is a diagram illustrating the relationship between the minute surface dS ₁ of the subject and the minute surface dS ₂ of the image sensor of the far-infrared camera when heat exchange is performed. In the far-infrared camera, heat exchange given by the following equation (3) is performed between the subject and the sensor, and a thermal image is generated by an electrical signal corresponding to this radiation exchange amount.

_{q 12 = ψ 12 ε 1 ε} 2 σ (T 1 4 -T 2 4) [W / m 2] ··· (3)
here,
q ₁₂ : Total radiation exchange amount [W / m ₂ ]
ψ ₁₂ : form factor of surface S ₂ viewed from minute surface dS ₁ ε ₁ : radiation rate 0 to 1.0 of object 1 (subject)
ε ₂ : The radiation rate of the object 2 (sensor) T ₁ : The temperature of the object 1 (subject) T ₂ : The temperature of the object 2 (sensor).

In a normal far-infrared camera, the substrate temperature T ₂ is kept constant, and in addition, since the emissivity of the sensor cannot be changed, ε ₂ is also fixed. As can be seen from the equation (3), the radiation exchange amount may be the same even for an object having a different emissivity and temperature as in the following equation (4).

_{q 12 = ψ 12 ε 1a ε} 2 σ (T 1a 4 -T 2 4)
= Ψ ₁₂ ε _1b ε ₂ σ (T _1b ⁴ -T ₂ ⁴ )
∴
ε _1a (T _1a ⁴ −T ₂ ⁴ ) = ε _1b (T _1b ⁴ −T ₂ ⁴ ) (4)
In addition,
ε _1a : emissivity of object A T _1a : temperature of object A ε _1b : emissivity of object B T _1b : temperature of object B In other words, objects having different emissivities and temperatures may have the same brightness on the thermal image. A specific example in the case where the sensor emissivity is 1.0 and the element substrate temperature is 295K is shown below.

  FIG. 3 is a diagram showing the temperature of an object having an emissivity of 0.98 and the same brightness as that of a human face having a temperature of 35 ° C. (308 K) and an emissivity different from the emissivity of 0.98.

  FIG. 4 is a diagram showing a thermal image with good contrast, and FIG. 5 is a diagram showing a thermal image with reduced contrast. As described above, the conventional far-infrared camera has a problem in that even objects having different emissivities and temperatures are assimilated to the same luminance and cannot be distinguished from each other.

  FIG. 6 shows a temperature at which the image pickup element is held at −3 ° C. (270 K), which has an emissivity of 0.98 and the same brightness as that of a human face at 35 ° C., and an emissivity different from the emissivity of 0.98. It is a figure which shows the temperature of the object which becomes. As shown in FIGS. 5 and 6, by changing the image sensor temperature, the temperature of an object with different emissivity, which has the same brightness as a human face with an emissivity of 0.98 and 35 ° C. (308 K), can be freely changed. be able to. That is, by changing the temperature of the image sensor, it is possible to change the conditions under which the objects A and B having different emissivities and temperatures have the same luminance.

  For example, even if the object is not identified because the contrast with the background is lowered (analyzed) at the image sensor temperature of 295K, the contrast can be improved and identified by setting the image sensor temperature to 270K.

  In this way, by using two or more far-infrared cameras that modulate the image sensor temperature or use two or more different image sensor temperatures, two objects that could not be identified due to a decrease in contrast even at different emissivities at different temperatures. Can be made identifiable.

  FIG. 7 shows various temperatures (0.8, 0.9, 0.95,... 1) when the substrate temperature is changed. It is a figure which shows the object temperature (anabolic temperature) which is a radiation rate of 0). As shown in the figure, as the substrate temperature is farther from the object to be detected (35 ° C.), the temperature at which the same luminance is obtained at different emissivities becomes farther from the object to be detected. The obstacle detection apparatus shown below utilizes the principle of the present invention.

[Specific Embodiment]
FIG. 8 is a diagram showing the configuration of the obstacle detection apparatus according to the embodiment of the present invention. The obstacle detection device includes a far-infrared lens 10 that collects far-infrared light, a far-infrared image sensor 20 that generates a thermal image corresponding to the temperature of the object to be imaged, and image processing based on the far-infrared image. And an image processing unit 30 for detecting an obstacle.

  The far-infrared image sensor 20 controls the temperature of the far-infrared image sensor substrate 21 in a state of being attached to the far-infrared image sensor substrate 21 and the far-infrared image sensor substrate 21 that generates an image according to light rays from the subject. And a temperature control device 22 for performing the operation.

  The temperature control device 22 can control the substrate temperature of the far-infrared image sensor substrate 21 in two or more stages, and is composed of, for example, a Peltier element. The temperature control device 22 is not limited to a Peltier element, and may be a Stirling cooler, a heater, or the like. In the present embodiment, the temperature control device 22 first controls the substrate temperature of the far-infrared image sensor substrate 21 to Ta, and controls the substrate temperature to Tb when no obstacle is detected.

  The far-infrared image sensor 20 further includes a vacuum-sealed package 23 that seals the periphery of the far-infrared image sensor substrate 21 in a vacuum state so as not to be affected by external temperature, and an opening of the vacuum-sealed package 23. A far-infrared transmitting window 24 that is provided and transmits far-infrared rays from the subject and enters the light-receiving surface of the far-infrared image sensor substrate 21.

  Instead of the far-infrared transmitting window 24, as shown in FIG. 9, a far-infrared transmitting window 24a that is a lens that collects far-infrared rays and enters the light-receiving surface of the far-infrared image sensor substrate 21 may be used. . Further, the following far infrared image sensor may be used.

  FIG. 10 is a diagram illustrating a configuration of an obstacle detection apparatus using the far-infrared image sensor 20a. The far-infrared image sensor 20a includes a far-infrared image sensor substrate 21a using wafer level sealing, and a temperature controller 22 that controls the temperature of the far-infrared image sensor substrate 21a. That is, the far-infrared image sensor 20a has a configuration without the vacuum sealed package 23 shown in FIG. 8 or FIG.

  The image processing unit 30 includes a CPU, a ROM, a RAM, and the like (not shown), and obstacles are generated based on an image (hereinafter referred to as a thermal image) generated by the far-infrared image sensor 20 at a predetermined substrate temperature. To detect.

  FIG. 11 is a flowchart showing an obstacle detection routine by the image processing unit 30. In this embodiment, a case where a pedestrian is detected as an obstacle will be described as an example.

  In step S1, the image processing unit 30 detects a pedestrian based on the thermal image generated by the far-infrared image sensor 20 in the state of the substrate temperature Ta as shown in FIG.

  For example, first, the image processing unit 30 performs a filtering process on the thermal image, and extracts a luminance range corresponding to a pedestrian in the thermal image. Next, the image processing unit 30 performs template matching using the pedestrian template indicating the size and shape of the pedestrian in the extracted luminance range, and detects the pedestrian. Note that template matching may not be performed accurately when the brightness difference between the pedestrian and the background is small and the contrast is low.

  In step S2, the image processing unit 30 detects whether or not a pedestrian has been detected. When the determination is affirmative, there is no decrease in contrast, so the pedestrian can be detected, pedestrian detection is terminated, and a negative determination is made. If so, go to Step S3.

  In step S3, as shown in FIG. 12, the image processing unit 30 detects a pedestrian based on a thermal image when the substrate temperature of the far-infrared image sensor 20 becomes Tb (for example, Tb <Ta). Proceed to step S4.

  In step S4, the image processing unit 30 detects whether or not a pedestrian has been detected. If the determination is affirmative, the image processing unit 30 ends the pedestrian detection because there is no decrease in contrast. The pedestrian detection is terminated when it is determined that it is not within the field of view.

  FIG. 13 is a diagram showing the relationship between the substrate temperature and the contrast reduction. According to the figure, when a pedestrian is detected regardless of whether the substrate temperature is Ta or Tb, no contrast reduction occurs. However, when a pedestrian is detected when the substrate temperature is Ta, but no pedestrian is detected when the substrate temperature is Tb, a decrease in contrast occurs at the substrate temperature Tb. Conversely, if a pedestrian is detected when the substrate temperature is Tb, but no pedestrian is detected when the substrate temperature is Ta, a decrease in contrast occurs at the substrate temperature Ta. If no pedestrian is detected at any of the substrate temperatures Ta and Tb, it is considered that there is no pedestrian as a detection target in the field of view (contrast reduction cannot be determined).

  Here, in the above example, the image processing unit 30 determines the presence or absence of the obstacle detection process at the substrate temperature Tb based on the result of the obstacle detection process at the substrate temperature Ta. Regardless of the result of the object detection process, the obstacle detection process at the substrate temperature Tb may be performed, and the substrate temperature may be selected and set based on the relationship of FIG.

  As described above, the obstacle detection device according to the embodiment of the present invention is configured so that when the image is taken at a predetermined substrate temperature, there is not much difference in luminance between the obstacle and the background, and the contrast is reduced. By changing the temperature, the brightness difference between the obstacle and the background is increased to prevent a decrease in contrast. As a result, the obstacle detection device can prevent an obstacle from being detected due to a decrease in contrast between the obstacle and the background, and can reliably detect the obstacle.

  That is, the obstacle detection apparatus can appropriately increase the contrast by changing the substrate temperature when there is a decrease in contrast due to the environment. For this reason, even if an object emits the same or very close radiation energy at different emissivities and different temperatures, the object can be reliably detected.

  In addition, the road surface and the outer wall are not necessarily displayed with uniform brightness due to the influence of the material, the degree of sunlight, unevenness, and the like. Similarly, pedestrians are not reflected in uniform brightness due to the influence of clothing, the number of layers, the specific heat transmitted from the body, the sun exposure, the wind direction, etc. As reflected. In this way, the background and pedestrians do not necessarily have uniform brightness, but rather often have variations between each other.

  For this reason, contrast reduction (assimilation) between the background and the pedestrian does not occur in the entire pedestrian, but often occurs locally, such as only the head, only the feet, or only the hands. FIG. 14 is a diagram illustrating an example in which the foot and the background are assimilated. FIG. 15 is a diagram illustrating an example in which the hand and the background are assimilated.

  Therefore, the obstacle detection device may cut out and synthesize only a portion having high local contrast using thermal images generated with different substrate temperatures. As a result, a clear thermal image with high contrast that does not depend on actual radiant energy in the entire field of view can be generated, and obstacles can be reliably detected. FIG. 16 is a diagram illustrating a thermal image generated by combining high contrast portions.

  Note that the present invention is not limited to the above-described embodiment, and it is needless to say that the present invention can also be applied to a design modified within the scope of the claims.

  For example, the substrate temperatures Ta and Tb can be set freely. The substrate temperature is not limited to being set at two levels of Ta and Tb, and may be set at three or more levels. In this case, the image processing unit 30 may detect an obstacle using a thermal image at a substrate temperature with the least contrast reduction among a plurality of substrate temperatures. Further, the substrate temperature is fixed at Ta and Tb, respectively, but may be set to two or more values having different center values such as a sine wave, a square wave, and a Δ wave.

  In the above-described embodiment, a pedestrian has been described as an example of an obstacle. However, a vehicle, a falling object, a puddle, a frozen road surface, or the like may be used.

  The obstacle detection device may set an optimum substrate temperature different from Ta and Tb after performing the obstacle detection processing based on the thermal images at the substrate temperatures Ta and Tb. For example, when the obstacle is not detected when the substrate temperature is Ta and the obstacle is detected when Tb (<Ta), the substrate temperature may be controlled to Tc (<Tb).

  In the embodiment described above, the substrate temperature is modulated at any time. However, it is not necessary to perform the modulation at any time, and may be performed according to usage conditions such as hourly, daily, seasonal, etc. according to conditions. If the object to be photographed and the photographing environment are stable and an object having the same luminance and conditions for obtaining the same luminance such as date and time are known in advance, photographing at the substrate temperatures Ta and Tb and determination of contrast reduction are performed. Alternatively, the image may be taken at a substrate temperature at which both previously known objects do not have the same brightness.

  Moreover, although the obstacle detection apparatus mentioned above detected the obstacle using the template matching method, you may detect an obstacle using the density | concentration histogram of a thermal image, and may use the method which combined them. Good. Furthermore, an obstacle having a specific size and aspect ratio may be detected based on the shape feature, or a pedestrian that is one of the obstacles may be detected from partial features such as the head and shoulders.

  In the above-described embodiment, the obstacle detection apparatus is controlled in two stages of the substrate temperatures Ta and Tb of the far-infrared image sensor 20. However, as shown below, two far-infrared image sensors 20 are provided and each far-infrared image sensor 20 is provided. The substrate temperature of the image sensor 20 may be set to Ta or Tb.

FIG. 17 is a diagram illustrating a configuration of an obstacle detection apparatus using two far-infrared image sensors 20. The image processing unit 30, a substrate temperature of one of the far-infrared image sensor 20 executes an obstacle detection process when the T _a, when the obstacle is not detected, the substrate temperature of the other far-infrared image sensor 20 There executes an obstacle detection process when the T _b.

  As a result, the obstacle detection apparatus can omit the modulation of the substrate temperature by using the two far-infrared image sensors 20, and can determine the decrease in contrast at high speed. In addition, since the obstacle detection device can obtain a stereo image after setting the substrate temperature, even if one of the far-infrared image sensors 20 breaks down, the thermal image generated by the other far-infrared image sensor 20 can be obtained. Can be used to detect obstacles. The substrate temperature may be modulated as follows.

  FIG. 18 is a diagram illustrating changes in the substrate temperature of the two far-infrared image sensors 20 provided in the obstacle detection apparatus. That is, each temperature control device 22 may control the substrate temperature of each far-infrared image sensor substrate 21 based on the triangular wave having the same phase.

  Further, a far-infrared image sensor 20b which is a two-array mounted chip in which two far-infrared image sensors 20 are configured as one may be used.

  FIG. 19A is a front view of a far-infrared image sensor 20b which is a two-array mounting chip, and FIG. 19B is a sectional view thereof. The far infrared image sensor 20 b includes a far infrared image sensor substrate 25, a first light receiving element array 26 and a second light receiving element array 27 provided on the far infrared image sensor substrate 25.

  The far-infrared image sensor substrate 25 is formed thick in the regions 25a and 25c where the first light-receiving element array 26 and the second light-receiving element array 27 are provided, and the first light-receiving element array 26 and the second light-receiving element array 26 are formed. The region 25b between the light receiving element arrays 27 is formed thin.

On the surface opposite to the light receiving element array formation surface, a temperature control device 22a is provided in the region 25a, and a temperature control device 22b is provided in the region 25c. The temperature control device 22a controls the substrate temperature of the first light receiving element array 26 to T _a ° C., and the temperature control device 22b controls the substrate temperature of the second light receiving element array 27 to T _b ° C. The obstacle detection device provided with the far-infrared image sensor 20b having such a configuration can reduce the size of the device as compared with that shown in FIG.

It is a figure which shows an example of a radiation rate. It is a diagram showing the relationship between the infinitesimal surface dS ₁ and the minute surface dS ₂ of the far-infrared image sensor of the camera of the object when the heat exchange is taking place. It is a figure which shows the temperature of the object which is the temperature reflected in the same brightness | luminance as a human face of emissivity 0.98 and 35 degreeC (308K), and is emissivity different from emissivity 0.98. It is a figure which shows a thermal image with favorable contrast. It is a figure which shows the thermal image in which the contrast fell. When the imaging device is held at −3 ° C. (270 K), the temperature of an object having an emissivity of 0.98 and a brightness that is reflected in the same brightness as that of a human face at 35 ° C. It is a figure which shows temperature. Radiation of various values (0.8, 0.9, 0.95, 1.0) that is reflected in the same brightness as skin with an emissivity of 0.98 and 35 ° C when the substrate temperature is changed It is a figure which shows the body temperature (anabolic temperature) which is a rate. It is a figure which shows the structure of the obstruction detection apparatus which concerns on embodiment of this invention. It is a figure which shows the structure of the obstruction detection apparatus provided with the other far-infrared image sensor. It is a figure which shows the structure of the obstruction detection apparatus using another far-infrared image sensor. It is a flowchart which shows the obstruction detection routine by an image process part. It is a figure which shows the change of board | substrate temperature. It is a figure which shows the relationship between a substrate temperature and a contrast fall. It is a figure which shows the example which the foot | leg and the background were assimilated. It is a figure which shows the example in which the hand and the background were assimilated. It is a figure which shows the thermal image produced | generated by synthesize | combining a part with high contrast. It is a figure which shows the structure of the obstruction detection apparatus using two far-infrared image sensors. It is a figure which shows the change of the board | substrate temperature of the two far-infrared image sensors 20 provided in the obstruction detection apparatus. (A) is a front view of the far-infrared image sensor which is a 2 array mounting chip | tip, (B) is the sectional drawing.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Far-infrared lens 20, 20a, 20b Far-infrared image sensor 21, 25 Far-infrared image sensor board | substrate 22, 22a, 22b Temperature control apparatus 23 Vacuum sealing package 24 Far-infrared transmission window 30 Image processing part

Claims (8)

An image sensor substrate for generating an image according to far infrared rays from the subject;
Temperature control means for controlling the image sensor substrate to a different substrate temperature in time,
An imaging device comprising: An image sensor substrate on which first and second light receiving element arrays for generating an image corresponding to far infrared rays from a subject are mounted;
Temperature control means for controlling each region of the first and second light receiving element arrays on the imaging element substrate to different substrate temperatures;
An imaging device comprising: The image sensor according to claim 1 or 2,
Obstacle detection means for detecting an obstacle based on each image generated by the imaging device substrate in each state controlled to a different substrate temperature by the temperature control means;
Obstacle detection device comprising: The obstacle detection unit is configured to generate an image generated by the imaging device substrate at a second substrate temperature when no obstacle is detected based on an image generated by the imaging device substrate at a first substrate temperature. The obstacle detection device according to claim 3, wherein an obstacle is detected based on the obstacle. A first image sensor substrate that generates an image according to far infrared rays from the subject in a state of the first substrate temperature;
A second image sensor substrate that generates an image according to far infrared rays from the subject in a state of the second substrate temperature;
Obstacle detection means for detecting an obstacle based on each image generated by the first and second imaging element substrates;
Obstacle detection device comprising: The obstacle detecting means detects an obstacle based on the image generated by the second image sensor substrate when no obstacle is detected based on the image generated by the first image sensor substrate. The obstacle detection device according to claim 5 to detect. Detecting an obstacle based on an image generated by the imaging device substrate at the first substrate temperature;
When no obstacle is detected, the substrate temperature of the image sensor substrate is controlled to the second substrate temperature,
An obstacle detection method for detecting an obstacle based on an image generated by the imaging device substrate at a second substrate temperature. Detecting an obstacle based on an image generated by the first imaging device substrate at the first substrate temperature;
An obstacle detection method for detecting an obstacle based on an image generated by a second image sensor substrate controlled to a second substrate temperature when no obstacle is detected.