imaging device

Abstract

The photographing apparatus of the embodiment includes a photographing section, a vehicle information acquisition section, and a photographing parameter update section. The photographing section is provided in a vehicle and photographs a photographing target based on a photographing parameter related to a photographing condition. The vehicle information acquisition section acquires driving environment information of the vehicle in driving. The photographing parameter update section updates the photographing parameter based on an image photographed by the photographing section and the driving environment information acquired by the vehicle information acquisition section.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to an imaging device. BACKGROUND

[0002] A camera is provided in front of a vehicle for the purpose of monitoring. There is an obstacle detection system that detects an obstacle that has a possibility of colliding with the vehicle based on an image acquired by the camera and notifies the driver. From the viewpoint of ensuring the safety of travel, the obstacle detection system is also important for automatic operation of a railway. Furthermore, the obstacle detection system requires maintaining a high obstacle detection accuracy in the entire travel section.

[0003] However, for a camera image taken in a scene such as a tunnel, a platform, or the like where the illuminance changes significantly, the quality (for example, contrast, etc.) is sometimes significantly reduced. For such a reduced-quality image, it is difficult to detect an obstacle using an image recognition technique.

[0004] PRIOR ART DOCUMENTS

[0005] PATENT DOCUMENTS

[0006] Patent Document 1: Japanese Patent Application Publication No. 2018-037061

[0007] Patent Document 2: Japanese Patent Application Publication No. 2018-002007

[0008] Patent Document 3: Japanese Patent Application Publication No. 2009-027651

[0009] Patent Document 4: Japanese Patent Application Publication No. 2019-221115

[0010] Patent Document 5: Japanese Patent Application Publication No. 2018-181254 SUMMARY

[0011] PROBLEMS TO BE SOLVED BY THE INVENTION

[0012] An object is to provide an imaging device that can suppress a reduction in the quality of an image taken due to changes in the environment.

[0013] MEANS FOR SOLVING THE PROBLEMS

[0014] The imaging device of the present embodiment includes an imaging unit, a vehicle information acquisition unit, and an imaging parameter update unit. The imaging unit is provided in a vehicle and images an imaging target based on an imaging parameter related to an imaging condition. The vehicle information acquisition unit acquires travel environment information of the vehicle in travel. The imaging parameter update unit updates the imaging parameter based on an image imaged by the imaging unit and the travel environment information acquired by the vehicle information acquisition unit. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 is a diagram showing the configuration of a monitoring system including the photographing device of the first embodiment.

[0016] Figure 2 is a diagram showing the functional configuration of each device included in the monitoring system of the first embodiment.

[0017] Figure 3 is a flowchart of the operation of the monitoring system of the first embodiment.

[0018] Figure 4 is a diagram showing an example of the adjustment region and the reference value of the image feature quantity referred to from the photographing parameter dictionary.

[0019] Figure 5 is a diagram showing an example of the selection of the adjustment region of a railway vehicle traveling in the vicinity of a tunnel.

[0020] Figure 6 is a diagram showing an example in which obstacle detection is performed in traveling in the vicinity of a tunnel entrance and in the vicinity of a tunnel exit.

[0021] Figure 7 is a diagram showing the functional configuration of each device included in the monitoring system of the second embodiment. DETAILED DESCRIPTION

[0022] Hereinafter, an embodiment of the present application will be described with reference to the drawings. The present embodiment is not limiting of the present application. The drawings are schematic or conceptual, and the proportions of the parts, etc. are not necessarily the same as those in reality. In the specification and the drawings, the same elements as those already appearing in the drawings are denoted by the same reference numerals and a detailed description will be appropriately omitted.

[0023] (First Embodiment)

[0024] The driver is required to promptly find a cause that can become an obstacle to travel when the railway vehicle is in operation. When the driver finds an obstacle such as a flying object, an animal, etc. that can become an obstacle in the vicinity of the line in the traveling direction of the vehicle, for example, the driver promptly performs emergency stop of the vehicle. In order to mechanically perform such prompt finding of an obstacle, a photographing device such as a camera is used to perform obstacle detection, for example. The first embodiment shows an example in which an obstacle in the vicinity of the line in front of the vehicle that becomes an obstacle in traveling of the railway vehicle is detected with higher accuracy.

[0025] First, the configuration of the first embodiment will be described. Figure 1is a schematic view of a configuration of a monitoring system 100 including the photographing device 1 of the first embodiment. The monitoring system 100 is mounted on a railway vehicle RV that travels along a direction G on a pair of travel lines R configured. The monitoring system 100 is provided with the photographing device 1, a detection device 2, and an interface device 3. The photographing device 1 is provided, for example, near a driver's seat window, and can photograph a landscape in the travel direction G (in front of the railway vehicle) of the railway vehicle RV.

[0026] Figure 2 is a schematic view of a functional configuration of each device included in the monitoring system 100 of the first embodiment. The photographing device 1 is provided with a vehicle information acquisition section 1A, a photographing section IB, an image acquisition section 1C, a photographing parameter update section ID, and an image output section IE. The detection device 2 is provided with a detection section 2A and a detection result output control section 2B. The interface device 3 is provided with a detection result output section 3A.

[0027] The vehicle information acquisition section 1A included in the photographing device 1 has a function of acquiring information on the surroundings of the vehicle, and is provided with a weather information acquisition section 1A1, a running information acquisition section 1A2, a current position information acquisition section 1A3, and a headlamp information acquisition section 1A4. The weather information acquisition section 1A1 has a function of acquiring information on weather, such as the state of the weather, the speed, the travel direction, the line shape, the device information, and the position of the sun, at the time of travel or stop of the vehicle. The running information acquisition section 1A2 has a function of acquiring vehicle information, such as the current speed, the travel direction, the line shape, the device information, and the like, that can be obtained at the time of travel of the vehicle. The current position information acquisition section 1A3 has a function of acquiring current position information of the vehicle. The headlamp information acquisition section 1A4 has a function of acquiring information on the brightness, the irradiation angle, and the like, of a headlamp provided in front of the vehicle.

[0028] That is, the vehicle information acquisition section 1A acquires travel environment information of the railway vehicle RV (vehicle) in travel. The travel environment information includes, for example, at least one of the state of the weather, the speed, the travel direction, the line information, the device information, the position, and the state of the headlamp, of the surroundings of the railway vehicle RV, as described above. The vehicle information acquisition section 1A can acquire the travel environment information from various sensors provided in the railway vehicle RV, for example, or can acquire the travel environment information by communicating with the outside of the railway vehicle RV. The vehicle information acquisition section 1A transmits the vehicle environment information to the photographing parameter update section ID.

[0029] The photographing section IB included in the photographing device 1 is a photographing device that photographs based on the photographing parameters set by the photographing parameter decision section ID3.

[0030] That is, the imaging section 1B is provided to the railway vehicle RV, and images an imaging target based on an imaging parameter relating to an imaging condition. The image taken is acquired by the image acquisition section 1C. The imaging condition includes, for example, at least one of a shutter speed, a gain, an aperture value, an HDR (High Dynamic Range) parameter, and a field angle of the imaging section 1B. The imaging target includes, for example, at least one of a travel path (travel line R) of the railway vehicle RV, a periphery of the travel path, and a sign provided to the periphery. The imaging target can be changed according to a monitoring target or a monitoring purpose.

[0031] The image acquisition section 1C of the imaging device 1 acquires the image taken by the imaging section 1B. That is, the image acquisition section 1C acquires the imaging image taken by the imaging section 1B based on the imaging parameter. The image acquisition section 1C transmits the imaging image to the imaging parameter update section 1D. The imaging section 1B transmits the imaging image to the image acquisition section 1C and the image output section 1E, and receives the imaging parameter from the imaging parameter update section 1D.

[0032] The imaging parameter update section 1D of the imaging device 1 has an imaging parameter dictionary section 1D1, an imaging parameter adjustment region decision section 1D2, and an imaging parameter decision section 1D3. The imaging parameter dictionary section 1D1 is a reference dictionary section for acquiring a reference value of an adjustment region and an image feature amount, which are used as a reference for adjusting the imaging parameter, and the like. At this time, the value input to the reference dictionary is referred to based on the information obtained by the vehicle information acquisition section 1A. The imaging parameter adjustment region decision section 1D2 sets the adjustment region referred to by the above-described imaging parameter dictionary section 1D1. The imaging parameter decision section 1D3 sets the value of each imaging parameter based on the set adjustment region and the image obtained by the image acquisition section 1C so as not to cause a highlight blowout or a dark portion loss, and the like. The imaging parameter shown here is a parameter (value) of the imaging condition of the shutter speed, the gain, the aperture value, the HDR parameter, the field angle, and the like. At this time, in order to decide the imaging parameter, the information obtained by the above-described vehicle information acquisition section 1A can also be referred to.

[0033] That is, the imaging parameter update section 1D updates the imaging parameter based on the imaging image acquired by the image acquisition section 1C (the imaging image taken by the imaging section 1B) and the travel environment information acquired by the vehicle information acquisition section 1A. The imaging section 1B images the imaging target based on the updated imaging parameter. The image acquisition section 1C acquires the imaging image of the imaging target taken by the imaging section 1B based on the updated imaging parameter. Thus, the imaging device 1 can image the imaging target while suppressing a decrease in the quality of the imaging image (for example, a highlight blowout or a dark portion loss, and the like) caused by a change in the environment. The imaging parameter update section 1D receives the travel environment information and the imaging image from the vehicle information acquisition section 1A and the image acquisition section 1C, respectively, and transmits the imaging parameter to the imaging section 1B.

[0034] Further, the following will be described with reference to Figures 3 to 5 Details of the update of the shooting parameter using the adjustment region will be described.

[0035] The image output section IE of the shooting device 1 receives the photographed object obtained by the image obtaining section IC and outputs it to the detection section 2A of the detection device 2 described later.

[0036] The detection device 2 is provided with the detection section 2A and the detection result output control section 2B. The detection section 2A of the detection device 2 obtains the photographed object from the image output section IE described above and performs obstacle detection.

[0037] More specifically, the detection section 2A (detection device 2) detects an abnormality of a photographed object based on the image. The abnormality of the photographed object is not limited to an obstacle on the travel route R, and can be, for example, an abnormality of an animal approaching the vicinity of the travel route R, an abnormality of a sign, and the like.

[0038] The detection result output control section 2B of the detection device 2 outputs the result of the obstacle detection to the detection result output section 3A of the interface device 3 described later.

[0039] The interface device 3 is provided with the detection result output section 3A. The detection result output section 3A of the interface device 3 obtains the result of the obstacle detection from the detection result output control section 2B described above and notifies it through each interface device. The object of the notification is the driver, the passenger, the railway control station, the ordinary person in the vicinity of the travel, the animal, and the like. The method of the notification is not particularly specified as long as it is a form that the object of the notification can understand. The form that can be understood is, for example, a form in which, in the case where the object of the notification is located in the vehicle, the notification is made through a description on a display, a bulletin board, or the like in the vehicle or an attached speaker, or in the case where the object of the notification is located outside the vehicle, the notification is made through a sound of a siren, an external speaker, or the like of the vehicle; and the like. That is, the detection result output section 3A as a display device displays the image or the detection result of the detection device 2.

[0040] Figure 3 An action flowchart of the monitoring system 100 of the first embodiment. Figure 3 For example, the flowchart shows an action of one update of the shooting parameter. Figure 3 The action shown is repeatedly performed, for example, in the travel of the railway vehicle RV.

[0041] First, in step S001, the vehicle information acquisition section 1A acquires weather information, running information, current position, and headlamp information for deciding the photographing parameter. The weather information is, for example, information on the weather at the time of traveling or stopping of the vehicle, such as sunny, cloudy, rain, snow, hail, wind direction, wind speed, position of the sun, and the like. The running information is, for example, information on the line shape including the traveling direction, line branching, facilities such as tunnels, station platforms, viaducts, and railway crossings, and the like. The current position information is, for example, the current position of the railway vehicle RV using a position detection device called GNSS (Global Navigation Satellite System) or TG (Tachogenerator). The current position is recorded, for example, with latitude, longitude, and an odometer. Here, the odometer indicates the traveling position of the vehicle in units of 100 m from the starting point such as a departure station, and the latitude and longitude indicate the measurement values that can be acquired from the position detection device. The headlamp information is information on the brightness, irradiation angle, and the like of the headlamp provided in front of the railway vehicle RV.

[0042] Next, in step S002, the image acquisition section 1C acquires the image photographed by the photographing section IB. The image photographed in step S002 is an image before the photographing parameter is adjusted. For example, when the time at which the image for which the detection process is performed is set to t, the image photographed at time t-1 is used. At this time, the image photographed by adjusting the photographing parameter at time t-1 can also be used.

[0043] Next, in step S003, the photographing parameter update section ID acquires the reference values of the adjustment region and the image feature amount from the photographing parameter dictionary section IDI based on the acquired information. In addition, the reference values of the image feature amount are sometimes referred to as feature reference values hereinafter.

[0044] Figure 4 is a diagram indicating an example of the reference values of the adjustment region and the image feature amount referred to from the photographing parameter dictionary. Figure 4 (a) of is relationship information indicating the correspondence relationship between the running environment information, the adjustment region, and the feature reference values stored in the photographing parameter dictionary section IDI. The correspondence relationship is stored in the photographing parameter dictionary section IDI in a plurality of (for example, n in the example shown in (a) of Figure 4 The current position is indicated, for example, with latitude Xn° (n = 1, 2,... ) and longitude Yn° (n = 1, 2,... ). The headlamp angle is indicated with Zm° (m = 1, 2,... ). Furthermore, in each mode of the running environment information, the adjustment region An (n = 1, 2,...) and the feature reference value CVn (n = 1, 2,...) are associated. Figure 4 (b) of indicates the photographed image and the adjustment region in the photographed image.

[0045] Examples of the adjustment region candidate include an arbitrary grid on the captured image divided in an arbitrary size and shape such as a quadrangle. Further, by selecting a plurality of grids, selection of an adjustment region that can be enlarged or reduced, and selection of an adjustment region not limited to an adjacent grid can be performed. Figure 4 (b) of FIG. 10 indicates the adjustment regions Al, A2. The adjustment regions are sometimes changed in accordance with the travel environment information.

[0046] Figure 5 is a diagram indicating an example of selection of an adjustment region A in a railway vehicle RV traveling in the vicinity of a tunnel T. Figure 5 (a) of FIG. 10 is a diagram in which the railway vehicle RV is viewed from above. Figure 5 (b) of FIG. 10 indicates a captured image captured from the railway vehicle RV.

[0047] In Figure 5 In (b) of FIG. 10, a set of grids in the vicinity of the travel route R is acquired as the adjustment region A from adjustment region candidates AC divided into 64 grids in the captured image C. That is, even if the route is a curve, the adjustment region A is set in accordance with the shape of the route. Further, the reference value of the image feature amount can also use, for example, the edge amount in the acquired image in the adjustment region as a reference. The reference value is, for example, an edge amount in the adjustment region judged to be sufficiently sharp by the driver or the administrator. In this case, a reference value suitable for the edge amount in the adjustment region is obtained in accordance with each weather, operation information, current position, and headlamp. Based on this reference value, the captured parameter is dynamically decided in the subsequent step.

[0048] Next, in step S004, the captured parameter updating section ID acquires a prescribed image feature in the adjustment region from the image acquired in step S002 using image processing or signal processing. The image feature refers to an image feature set in the captured parameter dictionary, and is, for example, an edge amount. Note that the image feature amount is not limited to the edge amount, and can be another feature amount.

[0049] Next, in step S005, the captured parameter updating section ID determines whether or not updating of the captured parameter is required. For example, if the acquired image feature amount is within the range of the reference value of the image feature amount referred to in the captured parameter dictionary, updating of the captured parameter is not required. The captured parameter updating section ID proceeds to step S009 and continues the processing in the case where it is determined that updating is not required. On the other hand, if the acquired image feature amount is within the range of the feature reference value, the captured parameter updating section ID determines that the captured parameter is updated, and proceeds to step S006.

[0050] Next, in step S006, the photographing parameter updating section ID updates the photographing parameter. By updating the photographing parameter in accordance with the driving scene, a captured image suitable for detection taking into account the high light overflow, the dark portion loss, and the like is obtained. Here, the photographing parameter refers to a parameter (value) of a photographing condition such as shutter speed, gain, aperture value, HDR parameter, angle of view, and the like. As a method of changing the photographing parameter, for example, the photographing parameter can be determined in accordance with a dictionary that summarizes certain rules based on the insight that the photographing parameter of the gain is raised by one level or the like. The dictionary can include a correspondence relationship between the driving environment information and the photographing parameter. Further, an objective function can be set, and the photographing parameter can be determined in such a manner that the objective function is minimized.

[0051] That is, the photographing parameter updating section ID updates the photographing parameter based on the adjustment region set from within the captured image. In more detail, the photographing parameter updating section ID updates the photographing parameter so as to maintain or improve the image quality of the adjustment region.

[0052] In addition, the following will be described with reference to Figure 6 The details of the update of the photographing parameter related to steps S003 to S006 will be described.

[0053] Next, in step S007, the photographing parameter updating section ID transmits the adjusted photographing parameter to the photographing section IB.

[0054] Next, in step S008, the photographing section IB performs photographing again using the received adjusted photographing parameter. The image obtaining section IC obtains the image photographed again by the photographing section IB.

[0055] Next, in step S009, the image obtaining section IC transmits the image obtained using the adjusted photographing parameter to the image output section IE. Alternatively, when it is determined in step S005 that the update of the photographing parameter is not needed, the image obtaining section IC transmits the image obtained before the photographing parameter is adjusted (the image photographed using the unadjusted photographing parameter) to the image output section IE.

[0056] Next, in step S010, the image output section IE transmits the received image to the detection device 2.

[0057] Next, in step S011, the detection device 2 performs a detection process of detecting the object based on the received image. At this time, the detection result can be displayed on the image. Further, a method of not explicitly showing the detection result on the image but describing it in another file can be employed. As to the detection method, for example, a known example such as Patent Literature 4, Patent Literature 5 can be referred to.

[0058] Finally, in step S012, the detection result output section 3A of the interface device 3 notifies the detection result obtained in the above step S011. Here, in the case where the notification method is display or depiction of the detection output image by an image- outputable device such as a display, if the display or depiction update frequency is not properly set, it sometimes causes hindrance to the vehicle operation of the driver. For example, in the case where the detection result image is depicted on the image obtained from the image output section IE and notified to the driver by the display, due to the change in the brightness of the image accompanying the drastic change in the traveling environment, flicker of the screen, it is possible to cause hindrance to the vehicle operation of the driver. In this case, it is also possible to make the display or depiction update frequency of the detection result image depicted on the device of the interface device 3 the same degree as the frequency of the detection by the detection device 2 or reduce it.

[0059] Next, referring to Figure 6 to the specific implementation scenario related to the dynamic update of the shooting parameters of steps S003 to S006 of Figure 3 will be described.

[0060] In step S003 of Figure 3 , a dictionary that outputs the adjustment region and the reference value of the image feature amount of the shooting parameter by inputting each environmental condition at the time of setting the shooting parameter (refer to (a) of Figure 4 ). This dictionary can be previously obtained from the installed vehicle. Further, this dictionary can also be previously collected by simulation on software.

[0061] An example of the specific configuration of the dictionary is shown in (a) of Figure 4 . That is, the shooting parameter update section ID obtains the weather information, the running information, the current position, and the headlamp information from the vehicle information obtaining section IA of the shooting device 1 as the key, and obtains the adjustment region and the reference value of the image feature amount as the value. An example of each adjustment region is shown in the drawing.

[0062] Figure 6 is a drawing that describes an example in which the obstacle detection is implemented in the traveling in the vicinity of the entrance of the tunnel T and in the vicinity of the exit of the tunnel T. Figure 6 (a) of Figure 6 (b) is substantially the same as (b) of Figure 4 .

[0063] First, the case where the shooting condition ST1 just before passing through the tunnel T entrance is described. In the case where the detection device 2 detects an obstacle just before passing through the tunnel T entrance, since the area where the obstacle is detected is on the route inside the tunnel, it is necessary to dynamically decide the adjustment area and the shooting parameters. It is assumed that the weather information obtained from the vehicle information obtaining section 1A is sunny, the running information is that the next running route shape is a straight line, the facility information is a tunnel entrance, the current position is latitude X1°, longitude Y1°, and the headlamp is off. In this case, the shooting parameter updating section ID refers to the adjustment area and the reference value of the image feature amount from the shooting parameter dictionary section ID1 shown in (a), and obtains the adjustment area Al shown on the shooting screen Cl of (b). Figure 4 Figure 6

[0064] Next, the shooting parameter updating section ID decides the shooting parameters based on the adjustment area Al and the image information obtained by the image obtaining section IC. First, the shooting parameter updating section ID focuses on the inside of the adjustment area Al and obtains the image feature amount by image processing or signal processing. Here, the image feature amount refers to, for example, the edge amount or the like. In the case where this image feature amount is within the range of the reference value of the image feature amount obtained from the shooting parameter dictionary section ID1, the shooting parameter updating section ID determines that there is no need to update the shooting parameters. On the other hand, in the case where the image feature amount obtained from the adjustment area Al is outside the range of the reference value of the image feature amount, the shooting parameter updating section ID determines that there is a need to update the shooting parameters. In the case where there is a need to update the shooting parameters, the shooting parameter updating section ID decides the shooting parameters, for example, according to a dictionary that summarizes certain rules based on the insight of making the shooting parameters such as the gain rise by one level or the like. Since the adjustment area is on the route inside the tunnel T, the shooting parameter updating section ID sets the shutter speed to 1 / 30 s, sets the gain to be high, sets the aperture value (F value) to be low, obtains the HDR parameters at EV value 2 levels, and sets the field angle to be high, for example, so that the adjustment area is displayed in the center of the image. The shooting parameter updating section ID applies the decided shooting parameters to the shooting section IB, and the image obtaining section IC obtains the image again. After that, the image output section IE outputs the image to the detection device 2, and the detection device 2 performs the detection processing with the image that is more suitable for obstacle detection.

[0065] ​​That is, the shooting parameter update unit 1D sets (determines) a predetermined area (adjustment area) that includes at least the subject being photographed, based on the driving environment information obtained by the vehicle information acquisition unit 1A. More specifically, the shooting parameter update unit 1D applies the driving environment information obtained by the vehicle information acquisition unit 1A to a first relationship that establishes a relationship between the pre-obtained driving environment information and the adjustment area in the pre-obtained shooting image, thereby setting the adjustment area in the shooting image captured by the shooting unit 1B. Furthermore, the "pre-obtained driving environment information" may include, for example, past driving environment information of a railway vehicle RV previously obtained by the vehicle information acquisition unit 1A, or driving environment information obtained through simulation. The "pre-obtained shooting image" may include, for example, past shooting images previously obtained by the image acquisition unit 1C (past shooting by the shooting unit 1B), or shooting images obtained through simulation. Simulation, for example, simulates vehicle driving in software. In practice, obtaining driving environment information and shooting images on-site is costly, so the above-described simulation is sometimes used. The "driving environment information obtained by the vehicle information acquisition unit 1A" refers to the current (during driving) driving environment information. The current driving environment information may sometimes be used as past driving environment information for updating shooting parameters in the future. "Image captured by camera unit 1B" refers to the currently captured image (while driving). This currently captured image may sometimes be used as a past captured image for updating shooting parameters in the future. Furthermore, the first relationship is part of the shooting parameter dictionary unit 1D1. The adjustment area of ​​the shooting parameter dictionary unit 1D1 can be selected, for example, by the driver or administrator from pre-obtained captured images, or by determining the subject from pre-obtained captured images through image recognition.

[0066] Furthermore, the shooting parameter update unit 1D obtains feature reference values ​​of image feature quantities in the adjustment area based on the driving environment information obtained by the vehicle information acquisition unit 1A. The feature reference values ​​are image feature quantities in the adjustment area within the pre-obtained image. More specifically, the shooting parameter update unit 1D obtains feature reference values ​​by applying the driving environment information obtained by the vehicle information acquisition unit 1A to a second relationship that establishes a relationship between the pre-obtained driving environment information and the feature reference values. The second relationship is part of the shooting parameter dictionary unit 1D1.

[0067] In addition, Figure 4 In the example shown in (a), the shooting parameter dictionary unit 1D1, which includes the first relation and the second relation, establishes a relationship between the past driving environment information of the railway vehicle RV previously acquired by the vehicle information acquisition unit 1A and the adjustment area and feature reference value in the past shooting images captured by the shooting unit 1B.

[0068] Furthermore, the shooting parameter update unit 1D updates the shooting parameters whenever an adjustment area is set. More specifically, the shooting parameter update unit 1D determines whether to update the shooting parameters whenever the adjustment area is updated. In addition, the shooting parameter update unit 1D updates the shooting parameters based on a comparison between image feature quantities in the adjustment area set within the captured image and feature reference values. More specifically, the shooting parameter update unit 1D updates the shooting parameters when the image feature quantities in the adjustment area set within the captured image are outside a predetermined range including the feature reference values. The shooting parameter update unit 1D does not update the shooting parameters when the image feature quantities in the adjustment area set within the captured image are within a predetermined range. Therefore, when the image feature quantities are within the predetermined range, since the captured image is very useful for obstacle detection, the shooting parameter update process can be omitted.

[0069] Next, we will explain using the shooting situation ST2, which is about to pass through the exit of tunnel T, as an example. Unlike the shooting situation ST1, which is about to pass through the entrance of tunnel T, since the area where the obstacle is detected is on the track outside tunnel T, the possibility of dynamically determining and adjusting the area and shooting parameters, similar to shooting situation ST1, is higher. Assume that the weather information obtained from the vehicle information acquisition unit 1A is sunny, the driving information is that the next driving route is straight, the equipment information is tunnel exit, the current position is latitude X2°, longitude Y2°, the headlights are on, and the headlight angle is Z1°. In this case, the shooting parameter update unit 1D... Figure 4 The shooting parameter dictionary section 1D1, as shown in (a), refers to the baseline values ​​of the adjustment area and image feature quantities to obtain the shooting parameters. Figure 6 The adjustment area A2 is displayed on the shooting screen C2 of (b).

[0070] Next, the shooting parameter update unit 1D determines the shooting parameters based on the adjustment area A2 and the image information acquired by the image acquisition unit 1C. Similar to shooting condition ST1, the shooting parameter update unit 1D obtains image feature values ​​from the image containing image feature values ​​within the adjustment area through image processing or signal processing. The shooting parameter update unit 1D compares the obtained image feature values ​​with reference values ​​of image feature values ​​in the shooting parameter dictionary unit 1D1 to determine whether an update of the shooting parameters is needed. Assuming an update is needed, since the adjustment area is on the track outside tunnel T, the shooting parameter update unit 1D sets, for example, the shutter speed to 1 / 30s, the gain to a low value, the aperture value (F-number) to a high value, obtains HDR parameters at EV value 2 stops, and sets the field of view to a high value, so that the adjustment area is displayed in the center of the image. Afterwards, the detection device 2 performs obstacle detection in the same manner as in shooting condition ST1, detecting obstacles with higher accuracy.

[0071] That is, the photographing parameter updating section 1D sets (updates) the adjustment region within the photographed image again every prescribed section on the travel path of the railway vehicle RV. The prescribed section is, for example, within a range of 500 m around the photographing conditions ST1, ST2 shown in Figure 6 But, it is not limited thereto, and the prescribed section can be further provided between the photographing conditions ST1, ST2, in Figure 6 Figure 6 Thus, the photographing parameter is repeatedly updated in the travel of the railway vehicle RV.

[0072] As described above, according to the first embodiment, the photographing section 1B photographs the photographing object based on the photographing parameter related to the photographing condition. Further, the photographing parameter updating section 1D updates the photographing parameter based on the photographed image photographed by the photographing section 1B and the travel environment information acquired by the vehicle information acquiring section 1A. Further, the photographing section 1B photographs the photographing object based on the updated photographing parameter. Thereby, it is possible to change the photographing parameter according to the environmental change such as the change in the illuminance near the tunnel. As a result, it is possible to suppress the decrease in the quality of the photographed image due to the environmental change, and it is possible to improve the image quality. Thus, for example, it is possible to acquire an image useful for the higher-precision detection of the obstacle that becomes an obstacle near the line ahead of the vehicle. In addition, according to the kind of the travel environment information, the environmental change that can be coped with sometimes changes. The kind of the travel environment information is not limited to the above, and in addition, the number of the dictionaries that can also be used as needed can be limited. Further, the environmental change is not limited to the tunnel, and for example, it can be the change in the illuminance due to the sunset or the like.

[0073] Further, the photographing parameter updating section 1D updates the photographing parameter based on the adjustment region decided from within the photographed image.

[0074] Suppose that in the case where the photographing parameter that sharply photographs the entire photographed image is set, it can not be possible to sharply photograph the photographing object that is the monitoring object.

[0075] To the contrary, in the first embodiment, it is possible to focus on the adjustment region within the photographed image, and adjust the photographing parameter. Thereby, it is possible to improve the image quality of the photographing object, and the detection device 2 can easily recognize the photographing object from within the photographed image. As a result, it is possible to improve the detection precision of the detection device 2.

[0076] ​In addition, the adjustment region can be dynamically changed according to each situation during travel. Furthermore, the period in which the adjustment region is changed can also be arbitrary. For example, in a case where the frame rate of the camera (imaging section IB) acquired by the imaging device 1 is 30 fps, the imaging parameter update section ID can also update the adjustment region each time 1 frame is acquired. On the other hand, the imaging parameter update section ID can also update the adjustment region every 1 second regardless of the frame rate of the camera. That is, the imaging parameter update section ID can also set the adjustment region within the captured image again every prescribed time. In addition, the prescribed time is not limited to the frame rate described above and 1 second. Furthermore, the imaging parameter update section ID can also update the adjustment region in combination every prescribed interval and every prescribed time.

[0077] Furthermore, the imaging section IB can also capture an imaging target of at least one of the advancing direction (traveling direction G) of the railway vehicle RV and the direction opposite to the advancing direction. In this case, the imaging device 1 can also have, for example, an imaging section IB that captures an imaging target disposed behind the railway vehicle RV. In addition, the imaging targets can be different in the front and the rear, and furthermore, the update method of the imaging parameters can be the same. Furthermore, the imaging section IB can also not capture the imaging target in the front, but can capture the imaging target in the rear.

[0078] Furthermore, the determination based on the image feature amount of the steps S004 and S005 illustrated in FIG. 4 can also not be performed. Figure 3 That is, the imaging parameter update section ID does not omit the update of the imaging parameters each time the adjustment region is set again. In this case, the relationship establishment of the feature reference values in the imaging parameter dictionary section IDI is not required. Thus, the man-hours of the production of the imaging parameter dictionary section IDI can be reduced.

[0079] (Second Embodiment)

[0080] Figure 7 A schematic diagram that represents the functional configuration of each device included in the monitoring system 100 of the second embodiment is illustrated in FIG. 6. The second embodiment is different from the first embodiment in that the detection by the detection device 2 is not performed at times.

[0081] The interface device 3 also has an image display section 3B.

[0082] The image display section 3B as the display device displays an image. In more detail, the image display section 3B displays the image output by the image output section IE without passing through the detection device 2. This is because, at times, it is required to more clearly capture the imaging target without performing the detection of the obstacle. Thus, the imaging device 1 can set the imaging parameters, for example, so that a person easily recognizes the imaging target from the captured image.

[0083] The image output section 1E outputs the captured image captured by the imaging section 1B based on the updated imaging parameters to at least one of the detection section 2A and the image display section 3B.

[0084] In addition, in the Figure 7 In the present embodiment, the detection device 2 and the detection result output section 3A are shown, but they can not be provided.

[0085] The other configurations of the imaging device 1 and the monitoring system 100 of the second embodiment are the same as those of the corresponding configurations of the imaging device 1 and the monitoring system 100 of the first embodiment, and thus the detailed description thereof is omitted.

[0086] The imaging device 1 of the second embodiment can achieve the same effects as those of the first embodiment.

[0087] At least a part of the imaging device 1 of the present embodiment can be configured by hardware or by software. In the case of being configured by software, a program that realizes the function of at least a part of the imaging device 1 can be stored in a recording medium such as a floppy disk, a CD-ROM, or the like, and read and executed by a computer. The recording medium is not limited to a detachable medium such as a disk or an optical disk, and can be a fixed-type recording medium such as a hard disk device or a memory. In addition, the program that realizes the function of at least a part of the imaging device 1 can be distributed via a communication line such as the Internet (including wireless communication). Furthermore, the program can be encrypted or modulated, and distributed in a compressed state via a wired line or a wireless line such as the Internet or stored in a recording medium.

[0088] The several embodiments of the present application have been described, but these embodiments are presented as examples, and are not intended to limit the scope of the application. These embodiments can be implemented in other various ways, and various omissions, substitutions, and changes can be made within the scope of the gist of the application. These embodiments and modifications thereof are included in the scope and gist of the application, and are also included in the application described in the technical scope and the range equivalent thereto.

Claims

1. An imaging device comprising: an imaging section configured to be provided to a vehicle and configured to image an imaging target based on an imaging parameter related to an imaging condition; a vehicle information acquisition section configured to acquire driving environment information of the vehicle while the vehicle is driving; and an imaging parameter update section configured to update the imaging parameter based on an imaged image imaged by the imaging section and the driving environment information acquired by the vehicle information acquisition section, wherein the imaging parameter update section is configured to: set a predetermined area including at least the imaging target from the imaged image based on the driving environment information acquired by the vehicle information acquisition section, acquire a reference value of an image feature amount in the predetermined area based on the driving environment information acquired by the vehicle information acquisition section, and update the imaging parameter based on a comparison between the image feature amount in the predetermined area set from the imaged image and the reference value.

2. The imaging device according to claim 1, wherein the imaging parameter update section sets the predetermined area from the imaged image imaged by the imaging section by applying the driving environment information acquired by the vehicle information acquisition section to a first relationship in which the driving environment information acquired in advance and the predetermined area in the imaged image acquired in advance are related to each other.

3. The imaging device according to claim 1, wherein the reference value is the image feature amount in the predetermined area in the imaged image acquired in advance.

4. The imaging device according to claim 1, wherein the imaging parameter update section acquires the reference value by applying the driving environment information acquired by the vehicle information acquisition section to a second relationship in which the driving environment information acquired in advance and the reference value are related to each other.

5. The imaging device according to claim 1, wherein the imaging parameter update section is configured to: update the imaging parameter when the image feature amount in the predetermined area set from the imaged image is outside a predetermined range including the reference value, and not update the imaging parameter when the image feature amount in the predetermined area set from the imaged image is within the predetermined range.

6. The imaging device according to claim 1, wherein the imaging parameter update section sets the predetermined area in the imaged image again at predetermined intervals on a driving path of the vehicle.

7. The imaging device according to claim 1, wherein the imaging parameter update section sets the predetermined area in the imaged image again at predetermined time intervals.

8. The imaging device according to claim 1, wherein the imaging parameter update section updates the imaging parameter each time the predetermined area is set. ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​

Images