JP2023047644A - Road information collection device and road information collection system - Google Patents

Abstract

To provide a road information collection device capable of reducing the volume of image data transmitted from an own vehicle.SOLUTION: A road information collection device 30 acquires a plurality of image data from an imaging device 10 mounted on an own vehicle 1. The road information collection device 30 acquires positional information of the own vehicle 1. For the road information collection device 30, an exclusion condition as a condition on whether each of the plurality of image data acquired from is to be excluded from transmission to an accumulation server 40 has been set. The road information collection device 30 determines whether or not each image data satisfies the exclusion condition, and excludes the image data from the transmission to the accumulation server 40 when it is determined that the exclusion condition is satisfied. The road information collection device 30 transmits image data that has been determined not to satisfy the exclusion condition to the accumulation server 40 in association with the positional information.SELECTED DRAWING: Figure 1

Description

The present disclosure relates to a road information collection device and a road information collection system.

Falling objects left on the road, carcasses of animals, potholes on the road, cracks in the road surface, rubbing of road markings, etc. require removal of obstacles and maintenance. Conventionally, road information is collected through patrols by road administrators and reports from residents.

However, collecting information on countless roads by relying on patrols by road administrators and reports from residents lacks comprehensiveness. Therefore, the discovery of an abnormality occurring on the road is delayed.

On the other hand, in the conventional road information distribution system, in order to discover anomalies occurring on the road, equipment on the vehicle side images the surroundings of the vehicle with a camera. In addition, the vehicle-side device transmits the captured video information to the road information distribution server together with the positional information. The road information distribution server receives the video information transmitted from the vehicle side equipment. The road information distribution server compares the image information with the reference image information in the database. Then, when there is a difference between both images, the road information distribution server distributes the position information and the image information to other vehicle-side equipment (see, for example, Patent Literature 1).

JP 2006-318356 A

In the above-described conventional road information distribution system, the vehicle-side device transmits all captured image information to the road information distribution server. Therefore, the amount of image data to be transmitted becomes enormous.

The present disclosure has been made to solve the above problems, and aims to obtain a road information collection device and a road information collection system that can reduce the amount of image data transmitted from a vehicle. do.

The road information collection device according to the present disclosure includes an image data acquisition unit that acquires a plurality of image data from an imaging device mounted on the vehicle, a position information acquisition unit that acquires position information of the vehicle, and an image data acquisition unit. An exclusion condition is set as a condition for excluding each of a plurality of acquired image data from transmission to the storage server. is determined to be satisfied, the image data is excluded from transmission to the storage server; A road information transmission section is provided for transmission to the accumulation server. Further, the road information collection system according to the present disclosure includes an accumulation server that receives the road information collection device and the image data and position information transmitted from the road information collection device and accumulates the data. there is

According to the road information collection device and the road information collection system of the present disclosure, it is possible to reduce the amount of image data transmitted from the own vehicle.

1 is a block diagram showing a road information collection system according to Embodiment 1; FIG. FIG. 2 is a flowchart showing processing by the road information collection device of FIG. 1; FIG. 2 is a flowchart showing processing by the accumulation server of FIG. 1; FIG. 10 is a block diagram showing a road information collection system according to Embodiment 2; FIG. 5 is a flow chart showing processing by the road information collection device of FIG. 4; FIG. 2 is a configuration diagram showing a first example of a processing circuit that implements the road information collection device of Embodiments 1 and 2; FIG. 4 is a configuration diagram showing a second example of a processing circuit that implements the road information collection device of the first and second embodiments;

Hereinafter, embodiments will be described with reference to the drawings.

Embodiment 1.
FIG. 1 is a block diagram showing a road information collection system according to Embodiment 1. FIG.

In FIG. 1 , the road information collection system 100 includes an imaging device 10 , a position specifying device 20 , a road information collection device 30 and an accumulation server 40 . The photographing device 10 , the position specifying device 20 and the road information collecting device 30 are mounted on the own vehicle 1 .

The photographing device 10 photographs the surroundings of the own vehicle 1 . The photographing device 10 sends the photographed image data to the road information collecting device 30 . The “image data” may be still image data or moving image data.

In Embodiment 1, imaging device 10 includes front camera 11 and rear camera 12 . The front camera 11 photographs the front of the vehicle 1 . The rear camera 12 photographs the rear of the vehicle 1. - 特許庁Hereinafter, the image data obtained by photographing the front of the own vehicle 1 among the plurality of image data photographed by the photographing device 10 will be referred to as "front data". Further, among the plurality of image data captured by the image capturing device 10, the image data obtained by capturing the rear of the own vehicle 1 is referred to as "backward data".

The position specifying device 20 specifies the position of the own vehicle 1 . The position specifying device 20 has a GPS (Global Positioning System) sensor. The locating device 20 receives carrier waves output from GPS satellites. The carrier wave contains time information and trajectory information. The position specifying device 20 calculates the latitude and longitude of the own vehicle 1 using the time information and the trajectory information. Position information indicated by the calculated latitude and longitude is sent to the road information collection device 30 .

The sensor used in the position specifying device 20 may not be a GPS sensor as long as it can specify the position of the own vehicle 1 . The sensor used in the position specifying device 20 may be, for example, a TOA (Time Of Arrival) sensor. The TOA sensor measures the relative positions of the vehicle 1 and control equipment around the road.

The road information collection device 30 is a processing device such as SoC (System on a chip). The road information collection device 30 includes an image data acquisition section 31 , a position information acquisition section 32 , an exclusion determination section 33 and a road information transmission section 34 .

The image data acquisition section 31 acquires a plurality of image data from the imaging device 10 .

The position information acquisition unit 32 acquires position information of the own vehicle 1 from the position specifying device 20 .

The exclusion determination unit 33 determines whether to send each image data acquired by the image data acquisition unit 31 to the storage server 40 .

An exclusion condition is set in the exclusion determination unit 33 . The exclusion condition is a condition as to whether or not to exclude each of the plurality of image data acquired by the image data acquisition unit 31 from transmission to the storage server 40 .

The first exclusion condition of the first embodiment includes the presence of an approaching vehicle, which is another vehicle whose distance from host vehicle 1 is equal to or less than a set distance. In addition, the exclusion condition includes a second condition that each image data is data obtained by photographing at night. In addition, the exclusion conditions include a third condition that each image data is data obtained by photographing in a backlit state. In addition, the exclusion condition includes a fourth condition that the photographing device 10 is dirty. The exclusion conditions include a fifth condition that there is no abnormality on the road.

The exclusion determination unit 33 determines whether or not each image data satisfies the first to fifth conditions. Then, if any one of the first to fifth conditions is satisfied, the exclusion determining unit 33 determines that the exclusion condition is satisfied and transmits the image data to the storage server 40 . Exclude from Image data excluded by the exclusion determination unit 33 in this manner is not transmitted to the storage server 40 .

The exclusion determination unit 33 determines whether or not the image data satisfies each of the first to fifth conditions described above. Each judgment of abnormality judgment on the road is performed.

The approaching vehicle determination is a process of determining whether or not there is an approaching vehicle. The exclusion determining unit 33 determines that the image data determined to include an approaching vehicle satisfies the exclusion condition. The exclusion determination unit 33 then excludes image data satisfying the exclusion condition from transmission to the storage server 40 .

When another vehicle is approaching the own vehicle 1, the image data of the image taken in that direction shows the body of the other vehicle enlarged. That is, the image data is image data in which unnecessary information occupies the majority. Such image data is not suitable for sending to the storage server 40 for image recognition processing. Therefore, if there is an approaching vehicle, the exclusion determination unit 33 excludes image data captured in that direction from transmission to the accumulation server 40 .

Specifically, when there is an approaching vehicle in front of the own vehicle 1, the exclusion determination unit 33 excludes forward data from transmission. In addition, when there is an approaching vehicle behind the own vehicle 1, the exclusion determination unit 33 excludes rear data from transmission. If there are approaching vehicles both in front of and behind the own vehicle 1 , neither the front data nor the rear data are transmitted to the storage server 40 .

Note that the exclusion determination unit 33 may measure the distance from the own vehicle 1 to the other vehicle using a range sensor (not shown). Alternatively, the exclusion determination unit 33 may obtain the distance from the host vehicle 1 to the other vehicle based on the image data captured by the imaging device 10 .

Nighttime determination is a process of determining whether it is nighttime or not. As a result of the nighttime determination, the exclusion determination unit 33 determines whether or not each image data is data obtained by photographing at nighttime. On the other hand, the exclusion determination unit 33 determines whether or not each image data is front data obtained by photographing the front of the vehicle 1 . The exclusion determination unit 33 determines that the image data obtained by photographing at night and other than in front of the vehicle 1 satisfies the exclusion condition.

Image data taken at night is dark and unclear due to insufficient light. Therefore, even if the own vehicle 1 transmits the image data to the storage server 40, it is difficult for the storage server 40 to identify the object appearing in the image data. On the other hand, forward data is less clear at night because the headlights are on. Therefore, the exclusion determination unit 33 excludes the image data other than the front data, which is the rear data in the first embodiment, from being transmitted to the storage server 40 as unclear image data.

The exclusion determination unit 33 may determine whether it is nighttime using an illuminance sensor (not shown). Alternatively, the exclusion determination unit 33 may determine whether it is night time based on the image data obtained by the imaging device 10 . In addition to these, the exclusion determination unit 33 may determine whether it is night time based on the time information. Further, the exclusion determination unit 33 may determine whether it is night time based on whether the headlight lighting switch has been operated.

Backlight determination is a process of determining whether each image data is data obtained by shooting in a backlight state. In the case of backlighting, the entire image tends to become white, resulting in overexposure. Therefore, even if the own vehicle 1 transmits the image data to the storage server 40, it is difficult for the storage server 40 to identify the object appearing in the image data.

The exclusion determination unit 33 determines whether or not each image data is data obtained by shooting in a backlit state. Then, the exclusion determination unit 33 determines that image data determined to be data obtained by photographing in a backlight state satisfies the exclusion condition. The exclusion determination unit 33 excludes image data satisfying the exclusion condition from transmission to the storage server 40 .

Note that the exclusion determination unit 33 may determine whether or not the subject is backlit using an illuminance sensor (not shown). Alternatively, the exclusion determination unit 33 may determine whether the subject is in a backlight state based on the image data obtained by the photographing device 10 .

Dirt adhesion determination is processing for determining whether or not there is dirt adhering to the imaging device 10 . The exclusion determination unit 33 determines the presence or absence of dirt in each image data. Then, the exclusion determination unit 33 determines that the image data determined to include dirt satisfies the exclusion condition, and excludes the image data from transmission to the storage server 40 .

The part where the dirt appears is a difficult-to-identify area. Therefore, the exclusion determination unit 33 excludes image data with dirt of a certain size or larger from transmission to the storage server 40 .

If the same object continues to be photographed at the same position, there is a high possibility that this object is dirt adhering to the photographing device 10 . In addition, dirt adhering to the photographing device 10 is often photographed in an out-of-focus, blurry state. Therefore, the exclusion determination unit 33 determines that the object is dirty when the unclear object is continuously photographed at the same position. Image processing other than this may be used.

If it is determined that the image data of both the front data and the rear data are stained, neither image is transmitted to the storage server 40 .

Road abnormality determination is processing for determining the presence or absence of an abnormality on a road using image recognition processing. The exclusion determination unit 33 performs road abnormality determination for all image data that have passed the above-described approaching vehicle determination, nighttime determination, backlight determination, and dirt adhesion determination.

Incidentally, "abnormality on the road" means that the road surface is in an unfavorable state different from normal. Examples of abnormalities on the road include falling objects or dead animals left on the road, depressions in the road surface, cracks in the road surface, signs on the road surface that are rubbed and difficult to distinguish. There are some difficult situations. In addition to this, there are various states such as a state where the road is blocked due to submersion in water or falling rocks.

The exclusion determination unit 33 determines that the image data for which it is determined that there is no abnormality on the road satisfies the exclusion condition, and excludes the image data from transmission to the accumulation server 40 . The road information collection system 100 does not collect image data for which it is determined that there is no abnormality on the road and that there is no problem. Therefore, the image data for which it is determined that there is no abnormality on the road is not transmitted to the accumulation server 40 .

The exclusion determination unit 33 performs image processing using machine learning such as Hog (Histogram of Oriented Gradients) or deep learning as image recognition processing for determining abnormality on the road. Alternatively, the exclusion determination unit 33 may determine whether there is an abnormality on the road using rule-based image processing such as edge detection.

Note that the image recognition processing performed by the exclusion determination unit 33 is simpler than the image recognition processing performed by the storage server 40, which will be described later. That is, the detection accuracy of road anomalies in the exclusion determination unit 33 is lower than the accuracy of detection of road anomalies in the accumulation server 40 .

Also, when there is an abnormality on the road such as a crack on the road surface of the lane on which the vehicle 1 is traveling, the front camera 11 first captures an image of the crack. After the vehicle 1 has passed through the crack, the same crack is photographed by the rear camera 12. - 特許庁In this way, an abnormality on the road photographed by the front camera 11 is also photographed by the rear camera 12 after passing. Therefore, if an abnormality on the road is detected in only one of the forward data and the backward data, it is highly possible that it is an erroneous detection.

Therefore, the exclusion determination unit 33 in the first embodiment does not transmit image data to the storage server 40 even if an abnormality on the road is detected in the front data, unless an abnormality on the road is detected in the rear data. Further, even if an abnormality on the road is detected in the rear data, image data is not transmitted to the accumulation server 40 if no abnormality on the road is detected in the front data photographed before passing the abnormality on the road. .

That is, when it is determined that there is an abnormality on the road in both the front data and the rear data obtained by photographing the same portion of the road with the photographing device 10, the exclusion determination unit 33 transmits the data to the storage server 40. Treat as image data. On the other hand, if it is determined that only one of the front data and the rear data obtained by photographing the same part of the road with the photographing device 10 has an abnormality on the road, the exclusion determination unit 33 Both backward data are judged to satisfy the exclusion condition. In this case, neither forward data nor backward data is transmitted to storage server 40 .

The exclusion determination unit 33 determines whether the image data satisfies the exclusion condition by performing each of the determinations described above. Then, the exclusion determination unit 33 treats image data that does not satisfy the exclusion condition as image data to be transmitted, and transmits the image data to the road information transmission unit 34 .

The road information transmission unit 34 transmits the image data determined by the exclusion determination unit 33 as not satisfying the exclusion condition to the storage server 40 in association with the position information. The road information transmission unit 34 transmits image data and location information to the storage server 40 via a mobile phone network that communicates according to a wireless standard such as LTE or 5G, for example.

The image transmitted by the road information transmission unit 34 may be only the image frame determined to be transmitted, that is, still image data. Alternatively, the image data transmitted by the road information transmission unit 34 may be moving image data for a set number of seconds before and after the image frame determined to be transmitted. In addition, the file name of the image data to be sent is accompanied by supplementary information such as whether it is forward data or backward data, the shooting date and time, and whether it is still image data or moving image data. may

Next, the storage server 40 shown in FIG. 1 will be described. The accumulation server 40 is a computer that includes a road information reception section 41 , an image recognition processing section 42 , a position calculation section 43 and an accumulation section 44 .

The road information receiving unit 41 receives image data and position information transmitted from the road information collecting device 30 . The road information receiving section 41 then sends the received image data to the image recognition processing section 42 . The road information receiving section 41 also sends the received position information to the position calculating section 43 .

The image recognition processing section 42 acquires image data from the road information receiving section 41 . The image recognition processing unit 42 performs image recognition processing for determining an abnormality on the road on the acquired image data. The content of the image recognition processing performed here is the same as the abnormality determination on the road performed by the exclusion determination unit 33 described above. The image recognition processing unit 42 performs image processing using machine learning using abundant computer resources, such as HoG or deep learning, as image recognition processing for judging an abnormality on the road. The image recognition processing unit 42 may use rule-based image processing such as edge detection to determine whether there is an abnormality on the road.

When the image recognition processing unit 42 determines that there is an abnormality on the road as a result of the image recognition processing, the image recognition processing unit 42 sends the image data to the position calculation unit 43 .

The position calculator 43 calculates the absolute position of the point on the road where the abnormality occurs based on the image data and the position information. Hereinafter, the point on the road where the abnormality occurs will be referred to as an "abnormal point".

The position calculator 43 first calculates the relative position of the abnormal point with respect to the position of the own vehicle 1 . This relative position is calculated based on whether the image data is front data or rear data, where the road abnormality appears in the image data, and its size.

Then, the position calculator 43 adds the calculated relative position to the position information of the own vehicle 1 acquired from the road information receiver 41 . Accordingly, the position calculation unit 43 calculates the absolute position of the abnormal point based on the latitude and longitude.

Note that it may be difficult to calculate the relative position depending on the image quality of the image data. In this case, the position calculator 43 treats the position information of the own vehicle 1 as the absolute position of the abnormal point.

The accumulation unit 44 accumulates various data obtained so far. In Embodiment 1, the accumulation unit 44 accumulates at least the absolute positions calculated by the position calculation unit 43 . Note that the accumulation unit 44 may associate the image data and the position information of the own vehicle 1 with the absolute position and accumulate them. Further, the accumulation unit 44 may associate the absolute position with text data indicating the name or state of the object detected by the image recognition processing unit 42 and accumulate the text data. This text data is called a "recognition result".

FIG. 2 is a flow chart showing processing by the road information collection device 30 of FIG.

In step S<b>101 , the image data acquisition unit 31 acquires image data captured by the imaging device 10 . The image data acquisition unit 31 acquires front data captured by the front camera 11 and rear data captured by the rear camera 12 . Then, in step S<b>102 , the position information acquisition unit 32 acquires position information of the own vehicle 1 from the position specifying device 20 . Various data acquired in steps S101 and S102 are sent to the exclusion determination unit 33 .

In step S103, the exclusion determination unit 33 performs approaching vehicle determination. If the exclusion determining unit 33 determines that there is image data of an approaching vehicle, the process proceeds to step S121. The exclusion determination unit 33 then excludes the corresponding image data from transmission to the storage server 40 .

On the other hand, if the exclusion determination unit 33 determines in step S103 that there is no image data of an approaching vehicle, or if in step S121 the exclusion processing for excluding image data from transmission is performed, then step S104. is executed. In step S104, the exclusion determination unit 33 performs nighttime determination. When the exclusion determination unit 33 determines that the image data was captured at night, the process proceeds to step S122. Then, the exclusion determination unit 33 excludes images other than the front data, that is, the rear data from transmission to the storage server 40 .

If the exclusion determination unit 33 determines in step S104 that the image data was not captured at night, or after performing exclusion processing in step S122, the exclusion determination unit 33 performs backlight determination in step S105. If there is backlit image data, the exclusion determination unit 33 advances the process to step S123. The exclusion determination unit 33 then excludes the corresponding image data from transmission to the storage server 40 .

If there is no backlit image data in step S105, or after the exclusion process is performed in step S123, the exclusion determination unit 33 performs dirt adhesion determination in step S106. Exclusion determination unit 33 advances the process to step S<b>124 when dirt of a certain size or larger appears in the image data. The exclusion determination unit 33 then excludes the image data in question from transmission to the storage server 40 .

In step S106, the exclusion determination unit 33 selects the image data that has passed all of the above determinations in step S107 when dirt of a certain size or larger is not captured in the image data, or after the exclusion process is performed in step S124. determine if exists. If there is no image data that has passed each of the above determinations, the process proceeds to step S112.

If only one of the front data and the rear data remains in step S107, the exclusion determination unit 33 performs image recognition processing on the remaining image data in step S113. By this image recognition processing, the exclusion determination unit 33 determines whether or not there is an abnormality on the road. Then, when the exclusion determination unit 33 determines that there is an abnormality on the road in step S114, the process proceeds to step S111. When the exclusion determination unit 33 determines that there is no abnormality on the road, it excludes the image data in step S115. Then, the process proceeds to step S112.

If both the front data and the rear data remain in step S107, the exclusion determination unit 33 performs image recognition processing on both image data in step S108. If no abnormality on the road is detected in both the front data and the rear data in step S109, exclusion determination unit 33 removes both front data and rear data from transmission to storage server 40 in step S116. exclude.

Further, if in step S110 only one of the front data and the rear data is found to have an abnormality on the road, the exclusion determination unit 33 transfers both the front data and the rear data to the storage server in step S116. 40 from transmission. After that, the process proceeds to step S112.

Exclusion determination unit 33 sends either or both of the front data and rear data to road information transmission unit 34 in step S<b>110 when an abnormality on the road is detected in both front data and rear data.

In step S111, the road information transmission unit 34 associates the image data determined to have an abnormality on the road with the position information. Then, the road information transmission unit 34 transmits the image data and the position information to the accumulation server 40 in step S111. After that, the process proceeds to step S112.

In step S112, the road information collection device 30 determines whether the power is turned off. The road information collection device 30 terminates the flowchart of FIG. 2 when the power is turned off. Further, when the power is not turned off, the road information collection device 30 returns the processing to step S101.

FIG. 3 is a flow chart showing processing by the storage server 40 of FIG.

In step S201, the road information receiving unit 41 determines whether the image data and position information transmitted by the own vehicle 1 have been received. When receiving the image data and the position information, the road information receiving section 41 sends the received image data to the image recognition processing section 42 and sends the received position information to the position calculation section 43 . If the road information receiving unit 41 does not receive the image data and the position information, the process proceeds to step S207.

The image recognition processing unit 42 acquires image data from the road information receiving unit 41 in step S202. Then, the image recognition processing unit 42 performs image recognition processing on the acquired image data.

When the image recognition processing unit 42 determines in step S<b>203 that there is an abnormality on the road in the image data, the image recognition processing unit 42 sends the image data to the position calculation unit 43 . Also, the image recognition processing unit 42 sends the recognition result to the position calculation unit 43 .

On the other hand, when the image recognition processing unit 42 determines in step S203 that there is no abnormality on the road in the image data, the process proceeds to step S207.

In step S204, the position calculator 43 calculates the relative position of the abnormal point with respect to the own vehicle 1 based on the image data. Then, in step S205, the position calculation unit 43 calculates the absolute position of the abnormal point made up of the latitude and longitude based on the position information of the own vehicle 1 and the calculated relative position.

The accumulation unit 44 accumulates the absolute positions of the abnormal points calculated by the position calculation unit 43 in step S206. The storage unit 44 may store the image data, the position information of the own vehicle 1, the recognition result of the image recognition processing unit 42, etc. in association with each other. After step S206, the process proceeds to step S207.

The storage server 40 determines whether the power is turned off in step S207. Then, when the storage server 40 is powered off, the flow chart of FIG. 3 ends. If the storage server 40 does not turn off, the process returns to step S201.

In the exclusion determination unit 33 of the first embodiment, an exclusion condition is set to exclude a plurality of pieces of image data acquired by the image data acquisition unit 31 from being transmitted to the storage server 40 . Then, the exclusion determination unit 33 determines whether each image data satisfies the exclusion condition. When the exclusion determination unit 33 determines that each image data satisfies the exclusion condition, the exclusion determination unit 33 excludes the image data from transmission to the storage server 40 . Therefore, the amount of image data transmitted from the own vehicle 1 can be reduced. In addition, delay in transmission from own vehicle 1 to storage server 40 can be suppressed.

In addition, since the amount of image data transmitted from the own vehicle 1 is reduced, communication costs can be kept low. Also, the amount of processing in the storage server 40 is reduced. Therefore, the occurrence of processing delays in the storage server 40 can be suppressed. Also, inappropriate image data that satisfies the exclusion conditions is not transmitted from the own vehicle 1 . For this reason, the accumulation server 40 has less opportunity to analyze unclear images. Therefore, the accuracy of analysis by the storage server 40 can be improved.

In addition, the exclusion determination unit 33 determines whether or not there is an abnormality on the road in each image data by performing image processing on each image data. Then, the exclusion determination unit 33 determines that the image data for which it is determined that there is no abnormality on the road satisfies the exclusion condition. Therefore, the image data without any abnormality on the road will not be transmitted from the own vehicle 1 . Therefore, the amount of image data transmitted from the own vehicle 1 can be reduced.

Further, when it is determined that only one of the front data and the rear data obtained by photographing the same part of the road by the photographing device 10 has an abnormality on the road, the exclusion determination unit 33 excludes the front data and backward data are judged to satisfy the exclusion conditions. Therefore, image data that may have detected an abnormality on the road in error will not be transmitted from the own vehicle 1 . Therefore, the amount of image data transmitted from the own vehicle 1 can be reduced. In addition, appropriate image data will be transmitted from own vehicle 1 to storage server 40 . Therefore, the storage server 40 can accurately detect an abnormality on the road.

Further, the detection accuracy of road anomalies in the exclusion determination unit 33 is lower than the accuracy of road anomaly detection in the accumulation server 40 . Therefore, the load of image recognition processing on the own vehicle 1 can be reduced.

In addition, the exclusion determination unit 33 determines whether or not there is an approaching vehicle. Then, the exclusion determination unit 33 determines that the image data determined to include the approaching vehicle satisfies the exclusion condition. Therefore, image data inappropriate for image recognition processing is not transmitted from the own vehicle 1 . Therefore, the amount of image data transmitted from the own vehicle 1 can be reduced. In addition, appropriate image data will be transmitted from own vehicle 1 to storage server 40 . Therefore, the storage server 40 can accurately detect an abnormality on the road.

In addition, the exclusion determination unit 33 determines whether or not each image data is data obtained by photographing at night. The exclusion determination unit 33 also determines whether each image data is data obtained by photographing the front of the vehicle 1 . The exclusion determination unit 33 determines that the image data obtained by photographing at night and other than in front of the vehicle 1 satisfies the exclusion condition. Therefore, unclear image data other than forward data is not transmitted from the own vehicle 1 . Therefore, the amount of image data transmitted from the own vehicle 1 can be reduced. In addition, appropriate image data will be transmitted from own vehicle 1 to storage server 40 . Therefore, the storage server 40 can accurately detect an abnormality on the road.

In addition, the exclusion determination unit 33 determines whether each image data is data obtained by shooting in a backlit state. Then, the exclusion determination unit 33 determines that image data determined to be data obtained by photographing in a backlight state satisfies the exclusion condition. Therefore, unclear image data in the backlight state is not transmitted from the own vehicle 1 . Therefore, the amount of image data transmitted from the own vehicle 1 can be reduced. In addition, appropriate image data will be transmitted from own vehicle 1 to storage server 40 . Therefore, the storage server 40 can accurately detect an abnormality on the road.

In addition, the exclusion determination unit 33 determines whether or not there is dirt adhering to the photographing device 10 . Then, the exclusion determining unit 33 determines that the image data determined to include dirt of a certain size or more satisfies the exclusion condition. Therefore, image data that is partially blurred is not transmitted from the own vehicle 1 . Therefore, the amount of image data transmitted from the own vehicle 1 can be reduced. In addition, appropriate image data will be transmitted from own vehicle 1 to storage server 40 . Therefore, the storage server 40 can accurately detect an abnormality on the road.

Embodiment 2.
FIG. 4 is a block diagram showing a road information collection system 100 according to Embodiment 2. As shown in FIG. The photographing device 10 of the second embodiment has a right camera 13 and a left camera 14 in addition to the front camera 11 and the rear camera 12 of the first embodiment.

The right camera 13 captures the right side of the vehicle 1 with respect to the traveling direction. The left camera 14 photographs the left side with respect to the traveling direction of the own vehicle 1 . In this manner, the photographing device 10 of Embodiment 2 is capable of photographing the right side and the left side of the own vehicle 1 with respect to the traveling direction.

Further, the image data acquisition unit 31 of the second embodiment acquires each image data captured by the front camera 11, the rear camera 12, the right camera 13, and the left camera . Here, among the plurality of image data captured by the image capturing device 10, the image data obtained by capturing the right side of the own vehicle 1 is referred to as "right side data". Further, image data obtained by photographing the left side of the own vehicle 1 among the plurality of image data photographed by the photographing device 10 is referred to as "left side data".

The exclusion determination unit 33 of the second embodiment performs determination processing similar to that of the first embodiment on the right side data and the left side data in addition to the front data and the rear data.

Configurations other than the above are the same as those of the first embodiment.

FIG. 5 is a flow chart showing processing by the road information collection device 30 of FIG. The image data acquisition unit 31 acquires image data captured by the imaging device 10 in step S201. The image data acquisition unit 31 acquires front data and rear data from the imaging device 10 . The image data acquisition unit 31 also acquires right side data and left side data from the imaging device 10 .

Subsequent steps S202 to S206 and steps S221 to S224 are the same as steps S102 to S106 and steps S121 to S124 shown in FIG.

In step S207, the exclusion determination unit 33 determines whether or not there is image data that has passed all of the approaching vehicle determination, nighttime determination, backlight determination, and dirt adhesion determination. If no image data remains, the process proceeds to step S211.

On the other hand, if any of the front data, the rear data, the right data, and the left data remains without being excluded in step S207, the exclusion determination unit 33 performs image recognition on the remaining image data in step S208. process. The image recognition processing here is also the processing of determining whether or not there is an abnormality on the road, as in the first embodiment.

When the exclusion determination unit 33 determines in step S209 that there is an abnormality on the road as a result of the image recognition processing, the exclusion determination unit 33 sends image data determined to have an abnormality to the road information transmission unit 34 . On the other hand, when the exclusion determination unit 33 determines that there is no abnormality on the road, the image data is excluded from transmission to the accumulation server 40 in step S212. Then, the exclusion determination unit 33 sends the image data not excluded to the road information transmission unit 34 in step S212.

In step S210, the road information transmission unit 34 transmits the image data determined to have an abnormality on the road to the storage server 40 in association with the position information. After that, the process proceeds to step S211.

In step S211, the road information collection device 30 determines whether the power is turned off. The road information collection device 30 terminates the flowchart of FIG. 5 when the power is turned off. Further, when the power is not turned off, the road information collection device 30 returns the processing to step S201.

The processing of the storage server 40 in the second embodiment is the same as the processing of the storage server 40 in the first embodiment shown in FIG.

The road information collection device 30 of the second embodiment acquires image data from the imaging device 10 capable of imaging the right side and the left side of the own vehicle 1 . Therefore, the road information collection device 30 can acquire image data in which the left and right lanes other than the lane in which the vehicle 1 is traveling are captured in a wide range. Therefore, image recognition processing is performed on image data in which the left and right lanes are captured in a wide range. Therefore, highly accurate recognition results can be obtained for the left and right lanes.

Step S107 in FIG. 2 shown in the first embodiment and post-processing after step S107, and step S207 in FIG. 5 shown in the second embodiment and post-processing after step S207 are They may replace each other.

In Embodiments 1 and 2, the exclusion determination unit 33 performs all of the approaching vehicle determination, nighttime determination, backlight determination, dirt adhesion determination, and road abnormality determination, but the present invention is not limited to this. The exclusion determination unit 33 may perform one or more of approaching vehicle determination, nighttime determination, backlight determination, dirt adhesion determination, and road abnormality determination.

Further, the photographing device 10 according to Embodiment 1 uses two cameras to photograph the front and rear of the own vehicle 1 . The photographing device 10 of Embodiment 2 uses four cameras to photograph the front, rear, right, and left sides of the own vehicle 1 . On the other hand, the photographing device 10 may photograph the front, rear, right, or left side of the vehicle 1 using a single camera such as a 360-degree camera.

Moreover, the road information collection device 30 may be provided in a state including either or both of the photographing device 10 and the position specifying device 20 .

Also, each function of the road information collection device 30 of the first and second embodiments is implemented by a processing circuit. FIG. 6 is a configuration diagram showing a first example of a processing circuit that implements each function of the road information collection device 30 of the first and second embodiments. The processing circuit 50 of the first example is dedicated hardware.

The processing circuit 50 corresponds to, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a combination thereof. . Also, each function of the road information collection device 30 may be implemented by individual processing circuits 50 . Alternatively, each function of the road information collection device 30 may be collectively implemented by the processing circuit 50 .

FIG. 7 is a diagram showing a second example of a processing circuit that implements each function of the road information collection device 30 of the first and second embodiments. A processing circuit 60 of the second example comprises a processor 61 and a memory 62 .

In the processing circuit 60, each function of the road information collection device 30 is implemented by software, firmware, or a combination of software and firmware. Software and firmware are written as programs. The software and firmware are then stored in memory 62 . The processor 61 reads out and executes programs stored in the memory 62 to achieve the functions of each unit.

It can be said that the program stored in the memory 62 causes the computer to execute the procedures or methods of the respective units described above. Here, the memory 62 is, for example, a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable Read Only Memory), an EEPROM (Electrically Erasable and Programmable Read Only Memory), or a non-volatile memory such as an EEPROM (Electrically Erasable and Programmable Read Only Memory). volatile or volatile semiconductor memory. A magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, a DVD, etc. also correspond to the memory 62 .

Some of the functions of the units described above may be implemented by dedicated hardware, and some may be implemented by software or firmware.

Thus, the processing circuit can implement the functions of the above-described units by hardware, software, firmware, or a combination thereof.

1 own vehicle, 10 imaging device, 30 road information collection device, 31 image data acquisition unit, 32 position information acquisition unit, 33 exclusion determination unit, 34 road information transmission unit, 40 accumulation server, 100 road information collection system.

Claims (10)

an image data acquisition unit that acquires a plurality of image data from an imaging device mounted on the own vehicle;
a location information acquisition unit that acquires location information of the own vehicle;
An exclusion condition is set as a condition for excluding each of the plurality of image data acquired by the image data acquisition unit from transmission to the storage server, and each of the image data satisfies the exclusion condition. an exclusion determination unit that determines whether the image data is included in the image data, and excludes the image data from transmission to the storage server if it is determined that the exclusion condition is satisfied;
A road information collecting device, comprising: a road information transmission unit that associates image data determined by the exclusion determination unit as not satisfying the exclusion condition with the position information and transmits the image data to the accumulation server. The exclusion conditions include that there is no abnormality on the road,
The exclusion determination unit performs image processing on each of the image data to determine whether or not there is an abnormality on the road in each of the image data. 2. The road information collection device according to claim 1, wherein the road information collection device determines that The photographing device is capable of photographing the front and rear of the own vehicle,
Among the plurality of image data, image data obtained by photographing the front of the vehicle by the photographing device is defined as front data, and among the plurality of image data, the photographing device photographs the rear of the vehicle. When the image data obtained by
When it is determined that only one of the front data and the rear data obtained by photographing the same part of the road by the photographing device has an abnormality on the road, the exclusion determination unit 3. The road information collecting device according to claim 2, wherein it is determined that both the forward data and the backward data satisfy the exclusion condition. The photographing device is capable of photographing the right side and the left side with respect to the traveling direction of the own vehicle,
Among the plurality of image data, the image data obtained by photographing the right side of the own vehicle by the photographing device is defined as right side data, and among the plurality of image data, the photographing device photographs the left side of the own vehicle. When the image data obtained by
4. The road information according to claim 2 or 3, wherein the exclusion determination unit determines whether or not there is an abnormality on the road in the right data and the left data by performing image processing on the right data and the left data. collector. 5. The road information collecting device according to any one of claims 2 to 4, wherein the detection accuracy of the road anomaly in the exclusion determination unit is lower than the detection accuracy of the road anomaly in the accumulation server. . The exclusion conditions include the presence of an approaching vehicle that is another vehicle whose distance from the own vehicle is equal to or less than a set distance,
The exclusion determination unit determines whether or not the approaching vehicle is present, and determines that the image data including the approaching vehicle satisfies the exclusion condition. The road information collection device according to any one of claims 1 to 3. The exclusion conditions include that each of the image data is data obtained by photographing at night,
The exclusion determination unit determines whether each image data is data obtained by photographing at night and whether each image data is data obtained by photographing the front of the own vehicle. and image data obtained by photographing at night and determined to be data obtained by photographing an area other than the area in front of the vehicle is determined to satisfy the exclusion condition. The road information collection device according to any one of items 6 to 6. The exclusion conditions include that each of the image data is data obtained by photographing in a backlit state,
The exclusion determination unit determines whether or not each of the image data is data obtained by photographing in a backlight state, and for image data determined to be data obtained by photographing in a backlight state, 8. The road information collecting device according to any one of claims 1 to 7, wherein it is determined that an exclusion condition is satisfied. The exclusion conditions include that the photographing device is dirty,
The exclusion determining unit determines whether or not the image data includes the stain, and determines that the image data determined to include the stain satisfies the exclusion condition. 9. The road information collection device according to any one of items 8 to 8. The road information collection device according to any one of claims 1 to 9, and the accumulation server that receives the image data and the position information transmitted from the road information collection device and accumulates the data. road information collection system.

Images