JPH08219784A - Method for measuring visibility - Google Patents
Method for measuring visibilityInfo
- Publication number
- JPH08219784A JPH08219784A JP7030990A JP3099095A JPH08219784A JP H08219784 A JPH08219784 A JP H08219784A JP 7030990 A JP7030990 A JP 7030990A JP 3099095 A JP3099095 A JP 3099095A JP H08219784 A JPH08219784 A JP H08219784A
- Authority
- JP
- Japan
- Prior art keywords
- visibility
- standard deviation
- luminance
- video signal
- distance
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Abstract
Description
【０００１】[0001]
【産業上の利用分野】本発明は道路等に設置したテレビ
カメラの画像を処理して視程の定量計測を可能にした視
程計測方法に関するものである。BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a visibility measuring method capable of quantitatively measuring a visibility by processing an image of a television camera installed on a road or the like.
【０００２】[0002]
【従来の技術】人間が肉眼で物体を認識出来る距離すな
わち視程は空間に浮遊する霧や降雪等により大幅に変化
する。そのため、道路交通の安全確保の上で車両の速度
制限を行う場合等において、視程情報は重要な役割を果
たしている。従来、この視程を知るには以下のような方
法があった。第１の方法は、テレビ画像を人が目視する
ことにより、通常の視認画像に比べ見にくくなった程度
から定性的に視認距離が短くなったものと推定するもの
である。第２の方法は発光体と受光体を対向させて配置
し、霧、雪等で視程が短くなるにつれ受光レベルが低下
することに着目し、視程と受光の相関係数から定量的に
測定するものである。第３の方法はカメラから一定の距
離に配置した特定のパターンの標識をテレビカメラで撮
像し、標識のパターンのビデオ信号におけるコントラス
ト等が視程距離と相関があることを利用して定量計測す
るものである。2. Description of the Related Art The distance at which a human can recognize an object with the naked eye, that is, the visibility is drastically changed due to fog floating in space or snowfall. Therefore, the visibility information plays an important role when the speed of the vehicle is limited to ensure the safety of road traffic. Conventionally, the following methods have been used to know this visibility. The first method is to estimate that the viewing distance is qualitatively shortened from the extent that it becomes more difficult to see than a normal viewing image when a person views a television image. In the second method, the light emitter and the light receiver are arranged so as to face each other, paying attention to the fact that the light reception level decreases as the visibility becomes shorter due to fog, snow, etc., and quantitatively measured from the correlation coefficient between the visibility and light reception. It is a thing. The third method is that a sign of a specific pattern arranged at a certain distance from the camera is imaged by a television camera, and quantitative measurement is performed by utilizing the fact that the contrast in the video signal of the sign pattern is correlated with the visibility distance. Is.
【０００３】[0003]
【発明が解決しようとする課題】前述の従来の第２の方
法には、発光体、受光体を正確に対向させる必要から一
定の場所に固定する必要がある。単一光路であるため周
辺の状況と一致しない場合がある。また、単一光路、単
一波長であることから人の目の感覚と差がある等の欠点
があった。また、前述の従来の第３の方法には、テレビ
カメラと標識の距離を一定に保つことが必要であり、一
定の場所に双方を固定する必要があること、標識の照度
によって補正が必要であること等の欠点があった。本発
明の目的は、これらの欠点を除去し、従来から多数道路
に配置されているテレビカメラから得られるビデオ画像
を利用し、簡易、低コストの視程距離の定量計測方法を
提供することにある。In the above-mentioned second conventional method, it is necessary to fix the light-emitting body and the light-receiving body in a fixed place because they must be accurately opposed to each other. Since it is a single optical path, it may not match the surrounding conditions. In addition, there is a drawback in that there is a difference in the sense of the human eye due to the single optical path and the single wavelength. Further, in the above-mentioned third conventional method, it is necessary to keep the distance between the television camera and the sign constant, it is necessary to fix both at a fixed place, and it is necessary to correct by the illuminance of the sign. There were drawbacks such as being present. An object of the present invention is to eliminate these drawbacks and provide a simple and low-cost quantitative measurement method of a visibility distance by utilizing a video image obtained from a television camera conventionally arranged on many roads. .
【０００４】[0004]
【課題を解決するための手段】本発明は上記目的を達成
するため、道路等に設置したテレビカメラで道路面を含
む周辺の景色を予め種々の条件下で撮像し、それらのビ
デオ画像を解析して、各輝度階調の出現頻度分布を調
べ、輝度の標準偏差と視程の関係を求め、この関係を利
用して、ある時点における輝度の標準偏差から視程距離
を算出するようにしたものである。In order to achieve the above object, the present invention takes a picture of surrounding scenery including a road surface in advance under various conditions with a television camera installed on a road and analyzes the video images. Then, the appearance frequency distribution of each brightness gradation is investigated, the relationship between the standard deviation of brightness and the visibility is obtained, and this relationship is used to calculate the visibility distance from the standard deviation of the brightness at a certain time. is there.
【０００５】[0005]
【作用】その結果、同一場所においてはテレビ画像の輝
度の標準偏差を計算することにより直ちに視程を知るこ
とが出来る。As a result, in the same place, the visibility can be immediately known by calculating the standard deviation of the brightness of the television image.
【０００６】[0006]
【実施例】図１、図２はある地点における、それぞれ吹
雪時と晴天時のビデオカメラの瞬間画像を各画素につい
て２５６の輝度階調にわけた頻度分布を示す図である。
図１、図２は縦軸に頻度、横軸に輝度階調を示してい
る。図１の降雪時の場合、平均階調は１１４.３、標準
偏差は１６.６である。また、図２の晴天時の場合、平
均階調は１０５.３、標準偏差は３０.６である。なお、
上記輝度階調の標準偏差ｓは、対象画像の画素数＝ｎ、
各画素の輝度階調＝ｐi（ｉ＝１、２、３、…ｎ）、画
素の平均輝度階調＝ｐhとした場合、次式で表される。
（「品質管理入門」日科技連 １０３頁）DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIGS. 1 and 2 are diagrams showing frequency distributions obtained by dividing instantaneous images of a video camera at a certain point during snowstorm and fine weather into 256 luminance gradations for each pixel.
1 and 2, the vertical axis represents frequency and the horizontal axis represents luminance gradation. In the case of snowfall in FIG. 1, the average gradation is 114.3 and the standard deviation is 16.6. In the case of fine weather in FIG. 2, the average gradation is 105.3 and the standard deviation is 30.6. In addition,
The standard deviation s of the brightness gradation is the number of pixels of the target image = n,
Luminance gradation of each pixel = p i (i = 1,2,3, ... n), when the average luminance level = p h pixels, is expressed by the following equation.
("Introduction to Quality Control", page 103 of JST)
【０００７】[0007]
【数１】 [Equation 1]
【０００８】一方、従来知られている反射型視程計であ
る地点における図１（吹雪時）と図２（晴天時）の視程
を計測するとそれぞれ４０ｍ，２５０ｍであった。これ
らのデータから、降雪時と晴天時とでは輝度分布に顕著
な差があり、画像の輝度の標準偏差と視程とが相関があ
ることが確認できた。従って、上記実験の結果から、画
像の輝度の標準偏差を求めればそれに対応する視程を計
測することが可能である。図３は地点Ａにおける種々の
気象条件が含まれるビデオ画像（４８種類）について上
記と同様のデータを取得し、標準偏差と視程の関係を求
めたもので、双方の関係の程度を示す相関係数ｒは、ｒ
＝０.９２ である。相関係数ｒは、各測定における輝度
標準偏差＝ｘi 、各測定における視程を対数変換した値
＝ｙi 、標準偏差の平均値＝ｘh 、視程を対数変換した
値の平均値＝ｙh 、とした場合、次式で定義される。
（「統計的方法」日本規格協会 １９３頁）On the other hand, when the visibility of the conventionally known reflection type visibility meter at the point shown in FIG. 1 (during snowstorm) and FIG. 2 (during clear weather) was measured, it was 40 m and 250 m, respectively. From these data, it was confirmed that there is a significant difference in the luminance distribution between snowfall and fine weather, and there is a correlation between the standard deviation of the luminance of the image and the visibility. Therefore, if the standard deviation of the brightness of the image is obtained from the result of the above experiment, it is possible to measure the corresponding visibility. FIG. 3 shows the same data as above for video images (48 types) including various weather conditions at point A, and the relationship between standard deviation and visibility is obtained. The number r is r
= 0.92. The correlation coefficient r is the luminance standard deviation in each measurement = x i , the value obtained by logarithmically converting the visibility in each measurement = y i , the average value of standard deviation = x h , the average value in the logarithmic conversion of visibility = y h , Is defined by the following equation.
("Statistical Method" Japanese Standards Association, p. 193)
【０００９】[0009]
【数２】 [Equation 2]
【００１０】図４は他の地点Ｂで取得したデータを標準
偏差と視程距離の関係として整理したもので相関係数ｒ
は、ｒ＝０．８１である。すなわち、図３、図４の場合
とも標準偏差と視程（対数変換値）とは極めて強い相関
を示している。なお、これらのデータは、データの瞬間
的な偏り、外乱を避けるため、標準偏差は４０秒ごとの
駒撮りビデオ画像の１０分間の平均値から算出したもの
であり、視程は反射型視程計を用いて０．１秒ごとに１
０分間計測した値を平均したものである。FIG. 4 shows the data obtained at another point B arranged as the relationship between the standard deviation and the visibility distance.
Is r = 0.81. That is, the standard deviation and the visibility (logarithmic conversion value) show an extremely strong correlation in both cases of FIG. 3 and FIG. Note that these data were calculated from the average value of 10-minute frame-shooting video images every 40 seconds for the standard deviation in order to avoid momentary data deviation and disturbance. 1 every 0.1 seconds using
It is the average of the values measured for 0 minutes.
【００１１】図３、図４のように輝度標準偏差と対数変
換した視程との間に直線的な相関関係があれば、各デー
タを用いた最小自乗法又は分散分析によって、輝度標準
偏差をｘ、視程を対数変換した値をｙとした場合、ｙ＝
ａｘ＋ｂ（ａ，ｂは定数）なる関係式を得ることができ
る。図３、図４に示す直線はこのようにして得られたも
のであり、輝度標準偏差と視程とは１対１に対応させる
ことができる。例えば、地点Ｂにおいてビデオ画像の輝
度標準偏差が３０の場合、図４から視程は略２００ｍと
なる。すなわち、上記のような関係（データ）をテレビ
カメラ設置箇所について求めておくことにより、これら
のデータから、視程を測定したい地点のカメラのビデオ
信号の輝度出現頻度分布、輝度標準偏差を算出すること
により、その地点の、その時点の視程距離を測定できる
ことは明らかである。本発明の実施例では１０分間の平
均値をもって分析する例を示したが、１０分間に限るこ
とはなく、目的に応じ数秒から数１０分までの平均値を
分析することもできる。If there is a linear correlation between the brightness standard deviation and the logarithmically converted visibility as shown in FIGS. 3 and 4, the brightness standard deviation is x by the least squares method or the analysis of variance using each data. , Where y is the value obtained by logarithmically converting the visibility, y =
A relational expression of ax + b (a and b are constants) can be obtained. The straight lines shown in FIGS. 3 and 4 are obtained in this way, and the luminance standard deviation and the visibility can be made to correspond one to one. For example, when the brightness standard deviation of the video image at the point B is 30, the visibility is about 200 m from FIG. That is, by obtaining the relationship (data) as described above for the location where the TV camera is installed, the luminance appearance frequency distribution and the luminance standard deviation of the video signal of the camera at the location where the visibility is desired to be measured can be calculated from these data. It is clear that the visibility distance at that point can be measured by. In the embodiment of the present invention, an example in which the average value for 10 minutes is used for analysis is shown, but it is not limited to 10 minutes, and an average value for several seconds to several tens of minutes can be analyzed according to the purpose.
【００１２】最近のテレビカメラにおいては蓄積型の高
感度カメラが存在するが、同一視野の画像を加算して感
度をあげている。一般に日本におけるテレビカメラの標
準方式は毎秒３０枚の画像を撮像しているが、例えば、
１秒間の画像３０枚を加算し３０倍の感度にしている例
がある。３０倍の感度の場合は星明かり、月明かり等微
弱な明るさでも撮像可能であるので、数分程度の平均輝
度階調を分析する場合は数分間の画像を加算することに
なり、極めて微弱な明るさでも視程距離の算出が可能と
なる。In recent television cameras, there is a storage type high sensitivity camera, but the sensitivity is increased by adding images of the same field of view. Generally, the standard system of TV cameras in Japan captures 30 images per second.
There is an example in which 30 images for 1 second are added to increase the sensitivity 30 times. In the case of 30 times sensitivity, it is possible to image even in weak light such as starlight and moonlight, so when analyzing the average brightness gradation of several minutes, images of several minutes are added, which is extremely weak. The visibility distance can be calculated even with the brightness.
【００１３】視程計測の目的は道路上の自動車運転者に
視程距離から見た適正な運転速度の指示等の情報を提供
すること等であるため、人間の目に近いテレビカメラで
測定することが望ましいとされているが、この点からも
本発明の方法は適しているといえる。Since the purpose of the visibility measurement is to provide an automobile driver on the road with information such as an instruction of an appropriate driving speed from the visibility distance, it can be measured by a television camera close to human eyes. Although desirable, the method of the present invention is also suitable in this respect.
【００１４】[0014]
【発明の効果】本発明によれば対向する光源は不要であ
り、また、特定のパターンを有する標識も不要であるの
で、テレビカメラは光源あるいは標識からの距離の設定
も不要である。撮像対象によって相関直線の傾きを較正
しておけば、テレビカメラ及びビデオ信号分析装置を自
動車等に積載し複数の計測地点を移動することも可能で
あり、また従来の道路の混雑状況等を監視するために設
置したテレビカメラにビデオ信号分析装置を付加するこ
とにより視程計測機能を付加することも可能である。According to the present invention, the light sources facing each other are not necessary, and the sign having a specific pattern is also unnecessary. Therefore, the television camera does not need to set the distance from the light source or the sign. If the slope of the correlation line is calibrated according to the imaged object, it is possible to load a TV camera and a video signal analysis device on an automobile etc. to move to multiple measurement points, and also to monitor the conventional road congestion situation. It is also possible to add a visibility measurement function by adding a video signal analysis device to the television camera installed for this purpose.
【図１】本発明を説明するための、降雪時のビデオ信号
の輝度階調毎の出現頻度を示す図である。FIG. 1 is a diagram illustrating an appearance frequency for each luminance gradation of a video signal during snowfall for explaining the present invention.
【図２】本発明を説明するための、晴天時のビデオ信号
の輝度階調毎の出現頻度を示す図である。FIG. 2 is a diagram showing the appearance frequency of each luminance gradation of a video signal in fine weather for explaining the present invention.
【図３】本発明を説明するための、ある地点における輝
度分布の標準偏差と視程距離の関係を示す図である。FIG. 3 is a diagram illustrating a relationship between a standard deviation of a luminance distribution and a visibility distance at a certain point for explaining the present invention.
【図４】本発明を説明するための、図３とは別の地点に
おける輝度分布の標準偏差と視程距離の関係を示す図で
ある。FIG. 4 is a diagram illustrating a relationship between a standard deviation of a luminance distribution and a visibility distance at a point different from that of FIG. 3 for explaining the present invention.
Ｍｅａｎ 平均階調 Ｓｄ 標準偏差 Mean Average gradation Sd Standard deviation
───────────────────────────────────────────────────── フロントページの続き (71)出願人 000005429 日立電子株式会社 東京都千代田区神田和泉町１番地 (72)発明者 石本 敬志 北海道札幌市豊平区平岸１条３丁目 北海 道開発局 開発土木研究所内 (72)発明者 千葉 隆広 北海道札幌市豊平区平岸１条３丁目 北海 道開発局 開発土木研究所内 (72)発明者 加治屋 安彦 北海道札幌市豊平区平岸１条３丁目 北海 道開発局 開発土木研究所内 (72)発明者 堀川 和典 北海道札幌市東区北13条東４丁目30−３ さつでんビル スタジオムーン内 (72)発明者 駒形 悟 北海道札幌市東区北13条東４丁目30−３ さつでんビル スタジオムーン内 (72)発明者 尾田 鴻一郎 東京都新宿区高田馬場３丁目18番25号 日 立電子株式会社ビデオシステム事業所内 ─────────────────────────────────────────────────── ─── Continuation of the front page (71) Applicant 000005429 Hitachi Electronics Co., Ltd. 1 Izumi-cho, Kanda, Chiyoda-ku, Tokyo (72) Inventor Takashi Ishimoto 1-3, Hiragishi, Toyohira-ku, Sapporo, Hokkaido Kitakai Road Development Bureau Development Civil Engineering Research (72) Inventor Takahiro Chiba 1-3, Hiragishi, Toyohira-ku, Sapporo-shi, Hokkaido Civil Engineering Research Institute, North Kaido Development Bureau (72) Inventor Yasuhiko Kajiya 1-3-chome, Hiragishi, Toyohira-ku, Sapporo, Hokkaido Kita Kaido Development Bureau Development Civil Engineering Research In-house (72) Inventor Kazunori Horikawa 4-30-3 Kita 13-jo Higashi-ku, Sapporo-shi, Hokkaido Satsuden Building Studio Moon (72) Inventor Satoru Komagata 4-30-3 Kita 13-higashi, Higashi-ku, Sapporo, Hokkaido Tsuden Building, Studio Moon (72) Inventor Koichiro Oda, 3-18-25 Takadanobaba, Shinjuku-ku, Tokyo Ceremony Company Video System Office
Claims (2)
レビカメラのビデオ信号の輝度分布の標準偏差を求め、
予め当該地点で採取した種々の条件下における輝度分布
標準偏差と視程との関係を示すデータに基づく相関直線
又は関係式により、対応する視程を得るようにしたこと
を特徴とする視程計測方法。1. A standard deviation of a luminance distribution of a video signal of a television camera provided at a point where visibility is to be measured,
A visibility measuring method characterized in that a corresponding visibility is obtained by a correlation straight line or a relational expression based on data showing a relationship between a brightness distribution standard deviation and visibility under various conditions collected in advance at the point.
上記テレビカメラのビデオ信号の輝度分布は、所定期間
内の所定の複数画面の輝度分布であることを特徴とする
視程計測方法。2. The visibility measuring method according to claim 1,
A visibility measuring method, wherein the brightness distribution of the video signal of the television camera is a brightness distribution of a plurality of predetermined screens within a predetermined period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP7030990A JPH08219784A (en) | 1995-02-20 | 1995-02-20 | Method for measuring visibility |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP7030990A JPH08219784A (en) | 1995-02-20 | 1995-02-20 | Method for measuring visibility |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JPH08219784A true JPH08219784A (en) | 1996-08-30 |
Family
ID=12319060
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP7030990A Pending JPH08219784A (en) | 1995-02-20 | 1995-02-20 | Method for measuring visibility |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH08219784A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002163792A (en) * | 2000-11-27 | 2002-06-07 | Natl Inst For Land & Infrastructure Management Mlit | Obstacle collision prevention supporting system |
| JP4914365B2 (en) * | 2004-10-19 | 2012-04-11 | ラボラトワール セントラル デ ポンゼ シャッセ | Line-of-sight measurement equipment |
| WO2021053789A1 (en) * | 2019-09-19 | 2021-03-25 | 株式会社ウェザーニューズ | Visibility estimation device, visibility estimation method, and recording medium |
-
1995
- 1995-02-20 JP JP7030990A patent/JPH08219784A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002163792A (en) * | 2000-11-27 | 2002-06-07 | Natl Inst For Land & Infrastructure Management Mlit | Obstacle collision prevention supporting system |
| JP4914365B2 (en) * | 2004-10-19 | 2012-04-11 | ラボラトワール セントラル デ ポンゼ シャッセ | Line-of-sight measurement equipment |
| WO2021053789A1 (en) * | 2019-09-19 | 2021-03-25 | 株式会社ウェザーニューズ | Visibility estimation device, visibility estimation method, and recording medium |
| JP6901647B1 (en) * | 2019-09-19 | 2021-07-14 | 株式会社ウェザーニューズ | Visibility estimation device, visibility estimation method, and recording medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7693344B2 (en) | Method and device for image processing and a night vision system for motor vehicles | |
| JP4985394B2 (en) | Image processing apparatus and method, program, and recording medium | |
| CA2315188C (en) | Road pavement deterioration inspection system | |
| US7016045B2 (en) | Video camera-based visibility measurement system | |
| EP0986036A2 (en) | Method of updating reference background image, method of detecting entering objects and system for detecting entering objects using the methods | |
| US8077995B1 (en) | Infrared camera systems and methods using environmental information | |
| US6850642B1 (en) | Dynamic histogram equalization for high dynamic range images | |
| US20090016636A1 (en) | Image processing apparatus, and method, program and recording medium | |
| US6853453B2 (en) | Video camera-based visibility measurement system | |
| KR19990072061A (en) | Vehicle navigation system and signal processing method for the navigation system | |
| EP1383100A3 (en) | Local positioning apparatus and method therefor | |
| Kwon | Atmospheric visibility measurements using video cameras: Relative visibility | |
| JP2003032669A (en) | On-vehicle image processing camera device | |
| KR102267517B1 (en) | Road fog detecting appartus and method using thereof | |
| McCarley et al. | Visibility of road hazards in thermal, visible, and sensor-fused night-time imagery | |
| JP2006221467A (en) | Low visibility evaluating method and low visibility evaluating system | |
| JPH08219784A (en) | Method for measuring visibility | |
| CN109285187A (en) | A kind of farthest visible point detecting method based on traffic surveillance videos image | |
| CN107328777A (en) | A kind of method and device that atmospheric visibility is measured at night | |
| EP1476326B1 (en) | Device for automatically adjusting the brightness of a luminous beam emitted by the rear lighting equipment on a vehicle | |
| DE3801368A1 (en) | Visual range meter | |
| KR20200075692A (en) | Fog detection device using coordinate system and method thereof | |
| JP3538828B2 (en) | Transmittance measuring device and transmittance measuring system | |
| CN112365626A (en) | Vehicle-mounted camera picture processing method | |
| JP2002230558A (en) | Method of evaluating visual environment |