Apparatus and method for measuring visual range using geometrical information of an image and an image pattern recognition technique

Abstract

The present invention relates to a measurement system for calculating visual range using perspective and geometrical information of an image captured by a camera, CCTV, camcorder, or other imaging device and a method for measuring visual range from a non-linear function through pattern recognition on an image. The present measurement system and method may make up for uncertainty due to assumption that aerosol is evenly distributed and limit to measurement space in the optimal measurement method, thus offering more objective data.

Claims

1. An apparatus for measuring a visual range using geometrical information of an image and an image pattern recognition method, the apparatus comprising: an IP camera installed to transmit an image of a measurement site; an image storing unit selectively extracting a still image from a video image of the transmitted image; and a visual range computing unit recognizing a coordinate of a similar color region from an image visual range function and the geometrical information of the image, performs distance-mapping on an analytic region of the image received from the IP camera, determines the image visual range function from a zero distance, a coordinate visual range efficiency coefficient, a perspective coefficient, a correction factor of an image visual range, and a distance per y-coordinate of an object positioned furthermost in the analytic region of an image with the distance for the coordinate represented, and calculates the visual range using the image pattern recognition method.

2. A method for measuring a visual range using geometrical information of an image and an image pattern recognition method, the visual range measurement method comprising: receiving an image from an IP camera installed in a measurement site; performing distance-mapping on the analytic region of the received image; determining an image visual range function from the received image; applying an image pattern recognition method using ambient color information of an object from the received image; and calculating a visual range from a coordinate calculated by the image pattern recognition method, wherein determining the image visual range for calculating a visual range from a zero distance, a coordinate visual range, and a distance per y-coordinate of an object positioned furthermost in the analytic region of an image with the distance for the coordinate represented.

3. The method of claim 2, wherein performing the distance-mapping on the analytic region of the received image includes inputting geometrical information for a 2-dimensional plane coordinate from the received image to represent a distance.

4. The method of claim 2, wherein applying the image pattern recognition method using ambient color information of an object from the received image includes extracting RGB values of an ambient reference color of the object, setting a similar color tolerance range, and connecting similar colors in the tolerance range for the RGB values of the reference color to display an area.

Description

BRIEF DESCRIPTION OF DRAWINGS

(1) A more complete appreciation of the present disclosure and many of the attendant aspects thereof will be readily obtained as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings, wherein:

(2) FIG. 1 is a concept view illustrating a visual range measurement apparatus according to an embodiment of the present invention;

(3) FIG. 2 is a flowchart illustrating a visual range measurement method according to an embodiment of the present invention;

(4) FIG. 3 is a view illustrating a captured image according to an embodiment of the present invention;

(5) FIG. 4 is a view illustrating an image that has undergone distance-mapping on the analytic region of the received image according to an embodiment of the present invention;

(6) FIG. 5 is a view illustrating actual distances (d) per y-coordinate and an image visual range (IVR) per y-coordinate according to an embodiment of the present invention;

(7) FIG. 6 is a view illustrating a correction factor of an image visual range for a coordinate visual range efficiency coefficient and a perspective coefficient according to an embodiment of the present invention;

(8) FIG. 7 is a view illustrating an exemplary image analyzed by an image pattern recognition method according to an embodiment of the present invention; and

(9) FIG. 8 is a scatter plot illustrating comparison between an image visual range and visual range measured using a transmissometer, which is an optical visual range measurement, according to an embodiment of the present invention.

DETAILED DESCRIPTION OF EMBODIMENTS

(10) A visual range measurement apparatus using an image according to an embodiment of the present invention is subjected to an installation step and an operation step. In the installation step, an IP camera is installed in a place where an object remotely identifiable from a long distance is viewed. After installation, the IP camera is fixed. In the operation step, an analytic region is set in an image transmitted from a measurement site, geometrical information is input to determine an image visual range function, and a visual range is then computed by a pattern recognition method. The visual range is indicated as an image visual range.

(11) According to an embodiment of the present invention, results may offer various embodiments depending on image obtaining conditions. Some embodiments of the present invention are hereinafter described with reference to the accompanying drawings, but the present invention is not limited thereto.

(12) It will be understood that when an element or layer is referred to as being on, connected to, coupled to, or adjacent to another element or layer, it can be directly on, connected, coupled, or adjacent to the other element or layer, or intervening elements or layers may be present. As used herein, the singular forms a, an, and the are intended to include the plural forms as well, unless the context clearly indicates otherwise.

(13) Hereinafter, embodiments of the present invention are described in detail with reference to the accompanying drawings.

(14) FIG. 1 is a concept view illustrating a visual range measurement apparatus according to an embodiment of the present invention. FIG. 2 is a flowchart illustrating a visual range measurement method according to an embodiment of the present invention. According to an embodiment of the present invention, a visual range measurement apparatus and method are described with reference to FIGS. 1 and 2.

(15) Referring to FIG. 1, according to an embodiment of the present invention, the visual range measurement apparatus includes an IP camera 100, an image storing unit 200, a visual range computing unit 300, and a screen output unit 400.

(16) The IP camera 100 selects a measurement site that may represent visibility of a target area, captures a color image, and remotely transmits the image to the image storing unit 200 positioned in a monitoring center (A10). Here, the target area refers to an area where a long visual range may be measured. The IP camera 100 may refer to any device that, such as a CCD (Charge-Coupled Device) or CMOS (Complementary Metal-Oxide Semiconductor), may capture an image and that may be assigned an IP (Internet Protocol) address or domain address to transmit the image to a remote site.

(17) The image storing unit 200 may refer to any device that may store the image transmitted from the IP camera 100 in, e.g., a hard disc, in an electronic video file, and that may selectively extract a still image from the stored video image and transmit the extracted image to the visual range computing unit 300.

(18) FIG. 2 is a flowchart illustrating a visual range measurement method by a visual range computing unit 300 according to an embodiment of the present invention. The image (A10) received and obtained from the remote monitoring site undergoes distance-mapping on the analytic region (A20), determination of an image visual range function (A30), an image pattern recognition method (A40), and visual range calculation (A50), thereby producing a visual range.

(19) FIG. 3 is a view illustrating a captured image according to an embodiment of the present invention. The visual range computing unit 300 investigates the image A10 received from the image storing unit 200 shown in FIG. 3 and selects an analytic region for measuring a visual range.

(20) FIG. 4 is a view illustrating an image that has undergone distance-mapping on the analytic region of the image according to an embodiment of the present invention. The analytic region selected from the image A10 is subjected to distance-mapping on the analytic region A20 using geometrical information of the captured area as shown in FIG. 4. A distance for a 2D (two dimensional) plane coordinate in the selected analytic region of the image is expressed and processed as in Equation 1:
f(x,y)=d[Equation 1]

(21) where, d refers to the distance at a given coordinate.

(22) Distance per y-coordinate calculated by the distance-mapping on the analytic region (A20) are utilized in the step of the determination of an image visual range function (A30) as shown in FIG. 2. A visual range is expressed and processed with respect to the distances per y-coordinate as in Equation 2:
IVR=f(y)+d.sub.0[Equation 2]

(23) where, IVR refers to an image visual range, f(y) an image visual range function, and do a zero distance. The image visual range is calculated by summing the image visual range function and the zero distance. The zero distance refers to a distance at coordinate (x,0). The image visual range function that is a function of distance with respect to y coordinate is expressed and processed as in Equation 3:
f(y)=[{1y/(ymax+f/dmax)}.sup.1][Equation 3]

(24) where, is a coordinate visual range efficiency coefficient, and is a perspective coefficient. The coordinate visual range efficiency coefficient and the perspective coefficient are adjusted based on the distance per y-coordinate calculated by performing distance-mapping on the analytic region of the image to determine the coordinate visual range efficiency coefficient and the perspective coefficient so that the image visual range most complies with the distance per y-coordinate.

(25) Here, d.sub.max is the distance per y-coordinate of an object positioned furthermost in the analytic region of the image, and f is a correction factor of an image visual range. According to an embodiment of the present invention, an example is described in which the determination of an image visual range function (A30) applies to the visual range measuring apparatus using an image.

(26) FIG. 5 is a view illustrating relationship between a distance per y-coordinate (d) and an image visual range (IVR) according to an embodiment of the present invention. Referring to FIG. 5, distances per y-coordinate which are calculated by performing distance-mapping on the analytic region of the image are shown.

(27) FIG. 6 is a view illustrating correction factors of an image visual range for coordinate visual range efficiency coefficients and perspective coefficients according to an embodiment of the present invention.

(28) Referring to FIG. 5, the image visual range function that most complies with the distance per y-coordinate calculated by performing distance-mapping on the analytic region of the image is obtained when the coordinate visual range efficiency coefficients is 1.2, and the perspective coefficient is 1.2. In this case, referring to FIG. 6, the correction factors of the image visual range is determined to be 192.0. According to an embodiment of the present invention, in the step of distance-mapping on the analytic region (A20), d.sub.max is 127 km.

(29) Accordingly, the image visual range function A30 may be determined from the determined the coordinate visual range efficiency coefficient, the perspective coefficient, the correction factors of the image visual range, and the distance per y-coordinate of an object positioned furthermost in the analytic region of the image.

(30) FIG. 7 is a view illustrating an exemplary image analyzed by an image pattern recognition method according to an embodiment of the present invention.

(31) The image pattern recognition method utilizes RGB color information. An ambient color of an object viewed farthest in the analytic region of the image is selected as a reference color. The RGB values of the reference color are extracted, and a similar color tolerance range is set. Similar colors in the tolerance range for the RGB values of the reference color may be connected to one another to determine an ambient color region of the object viewed farthest in the analytic region of the image.

(32) Referring to FIG. 7, the upper image of FIG. 7 is observed further than the lower image of FIG. 7. It can be seen from FIG. 7 that an ambient color region of an object viewed farthest in the upper image is shown to be different from that in the lower image.

(33) In the visual range calculation step (A50), coordinates to be applied to the image visual range function are input from the RGB values per coordinate of the color region determined by the image pattern recognition method. In Equation 3 above, the minimum coordinate value of the color region is input as y coordinate to thereby calculate a visual range.

(34) FIG. 8 is a scatter plot illustrating comparison between an image visual range and visual range measured using a transmissometer, which is an optical visual range measurement, according to an embodiment of the present invention.

(35) Referring to FIG. 8, a slope is 0.938 and a correlation is 0.8738 between the visual range (image visual range) calculated by the step of distance-mapping on the analytic region (A20), the step of the determination of an image visual range function (A30), the image pattern recognition method step (A40), and the visual range calculation step (A50) and the visual range (optical visual range) measured by a transmissometer (not shown) which is an optical visual range measurement. Accordingly, it can be identified that the visual range calculated according to an embodiment of the present invention shows good correlation with the visual range measured by the transmissometer (not shown).

(36) The screen output unit 400 may display the image on the screen or store the image, may display the image color information per coordinate calculated in the image pattern recognition method (A40) in RGB and store the image color information per coordinate, and may display and store the visual range obtained from the visual range calculation unit 300.

(37) Embodiments of the present invention may be stored in a computer-readable medium that contains programming commands implementable by various computers. The computer-readable medium may include programming commands, local data files, local data structures, alone or in combinations thereof. The medium may be one that may be specially manufactured for the present invention or that may be known and available to one of ordinary skill in the computer software-related art.

(38) Although the present invention has been shown and described in connection with embodiments thereof, it should be understood that various changes in form and detail may be made thereto without departing from the scope of the present invention defined in the following claims.