Image processing system, image processing method and image processing program

Abstract

To provide an image processing system, an image processing method and an image processing program for processing image data so as to easily estimate a direction in an image having an azimuth mark.SOLUTION: A server 1, which is an image processing system, includes a direction processing part for: extracting an azimuth mark from an image acquired by an acquisition part on an object property; acquiring 2 or more positions in the image on the basis of the azimuth mark; and estimating a direction in the image on the basis of the 2 or more positions.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to an image processing device, an image processing method, and an image processing program for processing image data. [Background technology]

[0002] There are services that search for properties based on search conditions entered by a user. For example, the search conditions include conditions on attributes such as location (such as the area or the nearest station), rent, exclusive area, and floor plan type (such as 2LDK) that indicates the area configuration of the property. Such services search for one or more properties that match the search conditions based on tags that represent each attribute set for each property, and present one or more properties that match the search conditions to the user.

[0003] A user may narrow down a desired property based on various attributes other than those described above. For example, a user may narrow down a desired property based on a direction of the property. The direction of the property may be a direction in which a main window, such as a window leading to the outside, is located, a direction in which an entrance is located, or a direction in which each element of the property is located.

[0004] The direction of a property is set based on manual input by real estate companies. Real estate companies judge and input the direction of a property by comparing the compass symbol and the floor plan, but there are cases where input is omitted or judged incorrectly. Therefore, there is a demand for technology that can automatically estimate the direction in images that include compass symbols.

[0005] Patent Document 1 discloses a method for determining the orientation of an image from an orientation mark included in the image by recording the orientation mark patterns and the orientations indicated by each pattern in a memory. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] JP 2013-104854 A Summary of the Invention [Problem to be solved by the invention]

[0007] However, the technology disclosed in Patent Document 1 judges the orientation of an image by comparing an orientation mark included in the image with a plurality of orientation mark patterns recorded in a memory. Therefore, in order to improve the accuracy of judging the orientation of an image, many orientation mark patterns must be recorded in a memory. The number of orientation mark patterns increases as the orientation indicated by the orientation mark pattern becomes finer. Since there are many variations in the shape of the orientation mark, the number of orientation mark patterns increases according to the number of variations. Furthermore, comparing an orientation mark included in an image with many orientation mark patterns imposes a processing load, such as a large amount of calculation processing.

[0008] SUMMARY OF THE PRESENT DISCLOSURE An object of the present invention is to provide a technique that makes it possible to easily estimate a direction in an image that includes a directional symbol. [Means for solving the problem]

[0009] An image processing device according to an embodiment includes a direction processing unit that extracts a directional symbol from an image, determines two or more positions in the image based on the directional symbol, and estimates a direction in the image based on the two or more positions. Effect of the Invention

[0010] According to the embodiment, a technique can be provided that makes it possible to easily estimate a direction in an image that includes a directional symbol. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is a block diagram illustrating an example of the configuration of a server according to an embodiment. [Diagram 2] FIG. 2 is a diagram illustrating an example of an image according to the embodiment. [Diagram 3] FIG. 3 is a flowchart illustrating an example of image data processing by the server according to the embodiment. [Figure 4] FIG. 4 is a flowchart illustrating an example of a process of estimating a direction in an image by a server according to the embodiment. [Diagram 5] FIG. 5 is a diagram showing an example of an image of a direction symbol according to the embodiment. [Figure 6] FIG. 6 is a diagram illustrating an example of the position of characters indicating a direction according to the embodiment. [Figure 7] FIG. 7 is a diagram showing an example of an image of a direction symbol according to the embodiment. [Figure 8] FIG. 8 is a diagram illustrating an example of an image of a direction symbol according to the embodiment. [Figure 9] FIG. 9 is a diagram illustrating an example of an image of a direction symbol according to the embodiment. [Figure 10] FIG. 10 is a diagram illustrating an example of an image of a direction symbol according to the embodiment. [Figure 11] FIG. 11 is a diagram illustrating an example of an image of a direction symbol according to the embodiment. [Figure 12] FIG. 12 is a diagram illustrating an example of an image of a direction symbol according to the embodiment. [Figure 13] FIG. 13 is a diagram illustrating an example of an image of a direction symbol according to the embodiment. [Figure 14] FIG. 14 is a diagram illustrating an example of an image of a direction symbol according to the embodiment. [Figure 15] FIG. 15 is a diagram showing an example of an image of a direction symbol according to the embodiment. [Figure 16] FIG. 16 is a flowchart illustrating an example of a process of estimating a direction of a property by the server according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] [Embodiment] Hereinafter, some embodiments will be described with reference to the drawings. Note that the scale of each part in each drawing used in the following description of the embodiments may be changed as appropriate. Also, for the sake of explanation, each drawing used in the following description of the embodiments may omit configurations.

[0013] (Configuration example) FIG. 1 is a block diagram showing an example of the configuration of the server 1. The server 1 is an electronic device that processes image data. The server 1 is an example of an image processing device that processes image data. Here, the image data is assumed to be data that represents an image including an image of an object and an image of a compass symbol. The image represented by the image data is an image represented in a coordinate system of x and y axes. The direction along the x axis is also called the left-right direction or horizontal direction. The direction along the y axis is also called the up-down direction or vertical direction.

[0014] The property may include properties for rent in various buildings such as condominiums, apartments, and houses, or may include properties for sale. For example, the image of the property is an image of a floor plan of the property. The floor plan of the property may be a floor plan. The floor plan of the property is a diagram that includes multiple elements of the property. For example, the image of the property is an image of a rectangle that encloses the property with four sides facing up, down, left, and right.

[0015] The elements of the property may include one or more areas. The areas may include indoor areas of the property, or outdoor areas of the property. The indoor areas are spacious indoor elements such as a bath, toilet, washroom, closet, wardrobe, corridor, entrance, and room. The rooms are areas where people stay, such as Western-style rooms, Japanese-style rooms, and LDK (living, dining, and kitchen). The rooms may be DK (dining and kitchen) or LK (living and kitchen) instead of LDK. The Western-style rooms may include a service room. The outdoor areas are spacious outdoor elements such as a balcony, veranda, and garden.

[0016] The plurality of elements of the object can include one or more doors. The doors include the outer door provided at the entrance of the object. The doors include the interior doors that partition areas indoors in the object. The interior doors include hinged doors and sliding doors. The sliding doors include fusuma and shoji. A door is sometimes referred to as a "doorway".

[0017] The plurality of elements of the object can include one or more windows. The windows include windows through which people can pass from indoors to outdoors. The windows include windows through which people cannot pass from indoors to outdoors. The windows may be windows that can be opened and closed, or windows that cannot be opened and closed.

[0018] The plurality of elements of the object can include one or more walls.

[0019] The azimuth symbol is a symbol indicating a direction. The azimuth symbol is also referred to as an azimuth mark. The azimuth symbol includes a figure indicating a direction. For example, the figure indicating a direction is a figure in which the part indicating north is distinguishable. The azimuth symbol may include a character indicating a direction. For example, the character indicating a direction is a character such as "N" or "north" indicating north as a direction. The direction includes the meaning of azimuth.

[0020] Server 1 includes a processing circuit 11, a main memory 12, an auxiliary storage device 13, and a communication interface 14. The processing circuit 11, the main memory 12, the auxiliary storage device 13, and the communication interface 14 are connected to each other so as to be able to input and output signals via a bus or the like. In FIG. 1, the interface is described as "IF".

[0021] The processing circuit 11 corresponds to the central part of the server 1. The processing circuit 11 constitutes the computer of the server 1. The processing circuit 11 includes one or more circuits that execute multiple processes by multiple functions. For example, the circuit is a processor, an ASIC (Application Specific Integrated Circuit) or an FPGA (Field-Programmable Gate Array), but is not limited thereto. For example, the processor is a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit), but is not limited thereto. The processing circuit 11 loads an image processing program that is pre-stored in the main memory 12 or the auxiliary storage device 13 into the main memory 12. The image processing program is a program that can cause the processing circuit 11 to execute processes by each unit described later. The processing circuit 11 executes the image processing program loaded into the main memory 12, thereby enabling various processes to be executed.

[0022] The main memory 12 corresponds to the main storage portion of the server 1. The main memory 12 includes a nonvolatile memory area and a volatile memory area. The main memory 12 stores an operating system or a program in the nonvolatile memory area. The main memory 12 uses the volatile memory area as a work area where data is appropriately rewritten by the processing circuit 11. For example, the main memory 12 includes a ROM (Read Only Memory) as a nonvolatile memory area. For example, the main memory 12 includes a RAM (Random Access Memory) as a volatile memory area. The main memory 12 may store an image processing program.

[0023] The auxiliary storage device 13 corresponds to the auxiliary storage portion of the server 1. The auxiliary storage device 13 includes one or more storage devices. The storage devices are, but are not limited to, EEPROM (registered trademark) (Electric Erasable Programmable Read-Only Memory), HDD (Hard Disc Drive), SSD (Solid State Drive), flash memory, etc. The auxiliary storage device 13 stores the above-mentioned image processing program, data used by the processing circuit 11 in performing various processes, and data generated by the processing in the processing circuit 11.

[0024] The auxiliary storage device 13 includes an image storage area 131. The image storage area 131 stores image data of a plurality of images. For example, the image data of a plurality of images is data registered using terminals of a plurality of real estate companies. The data stored in the image storage area 131 may be updated each time image data is saved. The image storage area 131 is an example of a storage unit that stores image data.

[0025] The auxiliary storage device 13 includes a processing data storage area 132. The processing data storage area 132 stores processing data for each image data. The processing data is data obtained by processing the image data. For example, the processing data includes data on a direction in an image represented by the image data. The direction in an image is about which direction in an image corresponds to which direction. For example, the direction in an image is the north direction in the image. In this example, the data on the direction in an image is about which direction in an image corresponds to the north direction. For example, the processing data includes data on a direction in a property included in an image represented by the image data. The direction in a property is the direction in which each element of the property is arranged. The direction in which each element of the property is arranged includes the direction in which each element of the property faces. For example, the direction in a property is the direction in which a main window such as a window connecting to the outdoors is arranged or the direction in which an entrance is arranged, but is not limited to these. For example, the direction in which a main window is arranged includes the direction in which the main window faces. The direction in which the entrance is located includes the direction in which the entrance faces. The data stored in the processing data storage area 132 may be updated each time the processing data is saved. The processing data storage area 132 is an example of a storage unit that stores processing data.

[0026] The communication interface 14 includes various interfaces that communicatively connect the server 1 to other electronic devices using a communication protocol defined by the network. The network includes at least one of the Internet, a mobile network, and a LAN (Local Area Network), etc.

[0027] The hardware configuration of the server 1 is not limited to the above-mentioned configuration. The server 1 allows the above-mentioned components to be omitted or changed and new components to be added as appropriate.

[0028] Each unit realized by the processing circuit 11 will be described. The processing circuit 11 realizes an acquisition unit 111, a direction processing unit 112, and an object processing unit 113. Each unit realized by the processing circuit 11 can also be called each function. Each unit realized by the processing circuit 11 can also be called a control unit including the processing circuit 11 and the main memory 12.

[0029] The acquisition unit 111 acquires image data of a target object. The target object is an object from which a direction in an image and a direction at the object are to be obtained.

[0030] The direction processor 112 estimates the direction in the image based on the direction symbol included in the image represented by the image data. Estimating the direction in the image includes various aspects for obtaining the direction in the image, such as determining and identifying.

[0031] The object processing unit 113 estimates the direction of an object included in the image based on the direction in the image. Estimating the direction of the object includes various aspects for obtaining the direction of the object, such as determining and identifying.

[0032] (Example image) FIG. 2 is a diagram showing an example of an image. Image Im includes an image ImA of a property. The image of the property includes a bathroom, a toilet, a closet, a wardrobe, a hallway, a Japanese-style room, a Western-style room, a DK, an exterior door, a plurality of interior doors, and a plurality of windows as multiple elements of the property. Image Im includes an image ImB of a direction symbol. Image ImB of the direction symbol includes an image ImB1 of a figure indicating a direction. Image ImB of the direction symbol includes an image ImB2 of letters indicating a direction.

[0033] (Example of operation) Next, an example of the operation of the server 1 configured as above will be described. Note that the process steps described below are merely examples, and each process may be modified as much as possible. In addition, steps may be omitted, replaced, or added as appropriate for the process steps described below depending on the embodiment.

[0034] FIG. 3 is a flowchart showing an example of image data processing by the server 1.

[0035] The acquisition unit 111 acquires image data about a target property (step S1). In step S1, for example, the acquisition unit 111 acquires image data about an arbitrary property from the image storage area 131 as image data about the target property.

[0036] The direction processing unit 112 estimates a direction in the image Im based on a direction symbol included in the image Im represented by the image data acquired by the acquisition unit 111 (step S2). The process of step S2 will be described later.

[0037] The object processing unit 113 estimates the direction of the object included in the image Im based on the direction in the image Im estimated by the direction processing unit 112 (step S3). The process of step S3 will be described later.

[0038] FIG. 4 is a flowchart showing an example of the process of estimating the direction in the image Im by the server 1 in step S2.

[0039] The direction processing unit 112 extracts a direction symbol from the image Im represented by the image data acquired by the acquisition unit 111 (step S21). In step S21, for example, the direction processing unit 112 executes area determination by semantic segmentation based on the image data. For example, the semantic segmentation is FCN (Fully Convolutional Networks) or the like. Area determination by semantic segmentation may be a known technique. The direction processing unit 112 detects a direction symbol from the image Im represented by the image data based on the area determination by semantic segmentation. The direction processing unit 112 extracts the direction symbol from the image Im based on the detection of the direction symbol. Extracting the direction symbol from the image Im includes extracting an image org of the direction symbol from the image Im. The image org of the direction symbol is a rectangular image obtained by padding the periphery of the area to analyze the image within the area after the area of ​​the direction symbol is extracted by semantic segmentation.

[0040] The direction processing unit 112 determines two or more positions in the image Im based on the extracted directional symbol (step S22). For example, the positions in the image Im are coordinates in the image Im. Below, some examples of the positions in the image Im will be described, but the positions in the image Im may be determined based on the directional symbol and are not limited to those exemplified here.

[0041] The two or more positions in the image Im may include a position NC of the characters indicating the direction included in the azimuth sign. The azimuth processing unit 112 detects the characters indicating the direction from the azimuth sign image org by using a specific object detection technique in the image. The specific object detection technique in the image may be a neural network technique. The azimuth processing unit 112 detects the position NC of the characters indicating the direction based on the detection of the characters indicating the direction. The position NC of the characters indicating the direction is a position within the area of ​​the characters indicating the direction. The area of ​​the characters indicating the direction may be a convex polygonal area such as a rectangle including the characters indicating the direction, or may be a circular area. The area including the characters indicating the direction may be an area including the characters indicating the direction and having the smallest area. For example, the position NC of the characters indicating the direction is the center position of the area of ​​the characters indicating the direction. The position NC of the characters indicating the direction is an example of a position obtained based on the characters indicating the direction included in the azimuth sign. The characters indicating the direction are characters indicating north. Therefore, the position NC of the characters indicating the direction is an example of a position related to the north direction.

[0042] The two or more positions in the image may include a position ICC of the center of the inscribed circle of a convex polygon including the directional symbol. The direction processing unit 112 generates a direction symbol image bw by binarizing the direction symbol image org into black and white. The direction processing unit 112 calculates a convex polygon CP including a direction symbol that is different in color from the background color in the direction symbol image bw. The convex polygon including the direction symbol may be a convex polygon region that includes the direction symbol and has a minimum area. When the direction symbol includes a figure indicating a direction and a character indicating a direction, the convex polygon including the direction symbol may be a convex polygon including a figure indicating a direction excluding the character indicating a direction. The direction processing unit 112 calculates a position ICC of the center of the inscribed circle of the convex polygon CP based on the calculated convex polygon CP. The position ICC of the center is an example of a position that is geometrically calculated by the direction processing unit 112 based on the directional symbol. The position ICC of the center is an example of a center position for the directional symbol.

[0043] The two or more positions in the image may include a position GC of the center of gravity of a convex polygon including a direction symbol. The direction processing unit 112 determines the position GC of the center of gravity of the convex polygon CP based on the convex polygon CP calculated in the image bw. The position GC of the center of gravity is an example of a position geometrically determined by the direction processing unit 112 based on the direction symbol. Since figures indicating directions tend to have shapes in which the part indicating north protrudes, the position GC of the center of gravity tends to be closer to the part indicating north than the position ICC of the center. Therefore, the position GC of the center of gravity is an example of a position related to the direction of north.

[0044] The two or more positions in the image may include a position of the farthest point of the directional symbol from the central position ICC. The farthest point of the directional symbol from the central position ICC is the position of the directional symbol that is farthest from the central position ICC. The position of the farthest point of the directional symbol may be the position of the farthest point of a figure indicating a direction included in the directional symbol. The position of the farthest point is an example of a position that is geometrically determined by the direction processing unit 112 based on the directional symbol. Since a figure indicating a direction tends to have a shape in which the part indicating north protrudes, the position of the farthest point tends to be on the side of the part indicating north of the figure indicating the direction. Therefore, the position of the farthest point is an example of a position related to the direction of north.

[0045] A first example of the position of the farthest point will be described. The direction processing unit 112 detects and fills the outline of the white area excluding the outermost background white area in the image bw of the direction symbol, and performs expansion and contraction processing to generate the image fill of the direction symbol. The direction processing unit 112 determines the position FP_fill of the farthest point from the center position ICC in the image fill of the direction symbol. The position FP_fill of the farthest point is an example of a position geometrically determined by the direction processing unit 112 based on the direction symbol. The position FP_fill of the farthest point is an example of a position related to the direction of north.

[0046] A second example of the position of the farthest point will be described. The direction processing unit 112 generates an image bh of a direction symbol by inverting the image bw of the direction symbol (the background is black) and removing the area along the convex polygon CP that includes all the white areas. The direction processing unit 112 determines the position FP_bh of the farthest point from the center position ICC in the image bh of the direction symbol. The position FP_bh of the farthest point is an example of a position geometrically determined by the direction processing unit 112 based on the direction symbol. The position FP_bh of the farthest point is an example of a position related to the direction of north.

[0047] A third example of the position of the farthest point will be described. The direction processing unit 112 generates a direction symbol image bhe1 by shrinking the direction symbol image bh once. The direction processing unit 112 determines the position FP_bhe1 of the farthest point from the center position ICC in the direction symbol image bhe1. The position FP_bhe1 of the farthest point is an example of a position geometrically determined by the direction processing unit 112 based on the direction symbol. The position FP_bhe1 of the farthest point is an example of a position related to the direction of north.

[0048] A fourth example of the position of the farthest point will be described. The direction processing unit 112 generates a direction symbol image bhe2 by contracting the direction symbol image bh twice. The direction processing unit 112 obtains the position FP_bhe2 of the farthest point from the center position ICC in the direction symbol image bhe2. The position FP_bhe2 of the farthest point is an example of a position geometrically obtained by the direction processing unit 112 based on the direction symbol. The position FP_bhe2 of the farthest point is an example of a position related to the direction of north.

[0049] The two or more positions in the image may include the position of the farthest intersection of the directional sign from the center position ICC. The farthest intersection of the directional sign from the center position ICC is the position of the intersection farthest from the center position ICC among the intersection groups based on the multiple line components in the directional sign. The intersection group is a collection of multiple intersections based on the multiple line components in the directional sign. The intersection is an intersection where the multiple line components form an angle of a predetermined range. For example, the predetermined range is a range from 10 degrees to 80 degrees, but is not limited to this. The position of the farthest intersection of the directional sign may be the position of the farthest intersection of a figure indicating a direction included in the directional sign. The position of the farthest intersection is an example of a position geometrically determined by the direction processing unit 112 based on the directional sign. The position of the farthest intersection tends to be toward the part indicating north of the figure indicating the direction. Therefore, the position of the farthest intersection is an example of a position related to the direction of north.

[0050] A first example of the position of the farthest intersection point will be described. The direction processing unit 112 calculates multiple line components in the direction symbol in the image fill of the direction symbol. The direction processing unit 112 finds a group of intersection points at which the multiple line components form an angle within a predetermined range. The direction processing unit 112 finds the position CP_fill of the farthest intersection point from the center position ICC among the group of intersection points. The position CP_fill of the farthest intersection point is an example of a position geometrically found by the direction processing unit 112 based on the direction symbol. The position CP_fill of the farthest intersection point is an example of a position related to the direction of north.

[0051] A second example of the position of the farthest intersection point will be described. The direction processing unit 112 calculates a plurality of line components in the direction symbol in the image bh of the direction symbol. The direction processing unit 112 finds a group of intersection points at which the plurality of line components form an angle within a predetermined range. The direction processing unit 112 finds the position CP_bh of the farthest intersection point from the center position ICC among the group of intersection points. The position CP_bh of the farthest intersection point is an example of a position geometrically found by the direction processing unit 112 based on the direction symbol. The position CP_bh of the farthest intersection point is an example of a position related to the direction of north.

[0052] A third example of the position of the farthest intersection point will be described. The direction processing unit 112 calculates a plurality of line components in the direction symbol in the image bhe1 of the direction symbol. The direction processing unit 112 finds a group of intersection points at which the plurality of line components form an angle within a predetermined range. The direction processing unit 112 finds the position CP_bhe1 of the farthest intersection point from the center position ICC among the group of intersection points. The position CP_bhe1 of the farthest intersection point is an example of a position geometrically found by the direction processing unit 112 based on the direction symbol. The position CP_bhe1 of the farthest intersection point is an example of a position related to the direction of north.

[0053] A fourth example of the position of the farthest intersection point will be described. The direction processing unit 112 calculates a plurality of line components in the direction symbol in the image bhe2 of the direction symbol. The direction processing unit 112 finds a group of intersection points at which the plurality of line components form an angle within a predetermined range. The direction processing unit 112 finds the position CP_bhe2 of the farthest intersection point from the center position ICC among the group of intersection points. The position CP_bhe2 of the farthest intersection point is an example of a position geometrically found by the direction processing unit 112 based on the direction symbol. The position CP_bhe2 of the farthest intersection point is an example of a position related to the direction of north.

[0054] The direction processor 112 may obtain the ratio of intersections that are in the same direction as the position of the farthest intersection among the intersections included in the intersection group in the image of the direction symbol. For example, the same direction is within a predetermined small range of an orientation angle such as -15 degrees to +15 degrees, but is not limited to this.

[0055] A first example of the ratio will be described below. The direction processing unit 112 may obtain the ratio CPrt_fill of the intersections that are in the same direction as the position CP_fill of the farthest intersection among the intersections included in the intersection group in the image fill of the direction symbol.

[0056] A second example of the ratio will be described below. The direction processing unit 112 may obtain the ratio CPrt_bh of the intersections that are in the same direction as the position CP_bh of the farthest intersection among the intersections included in the intersection group in the image bh of the direction symbol.

[0057] A third example of the ratio will be described below. The direction processing unit 112 may obtain a ratio CPrt_bhe1 of intersections that are in the same direction as the position CP_bhe1 of the farthest intersection among a plurality of intersections included in the intersection group in the image bhe1 of the direction symbol.

[0058] A fourth example of the ratio will be described below. The direction processing unit 112 may obtain a ratio CPrt_bhe2 of intersections that are in the same direction as the position CP_bhe2 of the farthest intersection among a plurality of intersections included in the intersection group in the image bhe2 of the direction symbol.

[0059] The direction processing unit 112 estimates a direction in the image Im based on the two or more positions that have been obtained (step S23). In step S23, for example, the direction processing unit 112 estimates a direction in the image Im based on a geometric calculation using two or more positions. The direction processing unit 112 obtains one or more values, exemplified below, based on a geometric calculation using two or more positions. The direction processing unit 112 estimates a direction in the image Im according to a rule based on the one or more values ​​that have been obtained. For example, the rule is a weighting rule that generates a neural network. Here, the direction processing unit 112 estimates the direction of north in the image Im as the direction in the image Im. The direction processing unit 112 stores data related to the estimated direction in the image Im in the processing data storage area 132. Below, several examples of values ​​obtained by the direction processing unit 112 will be described, but the values ​​obtained by the direction processing unit 112 are not limited to those exemplified here.

[0060] When two or more positions include the center position ICC and the position NC of the characters indicating the direction, the direction processing unit 112 obtains the value of the angle between the center position ICC and the position NC of the characters indicating the direction. The angle between the center position ICC and the position NC of the characters indicating the direction is an angle formed by a vector from the center position ICC toward the position NC of the characters indicating the direction and a vector parallel to the y-axis, which is in the clockwise direction from the bottom to the top of the image Im, with the positive angle. Hereinafter, the vector parallel to the y-axis, which is in the bottom to the top of the image Im, is also referred to as a reference vector. The vector from the center position ICC toward the position NC of the characters indicating the direction tends to indicate the direction of north in the image Im. The direction of north in the image Im differs depending on the value of the angle between the center position ICC and the position NC of the characters indicating the direction.

[0061] When two or more positions include the center position ICC and the center position GC, the direction processing unit 112 calculates the angle between the center position ICC and the center position GC. The angle between the center position ICC and the center position GC is an angle between the vector from the center position ICC toward the center position GC and a reference vector, with the clockwise direction being positive. The vector from the center position ICC toward the center position GC tends to indicate the direction of north in the image Im. The direction of north in the image Im varies depending on the value of the angle between the center position ICC and the center position GC.

[0062] When two or more positions include the center position ICC and the farthest position FP_fill, the direction processing unit 112 calculates the value of the angle between the center position ICC and the farthest position FP_fill. The angle between the center position ICC and the farthest position FP_fill is an angle between the vector from the center position ICC toward the farthest position FP_fill and a reference vector, with the clockwise direction being positive. The vector from the center position ICC toward the farthest position FP_fill tends to indicate the direction of north in the image Im. The direction of north in the image Im differs depending on the value of the angle between the center position ICC and the farthest position FP_fill.

[0063] When two or more positions include the center position ICC and the farthest position FP_bh, the direction processing unit 112 calculates the angle between the center position ICC and the farthest position FP_bh. The angle between the center position ICC and the farthest position FP_bh is an angle between the vector from the center position ICC toward the farthest position FP_bh and a reference vector, with the clockwise direction being positive. The vector from the center position ICC toward the farthest position FP_bh tends to indicate the direction of north in the image Im. The direction of north in the image Im varies depending on the value of the angle between the center position ICC and the farthest position FP_bh.

[0064] When two or more positions include the center position ICC and the farthest position FP_bhe1, the direction processing unit 112 calculates the angle between the center position ICC and the farthest position FP_bhe1. The angle between the center position ICC and the farthest position FP_bhe1 is an angle between the vector from the center position ICC toward the farthest position FP_bhe1 and a reference vector, with the clockwise direction being positive. The vector from the center position ICC toward the farthest position FP_bhe1 tends to indicate the direction of north in the image Im. The direction of north in the image Im varies depending on the value of the angle between the center position ICC and the farthest position FP_bhe1.

[0065] When two or more positions include the center position ICC and the farthest position FP_bhe2, the direction processing unit 112 calculates the angle between the center position ICC and the farthest position FP_bhe2. The angle between the center position ICC and the farthest position FP_bhe2 is an angle between the vector from the center position ICC to the farthest position FP_bhe2 and the reference vector, with the clockwise direction being positive. The vector from the center position ICC to the farthest position FP_bhe2 tends to indicate the direction of north in the image Im. The direction of north in the image Im varies depending on the value of the angle between the center position ICC and the farthest position FP_bhe2.

[0066] When two or more positions include the center position ICC and the farthest intersection position CP_fill, the direction processing unit 112 obtains the value of the angle between the center position ICC and the farthest intersection position CP_fill. The angle between the center position ICC and the farthest intersection position CP_fill is an angle between the vector from the center position ICC toward the farthest intersection position CP_fill and the reference vector, with the clockwise direction being positive. The vector from the center position ICC toward the farthest intersection position CP_fill tends to indicate the direction of north in the image Im. The direction of north in the image Im varies depending on the value of the angle between the center position ICC and the farthest intersection position CP_fill.

[0067] When two or more positions include the center position ICC and the farthest intersection position CP_bh, the direction processing unit 112 obtains the value of the angle between the center position ICC and the farthest intersection position CP_bh. The angle between the center position ICC and the farthest intersection position CP_bh is an angle between the vector from the center position ICC toward the farthest intersection position CP_bh and the reference vector, with the clockwise direction being positive. The vector from the center position ICC toward the farthest intersection position CP_bh tends to indicate the direction of north in the image Im. The direction of north in the image Im varies depending on the value of the angle between the center position ICC and the farthest intersection position CP_bh.

[0068] When two or more positions include the center position ICC and the farthest intersection position CP_bhe1, the direction processing unit 112 obtains the value of the angle between the center position ICC and the farthest intersection position CP_bhe1. The angle between the center position ICC and the farthest intersection position CP_bhe1 is an angle between the vector from the center position ICC toward the farthest intersection position CP_bhe1 and the reference vector, with the clockwise direction being positive. The vector from the center position ICC toward the farthest intersection position CP_bhe1 tends to indicate the direction of north in the image Im. The direction of north in the image Im varies depending on the value of the angle between the center position ICC and the farthest intersection position CP_bhe1.

[0069] When two or more positions include the center position ICC and the farthest intersection position CP_bhe2, the direction processing unit 112 calculates the angle between the center position ICC and the farthest intersection position CP_bhe2. The angle between the center position ICC and the farthest intersection position CP_bhe2 is an angle between the vector from the center position ICC to the farthest intersection position CP_bhe2 and the reference vector, with the clockwise direction being positive. The vector from the center position ICC to the farthest intersection position CP_bhe2 tends to indicate the direction of north in the image Im. The direction of north in the image Im varies depending on the value of the angle between the center position ICC and the farthest intersection position CP_bhe2.

[0070] The direction processing unit 112 may use the value of the ratio CPrt_fill to estimate the direction in the image. The ratio CPrt_fill is a value calculated based on the position CP_fill of the farthest intersection, which is calculated based on the position ICC of the center. Therefore, the ratio CPrt_fill is an example of a value calculated based on a geometric calculation using two or more positions. The ratio CPrt_fill is a value estimating the degree of certainty that the vector from the position ICC of the center to the position CP_fill of the farthest intersection indicates the direction of north in the image Im. For example, in the case of a direction symbol having multiple vertices along the circumference of a circle, the intersections of the straight line components are dispersed with respect to the position ICC of the center, and the ratio of intersections that happen to be in the same direction as the intersection that is the farthest is low. The degree of certainty that the vector from the position ICC of the center to the position CP_fill of the farthest intersection indicates the direction of north in the image Im varies depending on the value of the ratio CPrt_fill.

[0071] The direction processing unit 112 may use the value of the ratio CPrt_bh to estimate the direction in the image. The ratio CPrt_bh is a value calculated based on the position CP_bh of the farthest intersection point, which is calculated based on the position ICC of the center. Therefore, the ratio CPrt_bh is an example of a value calculated based on a geometric calculation using two or more positions. The ratio CPrt_bh is a value estimating the degree of certainty that the vector from the position ICC of the center to the position CP_bh of the farthest intersection point indicates the direction of north in the image Im. The degree of certainty that the vector from the position ICC of the center to the position CP_bh of the farthest intersection point indicates the direction of north in the image Im varies depending on the value of the ratio CPrt_bh.

[0072] The direction processing unit 112 may use the value of the ratio CPrt_bhe1 to estimate the direction in the image. The ratio CPrt_bhe1 is a value obtained based on the position CP_bhe1 of the farthest intersection, which is a position obtained based on the position ICC of the center. Therefore, the ratio CPrt_bhe1 is an example of a value obtained based on a geometric calculation using two or more positions. The ratio CPrt_bhe1 is a value obtained by estimating the degree of certainty that the vector from the position ICC of the center to the position CP_bhe1 of the farthest intersection indicates the direction of north in the image Im. The degree of certainty that the vector from the position ICC of the center to the position CP_bhe1 of the farthest intersection indicates the direction of north in the image Im varies depending on the value of the ratio CPrt_bhe1.

[0073] The direction processing unit 112 may use the value of the ratio CPrt_bhe2 to estimate the direction in the image. The ratio CPrt_bhe2 is a value obtained based on the position CP_bhe2 of the farthest intersection, which is a position obtained based on the position ICC of the center. Therefore, the ratio CPrt_bhe2 is an example of a value obtained based on a geometric calculation using two or more positions. The ratio CPrt_bhe2 is a value obtained by estimating the degree of certainty that the vector from the position ICC of the center to the position CP_bhe2 of the farthest intersection indicates the direction of north in the image Im. The degree of certainty that the vector from the position ICC of the center to the position CP_bhe2 of the farthest intersection indicates the direction of north in the image Im varies depending on the value of the ratio CPrt_bhe2.

[0074] The direction processor 112 may obtain a ratio value between the vertical and horizontal components of the linear distance between the center position ICC and another position different from the center position ICC and the vertical and horizontal sizes of the image org. The ratio value is an index value indicating how far the other values ​​are from the center position ICC in the image org. The direction of north in the image Im varies depending on the ratio value.

[0075] When two or more positions include the center position ICC and the position NC of the character indicating the direction, the direction processing unit 112 calculates the value of the ratio of the vertical and horizontal components of the straight-line distance between the center position ICC and the position NC of the character indicating the direction to the vertical and horizontal size of the image org. When two or more positions include the center position ICC and the position GC of the center of gravity, the direction processing unit 112 calculates the value of the ratio of the vertical and horizontal components of the straight-line distance between the center position ICC and the position GC of the center of gravity to the vertical and horizontal size of the image org. When two or more positions include the center position ICC and the position FP_fill of the farthest point, the direction processing unit 112 calculates the value of the ratio of the vertical and horizontal components of the straight-line distance between the center position ICC and the position FP_fill of the farthest point to the vertical and horizontal size of the image org. When two or more positions include the center position ICC and the position FP_bh of the farthest point, the direction processing unit 112 calculates the value of the ratio of the vertical and horizontal components of the straight-line distance between the center position ICC and the position FP_bh of the farthest point to the vertical and horizontal size of the image org. When two or more positions include the center position ICC and the farthest position FP_bhe1, the direction processing unit 112 calculates the ratio of the vertical and horizontal components of the straight-line distance between the center position ICC and the farthest position FP_bhe1 to the vertical and horizontal size of the image org. When two or more positions include the center position ICC and the farthest position FP_bhe2, the direction processing unit 112 calculates the ratio of the vertical and horizontal components of the straight-line distance between the center position ICC and the farthest position FP_bhe2 to the vertical and horizontal size of the image org. Here, when the farthest point becomes smaller due to pre-processing of the image org of the compass symbol, the confidence that the farthest point indicates a northward direction tends to decrease. Therefore, the values ​​of the straight-line distance between the center position ICC and the farthest position FP_fill, the values ​​of the straight-line distance between the center position ICC and the farthest position FP_bh, the values ​​of the straight-line distance between the center position ICC and the farthest position FP_bhe1, and the values ​​of the straight-line distance between the center position ICC and the farthest position FP_bhe2 are used.

[0076] As described above, the server 1 can estimate a direction in the image Im based on two or more positions in the image Im. The server 1 can estimate the direction in the image Im regardless of the shape of the directional symbol because it only needs to use two or more positions in the image Im. The server 1 can estimate the direction in the image Im without increasing the amount of calculation processing because it only needs to use two or more positions in the image Im. Therefore, the server 1 can easily estimate the direction in the image Im that includes the directional symbol.

[0077] As described above, the server 1 can estimate a direction in the image Im based on a geometric calculation using two or more positions in the image Im. The server 1 can estimate the direction in the image Im regardless of the shape of the azimuth symbol because it only needs to perform geometric calculations using two or more positions in the image Im. The server 1 can estimate the direction in the image Im without increasing the amount of calculation processing because it only needs to perform geometric calculations using two or more positions in the image Im.

[0078] As mentioned above, the two or more locations in the image Im may include a center location for the north arrow. By taking the position of the center of the directional sign as a reference, the server 1 can estimate a direction in the image Im using the position of the center of the directional sign and other positions based on the directional sign.

[0079] As mentioned above, the two or more positions in the image Im may include a position relative to the direction of north. The position of the north direction based on the directional sign may be located approximately along the north direction indicated by the directional sign, based on the position of the center of the directional sign. Therefore, the server 1 can estimate the direction in the image Im by using the position of the north direction based on the directional sign.

[0080] An example of an image used in the above-mentioned direction estimation process will be described.

[0081] FIG. 5 is a diagram showing an example of a directional symbol image org. The north arrow image org is an image extracted from image Im.

[0082] FIG. 6 is a diagram showing an example of the position NC of the characters indicating a direction. The position NC of the directional characters is the center position of the area of ​​the directional characters.

[0083] FIG. 7 is a diagram showing an example of an image bw of a direction symbol. The north arrow image bw is a generated image based on the north arrow image org. The convex polygon CP is a convex polygon that contains the directional symbol in the image bw of the directional symbol. The center location ICC is the location of the center of the inscribed circle of the convex polygon CP. The center of gravity location GC is the location of the center of gravity of the convex polygon CP.

[0084] FIG. 8 is a diagram showing an example of the image fill of a direction symbol. The north arrow image fill is an image generated based on the north arrow image bw. The farthest point position FP_fill is the position of the north sign in the north sign image fill that is farthest from the center position ICC.

[0085] FIG. 9 is a diagram showing an example of the image bh of a direction symbol. The image bh of the north sign is an image generated based on the image bw of the north sign. The farthest point position FP_bh is the position of the direction sign that is farthest from the center position ICC in the direction sign image bh.

[0086] FIG. 10 is a diagram showing an example of the image bhe1 of a direction symbol. The image bhe1 of the directional symbol is an image generated based on the image bh of the directional symbol. The farthest point position FP_bhe1 is the position of the direction sign that is farthest from the center position ICC in the direction sign image bhe1.

[0087] FIG. 11 is a diagram showing an example of an image bhe2 of a direction symbol. The north sign image bhe2 is an image generated based on the north sign image bh. The farthest point position FP_bhe2 is the position of the direction sign that is farthest from the center position ICC in the direction sign image bhe2.

[0088] FIG. 12 is a diagram showing an example of the image fill of a direction symbol. The image fill of the directional symbol indicates a plurality of line components in the directional symbol and a group of intersections based on the plurality of line components in the directional symbol. Of the plurality of intersections included in the group of intersections, the position of the intersection farthest from the center position ICC is the position of the farthest intersection CP_fill.

[0089] FIG. 13 is a diagram showing an example of the image bh of a direction symbol. The image bh of the directional symbol shows a plurality of line components in the directional symbol and a group of intersections based on the plurality of line components in the directional symbol. Of the plurality of intersections included in the group of intersections, the position of the intersection farthest from the center position ICC is the position of the farthest intersection CP_bh.

[0090] FIG. 14 is a diagram showing an example of the image bhe1 of a direction symbol. The image bhe1 of the directional symbol shows a plurality of line components in the directional symbol and a group of intersections based on the plurality of line components in the directional symbol. Of the plurality of intersections included in the group of intersections, the position of the intersection farthest from the center position ICC is the position of the farthest intersection CP_bhe1.

[0091] FIG. 15 is a diagram showing an example of an image bhe2 of a direction symbol. The image bhe2 of the directional symbol shows a plurality of line components in the directional symbol and a group of intersections based on the plurality of line components in the directional symbol. Of the plurality of intersections included in the group of intersections, the position of the intersection farthest from the center position ICC is the position of the farthest intersection CP_bhe2.

[0092] FIG. 16 is a flowchart showing an example of the process of estimating the direction of the property by the server 1 in step S3.

[0093] The property processing unit 113 obtains the configuration of the property included in the image Im based on the image Im represented by the image data acquired by the acquisition unit 111 (step S31). In step S31, for example, the property processing unit 113 executes area determination by semantic segmentation based on the image Im represented by the image data. The property processing unit 113 creates an area determination bitmap based on the area determination by semantic segmentation. When the image Im represented by the image data is the first image of the property, the area determination bitmap is the second image of the property based on the image Im. The property processing unit 113 obtains the configuration of the property included in the image Im based on the area determination bitmap. The property processing unit 113 creates a floor plan graph representing the floor plan structure of the property based on the property configuration. The floor plan graph is an image showing each of the multiple elements of the property as a node and showing the nodes as edges. The floor plan graph includes the arrangement relationship between the elements associated with the edges.

[0094] The property configuration indicates how the property is configured. The property configuration includes a layout relationship between one or more sets of elements of the property. A set of elements is composed of two elements. The layout relationship is a layout relationship based on the direction between the elements in the image Im. The multiple elements include a first element and a second element. When the first element and the second element are considered as a set of elements, the layout relationship is a layout relationship based on the direction between the first element and the second element in the image Im.

[0095] In one example, the orientation-based arrangement relationship between a first element and a second element is the arrangement relationship of the second element relative to the first element. The arrangement relationship of the second element relative to the first element indicates the direction from the first element to the second element in the image Im. The direction from the first element to the second element in the image Im may be any of up, down, left, right, top right, bottom right, top left, or bottom left, or may be a direction other than these. In another example, the orientation-based arrangement relationship between a first element and a second element indicates a mutual direction between the first element and the second element. The mutual direction between the first element and the second element may be either vertical or horizontal, or may be a direction other than these. Vertical is also referred to as up and down. Horizontal is also referred to as left and right. The arrangement relationship between elements may be an arrangement relationship between elements of the same type, or may be an arrangement relationship between elements of different types. The type of element does not matter.

[0096] The arrangement relationship between one or more sets of elements may include one or both of an arrangement relationship between one or more sets of doors and an area and an arrangement relationship between one or more sets of windows and an area. The arrangement relationship between the door and an area is an arrangement relationship based on a direction between the door and an area. For example, the arrangement relationship between the door and an area is an arrangement relationship based on a direction from an area to a long side of the door between adjacent doors and areas. The arrangement relationship based on a direction between the door and an area includes an arrangement relationship based on a direction between a long side of the door and an area. The arrangement relationship based on a direction between the door and an area is an arrangement relationship of the door with respect to an area. The arrangement relationship of the door with respect to an area includes an arrangement relationship of the long side of the door with respect to an area. The arrangement relationship based on a direction between the door and an area indicates a direction from an area to the door. The direction from an area to the door includes a direction from an area to the long side of the door. The arrangement relationship between a window and an area is an arrangement relationship based on a direction between a window and an area. For example, the arrangement relationship between a window and an area is an arrangement relationship based on a direction from an area to a long side of the window between adjacent windows and areas. The orientation-based positional relationship between the window and the region includes an orientation-based positional relationship between a long side of the window and the region. The orientation-based positional relationship between the window and the region is a positional relationship of the window with respect to the region. The orientation relationship of the window with respect to the region includes a positional relationship of the long side of the window with respect to the region. The orientation-based positional relationship between the window and the region indicates a direction from the region to the window. The direction from the region to the window includes a direction from the region to the long side of the window.

[0097] The relationship between one or more elements may include a relationship between one or more rooms. For example, the relationship between the rooms may be a relationship between the centers of the rooms. The relationship between the centers of the rooms may indicate a mutual orientation between the rooms.

[0098] The positional relationship between one or more sets of elements is a positional relationship between one or more sets of a hallway and a room. For example, the positional relationship between a hallway and a room is a positional relationship between a center of the hallway and a center of the room. The positional relationship between the center of the hallway and the center of the room may indicate a direction from the hallway to the room or a direction from the room to the hallway.

[0099] The object processing unit 113 estimates a direction of the object included in the image Im based on the direction in the image Im (step S32). In step S32, for example, the object processing unit 113 estimates a direction of the object based on the direction in the image Im and the configuration of the object included in the image Im. The direction processing unit 112 stores data related to the estimated direction of the object included in the image Im in the processed data storage area 132.

[0100] In one example, the direction in the image Im is the north direction in the image Im. The configuration of the property includes the positional relationship between the window and the area. The positional relationship between the window and the area indicates the direction from the living / dining / kitchen area to the window leading to the outdoors. In this example, the direction processing unit 112 can estimate the direction in the property where the main window, such as the window leading to the outdoors, is located by comparing the positional relationship between the window and the area with the north direction in the image Im.

[0101] In another example, the direction in the image Im is the north direction in the image Im. The configuration of the property includes the layout relationship between the entrance and the rooms. The layout relationship between the entrance and the rooms indicates the direction from the living, dining, and kitchen area to the entrance. In this example, the direction processing unit 112 can estimate the direction in which the entrance is located as a direction in the property by comparing the layout relationship between the entrance and the rooms with the north direction in the image Im.

[0102] As described above, the server 1 can estimate the direction of an object included in the image Im based on the direction in the image Im. By referring to the direction in the image Im, the server 1 can estimate the direction of the property that cannot be estimated from the configuration of the property alone. For example, the direction of the property is information that users can use as a reference when selecting a property. The direction in which the main window, such as a window leading to the outdoors, is located is useful information for users who place importance on sunlight, such as a south-facing direction. The direction in which the entrance is located is useful information for users who place importance on feng shui.

[0103] [Other embodiments] In the above embodiment, an example has been described in which the image of the property is an image of a floor plan of the property, but the image of the property is not limited to this. The image of the property may be various images related to the property, such as an exterior view or a blueprint of a building including the property.

[0104] In the above embodiment, an example has been described in which the image data represents an image including an image of a property and an image of a directional sign, but the present invention is not limited to this. The image data may be data representing an image including at least an image of a directional sign. The image represented by the image data does not have to include an image of the property. Therefore, the above embodiment is applicable to fields other than the technical field of real estate. In this case, the process of step S3 is omitted.

[0105] In the above embodiment, an example in which the auxiliary storage device 13 of the server 1 stores various data has been described, but the present invention is not limited to this. A server different from the server 1 may store various data instead of the server 1. A plurality of servers different from the server 1 may store various data in a distributed manner instead of the server 1.

[0106] The expression "one or more" in the above-described embodiment intends one or more, and may be read as "one" or "multiple."

[0107] The expression "two or more" in the above-described embodiment intends two or three or more, and may be read as "two" or "three or more."

[0108] In the above embodiment, the arrangement relationship between rooms and the arrangement relationship between the entrance and the room have been described as examples of the arrangement relationship between areas, but the present invention is not limited to these. The arrangement relationship between areas may be an arrangement relationship between areas of the same type, or an arrangement relationship between areas of different types. The type of area is not important. The arrangement relationship between one or more sets of elements may include an arrangement relationship between one or more sets of areas.

[0109] Although the image processing device has been described by taking the server 1 as an example, the image processing device is not limited to this. The image processing device may be realized by an electronic device having the same functions as the server 1. The electronic device may be a terminal such as a PC (Personal Computer), a smartphone, or a tablet terminal.

[0110] The image processing device may be realized by one device such as the server 1 described in the above embodiment, or may be realized by a plurality of devices with distributed functions.

[0111] The above-described embodiment may be applied not only to the device but also to a method executed by the device. The above-described embodiment may be applied to a program capable of causing a computer of the device to execute each function. The above-described embodiment may be applied to a recording medium for storing the program.

[0112] Each of the one or more circuits constituting the processing circuit executes one or more processes among the multiple processes. When the processing circuit is composed of a single circuit, the single circuit executes all of the multiple processes. When the processing circuit is composed of multiple circuits, each of the multiple circuits executes a part of the multiple processes. The part of the multiple processes may be one of the multiple processes, or two or more of the multiple processes. When the processing circuit is composed of multiple circuits, the multiple circuits may be included in one device, or may be distributed across multiple devices.

[0113] The program may be transferred in a state where it is stored in the device according to the embodiment, or in a state where it is not stored in the device. In the latter case, the program may be transferred via a network, or in a state where it is recorded on a recording medium. The recording medium is a non-transitory tangible medium. The recording medium is a computer-readable medium. The recording medium may be in any form, such as a CD-ROM or a memory card, as long as it is capable of storing the program and is readable by a computer.

[0114] In short, this invention is not limited to the present embodiment as it is, and the components can be modified and embodied in the implementation stage without departing from the gist of the invention. In addition, various inventions can be formed by appropriately combining multiple components disclosed in this embodiment. For example, some components may be deleted from all the components shown in this embodiment. Furthermore, components from different embodiments may be appropriately combined.

[0115] Some of the above embodiments may be expressed as follows. [1] Extract north arrows from images, determining two or more locations in the image based on the north arrow; estimating a direction in the image based on the two or more positions; An image processing device comprising a direction processing unit. [2] The image processing device described in [1], wherein the direction processing unit estimates a direction in the image based on a geometric calculation using the two or more positions. [3] The image processing device according to [1] or [2], wherein the two or more positions include a center position of the compass symbol. [4] An image processing device described in any of [1] to [3], wherein the two or more positions include a position relative to the north direction. [5] The image processing device according to any one of [1] to [4], further comprising an object processing unit that estimates a direction of an object included in the image based on a direction in the image. [6] Extracting north arrows from images; determining two or more locations in the image based on the north arrow; estimating a direction in the image based on the two or more positions; An image processing method comprising: [7] An image processing program capable of causing a computer to execute the processing of each part of the image processing device described in any one of [1] to [5]. [Explanation of symbols]

[0116] 1...server, 11...processing circuit, 12...main memory, 13...auxiliary storage device, 14...communication interface, 111...acquisition unit, 112...direction processing unit, 113...object processing unit, 131...image storage area, 132...processed data storage area.

Claims

1. Extract north arrow from image, determining two or more locations in the image based on the north arrow; estimating a direction in the image based on the two or more positions; An image processing device comprising a direction processing unit.

2. The image processing device according to claim 1 , wherein the direction processing unit estimates the direction in the image based on a geometric calculation using the two or more positions.

3. The image processing device of claim 1 , wherein the two or more locations include a center location for the north arrow.

4. The image processing device according to claim 1 , wherein the two or more positions include a position relative to a north direction.

5. The image processing device according to claim 1 , further comprising an object processing unit that estimates a direction of an object included in the image based on a direction in the image.

6. Extracting a north arrow from the image; determining two or more locations in the image based on the north arrow; estimating a direction in the image based on the two or more positions; An image processing method comprising:

7. 6. An image processing program capable of causing a computer to execute processing of each unit included in the image processing device according to claim 1.

Images