Image generating device, image generating method, program, and image generating system

The image generation device uses three-dimensional smoke models to synthesize light attenuation and scattered light components with captured images, addressing high processing loads in fire detection systems and creating realistic smoke images efficiently.

JP2024157180APending Publication Date: 2024-11-07HOCHIKI CORP
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
JP2023071358
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-04-25
Publication Date
2024-11-07

AI Technical Summary

Technical Problem

Existing fire detection systems that use three-dimensional smoke models face high processing loads, making it difficult to generate realistic smoke images without requiring high-spec devices.

Method used

An image generation device that acquires a captured image and uses three-dimensional smoke models to generate light attenuation and scattered light components, synthesizing these with the captured image to create a composite image that realistically depicts smoke without excessive processing load.

Benefits of technology

The system effectively generates a smoke image that appears natural while reducing processing demands, allowing for efficient fire detection without the need for high-performance hardware.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024157180000001_ABST
    Figure 2024157180000001_ABST
Patent Text Reader

Abstract

To generate a smoke image which does not look unnatural while minimizing a processing load.SOLUTION: An image generating device comprises: an acquisition unit for acquiring an image in which a target region is imaged; a storage unit for storing an attenuation component image for representing an attenuation component of light emitted from behind a smoke model by using three-dimensional information of the smoke model having a three-dimensional shape, and a scattering light component image for representing a scattering light component resulting from light emitted to the smoke model and reflected by the smoke model; and a synthesis unit that generates a synthesis image by synthesizing a smoke image with the imaging image using the attenuation component image, the scattering light component image, and the imaging image.SELECTED DRAWING: Figure 2
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an image generating device, an image generating method, a program, and an image generating system. [Background technology]

[0002] There is a fire detection system that detects fires using images. For example, Patent Document 1 discloses a method of generating a composite image in which a non-fire image and a three-dimensional smoke model are placed in a virtual space, and the smoke model is rotated to create a smoke image that does not look unnatural. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-149378 Summary of the Invention [Problem to be solved by the invention]

[0004] However, because Patent Document 1 uses a three-dimensional shape model, the processing load is large, and it is difficult to generate a smoke image unless the device has high specifications. Therefore, there is a problem that the processing load increases in order to generate a smoke image that does not look unnatural.

[0005] The present invention has been made in consideration of the above circumstances, and its purpose is to provide an image generation device, an image generation method, and a program that can generate smoke images that do not appear unnatural while reducing the processing load. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems, one embodiment of the present invention is an image generating device that includes an acquisition unit that acquires an image of a target area, a memory unit that uses three-dimensional information of a smoke model having a three-dimensional shape to store an attenuation component image in which light irradiated from behind the smoke model is attenuated by the smoke model, and a scattered light component image in which the scattered light component of light irradiated to the smoke model is reflected by the smoke model, and a synthesis unit that uses the attenuation component image, the scattered light component image, and the captured image to generate a synthetic image by synthesizing an image of smoke with the captured image.

[0007] Moreover, one embodiment of the present invention is an image generation method performed by an image generating device that is a computer having a memory unit that uses three-dimensional information of a smoke model having a three-dimensional shape to store an attenuation component image in which light irradiated from behind the smoke model is attenuated by the smoke model, and a scattered light component image in which the scattered light component of light irradiated to the smoke model is reflected by the smoke model, in which an acquisition unit acquires an captured image in which a target area is captured, and a synthesis unit uses the attenuation component image and the scattered light component image stored in the memory unit, and the captured image acquired by the acquisition unit, to generate a synthetic image in which an image of smoke is synthesized with the captured image.

[0008] Moreover, one embodiment of the present invention is a program that uses three-dimensional information of a smoke model having a three-dimensional shape to cause an image generating device, which is a computer having a memory unit that stores an attenuation component image in which light irradiated from behind the smoke model is attenuated by the smoke model, and a scattered light component image in which the scattered light component of light irradiated to the smoke model is reflected by the smoke model, to acquire an image in which a target area is captured, and generates a composite image by combining an image of smoke with the captured image, using the attenuation component image and the scattered light component image stored in the memory unit, and the acquired captured image.

[0009] Moreover, one embodiment of the present invention is an image generation system that includes an acquisition unit that acquires an image of a target region, a generation unit that uses three-dimensional information of a smoke model having a three-dimensional shape to generate an attenuation component image in which light irradiated from behind the smoke model is attenuated by the smoke model, and a scattered light component image in which a scattered light component of light irradiated to the smoke model is reflected by the smoke model, and a synthesis unit that uses the attenuation component image, the scattered light component image, and the captured image to generate a synthetic image by synthesizing an image of smoke with the captured image. Effect of the Invention

[0010] As described above, according to the present invention, it is possible to generate a smoke image that does not look unnatural while suppressing the processing load. [Brief description of the drawings]

[0011] [Figure 1] 1 is a diagram illustrating an example of the configuration of a fire detection system 1 according to an embodiment. [Diagram 2] 1 is a block diagram showing an example of the configuration of an image generating device 10 according to an embodiment. [Diagram 3] 2 is a diagram for explaining a process performed by the image generating device 10 according to the embodiment. FIG. [Figure 4] 2 is a diagram for explaining a process performed by the image generating device 10 according to the embodiment. FIG. [Diagram 5] FIG. 2 is a diagram showing an example of a dimming component image A according to the embodiment. [Figure 6] FIG. 13 is a diagram showing an example of a scattered light component image B according to the embodiment. [Figure 7A] 2 is a diagram for explaining a process performed by the image generating device 10 according to the embodiment. FIG. [Figure 7B] 2 is a diagram for explaining a process performed by the image generating device 10 according to the embodiment. FIG. [Figure 7C] 2 is a diagram for explaining a process performed by the image generating device 10 according to the embodiment. FIG. [Figure 8] 4 is a diagram showing an example of information stored in a synthesis coefficient storage unit 104 according to the embodiment. FIG. [Figure 9] 4 is a flowchart showing the flow of processing performed by the image generating device 10 according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0013] (Fire Detection System 1) This is a system that monitors a target area T and detects a fire that has occurred in the target area T. The target area T is an area or object that is the target of monitoring by the fire detection system 1. The target area T may be set arbitrarily. For example, the target area T is a roadway, a tunnel premises, a building, a building premises, etc.

[0014] 1 is a diagram showing an example of the configuration of a fire detection system 1 according to an embodiment. The fire detection system 1 includes, for example, a camera CA, an image generating device 10, an operation control device 20, and a smoke detection device 30.

[0015] The camera CA and the image generating device 10 are communicatively connected. The camera CA and the smoke detection device 30 are communicatively connected via SW1. The image generating device 10 and the operation control device 20 are communicatively connected. The operation control device 20 and the smoke detection device 30 are communicatively connected via SW2.

[0016] Communication in the fire detection system 1 is performed, for example, via a communication network, short-distance wireless communication using a wireless LAN (Local Area Network), or a USB (Universal Serial Bus) cable. The camera CA and the image generating device 10 may be provided integrally. That is, in the fire detection system 1, the camera CA may be built into the image generating device 10. Similarly, in the fire detection system 1, the camera CA may be built into the smoke detection device 30.

[0017] The fire detection system 1 has a fire detection mode and an operation confirmation mode. The fire detection mode is a mode for detecting a fire in a target area T. In the fire detection mode, the camera CA and the smoke detection device 30 are connected by a switch SW1, and a captured image TG captured by the camera CA is output to the smoke detection device 30. The captured image TG is an image of the target area T. The smoke detection device 30 determines whether or not a fire has occurred in the target area T based on the captured image TG, and outputs the determination result as a detection result KK.

[0018] The operation check mode is a mode for checking whether the smoke detection device 30 operates correctly. In the operation check mode, the operation control device 20 and the smoke detection device 30 are connected by the switch SW1. The image generation device 10 acquires a captured image TG captured by the camera CA, and generates a composite image GG based on the acquired captured image TG. The composite image GG is an image in which a smoke image (an image showing smoke) is composited with the captured image TG. That is, the composite image GG is an image expressing a state in which smoke has occurred in the target area T. The image generation device 10 outputs the generated composite image GG to the smoke detection device 30 via the operation control device 20. The smoke detection device 30 determines whether a fire has occurred in the target area T based on the composite image GG, and outputs the determination result as a detection result KK. In the operation check mode, the operation control device 20 and the smoke detection device 30 are connected by the switch SW2, and the detection result KK output from the smoke detection device 30 is output to the operation control device 20. The operation control device 20 determines, based on the detection result output from the smoke detection device 30, whether or not the smoke detection device 30 has operated correctly.

[0019] The camera CA is an imaging device that captures an image of the target area T. The camera CA may capture moving images or still images.

[0020] The image generating device 10 is a computer, and is realized by, for example, a PC (Personal Computer), a server device, etc. The image generating device 10 acquires a captured image TG from a camera CA, and generates a composite image GG from the captured image TG in an operation check mode. A method in which the image generating device 10 generates the composite image GG will be described in detail later.

[0021] The operation control device 20 is a computer, and is realized by, for example, a PC (Personal Computer), a server device, or the like. The operation control device 20 controls the switches SW1 and SW2 depending on the mode (fire detection mode or operation check mode). Furthermore, the operation control device 20 determines whether the smoke detection device 30 has operated correctly in the operation check mode. Any method may be adopted as a specific method for the operation control device 20 to determine whether the smoke detection device 30 has operated correctly, and a detailed description thereof will be omitted in this embodiment.

[0022] The smoke detection device 30 is a computer, and is realized by, for example, a PC (Personal Computer), a server device, etc. In a fire detection mode, the smoke detection device 30 acquires a captured image TG from a camera CA, and determines whether or not something that may indicate the occurrence of a fire (here, smoke) is displayed in the target area T based on the acquired captured image TG. Any method may be adopted as a specific method for the smoke detection device 30 to detect a fire, and a detailed description thereof will be omitted in this embodiment.

[0023] (Regarding image generating device 10) 2 is a block diagram showing an example of the configuration of the image generating device 10 according to the embodiment. The image generating device 10 includes, for example, an image acquiring unit 100, a generating unit 101, a synthesizing unit 102, a three-dimensional smoke model information storage unit 103, a synthesis coefficient storage unit 104, and an image information storage unit 105.

[0024] The image acquisition unit 100 acquires image information of a captured image TG captured by a camera CA. The image acquisition unit 100 outputs the acquired image information to the generation unit 101.

[0025] The generating unit 101 generates a smoke image. The generating unit 101 generates the smoke image using a three-dimensional smoke model ML. The three-dimensional smoke model ML is a model showing a three-dimensional shape of smoke. The three-dimensional smoke model ML is presumed to be created in advance, and for example, three-dimensional information of the three-dimensional smoke model ML is stored in the three-dimensional smoke model information storage unit 103. The method of generating the three-dimensional smoke model ML is arbitrary. For example, the three-dimensional smoke model ML may be generated by forming a virtual three-dimensional shape of smoke using a known image processing device or image processing software, or three-dimensional information showing the shape of smoke may be acquired by downloading an existing three-dimensional model.

[0026] The generating unit 101 generates, as smoke images, an attenuation component image A and a scattered light component image B. The attenuation component image A is an image showing a state in which light is attenuated by being blocked by the three-dimensional smoke model ML. The attenuation component image A can express the shading and three-dimensional depth of the smoke in two dimensions. The scattered light component image B is an image showing a state in which light irradiated on the three-dimensional smoke model ML is reflected (scattered). The scattered light component image B can express the texture of the surface of the smoke in two dimensions.

[0027] By using both the light attenuation component image A and the scattered light component image B as a smoke image, it is possible to express the shading, three-dimensional depth, and texture of the smoke, making it possible to express smoke that is realistic and does not look unnatural. Moreover, since an already created three-dimensional model is used, there is no need to create a new three-dimensional smoke model ML or generate a three-dimensional space in which to place the three-dimensional smoke model ML. Therefore, the fire detection system 1 of this embodiment can generate a smoke image that does not look unnatural while suppressing the processing load.

[0028] Here, a method for generating a smoke image by the generation unit 101 will be described with reference to Fig. 3 and Fig. 4. Fig. 3 and Fig. 4 are diagrams for explaining the process performed by the image generation device 10 according to the embodiment.

[0029] Fig. 3 shows a schematic diagram of a virtual space for generating the dimming component image A. As shown in Fig. 3, a virtual camera VC, a three-dimensional smoke model ML, and a virtual surface VS1 are arranged in this order in the virtual space. The virtual surface VS1 is a virtual surface that irradiates uniform light, and is, for example, a virtual surface light source.

[0030] The generating unit 101 causes light from a virtual surface VS1 to be irradiated from behind the three-dimensional smoke model ML in a virtual space. The generating unit 101 captures a dimming component image A by having the virtual camera VC receive light that is irradiated from the virtual surface VS1 and attenuated by the three-dimensional smoke model ML. The generating unit 101 stores image information of the captured dimming component image A in, for example, the image information storage unit 105. In this way, the generating unit 101 generates the dimming component image A.

[0031] Fig. 4 shows a schematic diagram of a virtual space for generating the scattered light component image B. As shown in Fig. 4, a virtual camera VC, a virtual light source VL, a three-dimensional smoke model ML, and a virtual surface VS2 are arranged in this order in the virtual space. The virtual light source VL is a virtual light source that reproduces a lighting environment equivalent to that of the target area T. The virtual light source VL is arranged, for example, at a position and in an orientation equivalent to that of the lighting in the target area T, and has a directivity equivalent to that of the lighting in the target area T. The virtual surface VS2 is a virtual surface that does not reflect light.

[0032] The generation unit 101 illuminates the three-dimensional smoke model ML in the virtual space with the same lighting environment as the target area T. In this way, only the light reflected by the three-dimensional smoke model ML reaches the virtual camera VC, and the light reflected by objects other than the three-dimensional smoke model ML does not reach the virtual camera VC. The generation unit 101 captures a scattered light component image B by having the virtual camera VC receive only the light reflected by the three-dimensional smoke model ML. The generation unit 101 stores image information of the captured scattered light component image B in, for example, the image information storage unit 105. In this way, the generation unit 101 generates the scattered light component image B.

[0033] Fig. 5 is a diagram showing an example of a dimming component image A according to an embodiment. As shown in Fig. 5, the dimming component image A is an image expressing, for example, a silhouette of smoke. However, in the dimming component image A, the pixel values ​​of the pixels forming the smoke are not necessarily uniform, but have different pixel values.

[0034] Fig. 6 is a diagram showing an example of a scattered light component image B according to an embodiment. As shown in Fig. 6, the scattered light component image B is an image in which, for example, a part of the curved surface of the smoke is expressed. Note that, although the outline of the smoke is shown by a dotted line in Fig. 6, this is for the purpose of making it easier to understand that the white part in the scattered light component image B is a part of the smoke. It does not suggest that the outline of the smoke is shown by a dotted line in the scattered light component image B.

[0035] The generating unit 101 stores image information of the generated light attenuation component image A and scattered light component image B in an image information storage unit 105 (an example of a storage unit). The generating unit 101 may store image files integrated into one file in a format such as a png (Portable Network Graphics) file with a transmission channel as the light attenuation component image A and scattered light component image B. The generating unit 101 may store a collection of png images consisting of a plurality of images in time series as the light attenuation component image A and scattered light component image B. The generating unit 101 may store moving image files compressed by applying a standard such as H.264 as the light attenuation component image A and scattered light component image B, provided that the environment is tolerant of image quality degradation due to compression processing. Moreover, the synthesis unit 102 acquires image information of the light attenuation component image A and the scattered light component image B by referring to the image information storage unit 105 .

[0036] Returning to the explanation of FIG. 2, the synthesis unit 102 generates a synthetic image GG. The synthesis unit 102 generates the synthetic image GG by synthesizing a smoke image (light attenuation component image A and scattered light component image B) with a captured image TG captured by a camera CA. This makes it possible to generate an image (synthetic image GG) that looks as if a fire is occurring in the target region T.

[0037] Here, a method for the synthesis unit 102 to generate a synthetic image GG will be described with reference to Fig. 7 (Figs. 7A to 7C). Fig. 7 is a diagram for explaining the process performed by the image generating device 10 according to the embodiment.

[0038] 7A shows an example of a captured image TG. In the captured image TG, an object in a target region T is captured.

[0039] 7B, first, the synthesis unit 102 generates a synthesis image G1 by synthesizing the captured image TG with the shading component image A. Specifically, the synthesis unit 102 synthesizes the shading component image A by using formula (1) or formula (2).

[0040] G1(x, y)=Img(x, y)-(Imax-IA(x, y))×k1 …(1) however, G1(x, y) is the pixel value at coordinates (x, y) in the composite image G1. Img(x, y) is the pixel value at coordinates (x, y) in the captured image TG. Imax is the maximum value that a pixel value can take (for example, 255). IA(x, y) is the pixel value at coordinates (x, y) in the light attenuation component image A. k1 is a coefficient.

[0041] G1(x, y)=Img(x, y)×IA(x, y) / Imax / k1 …(2) however, G1(x, y) is the pixel value at coordinates (x, y) in the composite image G1. Img(x, y) is the pixel value at coordinates (x, y) in the captured image TG. Imax is the maximum value that a pixel value can take (for example, 255). IA(x, y) is the pixel value at coordinates (x, y) in the light attenuation component image A. k1 is a coefficient.

[0042] 7C, next, the synthesis unit 102 generates a synthetic image G2 by synthesizing the synthetic image G1 with the scattered light component image B. Specifically, the synthesis unit 102 synthesizes the scattered light component image B using equation (3).

[0043] G2(x, y)=G1(x, y)+k2×IB(x, y) …(3) however, G1(x, y) is the pixel value at coordinates (x, y) in the composite image G1. G2(x, y) is the pixel value at coordinates (x, y) in the composite image G2. IB(x, y) is the pixel value at coordinates (x, y) in scattered light component image B. k2 is a coefficient.

[0044] The synthesis unit 102 sets the synthetic image G2 as a synthetic image GG. The synthesis unit 102 stores image information of the generated synthetic image GG in, for example, the image information storage unit 105. Furthermore, the synthesis unit 102 outputs the synthetic image GG to the operation control device 20 in the operation check mode.

[0045] The synthesis unit 102 may perform clipping processing when the pixel value calculated using equations (1) to (3) or the like exceeds the upper limit (e.g., 255) or lower limit (e.g., 0) that the pixel can have. For example, when the calculated pixel value exceeds the upper limit, the synthesis unit 102 sets the pixel value as the upper limit. On the other hand, when the calculated pixel value exceeds the lower limit, the synthesis unit 102 sets the pixel value as the lower limit.

[0046] The synthesis unit 102 may also perform a rescaling process. For example, when the calculated pixel value exceeds an upper limit, the synthesis unit 102 sets the pixel value as the upper limit and recalculates the pixel value so that the distribution of all remaining pixel values ​​in the image is a distribution for the newly set upper limit. For example, when the calculated pixel value Ic (e.g., 300) exceeds the upper limit Imax (e.g., 255), the calculated pixel value Ic (300) is changed to the pixel value Imax (255). Also, for all remaining pixels in the image, the pixel value I is multiplied by a constant (Ic / Imax) (e.g., 255 / 300). This makes it possible to recalculate the pixel value so that the distribution of all pixel values ​​in the image is a distribution for the newly set upper limit.

[0047] Here, the coefficients k1 and k2 will be explained. The synthesis unit 102 may arbitrarily set the coefficient k1 in the formulas (1) and (2), and the coefficient k2 in the formula (3). For example, when the coefficient k1 is set to a large value, it is possible to synthesize thick smoke with a high concentration, and when the coefficient k1 is set to a small value, it is possible to synthesize thin smoke with a low concentration. When the coefficient k2 is set to a large value, it is possible to synthesize brightly lit smoke, and when the coefficient k2 is set to a small value, it is possible to synthesize dark smoke that appears in the dark.

[0048] However, it is troublesome to individually adjust the coefficients k1 and k2 every time the operation check mode is performed on-site. It is better to prepare (preset) in advance combinations of the coefficients k1 and k2 for each type of smoke to be synthesized. As a countermeasure, in this embodiment, a synthesis coefficient storage unit 104 is provided, and combinations of the coefficients k1 and k2 are stored for each type of smoke to be synthesized.

[0049] FIG. 8 is a diagram showing an example of information stored in the blending coefficient storage unit 104 according to the embodiment. In the horizontal direction of FIG. 8, three types of smoke thickness are set: "light", "medium", and "dark". "Light" indicates a type with a low smoke thickness. "Medium" indicates a type with a medium smoke thickness. "Dark" indicates a type with a high smoke thickness. In the vertical direction of FIG. 8, three types of lighting brightness in the target region T are set: "dark", "medium", and "light". "Dark" indicates a type with a low illuminance. "Medium" indicates a type with a medium illuminance. "Light" indicates a type with a high illuminance.

[0050] As shown in Fig. 8, the synthesis coefficient storage unit 104 stores the set values ​​of coefficients k1 and k2 for each cell corresponding to the combination of the thickness of the smoke in the horizontal direction and the brightness of the lighting in the vertical direction. For example, if the thickness of the smoke is "light" type and the brightness of the lighting is "dark" type, the set values ​​are indicated as coefficients k1 = 0.5 and k2 = 0.25. If the thickness of the smoke is "medium" type and the brightness of the lighting is "dark" type, the set values ​​are indicated as coefficients k1 = 1.0 and k2 = 0.5. If the thickness of the smoke is "light" type and the brightness of the lighting is "medium" type, the set values ​​are indicated as coefficients k1 = 0.5 and k2 = 0.5.

[0051] When the brightness of the lighting does not change with respect to the density of the smoke, that is, when viewed horizontally, the ratio of coefficients k1 and k2 is constant. Also, for example, when synthesizing smoke with the same lighting conditions as the virtual space used when generating the dimming component image A, k1=k2 is set.

[0052] Returning to the explanation of Fig. 2, the three-dimensional smoke model information storage unit 103 stores three-dimensional information of the three-dimensional smoke model ML. The synthesis coefficient storage unit 104 stores the set values ​​of coefficients k1 and k2 for each type of smoke to be synthesized. The image information storage unit 105 stores image information such as the captured image TG, the synthesis image G1, the synthesis image G2, and the synthesis image GG.

[0053] The storage unit of the image generating device 10 (including the 3D smoke model information storage unit 103, the synthesis coefficient storage unit 104, and the image information storage unit 105) is composed of a storage medium such as a hard disk drive (HDD), a flash memory, an electrically erasable programmable read only memory (EEPROM), a random access read / write memory (RAM), a read only memory (ROM), or a combination of these. The storage unit of the image generating device 10 stores programs for executing various processes in the image generating device 10, and temporary data used when performing various processes.

[0054] Furthermore, the functional units (including the image acquisition unit 100, the generation unit 101, and the synthesis unit 102) of the image generation device 10 are realized by causing a CPU (Central Processing Unit) provided as hardware in the image generation device 10 to execute a program.

[0055] (Processing flow) Here, the flow of processing performed by the image generating device 10 will be described with reference to Fig. 9. Fig. 9 is a flowchart showing the flow of processing performed by the image generating device 10 according to the embodiment. Here, it is assumed that the light attenuation component image A and the scattered light component image B have already been generated.

[0056] The image generating device 10 acquires image information of the captured image TG (step S10). The image generating device 10 sets the density of smoke to be synthesized into the acquired captured image TG (step S11). For example, the image generating device 10 sets which of three types of smoke density, "light", "medium", and "dark", the smoke density should be. The image generating device 10 sets the lighting conditions of the target area T in the acquired captured image TG (step S12). For example, the image generating device 10 sets which of three types of lighting brightness, "dark", "medium", and "light", the lighting brightness should be. The image generating device 10 refers to the synthesis coefficient storage unit 104 and determines the coefficients k1 and k2 according to the combination of the set smoke density and lighting conditions.

[0057] The image generating device 10 generates a composite image G1 by combining the light attenuation component image A with the captured image TG using the coefficient k1 (step S13). Next, the image generating device 10 generates a composite image G2 by combining the scattered light component image B with the composite image G1 using the coefficient k2 (step S14). The image generating device 10 stores the generated composite image GG as the composite image G2 in the image information storage unit 105 (step S15).

[0058] As described above, the image generating device 10 of the embodiment includes an image acquiring unit 100, a generating unit 101, and a synthesizing unit 102. The image acquiring unit 100 acquires a captured image in which a target region is captured. The generating unit 101 generates an attenuation component image A and a scattered light component image B using a three-dimensional smoke model ML (an example of "three-dimensional information of a smoke model"). The attenuation component image A is an image in which light irradiated from behind the three-dimensional smoke model ML is attenuated by the three-dimensional smoke model ML and the attenuation component is expressed as a pixel value. The scattered light component image B is an image in which light irradiated to the three-dimensional smoke model ML is reflected by the three-dimensional smoke model ML and the scattered light component is expressed as a pixel value. The synthesizing unit 102 generates a synthetic image GG using the attenuation component image A, the scattered light component image B, and the captured image TG. The synthetic image GG is an image in which a smoke image is synthesized with the captured image TG. As a result, in the image generating device 10 of the embodiment, in the composite image GG, it is possible to express the shading of the smoke using the light attenuation component image A, and to express the texture of the smoke using the scattered light component image B. Moreover, since the composite image GG is generated by image synthesis, it is not necessary to perform processing that increases the processing load, such as a rendering process for generating a three-dimensional model, and the processing load can be reduced. Therefore, it is possible to generate a smoke image that does not look unnatural while reducing the processing load.

[0059] In the image generating device 10 of the embodiment, the synthesis unit 102 calculates "G1(x, y)" using formula (1) or formula (2). The synthesis unit 102 calculates "G2(x, y)" using formula (3). Here, "G1(x, y)" is an example of a "first synthesis pixel value". In addition, "Img(x, y)" in formula (1) or formula (2) is an example of a "first pixel value". In formula (1), "(Imax-IA(x, y))" is an example of a "first conversion amount". In formula (2), "{(1-Imax-IA(x, y)) / Imax}" is an example of a "first conversion amount". In addition, "G2(x, y)" is an example of a "second synthesis pixel value". In addition, "IB(x, y)" in formula (3) is an example of a "third pixel value". As a result, the image generating device 10 of the embodiment can generate a composite image GG by performing arithmetic operations, and can generate a smoke image that does not look unnatural while suppressing the processing load.

[0060] In the image generating device 10 of the embodiment, the synthesis unit 102 adjusts the density of the smoke to be synthesized by using the coefficient k1 (an example of the "first coefficient") in the formula (1) or (2) and the coefficient k2 (an example of the "second coefficient") in the formula (3). Here, "(Imax-IA(x,y))×k1" in the formula (1) is an example of the "second conversion amount". "IA(x,y) / Imax / k1" in the formula (2) is an example of the "second conversion amount". "k2×IB(x,y)" in the formula (3) is an example of the "fourth pixel value". As a result, in the image generating device 10 of the embodiment, the density of the smoke to be synthesized can be adjusted, and it can be confirmed whether the smoke detection device 30 operates correctly depending on the density of the smoke that the smoke detection device 30 is to detect for various densities of smoke.

[0061] Furthermore, in the image generating device 10 of the embodiment, the synthesis unit 102 generates a synthetic image GG using a coefficient k1 (an example of a "first coefficient") and a coefficient k2 (an example of a "second coefficient") that are determined according to the density of the smoke to be synthesized. As a result, the image generating device 10 of the embodiment can synthesize smoke of various densities into the captured image TG. Therefore, the synthetic image GG, in which smoke of various densities appears to be generated, can be used to cause the smoke detection device 30 to perform fire detection, making it possible to more accurately check whether the smoke detection device 30 is operating correctly.

[0062] Furthermore, in the image generating device 10 of the embodiment, a composite image GG is generated using a coefficient k2 (an example of a "second coefficient") that is determined according to the brightness of the target area T. As a result, in the image generating device 10 of the embodiment, when the brightness of the target area T varies, a composite image GG in which smoke appears to be generated in the target area T can be generated, making it possible to check with greater accuracy whether the smoke detection device 30 is operating correctly.

[0063] Furthermore, the image generating device 10 of the embodiment generates a composite image GG using a coefficient k1 (an example of a "first coefficient") determined according to the density of the smoke to be synthesized, and a coefficient k2 (an example of a "second coefficient") determined according to a combination of the density of the smoke to be synthesized and the brightness of the target area T. As a result, the image generating device 10 of the embodiment can generate a composite image GG in which smoke of various densities appears to be generated when the brightness of the target area T varies, making it possible to more accurately check whether the smoke detection device 30 is operating correctly.

[0064] The components of the image generating device 10 may be distributed among a plurality of devices. For example, the generating unit 101 of the image generating device 10 may be provided in an external device different from the image generating device 10.

[0065] For example, the fire detection system 1 includes an image generating device 10 and an external device. The image generating device 10 and the external device are communicatively connected. The image generating device 10 includes an image acquiring unit 100, a synthesis unit 102, a synthesis coefficient storage unit 104, and an image information storage unit 105. The external device includes a generation unit 101 and a three-dimensional smoke model information storage unit 103. The external device generates a light attenuation component image A and a scattered light component image B by a generation unit 101. The image generating device 10 acquires the attenuation component image A and the scattered light component image B generated by the generation unit 101 of the external device via the image acquisition unit 100, and stores image information of the acquired images in the image information storage unit 105. The synthesis unit 102 of the image generating device 10 generates a synthetic image GG using the attenuation component image A and the scattered light component image B stored in the image information storage unit 105.

[0066] When generating the light attenuation component image A and the scattered light component image B from the three-dimensional smoke model ML, it is necessary to execute signal processing with a high processing load using a GPU (Graphics Processing Unit) etc. By providing the function of the generation unit 101 in the image generating device 10 in an external device, it is possible to reduce the processing load of the image generating device 10.

[0067] In such a fire detection system 1, an external device may be configured to function as a server. In this case, for example, the fire detection system 1 includes an external device and a plurality of image generating devices 10. The external device and the plurality of image generating devices 10 are communicably connected. The external device generates a light attenuation component image A and a scattered light component image B corresponding to each of the image generating devices 10, and outputs the generated light attenuation component image A and scattered light component image B to each of the image generating devices 10. According to this method, signal processing with a high processing load that requires a GPU or the like can be consolidated in the generating unit 101 in the external device. Therefore, the processing load of each of the image generating devices 10 can be reduced, and it becomes possible to use an inexpensive device with a low processing performance as the image generating device 10.

[0068] The fire detection system 1 and the image generating device 10 in the above-mentioned embodiment may be realized in whole or in part by a computer. In that case, a program for realizing the function may be recorded in a computer-readable recording medium, and the program recorded in the recording medium may be read into a computer system and executed to realize the function. The term "computer system" as used herein includes hardware such as an OS and peripheral devices. The term "computer-readable recording medium" refers to portable media such as flexible disks, optical magnetic disks, ROMs, and CD-ROMs, and storage devices such as hard disks built into a computer system. The term "computer-readable recording medium" may also include a medium that dynamically holds a program for a short period of time, such as a communication line when a program is transmitted via a network such as the Internet or a communication line such as a telephone line, and a medium that holds a program for a certain period of time, such as a volatile memory inside a computer system that is a server or client in that case. The above-mentioned program may be a program for realizing a part of the above-mentioned function, or may be a program that can realize the above-mentioned function in combination with a program already recorded in the computer system, or may be a program that is realized using a programmable logic device such as an FPGA.

[0069] Although an embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and designs that do not deviate from the gist of the present invention are also included. [Explanation of symbols]

[0070] 1. Fire detection system 10...Image generating device 100...Image acquisition unit (acquisition unit) 101...Generation section 102…Synthesis section 103…Three-dimensional smoke model information storage unit 104...synthetic coefficient memory unit 105...Image information storage unit (storage unit)

Claims

1. an acquisition unit that acquires an image of a target region; a storage unit that uses three-dimensional information of a smoke model having a three-dimensional shape to store an attenuation component image that expresses an attenuation component of light irradiated from behind the smoke model and attenuated by the smoke model, and a scattered light component image that expresses a scattered light component of light irradiated to the smoke model and reflected by the smoke model; a synthesis unit that generates a synthetic image by synthesizing an image of smoke with the captured image, using the light attenuation component image, the scattered light component image, and the captured image; An image generating device comprising:

2. the combining unit converts a first pixel value at a first coordinate in the captured image into a second combined pixel value obtained by adding a third pixel value at the position coordinate in the scattered light component image to a first combined pixel value calculated by reflecting a first conversion amount based on a second pixel value at a position coordinate corresponding to the first coordinate in the light attenuation component image to the first pixel value, and generates, as the combined image, an image obtained by converting the first pixel value into a second combined pixel value obtained by adding a third pixel value at the position coordinate in the scattered light component image to the first combined pixel value.

2. The image generating device of claim 1.

3. the synthesis unit converts the first pixel value into the second synthesis pixel value obtained by adding a fourth pixel value obtained by multiplying the third pixel value by a second coefficient to the first synthesis pixel value calculated by reflecting a second pixel amount obtained by multiplying the first conversion amount by a first coefficient on the first pixel value, and generates, as the synthetic image, an image; The first coefficient and the second coefficient are values ​​determined according to the density of the smoke to be synthesized into the captured image.

3. The image generating device of claim 2.

4. the synthesis unit generates, as the synthetic image, an image obtained by converting the first pixel value into the second synthetic pixel value obtained by adding a fourth pixel value obtained by multiplying the third pixel value by a second coefficient to the first synthetic pixel value calculated by reflecting the first conversion amount in the first pixel value; The second coefficient is a value determined according to the brightness illuminating the target area.

3. The image generating device of claim 2.

5. the synthesis unit converts the first pixel value into the second synthesis pixel value obtained by adding a fourth pixel value obtained by multiplying the third pixel value by a second coefficient to the first synthesis pixel value calculated by reflecting a second pixel amount obtained by multiplying the first conversion amount by a first coefficient on the first pixel value, and generates, as the synthetic image, an image; The first coefficient is a value determined according to a density of smoke to be synthesized into the captured image, The second coefficient is a value determined according to a combination of a density of the smoke to be synthesized into the captured image and a brightness of the light illuminating the target area.

3. The image generating device of claim 2.

6. 1. An image generating method performed by an image generating device that is a computer having a storage unit that uses three-dimensional information of a smoke model having a three-dimensional shape to store an attenuation component image in which a light irradiated from behind the smoke model is attenuated by the smoke model, and a scattered light component image in which a scattered light component of light irradiated to the smoke model is reflected by the smoke model, the image generating method comprising: The acquisition unit acquires a captured image of the target region, a synthesis unit generates a synthesis image by synthesizing an image of smoke with the captured image, using the light attenuation component image and the scattered light component image stored in the storage unit and the captured image acquired by the acquisition unit; Image generation method.

7. an image generating device that is a computer having a storage unit that uses three-dimensional information of a smoke model having a three-dimensional shape to store an attenuation component image in which a light irradiated from behind the smoke model is attenuated by the smoke model, and a scattered light component image in which a scattered light component of the light irradiated to the smoke model is reflected by the smoke model; Acquire an image of the target area; generating a composite image by combining an image of smoke with the captured image using the light attenuation component image and the scattered light component image stored in the storage unit and the captured image; program.

8. an acquisition unit that acquires an image of a target region; a generation unit that uses three-dimensional information of a smoke model having a three-dimensional shape to generate an attenuation component image that represents an attenuation component of light irradiated from behind the smoke model and attenuated by the smoke model, and a scattered light component image that represents a scattered light component of light irradiated to the smoke model and reflected by the smoke model; a synthesis unit that generates a synthetic image by synthesizing an image of smoke with the captured image, using the light attenuation component image, the scattered light component image, and the captured image; An image generating system comprising:

Citation Information

Patent Citations

  • Image generation system

    JP2021149378A