Imaging system

Abstract

An imaging system according to one embodiment of the present technology is provided with a light source, an imaging box, and a camera. The imaging box has a transmission screen on the surface facing the light source. The camera images a shadow of a subject projected on the transmission screen of the subject projected by the light source. Accordingly, since the shadow of the subject projected on the transmission screen of the subject projected by the light source is imaged, an imaging environment strong against disturbance can be easily maintained.

Description

Shooting System

[0001] The present technology relates to an imaging system that can be applied to image recognition and the like.

[0002] Patent Document 1 describes an imaging method in which a visible image is projected onto a screen surface from one side of the screen, and invisible light is projected onto the other surface of the screen from the opposite side of the screen, and the pattern of invisible light on the screen is imaged from one side of the screen. This makes it possible to obtain the pattern of an object or person between the rear surface of the screen and the invisible light projection means without being affected by the visible image projected onto the front surface of the screen, thereby enabling the detection and movement of an object or person behind the screen to be monitored and controlled (see, for example, paragraphs

[0021] to

[0034] , Figure 2 of the specification of Patent Document 1).

[0003] Special Publication No. 2013-541891

[0004] In order to improve the recognition accuracy, there is a demand for an imaging system that can easily maintain an imaging environment that is resistant to disturbances.

[0005] In view of the above circumstances, an object of the present technology is to provide an imaging system that can easily maintain an imaging environment that is resistant to disturbances.

[0006] In order to achieve the above object, an imaging system according to an embodiment of the present technology includes a light source, an imaging box, and a camera. The imaging box has a translucent screen on a surface facing the light source. The camera captures a shadow of a subject projected by the light source onto the translucent screen.

[0007] This photography system photographs the shadow of a subject projected by a light source onto a translucent screen, making it easy to maintain a photography environment that is resistant to external disturbances.

[0008] The photography box may be configured to reduce ambient light that interferes with photography by the camera.

[0009] The shooting box may be a hexahedron, and may be configured so that the surface having the translucent screen, the surface opposite the translucent screen on which the camera is placed, and the other four surfaces other than the surface on which the translucent screen and the camera are placed are not transparent to light.

[0010] The photographing system may further include an information processing device having a recognition unit that performs image recognition according to the shape of the shadow photographed by the camera.

[0011] The information processing device may include a playback unit that plays back content based on a recognition result from the recognition unit.

[0012] The content may include at least audio data or music data.

[0013] It is a schematic diagram of a configuration example of a photography system.It is a block diagram of a configuration example of an information processing device.It is a diagram showing a flowchart of content playback.

[0014] Hereinafter, embodiments of the present technology will be described with reference to the drawings.

[0015] [Photography System] Fig. 1 is a schematic diagram showing a configuration example of a photography system 100 according to the present technology. Fig. 1A is a schematic diagram showing an example of an arrangement of the photography system 100. Fig. 1B is a schematic diagram showing a photographed image.

[0016] 1, the photography system 100 includes a light source 1, a photography box 10, a camera 15, and an information processing device 20. In this embodiment, it is desirable to use a light source 1 that is more powerful than natural light, such as a spotlight.

[0017] The photography box 10 has a translucent screen 11 on the surface facing the light source 1. In this embodiment, the subject 3 (hand) is placed between the light source 1 and the photography box 10. That is, the shadow of the subject 3 is projected onto the translucent screen 11 of the photography box 10.

[0018] Furthermore, in this embodiment, the photography box 10 is configured to reduce ambient light that interferes with photography by the camera 15, which will be described later. For example, the photography box 10 is a cube, and is configured with a surface 12 having a translucent screen 11, a surface 13 on which the camera 15 is disposed, and four other surfaces 14 that are configured to block light. Note that in FIG. 1 , the surface 13 on which the camera 15 is disposed is the surface opposite the surface 12 having the translucent screen 11, but this is not limited thereto, and the camera 15 may be disposed in any position as long as it can photograph the shadow of the subject 3.

[0019] The shape and size of the photography box 10 are not limited. For example, the surface having the translucent screen 11 may be larger depending on the size of the subject 3 to be photographed (the size of the shadow to be projected). Alternatively, for example, the photography box 10 may be cylindrical, with the bottom surface having the translucent screen 11 and the sides configured to not transmit light.

[0020] The camera 15 captures the shadow of the subject 3 projected by the light source 1 onto the translucent screen 11. In this embodiment, the captured image (see FIG. 1B ) captured by the camera 15 is output to the information processing device 20.

[0021] The information processing device 20 performs image recognition based on the captured image. For example, if the subject 3 is a hand, the information processing device 20 recognizes the shape (gesture) of the hand, such as whether the hand is open or closed. In this embodiment, the information processing device 20 plays content corresponding to the shape of the hand based on the shape of the hand. The content will be described later.

[0022] Note that the disturbance in this embodiment may include factors that hinder image recognition by the information processing device 20 in addition to factors that hinder image capture by the camera 15 , i.e., factors that hinder capture of the shadow of the subject 3 .

[0023] FIG. 2 is a block diagram showing an example of the configuration of the information processing device 20.

[0024] As shown in FIG. 2, the information processing device 20 includes an acquisition unit 21, a recognition unit 22, a learning unit 23, and a content playback unit 24.

[0025] The acquisition unit 21 acquires a photographed image that shows the shadow of the subject 3 photographed by the camera 15. In this embodiment, the photographed image photographed by the camera 15 may be continuously acquired at a predetermined frame rate while the photographing system is running, or the shadow of the subject 3 may be acquired as a video.

[0026] The recognition unit 22 recognizes the shape of the subject 3 in the captured image. In this embodiment, the recognition unit 22 recognizes the shape of the subject 3 based on a shape recognition model output by a learning unit 23, which will be described later.

[0027] The learning unit 23 generates a shape recognition model that recognizes the shape of the subject 3 from the captured image. For example, a convolutional neural network (CNN) may be used as the network architecture. In this embodiment, the learning unit 23 performs learning by receiving captured images that show the shadow of the subject 3 as learning data.

[0028] This removes noise components that hinder learning and recognition by using the shadow of the subject 3. For example, if the subject 3 is a hand, wrinkles and lines on the surface are removed, reducing the amount of learning data, which can improve the accuracy of recognition.

[0029] The learning and recognition methods are not limited, and each of the information processing devices 20 of the present imaging system may perform learning individually and sequentially, or learning may be performed based on learning data collected online via the cloud or the like, or a shape recognition model that has been trained in advance may be used. Also, any learning method such as supervised learning or unsupervised learning may be performed.

[0030] The content playback unit 24 plays back the corresponding content based on the recognition result of the recognition unit 22. The content includes sounds of various musical scales, musical instruments, beats, animal sounds, music, sounds and voices corresponding to gestures, specific songs, voices, etc.

[0031] For example, if the user shapes the shadow of his / her hand into the shape of a dog, the recognition unit 22 recognizes the shape of the subject 3 as a dog, and the content of the dog barking is played back by the content playback unit 24. As a result, the user can change the shape of the subject 3 between the light source 1 and the photography box 10, thereby changing the content or continuously playing the same content.

[0032] Note that, in addition to a hand, objects other than a hand may be used as the subject 3, such as dolls resembling various animals or musical instruments, depending on the content.

[0033] FIG. 3 is a flowchart showing content playback.

[0034] 3, the flowchart is started by operating an application or a start switch that starts the photography system. In addition to this, various situations, such as the detection of illuminance (lux) above a predetermined value on the translucent screen 11 of the photography box 10, may be set as triggers for starting the photography system.

[0035] As shown in FIG. 3, the shadow of the subject 3 projected on the translucent screen 11 is photographed by the camera 15, and the photographed image is acquired by the acquisition unit 21 (step 101).

[0036] The recognition unit 22 recognizes the shape of the shadow of the subject 3 from the captured image (step 102). For example, the user places hand shape A, which corresponds to content that reproduces the sound "do," between the light source 1 and the capture box 10.

[0037] The recognition unit 22 also determines whether the shape of the subject 3 in the captured image has changed (step 103). Note that removing the subject 3 from between the light source 1 and the shooting box 10, i.e., changing the state in which the shadow of the subject 3 is not captured in the captured image, may also be included in the change in shape of the subject 3. Conversely, changing the state in which the shadow of the subject 3 is not captured in the captured image to a state in which the shadow of the subject 3 is captured in the captured image may also be included in the change in shape.

[0038] If the shape of the hand has changed (YES in step 103), the content playback unit 24 plays content corresponding to the shape based on the recognized shape of the shadow of the subject 3 (step 104). For example, if the user wants to play content that plays the sound "re" after content that plays the sound "do," the user changes the shape of the hand from A to B.

[0039] In this way, the user can change the content by repeating the process from step 101 to step 104 by changing the shape of his or her hand.

[0040] If the hand shape does not change (NO in step 103), the content being played is maintained. Note that the content can be stopped by turning off the power of the image capturing system 100, etc.

[0041] The flow for playing and stopping content may be set arbitrarily. For example, a special shape for the subject 3 may be set for stopping the content. For example, if the shape of the subject 3's shadow remains unchanged for N seconds or more, the content may be stopped. The content may also be played by changing the shape of the subject 3 in a specific order within M seconds. Of course, the content may also be stopped when the content's playback time is reached.

[0042] As described above, the photography system 100 according to this embodiment includes the light source 1, the photography box 10 having the translucent screen 11 on the surface 12 facing the light source 1, and the camera 15 that photographs the shadow of the subject 3 projected by the light source 1 onto the translucent screen 11. This makes it possible to easily maintain a photography environment that is resistant to external disturbances.

[0043] Conventionally, in devices that use images captured by a camera as input, the key to ensuring the quality of the input image is to maintain a constant shooting environment, including the distance from the subject, the angle of view, and lighting. While it is easy to maintain a constant environment, such as in a manufacturing plant where inspections can be performed, maintaining a constant environment is difficult for devices used in a variety of locations due to disturbances. While measures have been taken to maintain a constant quality of input images, such as using the same background pattern or adding contrast using an image processing device, and in the case of image recognition devices, improving the quality of the output results by improving the capabilities of the final information processing device, these measures are not easy.

[0044] In this technology, the shadow of a subject with little information content is projected onto a translucent screen mounted in a photography box that reduces disturbances such as natural light, and the shadow is photographed.

[0045] This makes it possible to easily maintain a constant shooting environment at the input stage, even when used in a location where a constant environment cannot be created, so less processing is required in the subsequent stages (image processing device or information processing device).In addition, since the input image itself is simplified, it is possible to increase the quality of the output result even if the final stage information processing device has low capabilities.

[0046] Furthermore, compared to photographing the subject directly, photographing the shadow projected onto a translucent screen eliminates disturbances, such as not capturing the background behind the subject, not being affected by the brightness of the shooting location if the light source is sufficient, and not being affected by the color temperature of the shooting location because the color components of the subject are removed. Furthermore, the removal of the color components of the subject reduces the load on downstream information processing devices, and if the subject is a hand, noise components such as wrinkles and lines on the surface are removed. Since the distance between the camera and the translucent screen is always constant, mechanisms such as autofocus are not required, making it possible to simplify and speed up the process. Furthermore, when a diverging light source is used, the projected shadow can be enlarged, enabling higher recognition accuracy.

[0047] Other Embodiments The present technology is not limited to the above-described embodiments, and various other embodiments can be realized.

[0048] In the above embodiment, the recognition unit 22 recognizes the shape of the projected shadow of the subject from the captured image. However, the present invention is not limited to this, and an image processing device that processes the captured image may be installed in the information processing device 20. For example, the image processing device may have a filter that removes information that becomes noise in image recognition, a saturation adjustment function, or the like.

[0049] In the above embodiment, the content playback unit 24 played back content according to the shape of the subject's shadow. This is not a limitation, and content may be played back according to the color of the light source. Of course, content may be played back according to a combination of the shape of the subject's shadow and the color of the light source. Furthermore, the number of subjects is not limited, and content may be played back according to the respective shapes of the left and right hands or the combination thereof.

[0050] The configurations of the recognition unit, learning unit, content playback unit, etc. described with reference to the drawings are merely one embodiment and can be modified as desired without departing from the spirit of the present technology. In other words, any other configurations, algorithms, etc. for implementing the present technology may be adopted.

[0051] It should be noted that the effects described in this disclosure are merely examples and are not limiting, and other effects may also be present. The description of multiple effects above does not necessarily mean that these effects are exhibited simultaneously. It means that at least one of the effects described above can be obtained depending on the conditions, etc., and of course, effects not described in this disclosure may also be exhibited.

[0052] It is also possible to combine at least two of the characteristic features of each embodiment described above. In other words, the various characteristic features described in each embodiment may be combined in any manner without distinguishing between the embodiments.

[0053] The present technology may also have the following configurations. (1) A photography system comprising: a light source; a photography box having a translucent screen on a surface facing the light source; and a camera that photographs a shadow of a subject projected by the light source and cast on the translucent screen. (2) The photography system described in (1), wherein the photography box is configured to reduce ambient light that interferes with photography by the camera. (3) The photography system described in (2), wherein the photography box is a hexahedron, and is configured so that light does not pass through on four surfaces other than the surface having the translucent screen, the surface opposite the translucent screen on which the camera is placed, and the surfaces on which the translucent screen and the camera are placed. (4) The photography system described in (1), further comprising: an information processing device having a recognition unit that recognizes an image according to the shape of the shadow photographed by the camera. (5) The photography system described in (4), wherein the information processing device has a playback unit that plays back content based on the recognition result of the recognition unit. (6) The photography system according to (5), wherein the content includes at least audio data or music data.

[0054] REFERENCE SIGNS LIST 1 light source 3 subject 10 photography box 11 translucent screen 15 camera 20 information processing device 22 recognition unit 24 content playback unit 100 photography system

Claims

1. A photography system comprising: a light source; a photography box having a translucent screen on a surface facing the light source; and a camera that photographs the shadow of a subject projected by the light source onto the translucent screen.

2. The photography system according to claim 1, wherein the photography box is configured to reduce ambient light that interferes with photography by the camera.

3. A photography system according to claim 2, wherein the photography box is a hexahedron and is configured so that light is not transmitted on four sides other than the side having the translucent screen, the side opposite the translucent screen on which the camera is placed, and the sides on which the translucent screen and the camera are placed.

4. The photography system according to claim 1, further comprising an information processing device having a recognition unit that performs image recognition according to the shape of the shadow photographed by the camera.

5. The photography system according to claim 4, wherein the information processing device has a playback unit that plays back content based on the recognition result of the recognition unit.

6. The photography system according to claim 5, wherein the content includes at least audio data or music data.

Images