Noise reduction device and program

Abstract

To provide a technique capable of efficiently improving the visibility of moving image containing radiation ray noise.SOLUTION: A noise reduction device comprises: a reconfiguration part for rearranging pixels of a prescribed number of time-series original images in order of luminance for each coordinate in the prescribed number of time-series original images, and reconfiguring a prescribed number of images from the rearranged pixels; a selection part for selecting from the prescribed number of images an image obtained by reconfiguring pixels with relatively low luminance; and a display part for displaying the selected image. The noise reduction device may comprise a determination part for determining at least one of an image reconfigured for each coordinate from pixels with which position of luminance from the lowest should be selected, or the prescribed number.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a noise reduction device and a program. [Background technology]

[0002] Some nuclear reactor facilities have environments with high levels of radiation. During decommissioning work, imaging devices may capture video (time-series raw images) of the high-radiation environment. Workers located away from the nuclear reactor facility perform decommissioning work remotely while checking the video of the high-radiation environment.

[0003] The imaging device captures images of the inside of a facility using a solid-state imaging element such as a CMOS (Complementary Metal Oxide Semiconductor). In environments with high radiation levels, a lot of radiation noise occurs in the solid-state imaging element. As a result, multiple radiation noises (bright spots) are randomly included in videos captured using the solid-state imaging element. The more radiation noises included in a video captured inside a facility, the lower the visibility of the video becomes (see Patent Documents 1 and 2). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 5602530 [Patent Document 2] Patent No. 6063753 Summary of the Invention [Problem to be solved by the invention]

[0005] However, there is a problem in that it is not possible to efficiently improve the visibility of moving images that include radiation noise.

[0006] The present invention has been made in consideration of the above-mentioned circumstances, and provides a technique that makes it possible to efficiently improve the visibility of moving images that include radiation noise. [Means for solving the problem]

[0007] One aspect of the present invention is a noise reduction device that includes a reconstruction unit that sorts, for each coordinate in a predetermined number of original images in a time series, each pixel of the predetermined number of original images in order of brightness and reconstructs the predetermined number of images from the sorted pixels; a selection unit that selects, from the predetermined number of images, an image reconstructed from the pixels with relatively low brightness; and a display unit that displays the selected image.

[0008] One aspect of the present invention is a program for causing a computer to execute the steps of: sorting each pixel of a predetermined number of original images in a time series in order of brightness for each coordinate in the predetermined number of original images; reconstructing the predetermined number of images from the sorted pixels; selecting an image reconstructed from the pixels with relatively low brightness from the predetermined number of images; and displaying the selected image. [Effects of the Invention]

[0009] According to the present invention, it is possible to efficiently improve the visibility of moving images containing radiation noise. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a noise reduction system according to an embodiment. [Figure 2] 10A and 10B are diagrams showing examples of two original images in time series and two images reconstructed from pixels sorted in order of brightness in an embodiment. [Figure 3] 10A and 10B are diagrams showing examples of a time series of original images including two original images and a time series of images including an image reconstructed from each pixel with the lowest brightness for each coordinate in an embodiment. [Figure 4]10A and 10B are diagrams showing examples of three original images in time series and three images reconstructed from pixels rearranged in order of brightness in an embodiment. [Figure 5] 10A and 10B are diagrams showing examples of a time series of original images including three original images and a time series of images including an image reconstructed from each pixel with the lowest brightness for each coordinate in an embodiment. [Figure 6] 10A and 10B are diagrams showing examples of a time series of original images including three original images and a time series of images including an image reconstructed from each pixel with the second lowest brightness for each coordinate in an embodiment. [Figure 7] 1 is a flowchart illustrating an example of the operation of the noise reduction system according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail with reference to the drawings. 1 is a diagram showing an example of the configuration of a noise reduction system 1 according to an embodiment. The noise reduction system 1 is a system that reduces random noise (bright spots) contained in a moving image (a time-series of original images). The random noise is, for example, radiation noise.

[0012] The noise reduction system 1 includes a mobile object 2, a communication line 3, and a noise reduction device 4. The noise reduction system 1 may include a storage device 5. The mobile object 2 includes a control device 21, a drive device 22, an imaging device 23, and a communication device 24. The noise reduction device 4 includes a storage device 41, a memory 42, an operation unit 43, a communication unit 44, a processing unit 45, and a display unit 46. The processing unit 45 includes an image processing unit 451, a determination unit 452, a reconstruction unit 453, and a selection unit 454.

[0013] The mobile body 2 is, for example, a robot or a small vehicle. The control device 21 controls the operation of each device provided in the mobile body 2. The control device 21 controls, for example, the imaging range (angle of view) and imaging direction of the imaging device 23. The drive device 22 is, for example, a motor. The drive device 22 drives legs (not shown) or wheels (not shown) for moving the mobile body 2 according to the control of the control device 21. This allows the mobile body 2 to move through a space in an environment with a high radiation dose (for example, inside a nuclear reactor facility) while capturing images of the space.

[0014] The imaging device 23 generates time-series original images (moving images). The size of the original images is not limited to a particular size, but may be, for example, a VGA (Video Graphics Array) size of 640 pixels x 480 pixels. The frame rate of the time-series original images is not limited to a particular frame rate, but may be, for example, 30 fps. The time-series original images may include images of the subject 10 together with a space in a high-radiation environment. The subject 10 may be, for example, a component within a nuclear reactor facility.

[0015] A time-series raw image captured using a solid-state imaging device may contain multiple random radiation noises. In the raw image, pixels where radiation noise occurs are pixels with relatively high brightness (bright spots). When electrons hit by radiation such as gamma rays hit surrounding electrons (by Compton scattering), the electron energy spectrum has multiple energy levels. For this reason, the color of pixels with relatively high brightness may be gray or white.

[0016] The number of pixels in which radiation noise occurs (noise generation rate) relative to the total number of pixels in the solid-state imaging device is, for example, about 1 / 10 to 1 / 100. Note that the higher the radiation dose, the greater the number of pixels in which radiation noise occurs.

[0017] The communication device 24 acquires a signal (hereinafter referred to as a "video signal") representing a time-series original image (video) from the imaging device 23. The video data may be compressed based on a predetermined video compression standard. The communication device 24 transmits the video signal to the communication unit 44. The communication device 24 may record the video signal in the storage device 5.

[0018] The storage device 41 stores a program in advance. When the noise reduction device 4 is started up, the program is loaded from the storage device 41 into the memory 42. The operation unit 43 is, for example, a keyboard, a touch panel, or a mouse. The operation unit 43 accepts operation input by the user. The operation unit 43 outputs a signal corresponding to the operation input to the processing unit 45. The communication unit 44 acquires a video signal from the communication device 24 or the storage device 5. The communication unit 44 records the video signal in the storage device 41.

[0019] The processing unit 45 is realized as software by a processor such as a CPU (Central Processing Unit) executing a program loaded from the storage device 41, which has a non-volatile recording medium (non-transitory recording medium), to the memory 42. The program is recorded on a computer-readable recording medium. The computer-readable recording medium is, for example, a non-transitory recording medium such as a solid state drive.

[0020] Part or all of the processing unit 45 may be realized using hardware including an electronic circuit (electronic circuit or circuitry) using, for example, an LSI (Large Scale Integrated circuit), an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array).

[0021] The image processing unit 451 acquires the moving image signal from the storage device 41. The image processing unit 451 may acquire the moving image signal from the storage device 5 using the communication unit 44. The image processing unit 451 performs image processing (decoding processing) on ​​the acquired moving image signal to reproduce original images (moving images) in time series.

[0022] The determination unit 452 determines the predetermined number of original images as a processing unit based on a predetermined number condition. This number condition is determined, for example, based on a signal corresponding to an operation input. For example, when the user operates the operation unit 43 and a signal corresponding to the operation input "2" regarding the number is input to the determination unit 452 from the operation unit 43, the determination unit 452 determines the predetermined number of original images as a processing unit to be two. When the determination unit 452 is to further enhance the noise reduction effect in the noise-reduced time-series images, the determination unit 452 may increase the predetermined number of original images as a processing unit. Furthermore, when the noise-reduced time-series images are to be reduced in terms of afterimages (blur), the determination unit 452 may decrease the predetermined number of original images as a processing unit.

[0023] The determination unit 452 determines, for each coordinate in a predetermined number of original images in time series, which pixel with the lowest luminance to select as the image reconstructed from, based on a predetermined luminance condition. This luminance condition is determined, for example, based on a signal corresponding to an operation input. For example, when the user operates the operation unit 43 and a signal corresponding to the operation input "1" related to luminance is input to the determination unit 452 from the operation unit 43, the determination unit 452 determines that the selection unit 454 should select the image reconstructed from the pixel with the lowest luminance.

[0024] The reconstruction unit 453 rearranges the pixels of a predetermined number of original images in chronological order in order of brightness for each coordinate in the predetermined number of original images. That is, the reconstruction unit 453 rearranges the pixels of the predetermined number of original images in order of brightness while maintaining the coordinates in the original images. The reconstruction unit 453 reconstructs a predetermined number of images from the rearranged pixels.

[0025] The selection unit 454 selects an image reconstructed from pixels with relatively low brightness from a predetermined number of images. The image reconstructed from pixels with relatively low brightness is an image with reduced bright spots (noise).

[0026] The display unit 46 is, for example, a liquid crystal display or an organic EL (Electro Luminescence) display. The display unit 46 displays, in time series, the images (images with reduced noise) selected by the selection unit 454. The display unit 46 may also display, based on a signal corresponding to an operation input, the original images in time series reproduced by the image processing unit 451.

[0027] Next, the processing unit 45 will be described in detail. 2 is a diagram showing an example of two original images 101 in time series and two images reconstructed from pixels sorted in order of brightness in an embodiment. Hereinafter, for ease of explanation, the x-direction size of the original image (image before noise reduction) is, for example, 6 pixels, and the y-direction size of the original image (image before noise reduction) is, for example, 4 pixels. Similarly, the x-direction size of the image after noise reduction is, for example, 6 pixels, and the y-direction size of the image after noise reduction is, for example, 4 pixels.

[0028] In the following description, time-series original images 101 include images of subject 10. In original images 101-1 and 101-2, the coordinates of each pixel constituting the image of subject 10 are, for example, coordinates (x, y) = (3, 1), (3, 2), (4, 1), and (4, 2).

[0029] The time-series original images 101 contain radiation noise as pixels with relatively high brightness. The coordinates of the pixels where radiation noise occurs are random for each original image 101. For example, in original image 101-1, radiation noise occurs in the pixels with coordinates (x, y) = (1, 2) and (4, 4). In contrast, in original image 101-2, radiation noise occurs in the pixels with coordinates (x, y) = (2, 2), (4, 4), and (6, 2).

[0030] The reconstruction unit 453 rearranges the pixels of the two original images 101 in order of brightness for each coordinate in the two original images 101 in time series. The reconstruction unit 453 reconstructs images 201a and 201b from the rearranged pixels. Image 201a is an image reconstructed from pixels with relatively low brightness (an image reconstructed from pixels with the first lowest brightness). Image 201b is an image reconstructed from pixels with relatively high brightness (an image reconstructed from pixels with the second lowest brightness).

[0031] The selection unit 454 selects the image 201a from the images 201a and 201b. Since pixels with relatively low brightness are concentrated in the image 201a, radiation noise (bright spots) is reduced in the image 201a. The display unit 46 displays the image 201a.

[0032] 3 is a diagram showing an example of a time series of original images 101 including two original images 101 and an example of a time series of images 201a including an image 201a reconstructed from each pixel with the lowest brightness for each coordinate in an embodiment. In the following, since the imaging ranges of original images 101-3 and 101-4 are changed from the imaging ranges of original images 101-1 and 101-2, the coordinates of each pixel constituting the image of subject 10 in original images 101-3 and 101-4 are, for example, coordinates (x, y) = (4, 1), (4, 2), (5, 1), and (5, 2).

[0033] The reconstruction unit 453 rearranges the pixels of the original images 101-1 and 101-2 in order of brightness for each coordinate in the original images 101-1 and 101-2. The selection unit 454 selects the reconstructed image 201a-1 from the pixel with the lowest brightness for each coordinate. The display unit 46 displays the selected image 201a-1.

[0034] The reconstruction unit 453 rearranges the pixels of the original images 101-2 and 101-3 in order of brightness for each coordinate in the original images 101-2 and 101-3. The selection unit 454 selects the reconstructed image 201a-2 from the pixel with the lowest brightness for each coordinate. The display unit 46 displays the selected image 201a-2.

[0035] The reconstruction unit 453 rearranges the pixels of the original images 101-3 and 101-4 in order of brightness for each coordinate in the original images 101-3 and 101-4. The selection unit 454 selects the reconstructed image 201a-3 from the pixel with the lowest brightness for each coordinate. The display unit 46 displays the selected image 201a-3. The same applies to the original images 101 in the time series after the original image 101-4.

[0036] 4 is a diagram showing an example of three original images 101 in time series and three images reconstructed from pixels sorted in order of brightness in an embodiment. In original image 101-1, radiation noise occurs in the pixels at coordinates (x,y)=(1,2) and (4,4). In original image 101-2, radiation noise occurs in the pixels at coordinates (x,y)=(2,2), (4,4), and (6,2). In original image 101-3, radiation noise occurs in the pixels at coordinates (x,y)=(2,3) and (5,4).

[0037] The reconstruction unit 453 rearranges the pixels of the three original images 101 in order of brightness for each coordinate in the three original images 101 in time series. The reconstruction unit 453 reconstructs images 301a, 301b, and 301c from the rearranged pixels. Image 301a is an image reconstructed from pixels with the lowest brightness. Image 301b is an image reconstructed from pixels with the second lowest brightness. Image 301c is an image reconstructed from pixels with the third lowest brightness.

[0038] The selection unit 454 selects the image 301a from among the images 301a, 301b, and 301c. Since the pixels with the lowest brightness are concentrated in the image 301a, radiation noise (bright spots) is reduced in the image 301a. The display unit 46 displays the image 301a.

[0039] 5 is a diagram showing an example of a time series of original images including three original images and a time series of images including an image reconstructed from each pixel with the lowest brightness for each coordinate in an embodiment. The reconstruction unit 453 rearranges each pixel in each of the original images 101-1 to 101-3 in order of brightness for each coordinate in each of the original images 101-1 to 101-3. The selection unit 454 selects an image 301a-1 reconstructed from each pixel with the lowest brightness for each coordinate. The display unit 46 displays the selected image 301a-1.

[0040] The reconstruction unit 453 rearranges the pixels in each of the original images 101-2 to 101-4 in order of brightness for each coordinate in each of the original images 101-2 to 101-4. The selection unit 454 selects the reconstructed image 301a-2 from the pixel with the lowest brightness for each coordinate. The display unit 46 displays the selected image 301a-2.

[0041] The reconstruction unit 453 rearranges the pixels in each of the original images 101-3 to 101-5 (not shown) in order of brightness for each coordinate in each of the original images from original image 101-3 to original image 101-5 (not shown). The selection unit 454 selects the reconstructed image 301a-3 from the pixel with the lowest brightness for each coordinate. The display unit 46 displays the selected image 301a-3. The same is true for the original images 101 in the time series from original image 101-4 onwards.

[0042] 6 is a diagram illustrating an example of a time series of original images including three original images and a time series of images including an image reconstructed from each pixel with the second lowest brightness for each coordinate in an embodiment. The reconstruction unit 453 rearranges each pixel in each of the original images 101-1 to 101-3 in order of brightness for each coordinate in each of the original images 101-1 to 101-3. The selection unit 454 selects an image 301b-1 reconstructed from each pixel with the second lowest brightness for each coordinate. The display unit 46 displays the selected image 301b-1.

[0043] The reconstruction unit 453 rearranges the pixels in each of the original images 101-2 to 101-4 in order of brightness for each coordinate in each of the original images 101-2 to 101-4. The selection unit 454 selects the reconstructed image 301b-2 from the pixel with the second lowest brightness for each coordinate. The display unit 46 displays the selected image 301b-2.

[0044] The reconstruction unit 453 rearranges the pixels in each of the original images 101-3 to 101-5 (not shown) in order of brightness for each coordinate in each of the original images 101-3 to 101-5 (not shown). The selection unit 454 selects the image 301b-3 reconstructed from the pixel with the second lowest brightness for each coordinate. The display unit 46 displays the selected image 301b-3. The same is true for the original images 101 in the time series from original image 101-4 onwards.

[0045] Next, an example of the operation of the processing unit 45 will be described. 7 is a flowchart showing an example of the operation of the noise reduction system 1 in an embodiment. The reconstruction unit 453 rearranges the pixels of a predetermined number of original images 101 in order of brightness for each coordinate of the original image 101 (step S101). The reconstruction unit 453 reconstructs a predetermined number of images from the rearranged pixels (step S102). The selection unit 454 selects images reconstructed from pixels with relatively low brightness (step S103). The display unit 46 displays the selected images (step S104).

[0046] The determination unit 452 determines whether or not to end the noise reduction process based on a signal corresponding to the operation input (step S105). If it is determined that the noise reduction process should be continued (step S105: NO), the processing unit 45 returns the process to step S101. If it is determined that the process should be ended (step S105: YES), the processing unit 45 ends the noise reduction process.

[0047] As described above, the reconstruction unit 453 rearranges the pixels of a predetermined number of original images 101 in order of brightness for each coordinate in the predetermined number of original images 101 in time series. The reconstruction unit 453 reconstructs a predetermined number of images (e.g., image 201a and image 201b) from the rearranged pixels. The selection unit 454 selects an image (e.g., image 201a) reconstructed from pixels with relatively low brightness from the predetermined number of images. The display unit 46 displays the selected images in time series. This makes it possible to efficiently improve the visibility of moving images including radiation noise.

[0048] The determination unit 452 may determine which image reconstructed from pixels with the lowest brightness for each coordinate should be selected (e.g., whether to select image 201a or image 201b). For example, if the overall brightness of image 201a is too low (if the overall brightness is equal to or lower than a threshold), the determination unit 452 may select image 201b, which has an overall brightness higher than that of image 201a.

[0049] The determination unit 452 may determine a predetermined number of original images 101 as a processing unit. The determination unit 452 may increase the predetermined number of original images 101 as a processing unit as the amount of noise generated in the original images 101 increases. That is, when the determination unit 452 wants to improve the effect of reducing noise in the selected images in time series, the determination unit 452 may increase the predetermined number of original images 101 as a processing unit. When the determination unit 452 wants to reduce afterimages in the images in time series, the determination unit 452 may decrease the predetermined number of original images 101 as a processing unit.

[0050] The determining unit 452 may detect feature points in each image (each image in which noise has been reduced) selected by the selecting unit 454. The determining unit 452 may align and superimpose the multiple images based on the positions of feature points common to the multiple images. Because noise has been reduced in each image selected by the selecting unit 454, the determining unit 452 can easily align and superimpose the multiple captured images.

[0051] Although an embodiment of the present invention has been described above in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and includes designs within the scope of the gist of the present invention. [Explanation of symbols]

[0052] 1... noise reduction system, 2... moving body, 3... communication line, 4... noise reduction device, 5... storage device, 10... subject, 21... control device, 22... driving device, 23... imaging device, 24... communication device, 41... storage device, 42... memory, 43... operation unit, 44... communication unit, 45... processing unit, 46... display unit, 451... image processing unit, 452... determination unit, 453... reconstruction unit, 454... selection unit, 101... original image, 201a, 201b... images, 301a, 301b... images

Claims

1. a reconstruction unit that rearranges pixels of a predetermined number of original images in order of luminance for each coordinate in the predetermined number of original images in time series, and reconstructs the predetermined number of images from the rearranged pixels; a selection unit that selects an image reconstructed from the pixels having relatively low luminance from among the predetermined number of images; a display unit that displays the selected image; A noise reduction device comprising:

2. 2. The noise reduction device according to claim 1, further comprising a determination unit that determines at least one of which of the pixels having the lowest luminance is to be used to select the image reconstructed from for each of the coordinates, and the predetermined number of images as a processing unit.

3. The noise reduction device according to claim 2 , wherein the determination unit increases the predetermined number when a noise reduction effect in the selected image is to be enhanced.

4. On the computer, a step of sorting pixels of a predetermined number of original images in time series in order of brightness for each coordinate in the predetermined number of original images, and reconstructing the predetermined number of images from the sorted pixels; selecting an image reconstructed from the pixels having relatively low brightness from among the predetermined number of images; displaying the selected image; A program to execute.

Images