Image processing circuit

Abstract

To change the intensity setting of each of plural times of image processing while using a neural network circuit.SOLUTION: Image processing circuits according to an embodiment include: a first processing circuit which executes noise reduction processing of reducing noise of an image; a second processing circuit which executes super-resolution processing of enhancing the resolution of the image; a calculation circuit which calculates the combining ratio indicating the ratio of the plurality of images to be combined on the basis of the intensity setting of each of the noise reduction processing and the super-resolution processing; a neural network circuit which executes the noise reduction processing or super-resolution processing on the basis of the intensity setting; a first combining circuit which combines the image applied with the noise reduction processing by the first processing circuit and the image applied with the image processing by the neural network circuit on the basis of the combining ratio; and a second combining circuit which combines the image applied with the super-resolution processing by the second processing circuit and the image applied with the image processing by the neural network circuit on the basis of the combining ratio.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] Embodiments of the present invention relate to an image processing circuit.

Background Art

[0002] There is a neural network circuit in which a neural network is implemented as a circuit. The neural network circuit performs, for example, noise reduction processing for reducing noise in an image. The neural network circuit reduces noise more effectively than a conventional noise reduction processing circuit in which the algorithm of the noise reduction processing is implemented as a circuit.

[0003] Therefore, the neural network circuit is desired to be used not only for noise reduction processing but also for various image processing such as super-resolution processing. Further, when performing a plurality of image processes, it is preferable that the intensity of the image process can be changed for each image process.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, since the neural network circuit has a large circuit scale, it is difficult to provide it for each image process. Further, since the neural network circuit has an enormous number of coefficient sets to be set for each neuron, it is difficult to prepare for each intensity setting of a plurality of image processes.

[0006] The problem to be solved by the present invention is to provide an image processing circuit that can change the intensity settings of a plurality of image processes while using a neural network circuit.

Means for Solving the Problems

[0007] The image processing circuit of the embodiment includes a first processing circuit that executes noise reduction processing, which is image processing for reducing noise in an image, a second processing circuit that executes super-resolution processing, which is image processing for increasing the resolution of an image, and a calculation circuit that calculates a synthesis ratio indicating the ratio of a plurality of images to be synthesized based on the intensity settings of each of the noise reduction processing and the super-resolution processing , front a neural network circuit formed by a neural network that executes the noise reduction processing or the super-resolution processing, a first synthesis circuit that synthesizes the image subjected to the noise reduction processing by the first processing circuit and the image subjected to the image processing by the neural network circuit based on the synthesis ratio, and a second synthesis circuit that synthesizes the image subjected to the super-resolution processing by the second processing circuit and the image subjected to the image processing by the neural network circuit based on the synthesis ratio A selection circuit that selects first setting information for causing the neural network circuit to execute the noise reduction process or second setting information for causing the neural network circuit to execute the super-resolution process based on the strength setting; and includes. Then, the neural network circuit executes the image processing of the first setting information or the second setting information selected by the selection circuit.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Embodiments for Carrying Out the Invention

[0009] Hereinafter, embodiments will be described in detail with reference to the drawings.

[0010] FIG. 1 is a diagram showing an example of the hardware configuration of a television apparatus 10 according to an embodiment. The television apparatus 10 executes image processing on an image input by broadcast or the like. Then, the television apparatus 10 displays the image on which the image processing has been executed.

[0011] As shown in FIG. 1, the television apparatus 10 includes an antenna 101, input terminals 102a to 102c, a tuner 103, a demodulator 104, a demultiplexer 105, an A / D (analog / digital) converter 106, a selector 107, a signal processing unit 108, a speaker 109, a display panel 110, an operation unit 111, a light receiving unit 112, an IP communication unit 113, a CPU (Central Processing Unit) 114, a memory 115, a storage 116, a microphone 117, and an audio I / F (interface) 118.

[0012] The antenna 101 receives a broadcast signal of digital broadcast and supplies the received broadcast signal to the tuner 103 via the input terminal 102a.

[0013] The tuner 103 selects a broadcast signal of a desired channel from the broadcast signal supplied from the antenna 101 and supplies the selected broadcast signal to the demodulator 104.

[0014] The demodulator 104 demodulates the broadcast signal supplied from the tuner 103 and supplies the demodulated broadcast signal to the demultiplexer 105.

[0015] The demultiplexer 105 separates the broadcast signal supplied from the demodulator 104 to generate a video signal and an audio signal, and supplies the generated video signal and audio signal to the selector 107.

[0016] The selector 107 selects one from a plurality of signals supplied from the demultiplexer 105, the A / D converter 106, and the input terminal 102c, and supplies the selected one signal to the signal processing unit 108.

[0017] The signal processing unit 108 performs predetermined signal processing on the video signal supplied from the selector 107 and supplies the processed video signal to the display panel 110. Also, the signal processing unit 108 performs predetermined signal processing on the audio signal supplied from the selector 107 and supplies the processed audio signal to the speaker 109. The signal processing unit 108 has an image processing circuit 1000 shown in FIG. 2.

[0018] The speaker 109 outputs sound or various sounds based on the audio signal supplied from the signal processing unit 108. Also, the speaker 109 changes the volume of the output sound or various sounds based on the control by the CPU 114.

[0019] The display panel 110 as the display unit displays videos such as still images and moving images, other images, and character information, etc. based on the video signal supplied from the signal processing unit 108 or based on the control by the CPU 114.

[0020] The input terminal 102b receives analog signals such as video signals and audio signals input from the outside. Also, the input terminal 102c receives digital signals such as video signals and audio signals input from the outside. For example, the input terminal 102c can receive digital signals from a recorder or the like equipped with a drive device for driving a recording medium for recording and playback such as a BD (Blu-ray (registered trademark) Disc).

[0021] The A / D converter 106 supplies the digital signal generated by performing A / D conversion on the analog signal supplied from the input terminal 102b to the selector 107.

[0022] The operation unit 111 receives the operation input of the user.

[0023] The light receiving unit 112 receives infrared rays from the remote controller 119.

[0024] The IP communication unit 113 is a communication interface for performing IP (Internet Protocol) communication via the network 40.

[0025] The CPU 114 controls the entire television apparatus 10.

[0026] The memory 115 is a ROM that stores various computer programs executed by the CPU 114, a RAM that provides a working area for the CPU 114, and the like.

[0027] The storage 116 is an HDD (Hard Disk Drive), an SSD (Solid State Drive), or the like. The storage 116 records, for example, a signal selected by the selector 107 as recording data.

[0028] The microphone 117 as a voice input unit acquires the voice spoken by the user and sends it to the audio I / F 118.

[0029] The audio I / F 118 performs analog / digital conversion on the voice acquired by the microphone 117 and sends it to the CPU 114 as a voice signal.

[0030] Next, the image processing circuit 1000 included in the signal processing unit 108 will be described.

[0031] FIG. 2 is a block diagram showing an example of the circuit configuration of the image processing circuit 1000. As shown in FIG. 2, the image processing circuit 1000 includes a level output circuit 1001, a coefficient selection circuit 1002, a synthesis ratio calculation circuit 1005, an NR (noise reduction) circuit 1006, a neural network circuit 1007, a first synthesis circuit 1008, an SR (Super Resolution) circuit 1009, and a second synthesis circuit 1010.

[0032] The image processing circuit 1000 performs multiple types of image processing on the input image input to the image processing circuit 1000 at their respective intensity levels. The image processing circuit 1000 can set, for example, four levels of intensity. As an example of the multiple image processes, the image processing circuit 1000 performs noise reduction processing and super-resolution processing. However, the image processing circuit 1000 may perform image processing different from the noise reduction processing or the super-resolution processing.

[0033] The level output circuit 1001 is a circuit that outputs the intensity levels of the noise reduction processing and the super-resolution processing respectively. The intensity level is an intensity setting indicating the intensity of each of the noise reduction processing and the super-resolution processing. Also, the intensity level may be a value determined by the user's setting, a value determined according to various settings, a calculated value, or a value other than these. Further, when the intensity level is input from outside the image processing circuit 1000, the image processing circuit 1000 may not have the level output circuit 1001.

[0034] The NR circuit 1006 is a circuit that performs noise reduction processing, which is an image processing for reducing noise in an image. The NR circuit 1006 is an example of the first processing circuit. More specifically, the NR circuit 1006 is a circuit in which the algorithm of the noise reduction processing is implemented as a circuit. Also, the NR circuit 1006 performs noise reduction processing according to the intensity level of the noise reduction processing output from the level output circuit 1001.

[0035] Specifically, the NR circuit 1006 performs noise reduction processing according to the intensity level of the noise reduction processing on the input image input to the image processing circuit 1000. Then, the NR circuit 1006 outputs the NR image, which is the image subjected to the noise reduction processing, to the neural network circuit 1007 and the first synthesis circuit 1008.

[0036] The SR circuit 1009 is a circuit that performs super-resolution processing, which is an image processing for enhancing the resolution of an image. The SR circuit 1009 is an example of a second processing circuit. For example, super-resolution processing is a process of lifting fine lines included in an image or emphasizing edges. More specifically, the SR circuit 1009 is a circuit in which the algorithm of super-resolution processing is implemented as a circuit. Also, the SR circuit 1009 performs super-resolution processing according to the intensity level of the super-resolution processing output from the level output circuit 1001.

[0037] Specifically, the SR circuit 1009 performs super-resolution processing according to the intensity level of the super-resolution processing on the first composite image synthesized by the first composite circuit 1008. Then, the NR circuit 1006 outputs the SR image, which is the image subjected to the super-resolution processing, to the second composite circuit 1010.

[0038] The coefficient selection circuit 1002 selects, based on the intensity level, the first coefficient set 1003 for causing the neural network circuit 1007 to perform noise reduction processing or the second coefficient set 1004 for causing the neural network circuit 1007 to perform super-resolution processing. The coefficient selection circuit 1002 is an example of a selection circuit. The first coefficient set 1003 is a set of coefficients to be set for each neuron of the neural network circuit 1007. For example, the first coefficient set 1003 is a set of coefficients for noise reduction processing when the intensity level is 4. Also, the first coefficient set 1003 is an example of the first setting information. The second coefficient set 1004 is a set of coefficients to be set for each neuron of the neural network circuit 1007. For example, the second coefficient set 1004 is a set of coefficients for super-resolution processing when the intensity level is 4. Also, the second coefficient set 1004 is an example of the second setting information.

[0039] More specifically, the coefficient selection circuit 1002 compares the intensity level of the noise reduction process with the intensity level of the super-resolution process. When the intensity level of the noise reduction process is higher than the intensity level of the super-resolution process, the coefficient selection circuit 1002 selects the first coefficient set 1003, which is a coefficient set for the noise reduction process. On the other hand, when the intensity level of the super-resolution process is higher than the intensity level of the noise reduction process, the coefficient selection circuit 1002 selects the second coefficient set 1004, which is a coefficient set for the super-resolution process. Then, the coefficient selection circuit 1002 outputs the selected first coefficient set 1003 or second coefficient set 1004.

[0040] Also, when the intensity level of the noise reduction process is the same as the intensity level of the super-resolution process, the coefficient selection circuit 1002 selects the first coefficient set 1003 that causes the neural network circuit 1007 to perform the noise reduction process with an intensity level of 4. Note that the coefficient selection circuit 1002 may select the second coefficient set 1004 that causes the neural network circuit 1007 to perform the super-resolution process, but it is preferable to select the first coefficient set 1003. The super-resolution process makes fine lines included in the image stand out and emphasizes edges. Also, the neural network circuit 1007 has a stronger image processing effect than the SR circuit 1009 or the NR circuit 1006. Therefore, when the super-resolution process acts more powerfully than the noise reduction process, there is a possibility that noise will be emphasized. Therefore, when the intensity level of the noise reduction process is the same as the intensity level of the super-resolution process, the coefficient selection circuit 1002 selects the first coefficient set 1003.

[0041] Note that the coefficient selection circuit 1002 shown in FIG. 2 stores the first coefficient set 1003 or the second coefficient set 1004. However, the coefficient selection circuit 1002 does not necessarily have to store the first coefficient set 1003 or the second coefficient set 1004. For example, the coefficient selection circuit 1002 may acquire the first coefficient set 1003 or the second coefficient set 1004 from a storage medium such as a RAM and output the acquired first coefficient set 1003 or second coefficient set 1004.

[0042] Further, the coefficient selection circuit 1002 has only one coefficient set with an intensity level of 4, rather than five coefficient sets with intensity levels from 0 to 4. In this way, the coefficient selection circuit 1002 selects the coefficient sets of some of the intensity levels among the multiple levels of intensity. As a result, the storage medium does not need to store the coefficient sets for all combinations of intensity levels of each image process. Here, since the coefficient sets are for setting each neuron of the neural network circuit 1007, the data volume is large. Therefore, if one were to prepare the coefficient sets for all combinations of intensity levels of each image process, a storage medium with a very large capacity would be required. By selecting the coefficient set from the first coefficient set 1003 or the second coefficient set 1004, the coefficient selection circuit 1002 does not need to have a storage medium with a very large capacity.

[0043] The neural network circuit 1007 is a circuit formed by a neural network that executes noise reduction processing or super-resolution processing based on the intensity levels of the noise reduction processing and super-resolution processing output from the level output circuit 1001. The neural network circuit 1007 is an example of a neural network circuit. More specifically, the neural network circuit 1007 executes the image processing of the first coefficient set 1003 or the second coefficient set 1004 selected by the coefficient selection circuit 1002 according to the intensity levels of the noise reduction processing and super-resolution processing.

[0044] That is, when the first coefficient set 1003 is output from the coefficient selection circuit 1002, the neural network circuit 1007 executes noise reduction processing with an intensity level of 4 indicated by the first coefficient set 1003. Also, when the second coefficient set 1004 is output from the coefficient selection circuit 1002, the neural network circuit 1007 executes super-resolution processing with an intensity level of 4 indicated by the second coefficient set 1004.

[0045] Further, the neural network circuit 1007 performs image processing according to the intensity level on the image subjected to noise reduction processing by the NR circuit 1006. That is, the neural network circuit 1007 performs noise reduction processing with an intensity level of 4 or super-resolution processing with an intensity level of 4. Then, the neural network circuit 1007 outputs the neural network image, which is the image subjected to image processing, to the first synthesis circuit 1008 and the second synthesis circuit 1010.

[0046] The first synthesis circuit 1008 synthesizes the NR image subjected to noise reduction processing by the NR circuit 1006 and the neural network image subjected to image processing by the neural network circuit 1007 based on the synthesis ratio calculated by the synthesis ratio calculation circuit 1005. The first synthesis circuit 1008 is an example of the first synthesis circuit. Also, the first synthesis circuit 1008 may synthesize the image by any method. For example, the first synthesis circuit 1008 may synthesize the image by multiplying the synthesis ratio as a weight coefficient. Then, the first synthesis circuit 1008 outputs the first synthesized image generated by synthesizing the NR image and the neural network image to the SR circuit 1009.

[0047] The second synthesis circuit 1010 synthesizes the SR image subjected to super-resolution processing by the SR circuit 1009 and the neural network image subjected to image processing by the neural network circuit 1007 based on the synthesis ratio calculated by the synthesis ratio calculation circuit 1005. The second synthesis circuit 1010 is an example of the second synthesis circuit. Also, the second synthesis circuit 1010 may synthesize the image by any method. For example, the second synthesis circuit 1010 may synthesize the image by multiplying the synthesis ratio as a weight coefficient. Then, the second synthesis circuit 1010 outputs the output image generated by synthesizing the SR image and the neural network image.

[0048] The composition ratio calculation circuit 1005 calculates a composition ratio indicating the ratio of a plurality of images to be composed based on the intensity levels of noise reduction processing and super-resolution processing respectively. The composition ratio calculation circuit 1005 is an example of a calculation circuit. More specifically, the composition ratio calculation circuit 1005 calculates a composition ratio for composing an image processed by the NR circuit 1006 or the SR circuit 1009 and a neural network image processed by the neural network circuit 1007 based on the difference between the intensity level of noise reduction processing and the intensity level of super-resolution processing.

[0049] FIG. 3 is a diagram for explaining an example of a method for calculating a composition ratio by the composition ratio calculation circuit 1005. The composition ratio calculation circuit 1005 calculates the composition ratio according to the graph shown in FIG. 3. The vertical axis of the graph indicates the intensity level of noise reduction processing. The horizontal axis of the graph indicates the intensity level of super-resolution processing. The first diagonal line R1, the second diagonal line R2, the third diagonal line R3, and the fourth diagonal line R4 are diagonal lines used to determine the composition ratio when the neural network circuit 1007 executes noise reduction processing. The fifth diagonal line R5, the sixth diagonal line R6, and the seventh diagonal line R7 are diagonal lines used to determine the composition ratio when the neural network circuit 1007 executes super-resolution processing.

[0050] In the graph shown in FIG. 3, the composition ratio calculation circuit 1005 detects an intersection point where the intensity level of noise reduction processing and the intensity level of super-resolution processing output from the level output circuit 1001 are orthogonal. When there is an intersection point on the first arrow L1 or the second arrow L2, the composition ratio calculation circuit 1005 plots the intersection point as the matching points P1 and P2 (see FIGS. 4 and 5). When there is no intersection point on the first arrow L1 or the second arrow L2, the composition ratio calculation circuit 1005 discriminates a diagonal line passing through the intersection point. That is, the composition ratio calculation circuit 1005 selects a diagonal line passing through the intersection point from the first diagonal line R1, the second diagonal line R2, the third diagonal line R3, the fourth diagonal line R4, the fifth diagonal line R5, the sixth diagonal line R6, and the seventh diagonal line R7.

[0051] Also, the composition ratio calculation circuit 1005 plots the points where the first arrow L1 or the second arrow L2 and the selected diagonal line are orthogonal as matching points P1 and P2 (see FIGS. 4 and 5). Then, the composition ratio calculation circuit 1005 determines the composition ratio based on the matching points P1 and P2 (see FIGS. 4 and 5) on the first arrow L1 or the second arrow L2.

[0052] Here, a specific example will be given using FIGS. 4 and 5 to explain the method for calculating the composition ratio by the composition ratio calculation circuit 1005. FIG. 4 is a diagram for explaining an example of the method for calculating the composition ratio by the composition ratio calculation circuit 1005. FIG. 4 shows a state where the intensity level of the noise reduction process is 3 and the intensity level of the super-resolution process is 2.

[0053] The composition ratio calculation circuit 1005 detects the intersection point where the intensity level of the noise reduction process and the intensity level of the super-resolution process are orthogonal. Since there is no intersection point detected on the first arrow L1 or the second arrow L2, the composition ratio calculation circuit 1005 detects the third diagonal line R3 passing through the intersection point. The composition ratio calculation circuit 1005 detects the point where the first arrow L1 and the third diagonal line R3 are orthogonal as the matching point P1. Then, the composition ratio calculation circuit 1005 determines the composition ratio of the image based on the matching point P1. Specifically, the matching point P1 divides the first arrow L1 into a ratio of 1 to 3. Therefore, the composition ratio calculation circuit 1005 determines that the ratio of the NR image is 75% and the ratio of the neural network image is 25%.

[0054] FIG. 5 is a diagram for explaining an example of a method for calculating a synthesis ratio by the synthesis ratio calculation circuit 1005. FIG. 5 shows a state where the intensity level of the noise reduction process is 1 and the intensity level of the super-resolution process is 3. The synthesis ratio calculation circuit 1005 detects an orthogonal intersection point where the intensity level of the noise reduction process and the intensity level of the super-resolution process are orthogonal. Since there is an intersection point detected on the first arrow L1 or the second arrow L2, the synthesis ratio calculation circuit 1005 detects the intersection point as the matching point P2. Then, the synthesis ratio calculation circuit 1005 determines the synthesis ratio of the image based on the matching point P2. Specifically, the matching point P2 divides the second arrow L2 in a one-to-one manner. Therefore, the synthesis ratio calculation circuit 1005 determines that the ratio of the NR image is 50% and the ratio of the neural network image is 50%.

[0055] As the matching points P1 and P2 (see FIGS. 4 and 5) approach the fourth diagonal line R4, the synthesis ratio calculation circuit 1005 reduces the proportion of the neural network image in the image synthesis. In other words, as the difference between the intensity level of the noise reduction process and the intensity level of the super-resolution process decreases, the synthesis ratio calculation circuit 1005 reduces the proportion of the neural network image processed by the neural network circuit 1007 among the NR image or SR image processed by the NR circuit 1006 or SR circuit 1009 and the neural network image processed by the neural network circuit 1007.

[0056] In the graph shown in FIG. 3, when the synthesis ratio calculation circuit 1005 plots the matching points P1 and P2 (see FIGS. 4 and 5) on the first arrow L1, as the matching points P1 and P2 (see FIGS. 4 and 5) approach the fourth diagonal line R4, the synthesis ratio calculation circuit 1005 reduces the proportion of the neural network image processed by the neural network circuit 1007 for noise reduction. On the other hand, when the synthesis ratio calculation circuit 1005 plots the matching points P1 and P2 (see FIGS. 4 and 5) on the second arrow L2, as the matching points P1 and P2 (see FIGS. 4 and 5) approach the fourth diagonal line R4, the synthesis ratio calculation circuit 1005 reduces the proportion of the neural network image processed by the neural network circuit 1007 for super-resolution.

[0057] In this way, the composition ratio calculation circuit 1005 can maintain the continuity of the image processing effect when the intensity level is changed by reducing the ratio of the neural network image according to the difference between the intensity level of the noise reduction process and the intensity level of the super-resolution process. Here, the neural network circuit 1007 has a higher image processing effect than the NR circuit 1006 and the SR circuit 1009. Therefore, when the content of the image processing executed by the neural network circuit 1007 is switched, the image processing effect changes.

[0058] For example, when the neural network circuit 1007 executes a noise reduction process, the first composition circuit 1008 generates a first composite image by composing the NR image and the neural network image. When the image processing executed by the neural network circuit 1007 is switched from the noise reduction process to the super-resolution process, the first composition circuit 1008 generates the first composite image from the NR image. In this case, since the neural network image is no longer used for image composition, if the ratio of the neural network image in the first composite image is high, the change in the effect of the noise reduction process becomes large. On the other hand, if the ratio of the neural network image in the first composite image is low, the change in the effect of the noise reduction process is small.

[0059] Also, when the difference in the setting level of the image processing is small, the content of the image processing executed by the neural network circuit 1007 is switched by a slight change in the setting level. Therefore, the composition ratio calculation circuit 1005 can maintain the continuity of the image processing effect by reducing the ratio of the neural network image as the difference in the setting level of the image processing becomes smaller.

[0060] Note that the composition ratio calculation circuit 1005 is not limited to the graph shown in FIG. 3 and may calculate the composition ratio. For example, the composition ratio calculation circuit 1005 calculates the difference between the intensity level of the noise reduction process and the intensity level of the super-resolution process. Then, the composition ratio calculation circuit 1005 calculates the composition ratio of the image based on the difference.

[0061] For example, when the difference is 0, the composition ratio calculation circuit 1005 sets the ratio of the image generated by the neural network circuit 1007 to 0% and the ratio of the image generated by the NR circuit 1006 or the SR circuit 1009 to 100%. Also, when the difference is 1, the composition ratio calculation circuit 1005 sets the ratio of the image generated by the neural network circuit 1007 to 25% and the ratio of the image generated by the NR circuit 1006 or the SR circuit 1009 to 75%. Also, when the difference is 2, the composition ratio calculation circuit 1005 sets the ratio of the image generated by the neural network circuit 1007 to 50% and the ratio of the image generated by the NR circuit 1006 or the SR circuit 1009 to 50%. Also, when the difference is 3, the composition ratio calculation circuit 1005 sets the ratio of the image generated by the neural network circuit 1007 to 75% and the ratio of the image generated by the NR circuit 1006 or the SR circuit 1009 to 25%. Also, when the difference is 4, the composition ratio calculation circuit 1005 sets the ratio of the image generated by the neural network circuit 1007 to 100% and the ratio of the image generated by the NR circuit 1006 or the SR circuit 1009 to 0%. Also, the ratio of each image may be arbitrarily changed.

[0062] As described above, the image processing circuit 1000 according to the embodiment includes an NR circuit 1006 that executes noise reduction processing, an SR circuit 1009 that executes super-resolution processing, and a neural network circuit 1007 that executes noise reduction processing or super-resolution processing based on the intensity level. Further, the composition ratio calculation circuit 1005 calculates a composition ratio indicating the ratio of a plurality of images to be composed based on the intensity levels of the noise reduction processing and the super-resolution processing respectively. The first composition circuit 1008 composes an NR image subjected to noise reduction processing by the NR circuit 1006 and a neural network image subjected to image processing by the neural network circuit 1007 based on the composition ratio. The second composition circuit 1010 composes an SR image subjected to super-resolution processing by the SR circuit 1009 and a neural network image subjected to image processing by the neural network circuit 1007 based on the composition ratio. Then, the image processing circuit 1000 outputs the neural network image composed by the second composition circuit 1010.

[0063] In this way, the image processing circuit 1000 causes the neural network circuit 1007 to execute noise reduction processing or super-resolution processing according to the intensity level. Further, the image processing circuit 1000 calculates a composition ratio based on the intensity level. Then, the image processing circuit 1000 composes the NR image and the neural network image according to the composition ratio, and composes the SR image and the neural network image according to the composition ratio. Therefore, the image processing circuit 1000 can change the respective intensity levels of a plurality of image processes while using the neural network circuit 1007.

[0064] Also, in the above-described embodiment, it has been described that the signal processing unit 108 has the image processing circuit 1000. However, members other than the signal processing unit 108 may have the image processing circuit 1000.

[0065] Also, in the above-described embodiment, the image processing circuit 1000 was described as being included in the television apparatus 10. However, the image processing circuit 1000 may be included in an apparatus other than the television apparatus 10. For example, a personal computer, a smartphone, a tablet terminal, a recorder that records images, or a display apparatus may include the image processing circuit 1000.

[0066] Although embodiments of the present invention have been described, these embodiments are presented by way of example and are not intended to limit the scope of the invention. The novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the gist of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are also included in the invention described in the claims and the equivalent scope thereof.

Description of Reference Numerals

[0067] 10…Television apparatus, 108…Signal processing unit, 1000…Image processing circuit, 1001…Level output circuit, 1002…Coefficient selection circuit, 1003…First coefficient set, 1004…Second coefficient set, 1005…Composite ratio calculation circuit, 1006…NR (noise reduction) circuit, 1007…Neural network circuit, 1008…First composite circuit, 1009…SR (Super Resolution) circuit, 1010…Second composite circuit, L1…First arrow, L2…Second arrow, P1, P2…Corresponding points, R1…First diagonal line, R2…Second diagonal line, R3…Third diagonal line, R4…Fourth diagonal line, R5…Fifth diagonal line, R6…Sixth diagonal line, R7…Seventh diagonal line.

Claims

1. A first processing circuit that executes noise reduction processing, which is an image processing for reducing noise in an image; A second processing circuit that executes super-resolution processing, which is an image processing for increasing the resolution of an image; A calculation circuit that calculates a synthesis ratio indicating the ratio of a plurality of images to be synthesized based on the intensity settings of the noise reduction processing and the super-resolution processing respectively; A neural network circuit formed by a neural network that executes the noise reduction processing or the super-resolution processing; A first synthesis circuit that synthesizes the image subjected to the noise reduction processing by the first processing circuit and the image subjected to the image processing by the neural network circuit based on the synthesis ratio; A second synthesis circuit that synthesizes the image subjected to the super-resolution processing by the second processing circuit and the image subjected to the image processing by the neural network circuit based on the synthesis ratio; A selection circuit that selects first setting information for causing the neural network circuit to execute the noise reduction processing or second setting information for causing the neural network circuit to execute the super-resolution processing based on the intensity setting; Comprising; The neural network circuit executes the image processing of the first setting information or the second setting information selected by the selection circuit; An image processing circuit.

2. The selection circuit selects the first setting information or the second setting information of some of the intensity settings of the multi-stage intensity settings; The image processing circuit according to claim 1.

3. The calculation circuit calculates the synthesis ratio for synthesizing the image subjected to the image processing by the first processing circuit or the second processing circuit and the image subjected to the image processing by the neural network circuit based on the difference between the intensity setting of the noise reduction processing and the intensity setting of the super-resolution processing; The image processing circuit according to claim 1 or 2.

4. As the difference between the intensity setting of the noise reduction processing and the intensity setting of the super-resolution processing decreases, the calculation circuit reduces the ratio of the image subjected to the image processing by the neural network circuit among the image subjected to the image processing by the first processing circuit or the second processing circuit and the image subjected to the image processing by the neural network circuit; The image processing circuit according to claim 3.

5. The first processing circuit executes the noise reduction process on the input input image. The neural network circuit executes the image processing according to the intensity setting on the image subjected to the noise reduction process by the first processing circuit. The first synthesis circuit synthesizes the image subjected to the noise reduction process by the first processing circuit and the image subjected to the image processing by the neural network circuit. The second processing circuit executes the super-resolution process on the image synthesized by the first synthesis circuit. The second synthesis circuit synthesizes the image subjected to the super-resolution process by the second processing circuit and the image subjected to the image processing by the neural network circuit. The image processing circuit according to any one of claims 1 to 4.

Images