JP2007026232A - Noise removal device, image processor and noise removal method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a noise removal device, an image processor and a noise removal method, capable of holding original image data as much as possible even while properly removing a noise component. <P>SOLUTION: This noise removal device having a noise removal means 36 removing the noise component included in the image data read by a solid-state imaging element has: a target pixel extraction means 33 sequentially extracting a notice pixel from the image data; a peripheral pixel characteristic amount calculation means 34 calculating a characteristic amount of a peripheral pixel on the basis of peripheral pixel data of the notice pixel; and a noise component influence degree decision means 35 deciding an influence degree of the noise component to the target pixel on the basis of target pixel data and the calculated characteristic amount of the peripheral pixel. When it is decided that the influence degree of the noise component to the target pixel is larger than a prescribed threshold value by the noise component influence degree decision means 35, the noise removal means 36 replaces the target pixel data with the peripheral pixel characteristic amount. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a noise removal device, an image processing device, and a noise removal method including a noise removal unit that removes a noise component included in image data read by a solid-state imaging device.

The image data read by the solid-state image sensor includes noise components such as shot noise caused by fluctuations in charge generated when the image data is read by the solid-state image sensor. For this reason, for example, as shown in Patent Document 1, the luminance component image data in the image data is subjected to predetermined filter processing with each pixel of the luminance component image data as a target pixel, and the entire luminance component is smoothed. Image data is generated, and the color difference component image data in the image data is subjected to another predetermined filtering process using each pixel of the color difference component image data as a target pixel to generate color difference component image data that is smoothed as a whole. Thus, a method for removing noise components from image data has been proposed.
JP 2005-167399 A

  However, when the smoothing process is uniformly performed on each pixel of the image data as in the conventional technique described above, the smoothing process is performed on the pixels not including the noise component, and the image data is blurred. In addition, the smoothed pixel data still contains a smoothed noise component, and when the influence of the noise component on the image data is large, the smoothed image data However, there is still a problem that the influence of noise components remains largely.

  The present invention provides a noise removing device, an image processing device, and a noise removing method capable of retaining original image data as much as possible while appropriately removing noise components in view of the above-described conventional problems. There is in point to do.

  In order to achieve the above-mentioned object, the first characteristic configuration of the noise removing apparatus according to the present invention is the noise included in the image data read by the solid-state imaging device as described in claim 1 of the claims. A noise removing apparatus including a noise removing unit that removes a component, wherein a target pixel extracting unit sequentially extracts a target pixel from the image data, and a feature amount of the peripheral pixel based on peripheral pixel data of the target pixel. Surrounding noise feature amount calculating means for calculating, and noise component influence degree determining means for determining an influence degree of a noise component on the attention pixel based on the calculated feature amount of the surrounding pixel and the attention pixel data, When the influence degree of the noise component with respect to the target pixel is determined by the component influence degree determination means to be greater than a predetermined threshold, the noise removal means Certain target pixel data in terms of the replacement process to the peripheral pixel features.

  With the above-described configuration, when the influence degree of the noise component with respect to the target pixel is large, the pixel data of the target pixel is replaced with the feature amount of the peripheral pixel in the target pixel, so that the noise component is extracted from the target pixel. In addition, when the influence degree of the noise component in the target pixel is small, it is possible to prevent the pixel data in the target pixel from being replaced with other data. That is, the original image data can be retained as much as possible while appropriately removing the noise component.

  As described in claim 2, the second characteristic configuration is more appropriate in addition to the first characteristic configuration described above, in that the feature amount of the peripheral pixel is an average density value of the peripheral pixel. This makes it possible to remove noise components.

  According to the third feature configuration, in addition to the first or second feature configuration described above, when the average density value of the image data is lower than a predetermined density threshold, the noise The removing means is that the pixel-of-interest data is replaced with the peripheral pixel feature amount.

The noise component included in the image data is more generated when the amount of received light is smaller when read by the solid-state imaging device, and the degree of influence on each pixel is larger. When the amount of light received by the solid-state image sensor is small, the average density value of the image data is low. That is, with the above-described configuration, the noise removal unit can remove the noise component only when the amount of noise component included in the image data is larger and the influence level is higher. It is. That is, unnecessary noise components can be prevented from being removed, and the original image data can be retained as much as possible while appropriately removing noise components.
In the fourth feature configuration, as described in claim 4, in addition to any of the first to third feature configurations described above, the peripheral pixel feature amount calculating unit may remove the noise from the peripheral pixels. A feature amount of the peripheral pixel is calculated based on the pixel data subjected to the replacement processing when a pixel whose pixel data has been replaced by the means is included, and noise components can be appropriately removed It becomes possible.

  In the fifth feature configuration, as described in claim 5, in addition to any of the first to fourth feature configurations described above, image data from which the noise removing means removes noise is read by a film scanner. The image data is in the point.

  In order to achieve the above object, the first characteristic configuration of the image processing apparatus according to the present invention is the density correction process or the contrast correction process based on the image data read by the solid-state imaging device as described in claim 6. An image processing apparatus including a correction processing unit that includes a noise removing device having any one of the first to fifth characteristic configurations described above, wherein the correction processing unit is denoised by the noise removing device. The image data is subjected to density correction processing or contrast correction processing.

  With the above configuration, it is possible to perform density correction processing and contrast correction processing based on image data that retains the original image data as much as possible while appropriately removing noise components. Therefore, it is possible to generate image data.

  In order to achieve the above object, the first characteristic configuration of the noise removal method according to the present invention is the noise removal for removing the noise component included in the image data read by the solid-state image sensor as described in claim 7. A noise removal method including a step, a target pixel extraction step of sequentially extracting a target pixel from the image data, and a peripheral pixel feature amount calculation that calculates a feature amount of the peripheral pixel based on peripheral pixel data of the target pixel And a noise component influence determination step for determining the influence degree of the noise component on the target pixel based on the calculated feature amount of the surrounding pixel and the target pixel data. In the noise component influence determination step, When it is determined that the degree of influence of the noise component on the target pixel is greater than a predetermined threshold, in the noise removal step, There the serial target pixel data in terms of the replacement process to the peripheral pixel features.

  As described above, according to the present invention, it is possible to provide a noise removal device, an image processing device, and a noise removal method capable of retaining original image data as much as possible while appropriately removing noise components. Can now.

  Hereinafter, an embodiment in which a noise removing apparatus or an image processing apparatus according to the present invention is applied to a photographic image processing apparatus will be described. As shown in FIG. 2, the photographic image processing apparatus includes a film scanner 1 that reads a photographic image from a negative film (hereinafter simply referred to as “photographic film”) F, etc., a monitor C1, a keyboard C2, and a mouse. Operation with C3, system controller C, and multiple drivers for various types of card-type memory such as smart media and compact flash (registered trademark), and magnetic and optical storage media such as CD-ROM and MO The setting unit OP and a printer 3 that exposes and develops photographic paper based on the photographic image data read by the film scanner 1 and outputs a photographic print P. The film scanner 1 includes an operation setting unit OP and The main body composed of the printer 3 is separated from the main body.

  As shown in FIG. 3, the printer 3 includes a paper magazine 50 in which roll-shaped photographic paper P is accommodated, a plurality of photographic paper conveyance rollers 52 that draws and conveys the photographic paper from the paper magazine 50, and conveyance rollers A motor 54 for driving 52, a fluorescent beam type print head 60 for exposing the photosensitive surface of the conveyed photographic paper P, and a development processing unit 62 for performing development, bleaching, and fixing processes on the exposed photographic paper. The image forming apparatus includes a drying unit 64 that conveys the developed photographic paper while drying, and a discharge unit 66 that discharges the dried photographic paper P as a final print. Note that the photographic paper P drawn out from the paper magazine 50 is cut into a predetermined print size by a cutter (not shown) disposed either before or after the development processing and is output to the discharge unit 66.

  The print head 60 includes three rows of a red light emitting block, a green light emitting block, and a blue light emitting block in which phosphor elements each having a lens and a color filter mounted on a phosphor whose light emission is controlled by adjusting a grid voltage are arranged in the main scanning direction. The photographic image is generated on the photographic paper by controlling the driving based on the R component, G component, and B component image data read by the film scanner 1.

  The system controller C includes a microcomputer system including a CPU, a ROM, a RAM, an I / F circuit, and the like, and is based on an input operation from the keyboard C2 in accordance with various information displayed on the monitor C1. Each block of the above-described film scanner 1 and printer 3 is controlled and operated according to a predetermined procedure.

  The film scanner 1 is recorded on the illumination optical system 10 that irradiates the photographic film F with reading illumination light, the film transport unit 20 that transports the photographic film F, and the photographic film F transported by the film transport unit 20. An image reading unit 30 for reading the frame image, and an image reading process control unit 40 for controlling an image reading process by the illumination optical system 10, the film transport unit 20, and the image reading unit 30. The image reading process control means 40 and the system controller C are connected by a control signal line L1 for controlling system operation, and the image reading means 30 and the system controller C are connected by an image signal line L2 for transmitting image data. It is connected.

  The illumination optical system 10 includes a halogen lamp 11, a dimming filter 12 that adjusts the color distribution of the light beam from the halogen lamp 11, a cylindrical lens 13 that condenses the light beam in a slit shape, and a uniform intensity distribution. A diffusing plate 14 and a narrow slit 15 are provided, and the photographic film is irradiated with slit light along the main scanning direction. The diffusion plate 14 and the slit 15 are attached to a film transport mechanism 20 described later.

  The film transport mechanism 20 includes a plurality of transport roller pairs 21 that transport the photographic film F toward the film projection unit directly below the slit 15, and a motor 22 that drives the transport roller pairs 21. The photographic film F is transport-controlled at a predetermined speed by the image reading process control means 40.

  The image reading unit 30 includes a condensing lens, a CCD line sensor as a solid-state image sensor, a sample hold circuit, an A / D converter, and the like. From the illumination optical system 10 that has passed through the photographic film F, The slit light is imaged on a CCD line sensor by a condenser lens, and an analog signal read by the CCD line sensor is converted into digital data by an A / D converter. The CCD line sensor is composed of three rows of CCD line sensors each provided with a color filter that selectively transmits light of R component, G component, and B component of transmitted light. As the photographic film F is conveyed, the upper frame images are separated into R component, G component, and B component color components and read.

  When a photographic film reading command is sent by the system controller C, the image reading process control means 40 turns on the halogen lamp 11 and then transports the photographic film F at a predetermined speed by the film transport mechanism 20. The frame image recorded on the photographic film is sequentially read by the image reading means 30, and the read image data is transmitted to the system controller C via the image signal line L2. The received image data is stored in the storage means 70.

  Then, after noise components included in the image data stored in the storage means 70 are removed, image processing such as density correction processing and contrast correction processing is further performed, and the print driven based on the image data is performed. The photographic paper P is exposed by the head 60 and developed by the development processing unit 62 to be output as a photographic print.

  As shown in FIG. 1, the functional block configuration relating to the removal of the noise component contained in the image data and the correction process for the image data from which the noise component has been removed is stored in the storage unit 70 as shown in FIG. An image data noise removing unit 31 that removes noise components from the image data read by the reading unit 30, and density correction processing, contrast correction processing, and the like for the image data from which noise components have been removed by the image data noise removing unit 31 The correction processing means 32 for performing the correction processing is provided. The image data noise removing unit 31 constitutes the main part of the noise removing apparatus of the present invention, and the image data noise removing unit 31 and the correction processing means 32 constitute the main part of the image processing apparatus of the present invention. ing.

  The image data noise removing unit 31 is a pixel-of-interest extraction unit 33 that sequentially extracts a pixel of interest from the image data, and a peripheral pixel feature amount calculation that calculates a feature amount of the peripheral pixel based on peripheral pixel data of the pixel of interest Means 34, noise component influence determination means 35 for determining the influence degree of the noise component on the target pixel based on the feature amount of the peripheral pixel calculated by the peripheral pixel feature amount calculation means 34 and the target pixel data, A noise removing unit 36 for replacing the pixel-of-interest data with the peripheral pixel feature amount when the noise-component influence determining unit 35 determines that the influence of the noise component on the pixel of interest is greater than a predetermined threshold; It is prepared for.

  As shown in FIG. 4, the pixel-of-interest extracting means 33 sequentially extracts the pixel of interest Pa from the image data Da according to the pixel arrangement, and extracts the peripheral pixel Pb (i) of the extracted pixel of interest Pa. (I = 1, 2,..., 8). Note that the extraction of the pixel of interest Pa is performed for each of the R component, the G component, and the B component, and the extraction of the peripheral pixel Pb (i) is performed for the color component having the same color as the extracted pixel of interest Pa. It is comprised so that it may be implemented.

  The peripheral pixel feature amount calculating unit 34 calculates the feature amount of the peripheral pixel Pb (i). For example, the peripheral pixel feature amount calculating unit 34 calculates the peripheral pixel Pb (i) from the density value Nb (i) of the peripheral pixel Pb (i). ) Average density value Nave.

  The noise component influence degree determination unit 35 is configured to calculate the target pixel Pa based on the average density value Nave of the peripheral pixel Pb (i) calculated by the peripheral pixel feature amount calculation unit 34 and the density value Npa of the target pixel Pa. For example, the ratio Rn = Npa / Nave between the density value Npa and the average density value Nave at the target pixel Pa is larger than a predetermined threshold value Nth. In some cases, it is determined that the degree of influence of the noise component on the target pixel Pa is large.

  The noise component influence determination unit 35 determines that the difference Vn = Npa−Nave between the density value Npa and the average density value Nave in the target pixel Pa is larger than a predetermined threshold value Vth. In addition, it may be configured to determine that the influence degree of the noise component on the target pixel Pa is large.

  The noise removing unit 36 determines the density value Npa of the target pixel as an average of the peripheral pixels when the noise component influence determining unit 35 determines that the influence level of the noise component on the target pixel Pa is large. The noise component is removed from the target pixel Pa by performing a replacement process with the density value Nave. The noise removing unit 36 does not perform the replacement process for the target pixel Pa when the noise component influence level determining unit 35 does not determine that the influence level of the noise component on the target pixel Pa is large. The density value Npa at the target pixel Pa is maintained as it is.

  The correction processing unit 32 performs density correction processing, contrast correction processing, and the like on the image data from which the noise component has been removed by the image data noise removal unit 31 based on a density correction function and a contrast correction function that are set in advance. The correction process is implemented.

  Hereinafter, the removal of the noise component contained in the image data and the operation of the correction process for the image data from which the noise component has been removed will be described with reference to the flowchart of FIG. When the image data read by the image reading unit 30 is stored in the storage unit 70 (SA1), the target pixel extraction unit 33 extracts the target pixel Pa from the image data Da (SA2), A peripheral pixel Pb (i) of the extracted target pixel Pa is extracted (SA3).

  When the peripheral pixel Pb (i) is extracted, the peripheral pixel feature amount calculating unit 34 calculates the average density value of the peripheral pixel Pb (i) from the density value Nb (i) of the peripheral pixel Pb (i). Nave is calculated (SA4).

  The noise component influence degree determination unit 35 compares the ratio Rn = Npa / Nave between the density value Npa and the average density value Nave in the target pixel Pa with a predetermined threshold value Nth set in advance. The influence degree of the noise component on the target pixel Pa is determined. That is, when the ratio Rn = Npa / Nave between the density value Npa and the average density value Nave in the pixel of interest Pa is larger than a predetermined threshold value Nth, the influence of the noise component on the pixel of interest Pa It is determined that the degree is large.

  When the noise component influence degree determination means 35 determines that the influence degree of the noise component with respect to the target pixel Pa is large (SA5), the noise removal means 36 determines the density value Npa of the target pixel as the peripheral value. By replacing the average density value Nave of the pixels (SA6), noise components are removed from the pixel of interest Pa.

  Further, when the noise component influence degree determination means 35 does not determine that the influence degree of the noise component with respect to the target pixel Pa is large (SA5), the noise removing means 36 performs a replacement process for the target pixel Pa. Without performing this, the density value Npa at the target pixel Pa is maintained as it is.

  Then, the pixel-of-interest extraction means 33 extracts a new pixel of interest Pa from the image data Da (SA2), and extracts a peripheral pixel Pb (i) of the extracted new pixel of interest Pa (SA3).

  The target pixel Pa is sequentially extracted from the image data Da according to the pixel arrangement by the target pixel extracting means 33, and the processing from step SA2 to step SA6 is completed for all the pixels in the image data Da. When the noise component is removed from the image data (SA7), the correction processing means 32 removes the noise component by the image data noise removing unit 31 based on the preset density correction function and contrast correction function. Correction processing such as density correction processing and contrast correction processing is performed on the obtained image data (SA8).

  The image data corrected by the correction processing unit 32 is temporarily stored in the storage unit 70 and then output to the print head 60.

  Here, it is preferable that the pixel feature amount calculating unit 34 calculates a feature amount based on peripheral pixel data from which noise components have not yet been removed by the noise removing unit 36, but the target pixel extracting unit 33 In the process of sequentially extracting the peripheral pixel Pb (i), when the peripheral pixel Pb (i) includes the peripheral pixel Pbh (i) from which the noise component has been removed by the noise removing unit 36, In the pixel data in the peripheral pixel Pbh (i), the pixel feature amount calculating unit 34 may calculate the feature amount using the pixel data after the noise component removal.

  Further, when the average density value Fave of the image data Da stored in the storage unit 70 is lower than a predetermined density threshold Fth, the noise removing unit 36 sets the density value Npa of the target pixel Pa to the average of the peripheral pixels. It is good also as a structure which substitutes to density value Nava. Mean density value calculating means for calculating the average density value Fave of the image data Da stored in the storage means 70 is provided, and the noise removing means 36 performs the replacement process based on a preset density threshold value Fth. What is necessary is just composition. It is possible to prevent unnecessary noise components from being removed, and it is possible to retain the original image data as much as possible while appropriately removing noise components.

  The embodiment described above is merely an example of the present invention, and it goes without saying that the specific configuration and the like of each block can be appropriately changed and designed within the scope of the effects of the present invention.

Explanatory drawing of the functional block structure concerning the removal process of the noise component contained in image data, and the correction process with respect to the image data from which the noise component was removed Explanatory drawing of the external configuration of the photographic image processing apparatus Explanatory drawing of the general structure of the photographic image processing device Explanatory diagram of relationship between target pixel and surrounding pixels Flow chart for explaining the operation of the removal of the noise component contained in the image data and the correction processing for the image data from which the noise component has been removed

Explanation of symbols

1: film scanner 30: image reading means 31: image data noise removing unit 32: correction processing means 33: attention pixel extracting means 34: peripheral pixel feature amount calculating means 35: noise component influence degree judging means 36: noise removing means 70: Storage means

Claims (7)

A noise removing device including a noise removing unit that removes a noise component included in image data read by a solid-state imaging device,
Pixel-of-interest extracting means for sequentially extracting the pixel of interest from the image data; peripheral pixel feature amount calculating means for calculating the feature amount of the peripheral pixel based on peripheral pixel data of the target pixel; and A noise component influence determination means for determining an influence degree of a noise component with respect to the target pixel based on a feature amount and the target pixel data;
When the noise component influence degree determining means determines that the influence degree of the noise component with respect to the target pixel is larger than a predetermined threshold, the noise removing means replaces the target pixel data with the peripheral pixel feature amount. To remove noise.   The noise removal apparatus according to claim 1, wherein the feature amount of the peripheral pixel is an average density value of the peripheral pixel.   3. The noise removing device according to claim 1, wherein when the average density value of the image data is lower than a predetermined density threshold, the noise removing unit replaces the target pixel data with the peripheral pixel feature amount. 4.   The peripheral pixel feature amount calculating unit includes a feature amount of the peripheral pixel based on the pixel data subjected to the replacement processing when the peripheral pixel includes a pixel whose pixel data is replaced by the noise removing unit. The noise removal device according to any one of claims 1 to 3, wherein   The noise removal apparatus according to any one of claims 1 to 4, wherein the image data from which the noise removal unit removes noise is image data read by a film scanner. An image processing apparatus comprising correction processing means for performing density correction processing or contrast correction processing based on image data read by a solid-state imaging device,
The noise removal device according to any one of claims 1 to 5 is provided,
An image processing apparatus in which the correction processing means performs density correction processing or contrast correction processing on the image data from which noise has been removed by the noise removal device. A noise removal method including a noise removal step of removing a noise component included in image data read by a solid-state imaging device,
A pixel-of-interest extraction step for sequentially extracting a pixel-of-interest from the image data; a peripheral pixel feature-amount calculating step for calculating a feature amount of the peripheral pixel based on peripheral pixel data of the pixel of interest; A noise component influence determination step for determining an influence degree of a noise component with respect to the target pixel based on a feature amount and the target pixel data;
When the noise component influence degree determining step determines that the influence degree of the noise component with respect to the target pixel is larger than a predetermined threshold value, the noise removing step replaces the target pixel data with the peripheral pixel feature amount. Noise removal method to process.

Images