JP2001043360A - Method and device for processing image - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily extract the specified object of an image and to faultlessly eliminate the extracted object by extracting and eliminating the elimination object image while utilizing the feature of the elimination object image included in the image. SOLUTION: The image of an original such as a film is read by an image input part 110, image data are generated and these image data are stored in a memory 130. In an image processing part 120, the elimination object image included in the image is extracted while utilizing the feature of the elimination object image. In this case, when the elimination object image is a character image, the feature of the elimination object image to be utilized for extraction signifies one of form, edge information, lightness, color and arrangement or combination thereof. The image processing part 120 samples the pixel values of pixels around the pixels of the extracted elimination object image. Concerning an area extracted as the elimination object image, the pixel values are replaced with a most frequent appearance value or average sampled around that area so that the elimination object image can be eliminated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing method and an image processing apparatus, and more particularly, to an image for extracting an image by using characteristics of an image to be removed contained in the image and removing the extracted image to be removed. The present invention relates to a processing method and an image processing device.

[0002]

2. Description of the Related Art Conventionally, there have been the following methods for extracting a portion (abnormal pixel) of an image to be removed.

A first method is to manually specify pixels of an image to be removed in an image displayed on a monitor by using photo retouching software. In addition, as a second method, there is a method in which an abnormal pixel is automatically extracted by setting a threshold value of a density change and a threshold value of a size of a pixel area by an operator.

As a third method, when reading image information from a film, there is a method in which infrared light transmitted through the film is read to obtain image information consisting only of information on dust and scratches (Japanese Patent Application Laid-Open (KOKAI)). 7-97402 and the like).

Further, as a fourth method, there is a method of extracting dust pixels by paying attention to the feature amount of dust (see Japanese Patent Application Laid-Open No. H10-111930).

[0006]

However, in the first method described above, not only is the operator required to have a high level of skill, but also when the number of pixels to be corrected is large, the operator has an excessive burden. There was a problem that it took.

Further, in the above second method, it is difficult to set the threshold value optimally, and in particular, there is a problem that erroneous extraction of abnormal pixels increases in a high-frequency portion. Further, the above third method has a problem that a device for reading infrared light is required separately at the time of image input.

Further, the fourth method is an extraction / removal technique corresponding to dust, and is limited to dust.
Therefore, it cannot respond to other objects. Further, in the case of dust, only the size is mentioned as a feature of the shape, so it has been difficult to apply it to other objects.

As described above, it has been relatively difficult to remove a specific object from an image. For this reason, it was troublesome to remove the date imprinted by the date function of the camera and the like.

The present invention has been made in view of the above problems, and an image processing method and an image processing apparatus capable of easily extracting a specific target of an image and reliably removing the extracted target. It is intended to provide.

[0011]

According to the present invention, an image to be removed is extracted using characteristics of the image to be removed included in an image.
An image processing method and an image processing apparatus, wherein the extracted removal target image is removed.

In this image processing, extraction is performed by using the characteristics of the image to be removed, and the extracted image to be removed is removed. Can be reliably removed.

In the present invention, the feature of the image to be removed is any one of shape, edge information, brightness, color, and arrangement, or any combination thereof. This makes it possible to reliably extract the image to be removed by a simple operation.

In the present invention, the image to be removed is a character image. As a result, the date and message imprinted by the date function and message function of the camera can be reliably extracted.

Further, in the present invention, the extraction using the feature of the image to be removed is performed by any one of a shape combining vertical or horizontal line segments, presence of an edge, and lighter than surrounding pixels. Based on the combination of
It is characterized by the following. This makes it possible to reliably extract a specific image to be removed having a predetermined characteristic.

In the present invention, the extraction using the feature of the image to be removed is performed based on the arrangement of any one of four corners of the image. As a result, it is possible to reliably extract a date or the like by the date function that is imprinted on a specific position.

In the present invention, the image to be removed is a horizontal streak (horizontal streak) generated when the silver halide film passes through the apparatus in any of photographing, developing, and reading. And In such image processing, it becomes possible to effectively extract and remove horizontal stripe flaws generated in the film.

In the present invention, the image to be removed is a muscle scar. In such image processing,
It is possible to effectively extract and remove lateral muscle flaws generated in the film.

In the present invention, the extraction using the feature of the image to be removed is performed by any one of a linear shape in a certain direction, the presence of an edge, and brightness different from surrounding pixels. , Based on the combination of This makes it possible to reliably extract a specific image to be removed having a predetermined characteristic.

Further, the present invention is characterized in that, in the above case, the parameter of the feature can be set or changed in order to absorb the variation of the feature. Through such image processing, the removal target image is reliably extracted, and fluctuations can be dealt with.

[0021]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a block diagram illustrating a configuration of an image processing apparatus 100 according to an embodiment of the present invention.

In FIG. 1, reference numeral 110 denotes an image input unit to which image data is input, and 120 denotes an image processing unit which receives image data from the image input unit and performs various types of image processing.
When the image input unit 110 reads an image on a film, the image input unit 110 itself may be an image sensor such as a film scanner, or may be an input interface for reading image data from an external film scanner. There is. Further, as the image processing unit 120, processors such as a CPU and a DSP and image processing software are used. Note that, in the present embodiment, the image processing unit 120 constitutes an extracting unit and a removing unit in order to perform extraction and removal processing.

Reference numeral 130 denotes a memory for temporarily storing an image input from the image input unit 110 and for storing various information. Reference numeral 140 denotes a user operation unit for inputting various operation commands and the like to the image processing unit 120. For example, a mouse, a joystick, a keyboard, or the like is used.

Reference numeral 150 denotes a user operation display unit which assists input and operation of the user operation unit 140.
An RT, a liquid crystal display, a plasma display, or the like is used. The operator inputs various control commands according to the display on the user operation display unit 150. When a CRT display or the like is used as an image output unit 160 described later, the user operation display unit 150 can be shared with the image output unit 160.

Reference numeral 160 denotes an output unit for outputting image data processed by the image processing unit, for example, an output interface for outputting to a display, a printer, or the like as an external device; And a display for displaying images.

In the above configuration, the image input unit 1
The input image data generated in 10 is input to the image input unit 110
Is an image sensor, the image input unit 110
Generated by When the image input unit 110 is an input interface, input image data is generated by an external film scanner (not shown).

The image output unit 160 displays an image on which image processing described below has been performed, or
Image data that has been subjected to image processing is sent to an external device.

The operation of the image processing apparatus and the image processing method according to the embodiment of the present invention will be described below with reference to the flowchart of FIG. 2 and the explanatory diagram of FIG. First, an image of a document such as a film (not shown) is input to the image input unit 110.
To generate image data, and store the image data in the memory 130 (S1 in FIG. 2).

Here, it is assumed that the image data of the image read by the image input unit 110 is as shown in FIG. Here, in the lower right part of the image, the date (character image) imprinted by the date function of the camera is present.

In this embodiment, the original image such as a character image of this kind or a message (character image) imprinted by a similar camera function does not exist. The image is treated as an image to be removed in order to remove the image. Also, streak-like scratches (skin wounds) on the film
Is treated as an image to be removed. in this case,
Muscle scratches occur when a silver halide film passes through a device in any of photographing, developing, and reading.

Then, the image processing unit 120 extracts the image to be removed included in the image by using the characteristics of the image to be removed (S2 in FIG. 2). Here, the characteristics of the image to be removed used for extraction are, when the image to be removed is a character image, any one of shape, edge information, brightness, color, and arrangement, or a combination of any of them. Or any combination of horizontal line segments, presence of edges, brighter than surrounding pixels, or any combination Arrangement at any of the four corners of the image (lower right is preferred) Numbers and characters If the image to be removed is a streak, one of a linear shape in a certain direction, the presence of an edge, and a brightness different from the surrounding pixels, or any combination Applicable.

As described above, the image processing unit 120 performs processing on all the pixels of the image data stored in the memory 130.
Alternatively, any of the four corners is prioritized, and a feature that satisfies the following conditions is detected by applying a known image processing technique.

Under the above conditions, a region having an edge amount that is at least half the value of the maximum edge amount is extracted as an image to be removed. In addition, since the number of the date imprinted by the date function of the camera is often a rectangle, the rectangle is preferentially detected as the shape. Also,
As for the color, a color close to orange is preferentially detected.

Then, when the extraction is completed, the image processing section 120 collects the pixel values of the peripheral pixels of the extracted pixel of the image to be removed (S3 in FIG. 2). In this case, it is detected whether or not a sudden change occurs in the pixel value, and sampling of the pixel value is not performed in a region where the sudden change occurs. In addition,
In this case, the most frequently occurring value and the average value are also obtained.

Then, the image to be removed is removed by replacing the region extracted as the image to be removed with the most frequently occurring value or the average value collected in the vicinity (FIG. 2S).
4). Note that if replaced with one pixel value, the trace of the image to be removed may be conspicuous, so it is desirable to randomly arrange pixels having the most frequent value, the average value, or some values close to them. In this case, if the distribution of pixel values around the image to be removed has a certain direction, the direction is also respected. Then, when such an image to be removed is removed, the area other than the image to be removed is kept at the original pixel value.

In the example of the image shown in FIG. 3, the area at the lower right of the date is extracted as an image to be removed, and is replaced with the pixel values of surrounding pixels. When sampling surrounding pixel values, pixel values on the wall surface of the building are not sampled.

The new image data in which the extraction and removal of the image to be removed has been completed as described above is stored in the memory 130. Then, the new image data is output to the image output unit 160.
To output image data, print, or output image data.

In the above embodiment, the lower right corner of the four corners of the screen is preferentially selected as a candidate for the extraction position. However, the edge amount at each of the four corners is detected by processing such as FFT, and the edge amount is maximized. May be determined as the area where the image to be removed exists.

In the above embodiment, the size of the character of the date as the image to be removed is not specified. However, the extraction may be performed in a state where the range of the size of the character is limited. Alternatively, the designation may be made from the user operation unit 140.

In the above-described embodiment, the case where the edge amount is fixed to half or more has been described. However, in order to increase the extraction accuracy, the edge amount may be selected from the user operation unit 140.

The image processing apparatus 100 described in the above embodiment may be constituted as a single image processing apparatus. However, by incorporating the image processing apparatus 100 in a film scanner, it is possible to remove the image processing before image processing such as gradation conversion. The target image can be removed, and more accurate processing can be realized. Also, the deterioration of the image quality is reduced.

It is also possible to perform the image processing of this embodiment at the same time as the other gradation conversion processing. Image quality can be maintained.

As described above, in the image processing according to the present embodiment, extraction is performed using the characteristics of the image to be removed.
Since the extracted removal target image is removed, a specific target of the image can be easily extracted, and the extracted target can be reliably removed.

Further, according to the present invention, the feature of the image to be removed is any one of shape, edge information, brightness, color, and arrangement, or a combination of any of them. This makes it possible to reliably extract the image to be removed by a simple operation.

In the image processing of the present embodiment, the date and message imprinted by the date function and the message function of the camera can be reliably extracted by using the image of characters and numbers as the image to be removed.

In the image processing according to the present embodiment, the extraction using the feature of the image to be removed is performed by any one of a shape combining vertical or horizontal line segments, the presence of an edge, and a lighter than surrounding pixels. Since it is performed based on one or any combination, a specific image to be removed having a predetermined characteristic can be reliably extracted.

Further, in the image processing of this embodiment, the extraction using the features of the image to be removed is performed based on the arrangement of any one of the four corners of the image. It is possible to reliably extract the date and the like by the date function which is imprinted on the position.

Further, in the image processing of this embodiment, the image to be removed is a muscle flaw, so that the horizontal muscle flaw generated in the film can be effectively extracted and removed. Become.

In the image processing of this embodiment, the image to be removed is a horizontal streak (horizontal streak) generated when the silver halide film passes through the apparatus during photographing, development, or reading. This feature makes it possible to effectively extract and remove lateral muscle flaws generated in the film. Further, in the image processing of the present embodiment, in the above case, the parameter of the feature amount can be set or changed in order to absorb the fluctuation of the feature amount. It is reliable and can handle fluctuations.

In the image processing according to the present embodiment, the extraction using the characteristics of muscle scars and horizontal streaks as the image to be removed is performed as follows.
Identification based on one or a combination of a linear shape in a certain direction, presence of edges, and brightness different from surrounding pixels Can be reliably extracted.

[0051]

As described above, according to the present invention, the following effects can be obtained. In the image processing of the present invention, extraction is performed using the characteristics of the image to be removed, and the extracted image to be removed is removed. It can be reliably removed. This makes it possible to reliably and effectively remove a date portion such as a photograph.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an embodiment of an image processing apparatus according to the present invention.

FIG. 2 is a flowchart illustrating an embodiment of an image processing method according to the present invention.

FIG. 3 is an explanatory diagram showing a state of extraction of an image to be removed in image processing according to the present invention.

[Explanation of symbols]

 REFERENCE SIGNS LIST 100 Image processing device 110 Image input unit 120 Image processing unit 130 Memory 140 User operation unit 150 User operation display unit 160 Image output unit

Claims (18)

[Claims] 1. An image processing method comprising: extracting a removal target image using characteristics of a removal target image included in an image; and removing the extracted removal target image. 2. The image processing apparatus according to claim 1, wherein the feature of the image to be removed is any one of shape, edge information, brightness, color, and arrangement, or a combination thereof. Method. 3. The image processing method according to claim 1, wherein the image to be removed is a character image. 4. The extraction using the feature of the image to be removed is performed by any one of a shape combining vertical or horizontal line segments, the presence of an edge, and lighter than surrounding pixels, or any combination thereof. 4. The image processing method according to claim 3, wherein the method is performed on the basis of: 5. The image processing method according to claim 3, wherein the extraction using the feature of the image to be removed is performed based on the arrangement of any one of four corners of the image. 6. The image processing method according to claim 1, wherein the image to be removed is a muscle scar. 7. The image processing method according to claim 1, wherein the image to be removed is a horizontal streak generated when the silver halide film passes through a device in any of photographing, developing, and reading. 8. The extraction using the feature of the image to be removed is based on one or a combination of a linear shape in a certain direction, the presence of an edge, and brightness different from surrounding pixels. The image processing method according to claim 6, wherein the image processing method is performed. 9. The image processing method according to claim 1, wherein a parameter of the feature amount can be set or changed for absorbing the variation of the feature amount. 10. An image processing apparatus comprising: extraction means for extracting a removal target image using characteristics of a removal target image included in an image; and removal means for removing the extracted removal target image. Image processing device. 11. The extraction means comprises: shape, edge information,
11. The image processing apparatus according to claim 10, wherein any one of brightness, color, and arrangement, or any combination thereof is extracted as a feature of the image to be removed. 12. The image processing apparatus according to claim 10, wherein the extraction unit extracts a character image as an image to be removed. 13. The extraction using the feature of the image to be removed by the extraction means may be any one of a shape combining vertical or horizontal line segments, the presence of an edge, and a lighter than surrounding pixels. The image processing apparatus according to claim 12, wherein the image processing is performed based on a combination of: 14. The image processing apparatus according to claim 12, wherein the extraction using the feature of the image to be removed by the extraction unit is performed based on an arrangement of any one of four corners of the image. 15. The image processing apparatus according to claim 10, wherein the extraction unit extracts a muscle scar as the image to be removed. 16. The apparatus according to claim 10, wherein the extracting means extracts, as the image to be removed, a horizontal streak generated when the silver halide film passes through a device in any of photographing, developing, and reading. The image processing apparatus according to any one of the preceding claims. 17. The extraction using the feature of the image to be removed by the extraction means includes a linear shape in a certain direction, the presence of an edge,
17. The image processing apparatus according to claim 15, wherein the image processing is performed based on any one of brightness different from surrounding pixels or a combination thereof. 18. The image processing apparatus according to claim 10, further comprising an operation unit capable of setting or changing a parameter of the feature amount for absorbing a variation of the feature amount. apparatus.