JP2000074846A - Picture image processing device and its method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect and modify, accurately in a short time, a defective picture image included in plural sheets of picture images. SOLUTION: In this step S204, a filter treatment for detecting a straight line in a picture image region including a defective picture image is executed, and a filter treatment result is accumulated in a corresponding data table, corresponding to whether the value shows a positive or a negative, and Hough transform is executed by using the data table to thereby detect two line segments having the same gradient, and pixels interposed between detected two line segments are recognized as a pixel composing the defective picture image. And then, in the step S207, radii of necessary median filters are obtained relative to each pixel composing the defective picture image, and the filter treatment is executed to each corresponding pixel by the median filter having the obtained radius.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus and method, and more particularly, to an image processing apparatus and method for detecting or correcting a flaw image.

[0002]

2. Description of the Related Art A fragile medium such as a film for recording an image is scratched linearly over a plurality of frames by a foreign matter mixed into a transport system of a camera or a developing device or a protrusion caused by a defect generated in a device. It is rarely done. The digital image read from the damaged film will include a flaw image corresponding to the flaw.

[0003]

When such a flaw image is corrected using a computer, it is necessary for the operator to manually fill the flaw image area from the surroundings using image correction software or the like. Become. This correction operation is troublesome, and it is difficult to accurately detect and correct a flaw image included in a plurality of images in a short time.

An object of the present invention is to solve the above-mentioned problem, and an object of the present invention is to provide an image processing apparatus and a method thereof capable of accurately detecting a flaw image included in an image in a short time. .

It is another object of the present invention to provide an image processing apparatus and method capable of easily correcting a flaw image included in an image in a short time.

[0006]

The present invention has the following configuration as one means for achieving the above object.

An image processing method according to the present invention is an image processing method for detecting a flaw image included in an image. The image processing method performs a filter process for detecting a substantially straight line in an image area including the flaw image, Depending on whether the value indicates positive or negative, the filter processing result is stored in a corresponding data table, Hough transform is performed using the data table, and two line segments having a predetermined slope are calculated. A pixel sandwiched between the detected and detected two line segments is set as a pixel constituting the flaw image.

An image processing method for correcting a flaw image included in an image, wherein a filter processing for detecting a substantially straight line is applied to an image area including the flaw image, and the value indicates positive or negative. , The filter processing result is stored in a corresponding data table, Hough transform is performed using the data table, and two line segments having a predetermined inclination are detected. The pixels sandwiched between the line segments are the pixels constituting the flaw image, the radius of the median filter required for each pixel constituting the flaw image is obtained, and the median filter of the obtained radius filters the corresponding pixels. Is performed.

An image processing apparatus according to the present invention is an image processing apparatus for detecting a flaw image included in an image, wherein the processing means performs a filtering process for detecting a substantially straight line in an image area including the flaw image. And a storage means for storing the filter processing result in a corresponding data table according to whether the value indicates positive or negative, and performing a Hough transform using the data table to have a predetermined slope. Detecting means for detecting two line segments and detecting a pixel sandwiched between the detected two line segments as a pixel constituting the flaw image.

An image processing apparatus for correcting a flaw image included in an image, a processing means for performing a filtering process for detecting a substantially straight line in an image area including the flaw image, Storage means for storing the filter processing result in a corresponding data table according to whether the data processing is performed, and performing a Hough transform using the data table to detect two line segments having a predetermined inclination. Then, detection means for detecting a pixel sandwiched between the two detected line segments as pixels constituting the flaw image, and a radius of a median filter required for each pixel constituting the flaw image were obtained and obtained. Correction means for performing filter processing on a corresponding pixel with a median filter having a radius.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an image processing apparatus according to an embodiment of the present invention will be described in detail with reference to the drawings.

[Configuration] FIG. 1 is a block diagram showing a schematic configuration of a digital image processing apparatus according to the present invention.

In FIG. 1, a computer device 108 includes:
CPU 101, ROM 102, R interconnected by internal bus 106
A storage device 104 such as an AM 103 and a hard disk is provided. CPU 101 follows the BIOS stored in ROM 103, etc.
The program stored in the storage device 104 is copied to the RAM 103, and processing such as image processing is executed according to the program copied to the RAM 103. In addition, the CPU 101 communicates with a device connected to the external bus 109 via the interface 105.

Various instructions and data including an instruction to instruct image processing from the operator of the apparatus are input to an operation unit 115 such as a keyboard and a pointing device. The data is sent to the CPU 101 via the internal bus 106. For example, operation unit 1
If the instruction input from 15 is an instruction to read the original image, the CPU 101 controls the reading device 114 such as a film scanner or an image reader to read the original image, and obtains the obtained image data and its image data. The type of the document is stored in the storage device 104 or the RAM 103. When an instruction to print an image is given, the CPU 101
The image data stored in 4 is sent to the printing device 117, and an image based on the image data is recorded on recording paper.

The CPU 101 displays the result of the processing and the progress of the processing on a display device 116 such as a CRT or an LCD to show the operator. The information displayed on the display device 116 includes input instructions and data, results of image processing of read images, and the like. Of course, according to instructions from the operator, unprocessed images and processed images may be displayed. it can.

FIG. 2 is a flowchart showing an example of a process for detecting and correcting a substantially linear flaw image according to an embodiment of the present invention. This is a process executed by the CPU 101 when instructed.

In step S201, the type of scratch is initially set according to the type of recording medium (eg, a negative film or a positive film) related to the image to be processed. For example, the correction target is set to a scratch image having a high brightness in the case of a negative film, and is initially set to a scratch image having a low brightness in the case of a positive film.

Next, in step S202, the type of the flaw image initially set in step S201 is displayed on the display device 116. If the type of the displayed flaw image is not correct, the operator inputs and corrects the type of the flaw image using the operation unit 115.

Next, in step S203, an image to be processed is displayed on the display device 116. Based on the displayed processing target image, the operator inputs and specifies the position of the wound image to be corrected using the operation unit 115. The position of the flaw image is specified by, for example, using a pointing device such as a mouse to specify the vicinity of both ends in the longitudinal direction of the flaw image.

Next, in step S204, pixels constituting the flaw image near the position of the specified flaw image are detected.
Although details will be described later, detection / correction image data for detecting and correcting the designated flaw image is created (step S20).
5) A Huff transform is performed on the obtained image data for detection / correction (step S206), thereby detecting a flaw pixel.

Next, in step S207, a damaged pixel is corrected. As will be described in detail later, the minimum median filter radius required to correct the flaw image is determined for each pixel detected as constituting the flaw image (step S208), and the median filter having the obtained radius is used. The applied filter processing is performed on the corresponding pixel (step S209).

Next, in step S210, the correction result, that is, the image on which the flaw image has been subjected to the correction processing is displayed on the display device 116, and the operator determines whether or not the desired correction result has been obtained. Subsequently, in step S211, if an instruction indicating “possible” is input from the operator, the process is terminated. After returning the value of the (detected pixel) to the value before correction, the process returns to step S202.
Therefore, the processing of steps S212 to S210 is repeated until an instruction indicating yes is input from the operator in step S211.

FIG. 3 is a diagram for explaining the creation of detection / correction image data for detecting and correcting a flaw image.

It is assumed that the position of the flaw image specified by the operator is a line segment L forming an angle θ with the x axis. This line segment
The length of L is expanded by e in both directions, and image data included in a rectangular area expanded by length d in a direction orthogonal to the line segment L is used as detection / correction image data Id. The angle of the scratch image to be detected by creating this image data Id is
It becomes almost zero degree in the image data Id.

[Detection of Flaw Image] FIG. 4 is a flowchart showing a detailed procedure example of the Hough transform (S206).

First, in step S501, a filtering process by a filter for detecting a horizontal straight line shown in FIG. 5 is performed by a lightness component (for example, an L * component of L * a * b * data) Io of the image data Id. And the output pixel value is recorded in image data If having the same size as the image data Io. Note that step
The type of the flaw image set through S201 and S202 uses the filter 401 shown in FIG. 5 when the flaw image has a high brightness, and when the flaw image has a low brightness, the sign of the coefficient is opposite to that of the filter 401. A certain filter 402 is used.

Next, in step S502, a locus as shown in FIG. 6 is obtained from all the pixel values (Xi, Yi) in the image data If by equation (5.1), and the pixel values (Xi, If Yi) is positive, it is stored in the AB plane data table Tdp.
In the data table Tdm of the AB plane different from Tdp, the pixel value (X
i, Yi). B = -Xi · A + Yi… (5.1) The above equation represents a straight line with slope -Xi and intercept Yi in AB plane

Here, since the angle between the line segment L corresponding to the flaw image in the image data Id and the X axis in the image data Id is substantially zero, the inclination A does not increase to infinity.

Next, in step S503, candidates BU and BL above and below the intercept B are obtained for each slope Aj. First, each inclination Aj in the area near Ao and Bo (shown by equations (5.2) and (5.3)) obtained from the designated position of the flaw image
, Let the slope and intercept having the smallest value among the pixel values (Aj, B) stored in Tdm be (Aj, Bmin). Ao-ΔA <A <Ao + ΔA… (5.2) Bo-ΔB <B <Bo + ΔB… (5.3)

Next, within Tdp, data is examined in the positive and negative directions of the B axis from the coordinates (Aj, Bmin), and in each direction, the pixel value (Aj, B) stored in Tdp indicates the maximum value. Are BU and BL, respectively.

Next, in step S504, an evaluation value SAj is set by equation (5.4). SAj = {Tdm (Aj, BU) + Tdm (Aj, BL)} / 2-Tdp (Aj, Bmin)… (5.4)

Next, in step S505, the obtained evaluation value SA
j and the maximum value Smax of SAj so far, SAj> Sm
In the case of ax, SAj is substituted for Smax. Also, A at that time
Substitute j, BU, and BL for Amax, BUmax, and BLmax, respectively.

Next, in step S507, equations (5.2) and (5.
It is determined whether or not the processing from steps S503 to S506 has been completed for all the inclinations A in the area indicated by 3). If the processing has been completed, the process proceeds to step S508; otherwise, the process returns to step S503.

Next, in step S508, the obtained Amax, BU
Let max and BLmax be the linear parameters representing the wound image.
Subsequently, in step S509, the number of pixels on the long side of the image data Id
Determine whether Ps is greater than a specific number of pixels Pf, Ps ≤ Pf
In step S511, the process jumps to step S511; otherwise, the process proceeds to step S510, where a margin proportional to the pixel number Ps is set.
Add to BUmax and BLmax. BUmax = BUmax + C (Ps-Pf)… (5.5) BLmax = BLmax-C (Ps-Pf)… (5.6) where C is a proportional coefficient

Finally, in step S511, a pixel sandwiched between two line segments having the inclination Amax, the intercept BUmax, and the intercept BLmax in the image data Id is set as a pixel constituting the flaw image.

[Correction of Scratch Image] FIG. 7 shows the modification of the scratch image (S20).
It is a flowchart which shows the detailed example of a procedure of 7). In addition,
Steps S601 to S606 correspond to step S208, and step S607 corresponds to step S209.

In step S601, the radius of the median filter
Set r to 1. Note that the median filter having a radius of 1 is a median filter that uses an area including a pixel of interest and a pixel in which a coefficient 1 is entered as shown in FIG.

Next, in step S602, the image data Ib in which the pixel constituting the flaw image is "1" and the other pixels are "0"
And the image data Ib is subjected to a filtering process using a median filter with a radius r in step S603. Subsequently, in step S604, the radius r of the median filter used in step S603 is added to the pixel position of the image data Im corresponding to the pixel position of the image data Ib whose pixel value has newly become “0” by the immediately preceding filtering process. Write. The image data Im
And the image data Ib have the same size.

Next, in step S605, the pixel value of the filtered image data Ib is checked. If all are "0", the process proceeds to step S607; otherwise, the process proceeds to step S606 to determine the radius r of the median filter. Is increased by “1”, and the process returns to step S602.

Finally, in step S607, a filtering process using a median filter having a radius r recorded in the image data Im is performed on the pixels of the corresponding image data Id, and in step S608, the corrected image data Id is set to θ The damaged pixel is corrected by reflecting the rotated image in the original image.

As described above, according to the present embodiment, the following effects can be obtained. (1) A linear scratch image can be detected, and only the detected scratch image can be corrected. (2) Since the image around the designated flaw image is extracted and the processing is performed only on the extracted image part, the detection range of the flaw image can be limited, the detection time can be shortened, and the detection accuracy can be improved. (3) Since the sign of the coefficient value of the filter for detecting a flaw image is changed depending on the level of the brightness of the flaw image, the flaw image can be detected regardless of the level of the brightness of the flaw image. (4) The detection error can be compensated by increasing the width of the line segment detected as the flaw image in proportion to the number of pixels of the original image. (5) By applying the filtering process by the median filter for each pixel to correct the flaw image, it is possible to correct the flaw image while preserving the edge of the original image and suppressing disturbance of the image. (6) Since the radius of the median filter required for each pixel forming the flaw image is determined, the radius of the median filter required for each pixel forming the flaw image can be set accurately.

[0042]

[Other Embodiments] Even if the present invention is applied to a system including a plurality of devices (for example, a host computer, an interface device, a reader, a printer, etc.), an apparatus (for example, a copying machine) Machine, facsimile machine, etc.).

Another object of the present invention is to provide a storage medium storing a program code of software for realizing the functions of the above-described embodiments to a system or an apparatus, and to provide a computer (or a CPU or MPU) of the system or the apparatus.
Needless to say, this can also be achieved by the PU) reading and executing the program code stored in the storage medium. In this case, the program code itself read from the storage medium implements the functions of the above-described embodiment, and the storage medium storing the program code constitutes the present invention. When the computer executes the readout program code, not only the functions of the above-described embodiments are realized, but also the OS running on the computer based on the instructions of the program code.
It goes without saying that an (operating system) performs a part or all of the actual processing, and the processing realizes the functions of the above-described embodiments.

Further, after the program code read from the storage medium is written into a memory provided in a function expansion card inserted into the computer or a function expansion unit connected to the computer, based on the instructions of the program code, It goes without saying that the CPU included in the function expansion card or the function expansion unit performs part or all of the actual processing, and the processing realizes the functions of the above-described embodiments.

[0045]

As described above, according to the present invention,
An image processing apparatus and a method for accurately detecting a flaw image included in an image in a short time can be provided.

Further, it is possible to provide an image processing apparatus and method for easily correcting a flaw image included in an image in a short time.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a schematic configuration of a digital image processing apparatus according to the present invention;

FIG. 2 is a flowchart illustrating an example of a process of detecting and correcting a linear scratch image,

FIG. 3 is a diagram for explaining creation of detection / correction image data for detecting and correcting a flaw image;

FIG. 4 is a flowchart showing a detailed procedure example of the Hough transform shown in FIG. 2,

FIG. 5 is a diagram showing an example of a filter for detecting a straight line;

FIG. 6 is a diagram for explaining Hough transform.

7 is a flowchart showing a detailed procedure example of the correction of the flaw image shown in FIG. 2,

FIG. 8 is a diagram for explaining a radius of a median filter.

 ──────────────────────────────────────────────────続 き Continued on the front page F-term (reference) 2G051 AA41 AA90 AB02 AC04 EA11 EA12 EA14 EA16 EB01 EC01 EC10 ED01 ED08 ED09 ED21 ED30 5B057 AA20 BA02 BA24 CC03 CE01 CE06 DA03 DA08 DC13 5C077 LL02 5 LL19 CA05 PP05 FA17 FA24 GA55 MA03

Claims (13)

[Claims] 1. An image processing method for detecting a flaw image included in an image, comprising: performing a filtering process for detecting a substantially straight line in an image area including the flaw image; , The filter processing result is stored in a corresponding data table, Hough transform is performed using the data table, and two line segments having a predetermined inclination are detected. An image processing method, wherein pixels sandwiched between book segments are pixels constituting the flaw image. 2. The image processing method according to claim 1, wherein an area including the flaw image is manually designated. 3. The image according to claim 1, wherein a sign of a coefficient value of a filter used in the filtering process is changed according to a level of brightness of a flaw image to be detected. Processing method. 4. The image according to claim 1, wherein the width of the two detected line segments is increased in proportion to the number of pixels of the original image. Processing method. 5. An image processing method for correcting a flaw image included in an image, wherein a filter processing for detecting a substantially straight line is performed on an image area including the flaw image, and the value indicates positive or negative. , The filter processing result is stored in a corresponding data table, Hough transform is performed using the data table, and two line segments having a predetermined inclination are detected. The pixels sandwiched between the line segments are pixels constituting the flaw image, the radius of the median filter required for each pixel constituting the flaw image is obtained, and the median filter of the obtained radius filters the corresponding pixels. Image processing method. 6. A median filter process is performed on image data in which the value of the pixels constituting the flaw image is 1 and the values of the other pixels are 0 while sequentially increasing the radius of the image data. 6. The image processing method according to claim 5, wherein a radius of a median filter for each pixel required for correction is obtained. 7. The image processing method according to claim 5, wherein an area including the flaw image is manually designated. 8. The filter according to claim 5, wherein a sign of a coefficient value of a filter used in the filtering process is changed according to a level of brightness of a flaw image to be detected. Image processing method. 9. The image according to claim 5, wherein the width of the two detected line segments is increased in proportion to the number of pixels of the original image. Processing method. 10. An image processing apparatus for detecting a flaw image included in an image, a processing means for performing filter processing for detecting a substantially straight line in an image area including the flaw image, Accumulating means for accumulating the filter processing result in a corresponding data table in accordance with whether the line processing is negative or positive, and performing a Hough transform using the data table to detect two line segments having a predetermined inclination An image processing apparatus comprising: a detection unit configured to detect a pixel sandwiched between the two detected line segments as a pixel constituting the flaw image. 11. An image processing device for correcting a flaw image included in an image, a processing means for performing a filter process for detecting a substantially straight line in an image area including the flaw image, Accumulating means for accumulating the filter processing result in a corresponding data table in accordance with whether the line processing is negative or positive, and performing a Hough transform using the data table to detect two line segments having a predetermined inclination Then, detection means for detecting a pixel sandwiched between the two detected line segments as pixels constituting the flaw image, and a radius of a median filter required for each pixel constituting the flaw image are obtained and obtained. Correction means for performing filter processing on a corresponding pixel with a median filter having a radius. 12. A recording medium on which is recorded a program code for image processing for detecting a flaw image included in an image, wherein the program code is used to detect a substantially straight line at least in an image area including the flaw image. And a code for storing the filtering result in a corresponding data table in accordance with whether the value indicates positive or negative, and A code of a step of performing conversion and detecting two line segments having a predetermined inclination, and a code of a step of detecting a pixel sandwiched between the detected two line segments as a pixel constituting the flaw image. A recording medium comprising: 13. A recording medium on which a program code for image processing for correcting a flaw image included in an image is recorded, wherein the program code is used to detect a substantially straight line at least in an image area including the flaw image. And a code for storing the filtering result in a corresponding data table in accordance with whether the value indicates positive or negative, and A code of a step of performing conversion and detecting two line segments having a predetermined inclination, and a code of a step of detecting a pixel sandwiched between the detected two line segments as a pixel constituting the flaw image, The code of the step of obtaining the radius of the median filter required for each pixel constituting the flaw image and the median filter of the obtained radius correspond. And a code for performing a filtering process on the pixels.