KR101092207B1 - Method and Apparatus for Correcting Balance of Digital Radiation Detection Image having Nonlinear Response Characteristics - Google Patents

Abstract

According to the present invention, in order to correct an image balance such that luminance characteristics are flattened for each pixel of a digital radiation detection image having a nonlinear response characteristic, a nonlinear correction method and a linear correction method are appropriately adapted to the response characteristics of each section according to the response characteristics of the image sensor. By using a mixture, the present invention relates to a method and apparatus for correcting a balance of an image in which the number of subdivision sections is reduced while reducing the complexity of correction and precise balance correction is possible. According to an aspect of the present invention, there is provided a correction method for balancing an image, comprising: dividing second slope information of response characteristic data of an image sensor indicating an output luminance value for each input luminance value into a plurality of sections divided by upper and lower threshold values, and Correcting the pixel data in a correction scheme corresponding to a corresponding section among the plurality of sections of pixel data received from the image sensor.

Radiographic image, nonlinear response characteristics, planarization, first-order slope, second-order slope

Description

Method and Apparatus for Correcting Balance of Digital Radiation Detection Image with Non-Linear Response Characteristics {Method and Apparatus for Correcting Balance of Digital Radiation Detection Image having Nonlinear Response Characteristics}

The present invention relates to a method and apparatus for correcting an image balance, and to correcting an image balance such that luminance is flattened for each pixel of a digital radiation detection image having a nonlinear response characteristic, the response characteristic of each section according to the response characteristic of the image sensor The present invention relates to a method and apparatus for correcting a balance of an image in which a nonlinear correction method and a linear correction method are suitably used, thereby reducing the complexity of the correction while reducing the number of subdivision sections and enabling precise balance correction.

In general, in order to diagnose a patient's cancer or other lesions, X-ray radiation is used to irradiate the lesion and acquire an image. X-ray imaging technology has been widely used, such as a general radiography (GR) method for acquiring still images on a general analog film, and a computed radiography (CR) method for digitizing an image by reading an X-ray irradiated onto an image plate with a predetermined reader. . In addition, recently, a digital radiography (DR) method of digitizing an image by directly detecting an X-ray using a semiconductor image sensor has been introduced.

Recently, in the DR method, various radiation diagnostic devices using semiconductor flat panel image sensors have been developed and commercialized. Radiation diagnostic method using such a semiconductor flat plate image sensor is expected to be widely used for diagnosing the condition of breast cancer or other lesions due to its convenience.

The semiconductor flat panel image sensor used in such various diagnostic equipment is composed of millions of pixels, and all the pixels have little brightness characteristics due to the characteristics of the sensor.

An image balance correction method that flattens an image by correcting slightly different luminance characteristics for each pixel during the preprocessing process.In the case of an image sensor having a completely linear response characteristic, the average luminance value of the white image obtained by giving a constant input to the image sensor is actually used. A method of multiplying by the luminance value of the measured image and dividing by the luminance value of the white image of each pixel may be used.

However, in the case of an image sensor having a nonlinear response characteristic, a balance image cannot be obtained for an arbitrary input by the above-described processing method for the linear response characteristic. In general, the balance correction of the luminance characteristics of an image sensor having a nonlinear response characteristic uses a method of performing the above-described general linear correction for each section on the assumption that the input section is segmented and the response characteristic is linear in each segmented section. In the case of having a response characteristic in the form of a multi-order curve, a section is subdivided to make a quadratic curve in each section, and then a correction procedure is performed by a quadratic polynomial. However, in such a linear correction scheme for each section, the number of detailed sections increases for accurate correction, and the process of obtaining a corrected image for each section takes a long time and is complicated. In addition, in the above polynomial correction method, the number of subdivision sections can be reduced to reduce the process of obtaining a corrected image, but the application formula of the corrected image is more complicated than the linear correction method.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and an object of the present invention is to adjust an image balance such that luminance is flattened for each pixel of a digital radiation detection image having a nonlinear response characteristic, according to a response characteristic of an image sensor. The present invention provides a method and apparatus for correcting a balance of an image in which a nonlinear correction method and a linear correction method are appropriately mixed according to the response characteristics of each section, thereby reducing the complexity of the correction while reducing the number of subdivision sections and enabling precise balance correction.

First, to summarize the features of the present invention, a method for correcting the balance of the image according to an aspect of the present invention for achieving the above object, 2 for the response characteristic data of the image sensor indicating the output luminance value for each input luminance value Dividing the difference slope information into a plurality of sections divided into upper and lower thresholds; And correcting the pixel data in a correction scheme corresponding to a corresponding section among the plurality of sections of pixel data received from the image sensor.

The pixel data may be a digital image signal of each pixel detected from an image sensor that acquires image data of radiation in a radiation diagnostic equipment.

The second slope information may be information obtained by calculating a slope again and taking an absolute value with respect to the slope information of the response characteristic data.

The threshold may be zero.

The correcting may include: performing nonlinear correction on a section above the threshold; And performing linear correction on the section smaller than the threshold.

The nonlinear correction is to correct the pixel data based on a quadratic or higher quadratic function fitted to the response characteristic data for the interval, and the linear correction is to fit the response characteristic data for the interval. Correcting the pixel data based on a linear function.

The nonlinear correction is a correction that multiplies the luminance average value of all pixels for the corresponding section determined by the multi-order function with the pixel data, and then divides the multiplied value by the luminance value corresponding to the average luminance of the pixels in the corresponding section, The linear correction is a correction that multiplies the luminance average value of all pixels in the interval determined by the linear function with the pixel data, and then divides the multiplied value by the luminance value corresponding to the average luminance of the pixels in the interval. Include.

According to another aspect of the present invention, an apparatus for correcting an image balance includes: dividing second slope information of response characteristic data of an image sensor indicating an output luminance value for each input luminance value into a plurality of sections divided by upper and lower thresholds; Setting unit; And a correction unit configured to correct the pixel data in a correction scheme corresponding to a corresponding section among the plurality of sections of pixel data received from the image sensor.

The correction device may include a first slope calculator configured to generate slope information of the response characteristic data; And a second slope calculator which obtains slope information on the slope information from the first slope calculator and calculates an absolute value to generate the second slope information.

The correction unit may perform nonlinear correction on a section above the threshold and perform linear correction on a section below the threshold.

The correction unit corrects the pixel data based on a quadratic or higher quadratic function fitted to the response characteristic data for the corresponding section as the nonlinear correction, and applies the linear characteristic to the response characteristic data for the section. The pixel data may be corrected based on a first-order function fitted with respect to each other.

The correction unit, in the nonlinear correction, multiplies the luminance average value of all pixels for the corresponding section determined by the multi-order function with the pixel data, and then multiplies the multiplied value by the luminance value corresponding to the average luminance of the pixels in the corresponding section. In the linear correction, a luminance average value of all pixels may be multiplied by the pixel data in the corresponding interval determined by the first order function, and then the multiplied value may be divided by a luminance value corresponding to the average luminance of the pixels in the corresponding interval. have.

According to the image balance correction method and apparatus according to the present invention, by using an appropriate correction method in consideration of the response characteristics of each section of the image sensor instead of the conventional image correction method, while reducing the accuracy of the correction image acquisition process and application complexity while increasing the accuracy of the correction As a result, the image quality and reliability of the final result of the radiation detection image can be obtained. In particular, when applied to a diagnostic system that uses low doses, such as the thermosynthesis system, the precision correction becomes more important than the conventional method.

In order to fully understand the present invention, the operational advantages of the present invention, and the objects achieved by the practice of the present invention, reference should be made to the accompanying drawings which illustrate preferred embodiments of the present invention and the contents described in the accompanying drawings.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

In the present invention, in the diagnostic equipment for diagnosing the condition of cancer or other lesions of the patient using X-rays and other radiation, the angle detected from the image sensor in the process of acquiring the image data using a semiconductor image sensor and the like and providing the corresponding image By correcting the digital image signal of the pixel using an appropriate correction method that takes into account the response characteristics of each section of the image sensor, the accuracy of the correction is increased while the complexity of the acquisition and application of the correction image is reduced, resulting in the final result of the radiation detection image. To improve the image quality and reliability.

In the present invention, the correction method for the balance (or flattening) of the image by correcting the digital image signal from the image sensor as described above, 2 for the response characteristic data of the image sensor as shown in FIG. Dividing the difference slope information (refer to FIG. 3) into a plurality of sections that are divided into upper and lower thresholds (for example, 0) and the pixel data received from the image sensor in consideration of the response characteristics of the sections as described above. Compensating the received pixel data in a correction method corresponding to the corresponding section of the plurality of divided sections. The second slope information as shown in FIG. 3 may be obtained by calculating a slope for the first slope information (see FIG. 2) with respect to the response characteristic data as shown in FIG. 1.

As shown in FIG. 1, the nonlinear response characteristic data of an image sensor indicating an output luminance value (eg, luminance 0 to 255 output digital image signal) for each input luminance value (eg, luminance 0 to 255 input digital image signal) may be viewed. As can be seen, when this is processed as a correction method according to the linear or quadratic multi-order function as a whole, a correction error occurs and there are disadvantages according to each method.

For example, it can be seen that the response characteristics data of FIG. 1 have different response characteristics for each section (A, B, C). That is, the section A having a low input luminance value can be seen as a response characteristic in the form of a quadratic curve, and the middle section B can be viewed as a linear response characteristic, and in the section C having a high input luminance value. It is again quadratic. Therefore, in the section A having a low input luminance value, since the response characteristic is in the form of a quadratic curve, a nonlinear correction method based on a second or more ordered quadratic function is applied. Based on the linear correction method, the last section (C) also applies the nonlinear correction method based on the quadratic or higher quadratic function, so that the number of sections is reduced and the calculation amount is appropriate when applied to the actual measured value while performing more accurate correction. Can be calculated.

At this time, the distribution of the appropriate section as described above may be made in the following way. First, inclination information (for example, inclination information for each input luminance value) for the response characteristic data as shown in FIG. 1 may be generated using a predetermined inclination calculation means (first order inclination information). As such, the slope information on the response characteristic data may be represented as shown in FIG. 2. That is, the slope information on the response characteristic data indicates a slope value that is continuously increased in the section A, a constant slope value in the section B, and the slope value decreases in the section C again.

Here again, the slope information (for example, the information of the slope value for each input luminance value) with respect to the slope information may be generated using the predetermined slope calculation means. In this case, since the inclination value has a negative value in the section C, when the absolute value is taken and displayed, the inclination information (secondary inclination information) at this time may be represented as shown in FIG. 3. The inclination information as shown in FIG. 3 reflects the change in response characteristics, and thus, the point at which the inclination information as shown in FIG. 3 exceeds a predetermined threshold value may be set as the interval division value. In addition, the threshold can be appropriately adjusted according to the desired precision. If the threshold value is set to 0, it can be divided into three sections (A, B, and C) as shown in FIG. 3, and in response to the section A (section 2 having a second slope greater than or equal to the threshold), response characteristics are in the form of a secondary curve. Since the nonlinear correction method based on the quadratic function of the quadratic or higher order is applied, the linear correction method based on the linear function is applied in the B section (the second slope is smaller than the threshold), and the last C section (the second slope). In the case where the value is greater than or equal to the threshold value, the nonlinear correction method based on the second-order or higher quadratic function can also be applied to reduce the number of intervals and to calculate the calculation amount when applied to the actual measured value while performing more accurate correction.

Here, the nonlinear correction is performed by applying a predetermined numerical method to response characteristic data for a corresponding section (section where the second inclination value is greater than or equal to the threshold value) and fitting it to a quadratic or higher quadratic function having a curve similar to the data. Acquiring a multi-order function, and correcting the pixel data received from the image sensor based on the obtained multi-order function.

For example, in order to nonlinearly correct pixel data, first, a multi-order function is determined for each pixel as described above, and the luminance average value of all pixels for a corresponding section of the luminance value represented by the pixel data is determined by a predetermined calculation means. The luminance average value of all the pixels of the corresponding interval of the luminance value indicated by the pixel data may be obtained using the average of the polynomial functions of the corresponding interval. The non-linear correction of the pixel data may be completed by multiplying the average value thus obtained by the received pixel data and dividing the multiplied value by the luminance value corresponding to the average luminance of the pixel in the corresponding section. Here, the average luminance of the pixels in the corresponding section is not the all pixels but the average luminance of the corresponding pixels in the corresponding section.

In addition, linear correction is performed by applying a predetermined numerical analysis method to response characteristic data for a corresponding section (section where the second slope value is smaller than the threshold value) and fitting it to a linear function having a curve similar to that data. The difference function is obtained, and the pixel data received from the image sensor is corrected based on the obtained function.

For example, in order to linearly correct pixel data, first, a linear function is determined for each pixel as described above, and the luminance average value of all pixels for a corresponding section of the luminance value represented by the pixel data is determined by a predetermined calculation means. The luminance average value of all the pixels of the corresponding interval of the luminance value represented by the pixel data may be obtained using the average of the linear function of the corresponding interval. After multiplying the obtained average value by the received pixel data, the linear correction of the pixel data may be completed by dividing the multiplied value by the luminance value corresponding to the average luminance of the pixel in the corresponding section. Here, the average luminance of the pixels in the corresponding section is not the all pixels but the average luminance of the corresponding pixels in the corresponding section.

Through such linear and nonlinear correction, the image balance may be corrected such that luminance characteristics are flattened for each pixel of the radiation detection image as shown in FIG. 4. 410 of FIG. 4 is a result of image output for a constant luminance value of all pixels before correction, and 420 of FIG. 4 is a result of image output after applying the linear and nonlinear correction schemes to the same luminance value.

The correction method for the image balance may be implemented by the image correction device 500 according to an embodiment of the present invention as shown in FIG.

Referring to FIG. 5, the image correcting apparatus 500 according to an embodiment of the present invention may include a first tilt calculator 510, a second tilt calculator 520, a section setting unit 530, and a corrector. 540.

The first slope calculator 510 receives response characteristic data of an image sensor indicating output luminance values for each input luminance value as shown in FIG. 1, generates corresponding slope information, and generates first slope information as shown in FIG. 2. The first slope information is information on a slope value of response characteristic data for each input luminance value.

The second slope calculator 520 obtains slope information of the first slope information, calculates an absolute value, and generates second slope information as illustrated in FIG. 3. The second slope information is information on a slope value of the first slope information for each input luminance value.

The section setting unit 530 generates information for identifying each section by dividing the second slope information calculated as above into a plurality of sections that are divided into upper and lower thresholds (for example, 0). FIG. 3 illustrates that secondary slope information is divided into three sections A, B, and C when the threshold is set to 0. When the threshold is set to another value, the second slope information divided into upper and lower thresholds is shown. The sections may be divided into more sections.

The correction unit 540 corrects the pixel data with respect to the pixel data received from the image sensor by a correction method corresponding to the corresponding section among the plurality of sections according to the section identification information set by the section setting unit 530. For example, the correction unit 540 may perform nonlinear correction on a section above the threshold, and may perform linear correction on a section below the threshold.

For example, in the case of FIG. 3, the correction unit 540 has a nonlinear correction based on the quadratic function of the quadratic or higher because the response characteristic is in the form of a quadratic curve for the section A (the quadratic slope is greater than or equal to the threshold). In the B section (secondary slope value is smaller than the threshold), the linear correction method is applied based on the linear function, and in the last C section (secondary slope value is above the threshold), the second order or more The received pixel data may be corrected by applying a nonlinear correction method based on a function.

Here, for nonlinear correction, the correction unit 540 applies a predetermined numerical analysis method to response characteristic data for a corresponding section (section where the second inclination value is greater than or equal to the threshold value) and has a quadratic or higher quadratic function having a curve similar to the data. The corresponding multi-order function may be obtained by fitting a function, and the pixel data received from the image sensor may be corrected based on the obtained multi-order function.

For example, in order to nonlinearly correct pixel data, first, the correction unit 540 determines the multi-order function for each pixel as shown in FIG. The luminance average value of all the pixels can be obtained. The luminance average value (average luminance in the average curve for the curves of each pixel) of all the pixels for the corresponding interval of the luminance value represented by such pixel data may be calculated using an average of the multi-order function of the corresponding interval. The correction unit 540 multiplies the average value thus obtained by the received pixel data and divides the multiplied value by the luminance value corresponding to the average luminance (average luminance in the curve of the pixel) of the pixel in the corresponding section. Nonlinear correction of the pixel data can be completed. Here, the average luminance of the pixels in the corresponding section is not the all pixels but the average luminance of the corresponding pixels in the corresponding section.

In addition, for linear correction, the correction unit 540 applies a predetermined numerical method to response characteristic data for a corresponding section (a section in which the second inclination value is smaller than the threshold value) to a linear function having a curve similar to the data. The first linear function may be acquired by fitting, and the pixel data received from the image sensor may be corrected based on the obtained function. For example, in order to linearly correct pixel data, first, the correction unit 540 determines the linear function of each pixel as shown in FIG. The luminance average value of all pixels for the corresponding interval of may be obtained. The luminance average value of all pixels of the corresponding interval of the luminance value represented by the pixel data may be calculated using the average of the linear function of the corresponding interval. The correction unit 540 multiplies the average value thus obtained by the received pixel data and divides the multiplied value by the luminance value corresponding to the average luminance (average luminance in the curve of the pixel) of the pixel in the corresponding section. Linear correction of the pixel data can be completed. Here, the average luminance of the pixels in the corresponding section is not the entire pixel but the average luminance of the corresponding pixels in the corresponding section.

Through such linear and nonlinear correction, an unflattened image of all the pixels before correction of the radiation detection image as shown in 410 of FIG. 4 is flattened and balanced in all the pixels of 420 of FIG. 4 after correction. It can be output as

As described above, optimal embodiments have been disclosed in the drawings and the specification. Although specific terms have been used herein, they are used only for the purpose of describing the present invention and are not intended to limit the scope of the present invention as defined in the claims or the claims. Therefore, those skilled in the art will understand that various modifications and equivalent other embodiments are possible from this. Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

1 is a graph showing an output signal with respect to an input signal for explaining the nonlinear response characteristics of radiographic image data.

2 is a graph illustrating the inclination of FIG. 1.

3 is a graph illustrating the inclination of FIG. 2.

4 is a view for explaining a result of image processing according to the image correction method according to an embodiment of the present invention.

5 is a block diagram of an image correction device according to an embodiment of the present invention.

6 is a graph illustrating a linear / nonlinear correction performed by a correction unit according to an embodiment of the present invention.

Claims (12)

Dividing the second slope information on the response characteristic data of the image sensor indicating the output luminance value for each input luminance value into a plurality of sections including a section above a threshold and a section below the threshold; And Correcting the pixel data in a correction scheme corresponding to a corresponding section among the plurality of sections of pixel data received from the image sensor. Correction method for the balance of the image comprising a. The method of claim 1, And the pixel data is a digital image signal of each pixel detected from an image sensor for obtaining image data of radiation in a radiation diagnostic equipment. The method of claim 1, And the second inclination information is information obtained by calculating an inclination again and taking an absolute value with respect to the inclination information with respect to the response characteristic data. The method of claim 1, wherein the threshold is zero. The method of claim 1, wherein the correcting step, Performing nonlinear correction on a section above the threshold; And Performing linear correction on a section smaller than the threshold Correction method for the balance of the image comprising a. The method of claim 5, The nonlinear correction is to correct the pixel data based on a quadratic or higher quadratic function fitted to the response characteristic data for a corresponding section, And the linear correction corrects the pixel data based on a linear function fitted to the response characteristic data for a corresponding section. The method of claim 6, The nonlinear correction is a correction that multiplies the luminance average value of all pixels for the corresponding section determined by the multi-order function with the pixel data, and then divides the multiplied value by the luminance value corresponding to the average luminance of the pixels in the corresponding section, The linear correction is a correction that multiplies the luminance average value of all pixels in the interval determined by the linear function with the pixel data, and then divides the multiplied value by the luminance value corresponding to the average luminance of the pixels in the interval. A correction method for balancing an image characterized by. A section setting unit dividing second gradient information on response characteristic data of an image sensor indicating output luminance values for each input luminance value into a plurality of sections including a section above a threshold and a section below the threshold; And A correction unit configured to correct the pixel data in a correction scheme corresponding to a corresponding section among the plurality of sections of pixel data received from the image sensor. Correction apparatus for the balance of the image, characterized in that it comprises a. The method of claim 8, A first slope calculator configured to generate slope information on the response characteristic data; And A second slope calculator which obtains slope information on the slope information from the first slope calculator and calculates an absolute value to generate the second slope information Correction apparatus for the balance of the image, characterized in that it further comprises. The method of claim 8, wherein the correction unit, And performing non-linear correction on the section above the threshold, and performing linear correction on the section below the threshold. The method of claim 10, wherein the correction unit, As the nonlinear correction, the pixel data is corrected based on a quadratic or higher quadratic function fitted to the response characteristic data for the interval, And correcting the pixel data based on a linear function fitted to the response characteristic data for the section as the linear correction. The method of claim 11, wherein the correction unit, In the nonlinear correction, a luminance average value of all pixels for the corresponding interval determined by the multi-order function is multiplied by the pixel data, and then the multiplied value is divided by the luminance value corresponding to the average luminance of the pixels in the corresponding interval, In the linear correction, a luminance average value of all pixels is multiplied by the pixel data in the corresponding interval determined by the first order function, and then the multiplied value is divided by the luminance value corresponding to the average luminance of the pixels in the corresponding interval. Correction device for the balance of the image.

Images