CN118301304A - Correction method for image color range conversion and image processing system - Google Patents

Abstract

A correction method for converting the color range of image includes such steps as counting the pixel values in image to obtain the actual color range of image, which can be full-color range or limited color range, when the actual color range is different from the color range recorded in image package, and executing a color range correction program. The method can convert the limited color range of all pixels in the image into the full color range according to a linear mapping method, and can gradually carry out the color range correction procedure in a multi-stage correction mode in the process so as to accurately and smoothly finish the conversion and execute the subsequent image processing procedure.

Description

Correction method for image color range conversion and image processing system

Technical Field

The present disclosure relates to a technique for correcting an image color range in a specific color space, and more particularly, to a method and an image processing system for performing conversion and correction according to the detected image color range.

Background

The image data input into the video and audio device is usually transmitted in a specific color coding mode, and when the image data is processed in the video and audio device, the image processing needs to be performed according to the information recorded in the image data packet, for example, the subsequent processing is performed according to the color range (color range).

Taking a YUV image as an example, the color range (color range) of the YUV image refers to the range of luminance and chrominance in the YUV color space, and the color range can be divided into a limited color range (narrow range or LIMITED RANGE, brightness range is [16,235], chrominance range is [16,240 ]) and a full color range (full range, range is [0,255 ]). For example, according to the itu.bt709/itu.bt2100 standard, the luminance limited range (luminance LIMITED RANGE) interval may be [16×2 ^n-8,235×2^n-8 ], the chrominance limited range (chrominance LIMITED RANGE) interval may be [16×2 ^n-8,240×2^n-8 ], where n is the number of image sequence bits, and the minimum should be 8 bits (n+.8).

Generally, in the input image data, a color range is recorded in each image sequence packet (e.g., a header), so that a processing circuit in the video/audio device performs image processing according to the color range, including converting the input image into an image conforming to a specific transmission specification (e.g., HDMI) or mixing (blending) the input image with a specific playing interface.

However, many image packets in image data lack color range information or the header records wrong color range information, which may cause errors in subsequent processing and generate image color distortion (distortion).

Therefore, the solution proposed in the prior art is that the processing circuit in the video and audio apparatus can obtain the correct color range according to the statistics data of the image pixels, and correct the image according to the correct color range, but the quality of the image watched by the user may be affected due to sudden change of real-time brightness or color correction when correcting the image.

Disclosure of Invention

In view of the influence of the lack of image color range information or the error of information in the image data, which requires to correct the image, or the sudden change generated when correcting the image, the specification provides a correction method and an image processing system for converting the image color range.

According to an embodiment, in a correction method for converting an image color range in an image processing system, an image is obtained first, data of the image is stored in a buffer, pixel values of the image in the buffer are counted to obtain an actual color range of the image, the actual color range can be a full color range or a limited color range, and if the actual color range does not match a color range recorded in an image packet, a color range correction procedure is performed to correctly process the image.

Further, in the method for counting the pixel values of the image, the system sets an upper limit threshold and a lower limit threshold, compares the pixel values of all or sampled pixels in the image with the upper limit threshold and the lower limit threshold, calculates the number of pixels with the pixel values larger than the upper limit threshold or smaller than the lower limit threshold, and compares the number of pixels with the upper limit threshold or the lower limit threshold, thereby judging the actual color range of the image.

Further, in the color range correction procedure, a linear mapping method is used to convert all pixels in the image from a limited color range to a full color range. In particular, the color range correction program may employ a progressive conversion method in which the color range conversion is set to be performed stepwise in a multistage correction manner, and one image is outputted by performing the color range conversion by adjusting one gradation at a time until the image is completely converted into a full color range.

For a further understanding of the nature and the technical aspects of the present invention, reference should be made to the following detailed description of the invention and the accompanying drawings, which are provided for purposes of reference only and are not intended to limit the invention.

Drawings

FIG. 1 is a diagram of an embodiment of an apparatus for applying a correction method for image color range conversion;

FIG. 2 is a main flow chart of a correction method for converting the color range of an image;

FIG. 3 is a diagram of an embodiment of an image processing system that performs a correction method for image color range conversion;

FIG. 4 is a diagram of an embodiment of a logic circuit for setting a threshold in an image processing system;

FIG. 5 is a diagram of an embodiment of a logic circuit for statistical operations in an image processing system;

FIG. 6 is a diagram of an embodiment of logic for screening images in an image processing system;

FIG. 7 is a diagram of an embodiment of logic for a linear mapping method in an image processing system;

FIG. 8 is a flow chart of an embodiment of a method for correcting image color range conversion; and

Fig. 9A to 9C are diagrams showing embodiments of performing the asymptotic color range correction method.

Detailed Description

The following embodiments of the present invention are described in terms of specific examples, and those skilled in the art will appreciate the advantages and effects of the present invention from the disclosure herein. The invention is capable of other and different embodiments and its several details are capable of modifications and various other uses and applications, all of which are obvious from the description, without departing from the spirit of the invention. The drawings of the present invention are merely schematic illustrations, and are not intended to be drawn to actual dimensions. The following embodiments will further illustrate the related art content of the present invention in detail, but the disclosure is not intended to limit the scope of the present invention.

It will be understood that, although the terms "first," "second," "third," etc. may be used herein to describe various components or signals, these components or signals should not be limited by these terms. These terms are used primarily to distinguish one element from another element or signal from another signal. In addition, the term "or" as used herein shall include any one or combination of more of the associated listed items as the case may be.

The image data can be transmitted to the video and audio device in a specific color coding mode, and a processing circuit in the video and audio device executes image processing according to the color range (color range) of the image data and then outputs the image processing to the display device in a specific transmission specification (such as HDMI), or outputs the image data after being mixed with a specific playing interface (blending). In general, images in video are transmitted and displayed in color coding modes such as YUV (Y represents brightness), U and V represent chromaticity (chrominance)) color coding, and the color range refers to the range of brightness and chromaticity in YUV color space, and can be divided into a limited color range (narrow range or LIMITED RANGE) and a full color range (full range). In 8-bit YUV color-coded video, the image pixel values for a limited color range fall within [16,235] and the image pixel values for a full color range fall within [0,255]. Generally, the color range of an image in a video is recorded in the header of each image sequence packet, and the processing circuit in the video and audio device converts the image into an image with a specific transmission specification according to the information of the color range, or mixes the image with a specific image in the same color range and outputs the mixed image.

However, the image packet may lack color range information, or even provide erroneous color range information, so that the subsequent processing circuit processes the image according to the erroneous color range, resulting in a problem of color distortion (color distortion) of the output video image. In order to overcome the defect, the present disclosure provides a correction method for converting an image color range and an image processing system, and under the inventive concept, the correction method for converting an image color range mainly uses image data in an actual statistical video to obtain a frame-by-frame image color range, and corrects a color range of a subsequent image if the image data is inconsistent with a color range recorded by an image packet. In another inventive concept, considering that the direct correction of an image performed at one time will form a significant change in brightness or color during video playing to affect the quality of a film watched by a user, it can be said that the correction method of image color range conversion proposed in the specification uses a multi-stage correction method to gradually perform color range conversion so as to correctly and smoothly perform conversion to execute a subsequent image processing procedure, wherein the multi-stage correction is used to replace the one-stage correction, thereby avoiding the influence on the picture quality due to overlarge brightness or color change generated during the correction process.

An application of the image processing system performing the correction method for image color range conversion may refer to an embodiment diagram of an application scenario architecture shown in fig. 1.

According to the example shown in fig. 1, an audio/video apparatus 100, which may be a built-in or external apparatus of a television system, includes an audio/video processing circuit cooperating with hardware in a software manner, and an image processing system 105 is mainly shown in addition to the audio processing circuit 101. The audio/video apparatus 100 receives audio/video data 10 from a data source through an input interface 103 (such as a wireless or wired network interface), wherein a sound portion is processed by an audio processing circuit 101, an image portion is processed by an image processing system 105, wherein a processing circuit 115 performs a correction method for converting an image color range for a case that color ranges are different, and a memory 117 is used for temporarily storing the image data before and after the processing. After the processing of the image and the audio is completed, the image is output through the output interface 107, wherein the image part is output to the display 110 for displaying.

The correction method for converting the color range of the image proposed in the present specification is implemented in the image processing system 105, wherein the main method steps are shown in the flowchart of the embodiment shown in fig. 2.

In general, the image processing system 105 of the video device 100 performs the subsequent processing according to the color range (limited color range or full color range) recorded in the image package, and if the color range is limited, the image is converted into the full color range and outputted in combination with the playing interface graphics provided by the video device. In the correction method for converting the color range of the image proposed in the specification, as shown in the flow chart of fig. 2, the image data is obtained from the buffer (step S201), the color range information recorded in the image packet is obtained (step S203), then the pixel values in the image are counted, and the actual color range of the image is determined according to the counted result of the pixel values of the image (step S205).

Next, judging whether the actual color range is consistent with the color range recorded in the image packet (step S207), if the actual color range is consistent with the color range recorded in the image packet, the image processing system 105 runs a general image processing program including necessary color range conversion and necessary processing before outputting the video (step S209); however, when the two are not coincident, it means that the image has been processed erroneously, a color range correction procedure is performed to enable correct processing of the image (step S211).

The above-mentioned method is implemented in an image processing system, one of the main purposes of the system is to convert an image with a limited color range into an image with a full color range, and the other purpose is to gradually convert the color range in a multi-stage correction manner so as to smoothly complete the video conversion, so that the problem of poor viewing experience caused by one-time correction can be avoided.

An exemplary embodiment of the circuit module for implementing the image processing system may refer to the circuit block diagram shown in fig. 3, in which several basic circuit components in the processing circuit for implementing the image processing system 30 are shown, including a control unit 300 and a plurality of circuit units electrically connected, where the control unit 300 controls the operations of the circuit units, such as setting a threshold, comparing, counting, determining a correction step, setting a comparison condition, and performing a progressive correction procedure according to the correction step.

Each circuit unit can be represented by a digital circuit formed by various logic gates and input/output terminals, and as shown in the figure, the control unit 300 sets various threshold values of the threshold unit 301 according to actual situations, determines the actual color range of the image according to the statistical results of the pixel statistical unit 302 and the frame image statistical unit 303, and when the color range needs to be corrected, determines the image to be subjected to color range conversion through the image screening unit 305, and then performs conversion through the linear mapping unit 307, and particularly performs progressive conversion.

When the image processing system 30 receives an input image 31, which may default to a video with a limited color range, pixel values of the input image 31 are obtained first, and pixel values of all pixels or sampled pixels in the input image 31 are compared by a threshold unit 301 in the image processing system 30 by using a threshold designed to distinguish between values of the limited color range and values of the full color range. Because the pixel values (brightness and chromaticity) of the limited color range and the full color range are mostly overlapped, the threshold is designed to be a value capable of distinguishing the two color ranges, the pixels screened by the threshold are judged to be the pixels belonging to the full color range, and the screened pixels are counted by the pixel statistics unit 302, so that the number of the screened pixels is used as the basis for judging the color range of each frame. The number of frames is then accumulated by the frame image statistics unit 303.

One embodiment of the threshold unit 301 may refer to an example diagram of a logic circuit for setting a threshold in the image processing system shown in fig. 4.

Fig. 4 shows that the input pixel value IN is compared with the lower threshold value l_thd and the upper threshold value u_thd by two comparators, respectively, and the pixel value may refer to each component value of the pixel IN a specific color space, such as a luminance value (Y component) and a chrominance value (U and V components) IN a YUV color space.

When the input pixel value IN is used for comparing the lower limit threshold value L_THD, the condition is set as IN < L_THD, and the comparison result meets the condition, namely, output 1, otherwise output 0; meanwhile, the input pixel value IN is used for comparing the upper limit threshold value U_THD, the condition is set to be IN > U_THD, and the comparison result meets the condition, namely, 1 is output, and otherwise 0 is output. Thus, the attribute of the color range of the input pixel value IN can be confirmed from the output value OUT being 0 or 1.

For example, the pixel value range of the full color range falls between 0 and 255 (2 ⁿ -1), the brightness pixel value of the limited color range falls between 16×2 ^n-8 and 235×2 ^n-8, the chrominance image pixel value falls between 16×2 ^n-8 and 240×2 ^n-8, and n is the number of image sequence bits, the minimum value is 8, and the pixels of the non-overlapping part of the two ranges can be used to determine the color range attribute of the image. In this way, the lower threshold l_thd may be set to 16×2 ^n-8, the upper threshold u_thd for brightness may be set to 235×2 ^n-8, and the upper threshold for chromaticity may be set to 240×2 ^n-8, where n is the number of image sequence bits, and the minimum value is 8, i.e., n+.8. Therefore, the output value OUT can determine whether the pixel falls in a range outside the limited color range [16×2 ^n-8,240×2^n-8 ], and the output value OUT is 1, which indicates that the pixel value falls in a range outside the limited color range, that is, the pixel can be determined to belong to the full color range; an output value OUT of 0 indicates that the pixel falls within a limited color range.

The designs of the pixel statistics unit 302 and the frame image statistics unit 303 can refer to the logic circuit embodiment diagram of the statistics operation in the image processing system shown in fig. 5.

The illustration uses an OR gate (OR gate) and an adder to realize a COUNTER, the output value of the threshold unit 301 becomes the input value of the pixel statistics unit 302 OR the frame image statistics unit 303, that is, the input value IN is 1 OR 0, and by the operation of the OR gate and the 1, only the output value IN is 1, the COUNTER is counted, the counter_out is output at the current time (t), and the counter_out at the previous time (t-1) is accumulated, and finally, the pixels belonging to the full color range can be counted for each frame, and the statistics can be performed for the frames belonging to the full color range IN the continuous frames IN the image.

As for the control unit 300, the color range attribute of each frame may be further determined, wherein when the number of pixels of which the pixel value of each frame is greater than the upper threshold or less than the lower threshold is calculated, the number of pixels may be compared with a number threshold to determine the actual color range of the image.

After the statistics of the pixel values of each frame are completed, the actual color range of the input image 31 is determined, for example, the actual color range is a limited color range, and the image processing system 30 will perform a correction procedure for converting the color range. The above-described operations of setting and counting the respective thresholds are performed by the control unit 300. When performing color range conversion, a linear mapping (LINEAR MAPPING) method is adopted to convert all pixels in the image into a full color range, the control unit 300 controls parameters thereof, and particularly in a progressive conversion method, the control unit 300 controls the step size of each adjustment in the linear mapping method to gradually perform color range conversion in a multi-stage correction mode.

In this example, the control unit 300 screens out the image in which the color range conversion is required to be performed in the input image 31 by the image screening unit 305, determines the gradation for each adjustment in the multi-stage correction, inputs the result to the linear mapping unit 307, and sequentially outputs the output image 32 obtained by the multi-stage correction color range conversion by the linear mapping unit 307.

Implementation of the image filtering unit 305 as shown IN the logic embodiment of fig. 6, the control unit 300 determines the input value IN of the image filtering unit 305 according to the above statistical method, obtains the values of the input image filtering unit 305, such as YUV component values (brightness and chromaticity), from the input image 31, and can filter the image output value OUT to be subjected to color range conversion.

The color range correction program adopts a linear mapping method, the implementation of the linear mapping unit 307 is shown IN the logic circuit embodiment of the linear mapping method IN the image processing system shown IN fig. 7, the input pixel value IN which is originally IN a limited color range is represented by Y, the luminance is represented by Y, the chrominance is represented by V, the adjustment multiplier Yscale, cscale such as the slope of a linear equation and the compensation value Yoffset, coffset are provided by the control unit 300, the progressive conversion method is implemented by the linear mapping unit 307, that is, the image processing system 30 is set to gradually perform the color range conversion IN a multi-stage correction manner by the operation of the processing circuit, such as the control unit of fig. 3, and each time the step is adjusted to perform the color range conversion to generate the output pixel value OUT, and the image after the color range conversion is sequentially output until the image is completely converted into the full color range.

With continued reference to fig. 8, a flowchart of an embodiment of a method for correcting image color range conversion is shown, wherein a detailed flow of an embodiment of a method for correcting image color range conversion based on the above-described image processing system and the various functional modules therein is shown, and the embodiment shown in fig. 1 is generally described as applied.

When the image processing system 105 receives an image (step S801), the processing circuit 115 stores the image data temporarily in the memory 117 (a register) for temporarily storing image information (step S803), obtains the color range information recorded in the image packet (step S805), executes the corresponding image processing procedure according to the color range recorded in the image packet (step S807), and under specific circumstances, after converting the image with the limited color range into the image with the full color range, combines the image with the image of the playing interface and outputs the combined image to the display 110 (step S809).

On the other hand, for the following image, the processing circuit 115 in the image processing system 105 executes the correction method of image color range conversion, and the main method process firstly obtains the image data from the memory 117, wherein the image may be each frame of image in the streaming video received frame by frame (step S811), the header (header) of the image packet records that the image sequence is an image with a limited color range, or the header lacks information of the color range. In this process, the processing circuit 115 performs statistics on the pixel values in the image (step S813) in order to obtain the actual color range of the image.

For example, the values mentioned in the examples are not intended to limit the applicable range of the method, and the color range generally defined can be divided into a limited color range with pixel values ranging from 16 to 235 and a full color range ranging from 0 to 255, so that it can be determined whether the image belongs to a limited color range image or a full color range image by setting the upper and lower thresholds of the pixel values, for example, setting the lower threshold to 16 and the upper threshold to 235.

The pixel value of each pixel (or may be sampled) in the resulting image is compared to the upper and lower thresholds and counted. The actual color range of the image can be determined according to the distribution of the pixel values (step S815), by calculating the number of pixels whose pixel values exceed the threshold (e.g., are greater than the upper threshold or are less than the lower threshold), and calculating the number of pixels whose pixel values are within the threshold range (e.g., between the upper threshold and the lower threshold). The detail is that a quantity threshold is set for avoiding noise from influencing judgment, if the quantity of pixels of which the pixel value exceeds the upper and lower limit thresholds in the image of the color range to be judged exceeds the quantity threshold, the image of the full color range is judged; otherwise, if the number of pixels in the image, of which the pixel values exceed the upper and lower limit thresholds, does not exceed the number threshold, the image is judged to belong to the image with the limited color range.

Next, judging whether the actual color range of the image obtained by statistics is consistent with the color range information obtained in the image packet (step S817), if so, executing the corresponding image processing procedure as described in step S807 for the subsequent image; on the contrary, if the two images do not match, a color range correction procedure is performed particularly for an image whose actual color range is a limited color range, so that the image can be processed correctly, and a progressive conversion method (step S819) is particularly adopted, and the progressive conversion method sets a step-by-step color range conversion in a multi-stage correction manner in the linear mapping method described in the above embodiment, and after the image is completely converted to an image with full color range, step S807 is performed, and the image of the playing interface are combined and output to the device for playing video.

By means of the asymptotic conversion method, smoother video can be output, and reference can be made to the embodiment of the method for performing the asymptotic color range correction shown in fig. 9A to 9C.

According to the example shown in fig. 9A, the color range obtained from the image packet is consistent with the actual color range, and if both are full color ranges, the processing circuit in the image processing system can directly execute the necessary image processing flow of the output video, including the output video after being combined with the specific playing interface graphics; if the color range obtained from the image packet and the actual color range are both limited color ranges, the processing circuit in the image processing system performs conventional color range conversion, so as to output video after being combined with the specific playing interface graphics. This is shown, for example, in fig. 9A, which shows a video with consecutive images that is output consistently without correction.

According to the example shown in fig. 9B, when the color range obtained from the image packet is inconsistent with the actual color range, the processing circuit of the image processing system will generate distortion problem when the processing circuit processes the image according to the color range obtained from the image packet, so the processing circuit will perform necessary correction on the subsequent image, i.e. correct the image output by the image processing system according to the actual color range obtained from the temporarily stored image. While fig. 9B shows that the immediate correction results can yield a correct output video, the process of correcting the change may occur with a sharp change in brightness or color.

According to the example shown in fig. 9C, a method of using progressive conversion in the correction process is shown, and according to the number of corrections (N > 2) set by the system, starting from the first frame 901 (frame n=0) of the start image, N progressive corrections are performed, this example shows the number of corrections being 3, obtaining the next frame image, i.e. the second frame 902 (frame n=1), after the first phase correction, obtaining the third frame 903 (frame n=2) from the second phase correction, and performing the third phase correction, and obtaining the fourth frame 904 (frame n=3) of the last completed correction. Therefore, the video with smoother brightness or color change can be corrected gradually by the progressive conversion method. It should be noted that, the example shown in fig. 9C is not intended to limit the implementation of the present invention, and the larger the number of correction times, i.e., the larger N, the smaller the corrected feeling of the display screen.

In summary, according to the embodiments of the correction method and the image processing system for converting the color range of the image described above, the main inventive concept is that the actual color space of the image can be detected by hardware, and then the adjustment number is gradually filled in by software to realize the asymptotic conversion. One of the main problems to be solved by the method is that when the color range is not consistent with the actual color range, it is obvious that the image processing system performs the wrong image processing, so that on the subsequent image, the image processing system will perform the color range correction procedure, and the color range conversion is gradually performed in a multi-stage correction manner during the correction procedure, so that the conversion can be completed correctly and smoothly and the subsequent image processing procedure can be performed.

The foregoing disclosure is only illustrative of the preferred embodiments of the present invention and is not to be construed as limiting the scope of the invention, which is defined by the appended claims and their equivalents.

[ Symbolic description ]

100 Video and audio device

101 Audio processing circuit

105 Image processing system

103 Input interface

10 Video and audio data

115 Processing circuit

117 Memory

107 Output interface

110 Display

30 Image processing System

300 Control unit

301 Threshold unit

302 Pixel statistics unit

303 Frame image statistics unit

305 Image screening unit

307 Linear mapping Unit

31 Input image

32 Output image

IN: input pixel value

L_THD lower threshold

U_THD upper threshold

OUT: output value

IN: input value

Counter_OUT count value

OUT: image output value

901 First frame

902 Second frame

903 Third frame

904 Fourth frame

Correction embodiment flow of step S201-S211 image color Range conversion

Steps S801 to S819 are flow of correction embodiments of image color range conversion.

Claims (10)

1. A correction method for image color range conversion, comprising:

acquiring an image, wherein data of the image is stored in a register;

And counting the pixel values of the image in the register to obtain an actual color range of the image, and executing a color range correction program to correctly process the image if the actual color range is not consistent with the color range recorded in the image packet.

2. The method of claim 1, wherein the actual color range of the image is determined according to statistics of pixel color ranges of the image, and the actual color range is a full color range or a limited color range.

3. The correction method for image color range conversion according to claim 2, wherein the method of counting pixel values of the image comprises:

Comparing an upper threshold with a lower threshold by pixel values of all or sampled pixels in the image;

calculating a number of pixels having a pixel value greater than the upper threshold or less than the lower threshold; and

The number of pixels is compared with a number threshold to determine the actual color range of the image.

4. A method of correcting an image color range conversion as claimed in claim 3, wherein the limited color range of bright image pixel values falls between 16 x2 ^n-8 and 235 x2 ^n-8 and the chroma image pixel values fall between 16 x2 ^n-8 and 240 x2 ^n-8; the image pixel value of the full color range falls between 0 and 2 ⁿ -1, i.e., the lower threshold is set to be 16×2 ^n-8, the upper threshold for brightness is 235×2 ^n-8, and the upper threshold for chromaticity is 240×2 ^n-8, where n is the number of image sequence bits, and the minimum value is 8, i.e., n+.8.

5. The method of claim 2, wherein the color range recorded by a header of the image packet is the full color range or the limited color range, wherein when the color range recorded by the image packet is the limited color range and the actual color range is the full color range, the color range correction procedure is performed on the image inputted after the image.

6. The method of claim 5, wherein the color range correction procedure uses a linear mapping method to convert all pixels in the image from the limited color range to the full color range.

7. The method of claim 6, wherein the color range correction procedure employs a progressive conversion method in which the linear mapping method is configured to gradually perform color range conversion in a multi-stage correction manner, wherein each time the step is adjusted by one step to perform color range conversion to output an image until the full color range is completely converted.

8. An image processing system that performs image color range conversion, comprising:

a processing circuit and a buffer, wherein the processing circuit performs a correction method for converting an image color range, comprising:

acquiring an image, wherein data of the image is stored in the register;

And counting the data of the image in the register to obtain an actual color range of the image, and executing a color range correction program to correctly process the image if the actual color range is not consistent with the color range recorded in the image packet.

9. The image processing system of claim 8, wherein the image processing system is applied to an audio-visual device, and the image is output after the image is combined with a graphic of a playing interface after the image processing system performs color range conversion.

10. The image processing system of claim 8, wherein the actual color range of the image is determined based on statistics of pixel color ranges of the image, the actual color range being a full color range or a limited color range, wherein the method of counting pixel values of the image comprises:

Comparing an upper threshold with a lower threshold by pixel values of all or sampled pixels in the image;

calculating a number of pixels having a pixel value greater than the upper threshold or less than the lower threshold; and

The number of pixels is compared with a number threshold to determine the actual color range of the image.