Patent Application
20220351701
The present disclosure relates to the field of image communication, and more particularly, to a method, device, and readable storage medium for adjusting image quality.
High dynamic range (HDR) imaging is an image displaying technology emerging in recent years. Compared with normal imaging technologies, the HDR imaging technology can provide more dynamic ranges and image details and better reflect visual effects of real environments. The HDR imaging technology may meet users' demands for viewing experience of watching high-quality movies and TVs. However, the existing HDR technologies cannot fully satisfy users' demands for colorful and exquisite images with great sense of level, because different users have different concerns or areas of interest for different scenes or images. Unfortunately, the existing HDR technologies adjust images mechanically, not intelligently according to different environments or scenes.
Therefore, current technologies still require improving and developing.
In view of the above-mentioned shortcomings of the prior arts, the present disclosure provides a method for intelligently adjusting image quality that can adjust a current display image to partially enhance image quality according to personal interests.
The technical solution used in the present application to solve the above technical problems is detailed in the following:
As a further improved technical solution, the focus area of the target image selected by the display device system is a central area in an image display of the target image.
As a further improved technical solution, when resolution of a current image is 3840×2160, the pixels in the corner areas of the selected central area are generally located in pixel ranges 1600<x<2260 and 900<y<1260, where x is the coordinate in the length direction of the image, y is the coordinate in the width direction of the image, and the origin of coordinates of the image display is at the upper left corner.
As a further improved technical solution, the focus area of the target image selected by the display device system is an image of interest.
As a further improved technical solution, the image area of interest is obtained through artificial intelligence training.
As another further improved technical solution, the image area of interest is an image area that is recognized through big data collection.
As another further improved technical solution, the outputting the mapping relationship curve of the input luminance value and the output luminance value of the target image comprises calculating using a linear Bezier curve to initially describe the mapping relationship curve of the input luminance value and the output luminance value.
As another further improved technical solution, the calculating the input luminance value range of the pixels in the focus area of the target image and proportionally adjusting the upper and lower edge coefficients of the input luminance value range to obtain the new upper and lower edge coefficients comprises:
As another further improved technical solution, the linear Bézier curve is preferably implemented by a fifth-order linear Bézier curve.
As another further improved technical solution, the 5th-order linear Bézier curve is described as Lt=Ltmax×ΣC(N,k)×L0{circumflex over ( )}k×(1−L0){circumflex over ( )}(N−k)×pk, wherein C(N,k) indicates permutations of k elements among N (N=5) elements, Ltmax represents a maximum output luminance supported by a current display device, L0 represents a ratio of the input luminance value to the maximum input luminance value, k iterates from 1 to 5, and pk represents a sequence of coefficients, initially, pk=k/5, the input is correlated with the output in a linear relationship, and the luminance ratio L0 is divided into five equal intervals.
As another further improved technical solution, the interval upper and lower edge coefficients are scaled up or down proportionally based on an adjustment factor which is selected based on conditions of images.
As another further improved technical solution, by applying the above adjusted upper and lower edge coefficients to the mapping relationship between the input luminance and the output luminance of the focus area of the image expressed by the fifth-order linear Bézier curve, luminance and contrast of the image can be adjusted in real-time.
As another further improved technical solution, the image area that is recognized through the artificial intelligence training or the big data collection is customized by a user according to specific requirements for image recognition.
The present application also provides a device for adjusting image quality that comprises a processor and a memory communicatively connected to the processor. The memory stores a computer program, and the computer program is executed by the processor to perform the steps of the above method.
The processor is configured to call the computer program in the memory to perform the steps of the method for adjusting image quality.
The present application further provides a readable storage medium. The readable storage medium stores a program for adjusting image quality. The program for adjusting image quality, when being executed by a processor, realizes the steps of the method for adjusting image quality.
Compared with current HDR image quality adjustment technologies, the present disclosure uses artificial intelligence technologies to set an image area that a user is interested in as the focus area of the target image, to represent the input luminance value range of the pixels in the focus area of the target image using a linear Bézier curve, and to scale up or down the contrast level of the image through adjusting the order of the linear Bézier curve or changing the adjustment factor for the upper and lower edge coefficients of the input luminance value range. The linear Bécurve is utilized to ensure smoothness and continuity in luminance adjustment and also ensure coherence of image display after the image adjustment. By improving the contrast level of the specific area of the image, the image quality becomes brighter and more colorful. The method is simple while can meet the requirements of real-time adjustment, and does not require additional hardware, reducing the cost.
Detailed description of the embodiments of the present application is given in the following with accompanying drawings.
To make the object, technical solution, and advantages of the present disclosure clearer and more comprehensible, the present disclosure is described in further detail hereinafter with reference to the accompanying drawings and by way of example. It should be understood that the specific embodiments described herein are intended only to explain not limit the present disclosure.
The present disclosure provides an image recognition technology using artificial intelligence to adjust a specific area of an image according to the features recognized by the artificial intelligence, thereby enhancing exquisiteness and sense of organization of a part of the image and improving the image quality. For example, in an application where image quality of movies or TV is enhanced, a certain portion of an image needs to be emphasized, or in the criminal investigation of the police, a portrait of one person needs to be partially highlighted. For these specific application environments requiring image quality enhancement, users may perform a setting operation on some specific images of interest in advance. The setting operation can be achieved through big data or artificial intelligence in the related art. Then, image display of the specific images can be enhanced through the method of the present disclosure during video playback or image browsing, so as to achieve the purpose of intelligently enhancing image quality of a specific area.
Step 100: selecting by a display device system a focus area of an image.
Specifically, the focus area of the image selected by the display device system is the central area or an image area of interest in image display of the image. The adjustment of the image quality mainly comprises adjusting the luminance and the contrast level. The image area of interest in the technical solution of the present disclosure is an image area that is recognized through artificial intelligence training or big data collection. For example, in a certain image, the central area of the image display generally serves as the focus area of the image. When the resolution of the current image is 3840×2160, the pixels in the corner areas of the selected central area are generally located in pixel ranges 1600<x<2260 and 900<y<1260, where x is the coordinate in the length direction of the image, y is the coordinate in the width direction of the image, and the origin of coordinates of the image display is at the upper left corner. Of course, the central area of the image display here is only an example. Any modification can be appropriately performed in the range according to the resolution of the image display, so that the central area is suitable for images with different resolutions. In addition, the image area of interest can be set by means of artificial intelligence training or collected big data. For example, in the field of the criminal investigation of the police, facial features can be extracted and learned as the basis for subsequent image adjustments. The schemes and methods of artificial intelligence learning belong to state of the art technology, and are not repeated here.
Step S200: outputting a mapping relationship curve of an input luminance value and an output luminance value of the target image.
Specifically, the step of outputting the mapping relationship curve of the input luminance value and the output luminance value of the target image is performed by calculating using a linear Bécurve to initially describe the mapping relationship curve of the input luminance value and the output luminance value.
In the technical solution of the present disclosure, the linear Bézier curve is preferably implemented by a fifth-order linear Bézier curve. Of course, the order of the linear Bézier curve can be selected based on requirements, which can be determined according to different conditions of the image. For example, the 5th-order linear Bézier curve is described as Lt=Ltmax×ΣC(N,k)×L0{circumflex over ( )}k×(1−L0){circumflex over ( )}(N−k)×pk, wherein C(N,k) indicates permutations of k elements among N (N=5) elements, Ltmax represents the maximum output luminance supported by a current display device, L0 represents a ratio of the input luminance value to the maximum input luminance value, k iterates from 1 to 5, and pk represents the sequence of coefficients, initially, pk=k/5. There is a linear relationship between the input and the output, and the luminance ratio L0 is divided into five equal intervals.
Step S300: calculating the input luminance value range of the pixels in the focus area of the target image and proportionally adjusting the upper and lower edge coefficients of the input luminance value range to obtain new upper and lower edge coefficients.
Specifically, the step of calculating the input luminance value range of the pixels in the focus area of the target image and proportionally scaling up or down the upper and lower edge coefficients of the input luminance value range to obtain new upper and lower edge coefficients comprise the following steps:
For example, an input luminance range L1<=L0<=L2 of the focus area of the image is obtained statistically, and then it is determined that L1 and L2 are in the five equal intervals determined in Step S200, wherein it is assumed that L1 and L2 are set as: 1/5<=L1<=2/5, 2/5<=L2<=3/5. The interval upper and lower edge coefficients determined by L1 and L2 are scaled up or down proportionally according to a scale (referred to as adjustment factor) to obtain new upper and lower edge coefficients. In this example, L1 corresponds to a lower limit coefficient P1, and L2 corresponds to an upper limit coefficient P3. It is assumed that the adjustment factor is set as 1.2. Scaling down P1 by the factor of 1.2 yields P1=0.16, and scaling up P3 by the factor of 1.2 yields P3=0.72. Accordingly, the corresponding new upper and lower edge coefficients can be obtained according to the adjustment factor.
The interval upper and lower edge coefficients are scaled up or down based on the coefficient adjustment factor that may be selected based on conditions of the image.
Step S400: adjusting the mapping relationship curve according to the new upper and lower edge coefficients to adjust the focus area of the target image.
Specifically, the luminance and contrast of the image can be adjusted in real-time by applying the above adjusted upper and lower edge coefficients to the mapping relationship between the input luminance and the output luminance of the focus area of the image expressed by the fifth-order linear Bézier curve. Smooth and continuous luminance changes in the adjusted image and coherence of image display before and after the image adjustment can be obtained without inconsistent appearance, and the image quality becomes brighter and more colorful, which greatly improves the contrast level of the specific areas of the image.
The present disclosure also provides a device for adjusting image quality.
Other specific steps performed by the program 22 for adjusting image quality are the same as the steps in the preferred embodiment of the foregoing method and can be referred t to the steps in the preferred embodiment of the foregoing method, which is not repeated here. The effect achieved by adjusting image quality through the device can be referred to the preferred embodiment of the above method, and is not repeated here.
The processor 20 may be implemented by a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or another programmable logic device, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or the processor may be any conventional processor or the like.
The memory 21 may be an internal storage unit of the device for computing a video code rate, for example, a hard disk or a memory of the device for computing a video code rate. The memory 21 may also be an external storage device of the device for computing a video code rate, for example, a plug-in hard disk equipped on the device for computing a video code rate, a smart memory card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory 21 may also include both an internal storage unit and an external storage device of the device for computing a video code rate. The memory 21 is used to store the computer program and other programs and data required by the device for computing a video code rate. The memory 21 may be used further to temporarily store data that has been output or will be output.
The present disclosure also provides a readable storage medium that stores a program for adjusting image quality. When the program for adjusting image quality is executed by a processor, the program for adjusting image quality realizes the steps of the aforementioned method for adjusting image quality. The specific steps are the same as the steps of the aforementioned method for adjusting image quality, and the resulted program for adjusting image quality is also the same as that described in the preferred embodiment of the aforementioned method. The related description will not be repeated here.
It should be understood that the above descriptions are related to only preferred embodiments of the present disclosure and not intended to limit the technical solutions of the present disclosure. For those of ordinary skill in the art, addition/reduction, replacement, modification, or improvement is allowed for the technical solutions within the spirit and principle of the present disclosure. The technical solutions after the addition/reduction, replacement, modification, or improvement should be covered by the protected scope of the appended claims of the present disclosure.
| Number | Date | Country | Kind |
|---|---|---|---|
| 20191056986.X | Jun 2019 | CN | national |
The PCT application claims priority to Chinese patent application No. 201910569860.X, filed on Jun. 27, 2019, of which technical solutions are incorporated by reference in the present application in its entirety.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/CN2020/090597 | 5/15/2020 | WO |
