Image quality adjusting device and image quality adjusting method
By identifying the target object and its area in the image and determining different picture quality parameters based on the area area ratio, the problem that unified adjustment in the prior art cannot adapt to multi-object images is solved, and a better image display effect is achieved.
Patent Information
- Application Number
- CN202411595005.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-08
- Publication Date
- 2025-02-25
AI Technical Summary
When the existing image quality adjustment method processes images containing multiple different objects, unified adjustment cannot adapt to the needs of all objects, resulting in color distortion of some objects.
By identifying the target object in the current frame image and its area in the image, if the target area area ratio exceeds the threshold, different picture quality parameters of the target area and the non-target area are determined, and personalized picture quality adjustment is performed.
It realizes careful adjustment of different areas in the image, avoids color distortion caused by unified adjustment, and improves the display effect of the image.
Smart Images
Figure CN119520882A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of image processing, and in particular to an image quality adjustment device and an image quality adjustment method. Background Art
[0002] The existing image quality adjustment method usually performs a unified adjustment on the entire image when adjusting the image quality. Although the unified adjustment method has a fast processing speed, when the current image includes multiple different objects, the unified adjustment may not be able to meet the needs of all objects, resulting in color distortion of some objects. Summary of the invention
[0003] The present application provides an image quality adjustment device and an image quality adjustment method to solve the above problems.
[0004] In a first aspect, a method for adjusting image quality is provided, comprising: acquiring a current frame image; identifying a target object in the current frame image and confirming a target area occupied by the target object in the current frame image; when an area ratio of the target area in the current frame image exceeds a threshold, determining a target area quality parameter of the target area and a non-target area quality parameter of a non-target area in the current frame image, wherein the target area quality parameter and the non-target area quality parameter are different, and the non-target area is at least a partial area outside the target area in the current frame image.
[0005] In a second aspect, an image quality adjustment device is also provided, comprising a DMA circuit, a neural network processor, an analysis circuit and a control circuit. The DMA circuit is used to obtain a current frame image. The neural network processor is used to identify a target object in the current frame image. The analysis circuit is used to determine a target area occupied by the target object in the current frame image. The analysis circuit is used to identify a target object in the current frame image and a target area occupied by the target object in the current frame image. When the area ratio of the target area in the current frame image exceeds a threshold, the control circuit determines a target area quality parameter of the target area and a non-target area quality parameter of a non-target area in the current frame image, wherein the non-target area quality parameter is different from the target area quality parameter, and the non-target area is at least a partial area outside the target area in the current frame image.
[0006] The image quality adjustment method and the image quality adjustment device in the present application identify the target object in the image and, under specific conditions, apply different image quality parameters to the target area and non-target area where the target object is located to adjust the image so as to more effectively improve the image display effect. BRIEF DESCRIPTION OF THE DRAWINGS
[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the background technology, the drawings required for use in the embodiments of the present application or the background technology will be described below.
[0008] Figure 1 This is a structural block diagram of an image quality adjustment device in some embodiments of the present application.
[0009] Figure 2 This is a flowchart of an image quality adjustment method in some embodiments of the present application.
[0010] Figure 3 In some embodiments of this application Figure 2 Sub-flow chart of step S25 in . DETAILED DESCRIPTION
[0011] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0012] In the description of this application, unless otherwise clearly specified and limited, the term "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection, an indirect connection through an intermediate medium, or the internal connection of two components; it can be a communication connection; it can be an electrical connection. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to the specific circumstances. Among them, in this application, the term "A and / or B" includes: any one of "A", "B", "A and B". Among them, the term "A / B" generally refers to "A" or "B". In the description of the embodiments of the present invention, the terms "first", "second", etc. are not specific, but are used to distinguish objects with the same name. If there is a description in the specification, the objects with the same name referred to by the terms "first", "second", etc. may be the same object.
[0013] See also Figure 1, is a structural block diagram of an image quality adjustment device in some embodiments of the present application. The image quality adjustment device 100 includes a DMA (direct memory access) circuit 1, an analysis circuit 2, a quality acquisition circuit 3, a neural network processor 4, a quality adjustment circuit 5, an internal memory 6 and a control circuit 7. The image quality adjustment device 100 adjusts the image quality stored in the external memory 200 in response to the received image display request Req. The external memory 200 stores video images or streaming media images to be displayed. In one embodiment, the external memory 200 may be a dynamic random access memory.
[0014] The following will be combined Figure 2 The above-mentioned circuits of the image quality adjustment device 100 are described in detail.
[0015] See also Figure 2 , is a flow chart of an image quality adjustment method in some embodiments of the present application. The image quality adjustment method can be run on the aforementioned image quality adjustment device 100, and includes the following steps.
[0016] Step S21: DMA circuit 1 obtains the current frame image.
[0017] The DMA circuit 1 obtains the current frame image from the external memory 200 in response to the received read command. Specifically, the control circuit 7 generates a read command upon receiving the image display request Req and sends it to the DMA circuit 1. In one embodiment, the image display request Req is generated in response to a user operation. The DMA circuit 1 obtains the current frame image from the video stream stored in the external memory 200 in response to the read command issued by the control circuit 7, and stores the obtained current frame image in the internal memory 6. The internal memory 6 may be a static random access memory.
[0018] Step S22: The neural network processor 4 identifies the target object in the current frame image.
[0019] Specifically, the neural network processor 4 can perform object recognition on the current frame image based on the pre-trained object recognition model to generate an image object distribution map of the current frame image. The image object distribution map includes object labels corresponding to each object in the current frame image, and the positions of each object are marked with their corresponding object labels. For example, when two objects, a face and a sky, are recognized in the image, the first object label such as M1 is added to the position of the face in the image, and the second object label such as M2 is added to the position of the sky in the image, thereby generating an image object distribution map.
[0020] In one embodiment, the neural network processor 4 can identify the target object in the current frame image and mark the position of the target object in the current frame image to obtain a target object distribution map.
[0021] The neural network processor 4 can also identify the scene of the current frame image based on the trained scene model to obtain the image scene. The neural network processor 4 can store the acquired image object distribution map (or target object distribution map) and scene information into the internal memory 6.
[0022] It is understandable that the neural network processor 4 of the present case may include an object recognition engine and a scene recognition engine when implemented.
[0023] Step S23: the image quality acquisition circuit 3 obtains the image quality information of the current frame image.
[0024] In one embodiment, the image quality information includes at least one of image brightness, image chromaticity, color saturation and exposure. Specifically, the image quality acquisition circuit 3 can perform statistics on the pixel information of the current frame image to obtain the image quality information of the current frame image, such as statistically analyzing the pixel gray value, the distribution range and proportion of the pixel hue (such as blue, green), the distribution of the pixel saturation value and the brightness histogram to obtain the image brightness, image chromaticity, color saturation and exposure.
[0025] It can be understood that in order to speed up the entire execution speed of the image quality adjustment device 100, steps S22 and S23 can be executed synchronously.
[0026] Step S24: The analysis circuit 2 determines the target area occupied by the target object in the current frame image, and determines whether the area ratio of the target area in the current frame image exceeds a threshold. If yes, proceed to step S25; if no, proceed to step S26.
[0027] Specifically, the analysis circuit 2 can determine the target area occupied by the target object in the current frame image according to the target object distribution map or the image object distribution map, determine the area ratio of the target area in the current frame image, and compare the area ratio with the threshold. If the area ratio of the target area in the current frame image exceeds the threshold, the analysis circuit 2 generates a first determination result; if the area ratio of the target area in the current frame image does not exceed the threshold, the analysis circuit 2 generates a second determination result.
[0028] The analysis circuit 2 further determines a non-target area in the current frame image according to the target area, and the non-target area is at least a portion of the area outside the target area in the current frame image. In one embodiment, all areas outside the target area in the current frame image are non-target areas. In another embodiment, the non-target area is a portion of the area outside the target area in the current frame image rather than all areas outside the target area. The details of the non-target area will be described in conjunction with the following Figure 3 Provide explanation.
[0029] Step S25: the control circuit 7 determines the target area image quality parameters of the target area and the non-target area image quality parameters of the non-target area in the current frame image, and the target area image quality parameters and the non-target area image quality parameters are different.
[0030] In one embodiment, the image quality acquisition circuit 3 may acquire the image quality information of the target area based on the first determination result sent by the analysis circuit 2. Specifically, the image quality acquisition circuit 3 further acquires the target object distribution map from the internal memory 6 and acquires the image quality information of the target area accordingly, for example, by performing statistical analysis on all pixel values of the target area to acquire the image quality information of the target area. The control circuit 7 acquires the image quality information of the target area from the image quality acquisition circuit 3, and determines the image quality parameters of the target area according to the image quality information of the target area.
[0031] In an embodiment of the present case, the image quality parameters include at least one of brightness, sharpness, color saturation, and hue. The target area image quality parameters and the non-target area image quality parameters are different, and specifically, the parameter types and / or parameter values of the target area image quality parameters and the non-target area image quality parameters are different. For example, the target area image quality parameter may be sharpness, and the non-target area image quality parameter may be brightness. For another example, the target area image quality parameter and the non-target area image quality parameter both include sharpness, and the values of sharpness of the target area image quality parameter and the non-target area image quality parameter are different.
[0032] For example, taking the target object as a human face, when the area ratio of the human face in the current frame image exceeds the corresponding threshold, such as 10%, it can be determined that a separate image quality adjustment is required for the human face, for example, enhancing the sharpness of the face, that is, the image quality parameters of the target area include sharpness, and the sharpness value is relatively high; and in order to highlight the main effect of the human face, it is necessary to weaken other parts of the image, and the sharpness of the non-target area can be set to a lower level, for example, the sharpness can even be 0, which can create a depth of field effect and highlight the visual effect of the human face as the main body.
[0033] The control circuit 7 may determine the image quality parameters of the non-target area based on the image quality parameters of the target area to further improve the display effect of the image.
[0034] In some cases, the non-target area occupies a larger area in the current frame image than the target object, that is, the non-target area is equivalent to the background area of the current frame image, and therefore has a greater correlation with the global image quality information and the scene information. At this time, the control circuit 7 can determine the non-target area image quality parameters according to the target area image quality parameters, the global image quality information and the scene information to obtain more suitable image quality adjustment parameters for the non-target area. In this way, the target area and the non-target area will each use different parameters to adjust the image quality, avoiding the situation of image quality distortion or poor effect caused by using unified parameters.
[0035] Step S26: the control circuit 7 determines the global image quality parameters of the current frame image.
[0036] Among them, the control circuit 7 can determine the global image quality parameters of the current frame image in response to the second determination result. Specifically, the area ratio of the target area of the target object in the current frame image does not exceed the threshold, including the case where the target object is included in the current frame image, but the area ratio does not exceed the corresponding threshold, and also includes the case where the target object is not included in the current frame image (that is, its corresponding area ratio is 0). When the area ratio of the target area of the target object in the current frame image does not exceed the threshold, there is no object in the current frame image that needs to be adjusted separately, and the global image quality parameters applicable to the entire current frame image can be determined to perform global image quality adjustment on the current frame image.
[0037] In some embodiments, the global image quality parameters are the same as the non-target area image quality parameters. In other embodiments, the global image quality parameters are different from the non-target area image quality parameters. Specifically, the global image quality parameters can be determined to be the same or different from the non-target area image quality parameters according to the scene or user needs.
[0038] In summary, the image quality adjustment device 100 in the present application can determine the target area quality parameters of the target area and determine the non-target area quality parameters of the non-target area in the current frame image when it is identified that the area ratio of the target area occupied by the target object in the current frame image exceeds the threshold, and the non-target area quality parameters are different from the target area quality parameters. Therefore, different image quality parameters can be applied to different areas in the image to adjust the image more finely, so as to obtain an image with better display effect.
[0039] See also Figure 3 , for some embodiments of this application Figure 2 The sub-flow chart of step S25 in FIG. Step S25 includes the following steps:
[0040] Step S251: Determine whether the current frame image includes an associated object of the target object. If not, execute step S252; if yes, execute step S253.
[0041] Specifically, the neural network processor 4 can identify multiple objects in the current frame image (including but not limited to the target object and its associated objects), and mark their positions in the current frame image to obtain an image object distribution map, that is, the image distribution map includes all objects identified by the neural network processor 4 and the position marks of each of these objects in the current frame image.
[0042] The analysis circuit 2 can determine whether the current frame image also includes the associated object of the target object according to the image object distribution map. If the current frame image includes the associated object, the analysis circuit 2 determines the associated area of the associated object according to the image object distribution map, and regards the area outside the target area and the associated area in the current frame image as the non-target area.
[0043] In one embodiment, the target object and the associated object are both pre-set, for example, pre-set according to user needs. In a specific example, the target object is a blue sky and the associated object is a blue dress. It can be understood that the target object and the associated object have a certain degree of correlation in image quality effect. When the analysis circuit 2 determines that the current frame image includes an associated object of the target object, the control circuit 7 also determines an associated area adjustment parameter, which is different from the target area image quality parameter.
[0044] In some embodiments, the analysis circuit 2 may also determine whether there is an associated object of the target object in the current frame image according to the degree of association between the target object and other objects in the current frame image. The associated object of the target object may be an object that has a certain degree of association with the target object. For example, if the image quality similarity between an object and the target object is greater than a preset similarity, the analysis circuit 2 determines that the object is an associated object of the target object. For another example, if the image quality similarity between an object and the target object is greater than a preset similarity and the area ratio of the region where the object is located in the current frame image reaches a preset ratio (not the aforementioned), the analysis circuit 2 determines that the object is an associated object of the target object. In this scenario, in addition to calculating the image quality similarity between the object and the target object, the analysis circuit 2 also needs to calculate the area ratio of the region where the object is located in the current frame image. If both conditions are met at the same time (i.e., the image quality similarity is greater than the preset similarity value and the area ratio of the region where the object is located reaches a preset ratio), the object is determined to be an associated object of the target object.
[0045] Step S252: Determine that all areas outside the target area in the current frame image are non-target areas, and jump to step S255.
[0046] Step S253: Determine the associated area occupied by the associated object in the current frame image.
[0047] The specific content of determining the associated area by the analysis circuit 2 can refer to the detailed content of determining the target area by the analysis circuit 2, so it is not repeated here.
[0048] Step S254: Determine, according to the target area and the associated area, that an area outside the target area and the associated area in the current frame image is a non-target area.
[0049] Step S255 , determining a target area image quality parameter for adjusting the image quality of the target area and a non-target area image quality parameter for adjusting the image quality of the non-target area.
[0050] In some embodiments, step S254 further includes: determining an associated region image quality parameter of the associated region, wherein the associated region image quality parameter is different from the target region image quality parameter. Specifically, based on the fact that the associated object and the target object have a certain degree of correlation in image quality, such as image quality similarity, the associated region image quality parameter can be determined specifically according to the target region image quality parameter. In this way, the image quality correlation between the associated object and the target object can be increased or decreased according to the situation or actual needs.
[0051] To illustrate with a specific example, if the target object in the current frame image is a blue sky, and the associated object is preset as blue clothing, since human vision often prefers a blue sky with a higher saturation, when the proportion of the target area (i.e., the sky area) exceeds the threshold, the sky area is enhanced with a stronger adjustment intensity, such as setting the sky enhancement coefficient Sat to 2.0. Specifically, the enhanced sky chromaticity component Cb of the sky can be obtained based on the chromaticity components Cb and Cr of the sky area in the current frame image and the following formula: sat Cr sat The value of
[0052]
[0053] As for the associated object (blue clothing area), people rarely wear high-saturation clothing in daily life, so there is no need to configure saturation enhancement for the blue clothing area, such as setting the enhancement coefficient Sat of the blue clothing area to 1.0. It can be seen that in this example, both the target area image quality parameters and the associated area image quality parameters include saturation, but the values of the two are different. In this way, the blue sky and blue clothes in the image after image quality adjustment can have different saturations, avoiding distortion of the blue color of clothes when adjusting the image quality, especially adjusting the blue sky. In addition, the saturation enhancement coefficient Sat in the image quality parameters of other areas of the image, i.e., non-target areas, can also be set to 1.0, or less than 1.0, and is set according to actual needs.
[0054] In this way, when the current frame image also includes an associated object of the target object, the associated area occupied by the associated object can be adjusted differently from the target object, and the non-target area can be adjusted differently from the target area, and the same as or different from the associated area, so as to further improve the display effect of the adjusted image.
[0055] It should be noted that the above steps S251 to S254 are executed by the analysis circuit 2 , and step S255 is executed by the control circuit 7 .
[0056] It can be understood that, as mentioned above, the current frame image in the present application is an image in a video or streaming media. Therefore, the current area image quality parameters, non-target area image quality parameters, global image quality parameters, and associated area image quality parameters determined by the control circuit 7 can be applied to adjust the next frame image in the video or streaming media, and the next frame image is the image in the video or streaming media that is located after the current frame image.
[0057] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0058] The above description is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application; in the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.
Claims
1. A method for adjusting image quality, characterized in that: include: Get the current frame image; Identifying a target object in the current frame image and confirming a target area occupied by the target object in the current frame image; as well as When the area ratio of the target area in the current frame image exceeds a threshold, the target area image quality parameters of the target area and the non-target area image quality parameters of the non-target area in the current frame image are determined, wherein the target area image quality parameters and the non-target area image quality parameters are different, and the non-target area is at least a partial area outside the target area in the current frame image.
2. The image quality adjustment method according to claim 1, characterized in that: The determining of the target area image quality parameter of the target area specifically includes: The image quality information of the target area is acquired, and the image quality parameter of the target area is determined according to the image quality information of the target area.
3. The image quality adjustment method according to claim 1, characterized in that: The determining of the non-target area image quality parameter of the non-target area in the current frame image specifically includes: Acquiring global image quality information and scene information of the current frame image; and The non-target area image quality parameter is determined according to the target area image quality parameter, the global image quality information and the scene information.
4. The image quality adjustment method according to claim 1, characterized in that: Also includes: Determining whether the current frame image includes an associated object of the target object; When the associated object is not included in the current frame image, determining the non-target area to be all areas outside the target area in the current frame image; When the associated object is included in the current frame image, determining an associated area occupied by the associated object in the current frame image; as well as The area outside the target area and the associated area is determined as the non-target area.
5. The image quality adjustment method according to claim 4, characterized in that: The associated object has a correlation with the image quality of the target object, and the image quality adjustment method further includes: An associated region image quality parameter of the associated region is determined according to the target region image quality parameter, and the associated region image quality parameter is different from the target region image quality parameter.
6. The image quality adjustment method according to claim 1, characterized in that: Also includes: When it is determined that the area ratio of the target area of the target object in the current frame image does not exceed the threshold, the global image quality parameter of the current frame image is determined, and the global image quality parameter is the same as or different from the non-target area image quality parameter.
7. The image quality adjustment method according to claim 1, characterized in that: The obtaining of the current frame image specifically comprises obtaining the current frame image from a video stream; and The image quality of a next frame image in the video stream is adjusted based on the target area image quality parameters and the non-target area image quality parameters, and the next frame image is an image in the video stream that is located after the current frame image.
8. An image quality adjustment device, characterized in that: include: DMA circuit, used to obtain the current frame image; A neural network processor, used for identifying a target object in the current frame image; An analysis circuit, used for determining a target area occupied by a target object in the current frame image; A control circuit is used to determine a target area quality parameter of the target area and a non-target area quality parameter of a non-target area in the current frame image when the area ratio of the target area in the current frame image exceeds a threshold, wherein the non-target area quality parameter is different from the target area quality parameter, and the non-target area is at least a partial area outside the target area in the current frame image.
9. The image quality adjustment device according to claim 8, characterized in that: Also includes: An image quality acquisition circuit, used for acquiring image quality information of the target area; The control circuit determines the target area image quality parameter according to the image quality information of the target area.
10. The image quality adjustment device according to claim 8, characterized in that: The neural network processor also identifies scene information of the current frame image; An image quality acquisition circuit, used for acquiring image quality information of the target area and global image quality information of the current frame image; as well as The control circuit determines the non-target area image quality parameter according to the target area image quality parameter, the global image quality information and scene information.
11. The image quality adjustment device according to claim 8, characterized in that: The analysis circuit is also used for: Determining whether the current frame image includes an associated object of the target object; When the associated object is not included in the current frame image, determining the non-target area to be all areas outside the target area in the current frame image; When the associated object is included in the current frame image, an associated region occupied by the associated object in the current frame image is determined, and the target region and a region outside the associated region are determined as the non-target region.
12. The image quality adjustment device according to claim 11, characterized in that: The associated object has a correlation with the image quality of the target object, and the control circuit determines an associated region image quality parameter of the associated region according to the target region image quality parameter, and the associated region image quality parameter is different from the target region image quality parameter.