Color enhancement method and system

Abstract

The invention discloses a color enhancement method, which comprises the following steps of: firstly, determining scene classification credibility of content currently displayed on a display screen, and then determining a color adjustment value according to the scene classification credibility and based on a preset RGB color mapping curve of a scene. The method is high in real-time performance, color enhancement can be dynamically carried out according to the current display content, and the display effect of the display screen can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of display screens, and in particular to a color enhancement method and system. Background Art

[0002] With the development of the smart car industry, users have an increasing demand for smart cockpits, and the cockpit display is an extremely important component of the smart cockpit. On the one hand, it can display the status of the car, such as driving speed, fuel consumption, temperature inside the car, etc. On the other hand, it can also meet users' needs for infotainment, etc. At present, the cockpit display is developing towards large screens and multi-functions.

[0003] During the driving process of the car, the lighting and other conditions inside the car may change at any time. In addition, the display effects required by the cockpit display screen are also different for different functional requirements. In order to achieve better display effects and user experience, the color enhancement effect of the cockpit display screen can be dynamically adjusted. However, the display of existing car cockpit display screens is mostly based on traditional image enhancement methods, and the display color effect is mainly adjusted based on subjective experience. This makes the debuggers set different color enhancement schemes according to different test streams and pictures, and finally meet the corresponding needs by seeking a balance point, but it is difficult to dynamically adjust the color to the optimal effect according to the image content. Summary of the invention

[0004] In view of some or all of the problems in the prior art, the present invention provides a color enhancement method in a first aspect, comprising:

[0005] Determine the scene classification confidence of the content currently displayed on the display screen;

[0006] Determining a color adjustment value of the content according to the scene classification confidence and based on a preset RGB color mapping curve of the scene; and

[0007] The color enhancement processing is performed on the currently displayed content according to the color adjustment value.

[0008] Furthermore, if the content currently displayed on the display screen is a video, the scene classification confidence of each frame of the picture is determined frame by frame.

[0009] Furthermore, determining the scene classification trust of the content currently displayed on the display screen includes:

[0010] The content is identified through a scene-based classification model, and scores of the content corresponding to N preset scenes are determined based on the identification results as scene classification confidence levels of the content.

[0011] Furthermore, the sum of the scores of the content corresponding to the preset N scenarios is 100.

[0012] Furthermore, the value of N is 10, and the N scenes include: distant views of natural scenery, close views of natural scenery, urban buildings, street vehicles, close-ups of people, animals, fruits and vegetables, cartoon images, instrument displays, and industrial films.

[0013] Furthermore, the RGB color mapping curve is obtained based on historical experience and is preset in the system.

[0014] Furthermore, the contrast mapping curve includes three mapping curves of red, blue and green, wherein each curve includes M anchor points.

[0015] Furthermore, the value of M is 9.

[0016] Further, determining the color adjustment value of the content includes:

[0017] The scene classification confidence is used as a weight to perform weighted processing on the RGB color mapping curve of the preset scene.

[0018] A second aspect of the present invention provides a system for color enhancement, comprising:

[0019] an identification module, which is used to determine the scene classification confidence of the content currently displayed on the display screen; and

[0020] The color adjustment module is used to determine the color adjustment value of the content according to the scene classification trust and based on the RGB color mapping curve of the preset scene.

[0021] The present invention provides a color enhancement method and system, which dynamically adjusts the color enhancement effect based on the currently displayed image content to achieve the best display experience. The method has strong real-time performance and can effectively improve the display effect of the display screen. Moreover, it can be implemented directly using existing hardware in intelligent driving without the need to design additional hardware modules, which reduces the development difficulty and has obvious cost advantages. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] To further illustrate the above and other advantages and features of various embodiments of the present invention, a more specific description of various embodiments of the present invention will be presented with reference to the accompanying drawings. It will be understood that these drawings only depict typical embodiments of the present invention and are therefore not to be considered as limiting the scope thereof. In the accompanying drawings, for clarity, identical or corresponding parts will be represented by identical or similar reference numerals.

[0023] Figure 1 A schematic flow chart showing a color enhancement method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In the following description, the present invention is described with reference to various embodiments. However, those skilled in the art will recognize that the various embodiments can be implemented without one or more specific details or with other replacement and / or additional methods or components. In other cases, well-known structures or operations are not shown or described in detail to avoid blurring the inventive point of the present invention. Similarly, for the purpose of explanation, specific numbers and configurations are set forth to provide a comprehensive understanding of the embodiments of the present invention. However, the present invention is not limited to these specific details. In addition, it should be understood that the various embodiments shown in the drawings are illustrative representations and are not necessarily drawn in correct proportions.

[0025] In this specification, reference to "one embodiment" or "the embodiment" means that a particular feature, structure, or characteristic described in conjunction with the embodiment is included in at least one embodiment of the present invention. The phrase "in one embodiment" appearing in various places in this specification does not necessarily all refer to the same embodiment.

[0026] It should be noted that the embodiments of the present invention describe the method steps in a specific order, but this is only for the purpose of illustrating the specific embodiment, rather than limiting the order of the steps. On the contrary, in different embodiments of the present invention, the order of the steps can be adjusted according to actual needs.

[0027] In the present invention, each module of the system according to the present invention can be implemented using software, hardware, firmware or a combination thereof. When the module is implemented using software, the function of the module can be implemented by a computer program flow, for example, the module can be implemented by a code segment (such as a code segment of a language such as C, C++) stored in a storage device (such as a hard disk, a memory, etc.), wherein the corresponding function of the module can be implemented when the code segment is executed by a processor. When the module is implemented using hardware, the function of the module can be implemented by setting a corresponding hardware structure, for example, the function of the module can be implemented by hardware programming a programmable device such as a field programmable gate array (FPGA), or the function of the module can be implemented by designing an application-specific integrated circuit (ASIC) including electronic devices such as a plurality of transistors, resistors and capacitors. When the module is implemented using firmware, the function of the module can be written into a read-only memory such as an EPROM or EEPROM of the device in the form of a program code, and the corresponding function of the module can be implemented when the program code is executed by a processor. In addition, some functions of the module may need to be implemented by separate hardware or by collaboration with the hardware, for example, the detection function is implemented by corresponding sensors (such as proximity sensors, acceleration sensors, gyroscopes, etc.), the signal transmission function is implemented by corresponding communication devices (such as Bluetooth devices, infrared communication devices, baseband communication devices, Wi-Fi communication devices, etc.), the output function is implemented by corresponding output devices (such as displays, speakers, etc.), and so on.

[0028] In order to dynamically adjust the color enhancement effect based on the current image content and achieve better display effects, the present invention provides a color enhancement method and system, which is implemented based on the existing embedded neural network processor NPU hardware and classification model in intelligent driving without the need for additional hardware modules, thereby simplifying the overall structure of the intelligent driving system and having obvious cost advantages.

[0029] The scheme of the present invention is further described below in conjunction with the accompanying drawings of the embodiments.

[0030] Figure 1 FIG. 2 is a flow chart showing a color enhancement method according to an embodiment of the present invention. Figure 1 As shown, a color enhancement method comprises:

[0031] First, in step 101, scene recognition. The scene of the content currently displayed on the display screen is determined for recognition, and its scene classification confidence is determined. In one embodiment of the present invention, the scene classification confidence of the content currently displayed on the display screen is determined mainly by identifying the content through a scene-based classification model, and determining the score of the content corresponding to the preset N scenes based on the recognition result as the scene classification confidence of the content. In order to facilitate subsequent calculations, in one embodiment of the present invention, the sum of the scores of the content corresponding to the preset N scenes is 100. In one embodiment of the present invention, the N scenes may include, for example, 10 scenes such as: distant views of natural scenery, close views of natural scenery, urban buildings, street vehicles, close-ups of people, animals, fruits and vegetables, cartoon images, instrument displays, and industrial movies. In one embodiment of the present invention, the scene-based classification model directly uses the existing NPU hardware and classification model in intelligent driving, specifically a model based on TensorFlowRT, to identify the current scene situation. In one embodiment of the present invention, the scene classification trust w[0], w[1], ....., w[9] is converted according to the reasoning output result of the NPU, and the reasoning output result of the NPU is a one-dimensional vector, in which each item corresponds to the possibility value of a scene. Taking the 10 scenes mentioned above as an example, if the inference output result is [0.91, 0.01, 0.01, 0.01, 0.01, 0.01, 0.01, 0.01, 0.01, 0.01], it can be converted into the trust of the 10 scenes as 91, 1, 1, 1, 1, 1, 1, 1, 1, that is, the trust of the scene classification of the content corresponding to the distant view of natural scenery, the close view of natural scenery, urban buildings, street vehicles, close-ups of people, animals, fruits and vegetables, cartoon images, instrument displays, and industrial movies is 91, 1, 1, 1, 1, 1, 1, 1, 1, 1, respectively. It can be seen that its trust as the distant view of natural scenery is the largest and exceeds 90. In one embodiment of the present invention, the current display content of the display screen may be a picture or a video. If it is a picture, scene recognition is directly performed. If it is a video, the scene classification trust of each frame of the picture is determined frame by frame;

[0032] Next, in step 102, the color adjustment value is determined. According to the scene classification trust, based on the preset scene RGB color mapping curve, the color adjustment value of the content is determined. In one embodiment of the present invention, the RGB color mapping curve is drawn based on the color mapping values ​​of M anchor points (Po i nt), including three curves of red, green and blue, wherein each curve includes M anchor points. The value of M may be 9, for example. In one embodiment of the present invention, the RGB color mapping curve is obtained based on historical experience, specifically, it is the result of obtaining the highest score through subjective and objective evaluation according to the best display effect adjusted by the expert system according to the scene, and is set as curve data, and the contrast mapping curve is preset in the system. In one embodiment of the present invention, the color adjustment value of the content is determined mainly by taking the scene classification trust as the weight and performing weighted processing on the preset scene RGB color mapping curve. Taking the 10 scenes mentioned above as an example, the 10 preset RGB color mapping curves are defined as Sr1, Sg1, Sb1, Sr2, Sg2, Sb2, ..., Sr10, Sg10, Sb10 in sequence. Taking the red mapping curve of the first group as an example, the anchor points on each curve are defined as Sr1[0]Sr1[1]...Sr1[M-1]. Then, the color adjustment value of the content is determined by taking the scene classification trust obtained in the above steps as a weight to obtain a weighted average value as the current point position:

[0033] Sr[0]=(w[0]*Sr1[0]+w[1]*Sr2[0]+.....+w[9]*Sr10[0]) / 100,

[0034] After the adjustment values ​​of the M points are calculated in sequence, a color mapping curve corresponding to the content can be obtained, that is, a color enhancement value of the content currently displayed on the display screen can be obtained; and

[0035] Finally, in step 103, color enhancement is performed. The image is processed according to the color mapping curve, for example, contrast is adjusted, and finally displayed. In one embodiment of the present invention, the image is processed by an interpolation processing method, that is, the pixel value is determined in the X-axis direction between two anchor points by the contrast mapping curve, and then a new Y-axis value is obtained by a bilinear interpolation method and output.

[0036] In order to realize the dynamic adjustment method as described above, the present invention also provides a system for color enhancement, which includes a recognition module and a color adjustment module, wherein the recognition module is used to determine the scene classification trust of the content currently displayed on the display screen; and the color adjustment module is used to determine the color adjustment value of the content based on the RGB color mapping curve of the preset scene according to the scene classification trust, and obtain the color mapping curve. As mentioned above, the recognition module can directly use the existing NPU hardware and classification model in intelligent driving. The color adjustment module is pre-set with RGB color mapping curves of N different scenes, and the RGB color mapping curve is obtained based on historical experience. Specifically, it is the best display effect adjusted by the expert system according to the scene, and the result of obtaining the highest score after subjective and objective evaluation is set as curve data. In one embodiment of the present invention, the system also includes a post-processing module, which processes the current display content based on the color mapping curve and outputs it to the display screen.

[0037] Although various embodiments of the present invention are described above, it should be understood that they are presented as examples only and not as limitations. It is obvious to those skilled in the relevant art that various combinations, modifications and changes can be made thereto without departing from the spirit and scope of the present invention. Therefore, the breadth and scope of the present invention disclosed herein should not be limited by the exemplary embodiments disclosed above, but should only be defined according to the attached claims and their equivalents.

Claims

1. A color enhancement method, characterized in that: Includes steps: Determine the scene classification confidence of the content currently displayed on the display screen; Determining a color adjustment value of the content according to the scene classification confidence and based on a preset RGB color mapping curve of the scene; as well as The contrast is adjusted according to the color adjustment value.

2. The color enhancement method according to claim 1, characterized in that: If the content currently displayed on the display screen is a video, the scene classification confidence of each frame of the picture is determined frame by frame.

3. The color enhancement method according to claim 1, characterized in that: Determining the scene classification confidence of the content currently displayed on the display includes the steps of: The content is identified through a scene-based classification model, scores of the content corresponding to preset N scenes are determined based on the identification results, and the scores are used as scene classification confidence levels of the content.

4. The color enhancement method according to claim 3, characterized in that: The total score of the content corresponding to the preset N scenarios is 100.

5. The color enhancement method according to claim 3, characterized in that: The value of N is 10, and the N scenes include: distant views of natural scenery, close views of natural scenery, urban buildings, street vehicles, close-ups of people, animals, fruits and vegetables, cartoon images, instrument displays, and industrial films.

6. The color enhancement method according to claim 1, characterized in that: The RGB color mapping curve is obtained based on historical experience and is preset in the system.

7. The color enhancement method according to claim 1, characterized in that: The RGB color mapping curve includes a red mapping curve, a green mapping curve and a blue mapping curve, wherein each curve includes color mapping values ​​of M anchor points.

8. The color enhancement method according to claim 7, characterized in that: The value of M is 9.

9. The color enhancement method according to claim 1, characterized in that: Determining the color adjustment value of the content comprises the steps of: The scene classification confidence is used as a weight to perform weighted processing on the RGB color mapping curve of the preset scene.

10. A system for color enhancement, characterized in that The color of a display screen is adjusted according to the color enhancement method according to any one of claims 1 to 9, and the system comprises: an identification module configured to determine a scene classification confidence level of content currently displayed on the display screen; and The color adjustment module is configured to determine the color adjustment value of the content according to the scene classification confidence and based on a preset RGB color mapping curve of the scene.

Images