CN-115760626-B - Image tone mapping adjustment method, system, storage medium and terminal

CN115760626BCN 115760626 BCN115760626 BCN 115760626BCN-115760626-B

Abstract

The invention provides an image tone mapping adjustment method, system, storage medium and terminal, which comprise the steps of determining first characteristic information of an image with high bit width according to pixel values and image related information, dividing the image with high bit width into a plurality of characteristic areas, performing image enhancement, performing preliminary tone mapping on the characteristic areas to obtain mapping gain of pixel points in each characteristic area, obtaining brightness information and detail information of each layer of image in the characteristic areas, reconstructing second characteristic information based on original resolution of the image with high bit width according to dividing relations of the brightness information, the image information and the characteristic areas, compositing mapping gain of each characteristic area according to fusion weights to obtain composite gain of each characteristic area, and processing the image with high bit width according to the composite gain to obtain a low bit width output image. The invention can map the high-bit-width image to the displayable brightness range of the display and improve the image quality.

Inventors

LI JIAKUN
SONG BO
WANG YONG
WEN JIANXIN

Assignees

成都微光集电科技有限公司

Dates

Publication Date: 20260505
Application Date: 20221124

Claims (9)

1. A method for adjusting tone mapping of an image, comprising: Acquiring pixel values and image related information of an input high-bit-width image, and determining first characteristic information of the high-bit-width image according to the pixel values and the image related information, wherein the first characteristic information comprises current scene information, exposure information, image brightness information and equipment information of the high-bit-width image; dividing the high-bit wide image into a plurality of characteristic areas according to the first characteristic information, and carrying out image enhancement processing on the characteristic areas; performing preliminary tone mapping on each characteristic region of the high-bit-width image according to the first characteristic information to obtain mapping gain of pixel points in each characteristic region; Acquiring brightness information and detail information of each layer of image in the characteristic region, and reconstructing second characteristic information based on the original resolution of the high-bit-width image according to the brightness information, the detail information and the division relation of the characteristic region; Calculating the fusion weight of the feature areas according to the second feature information, and compositing the mapping gain of each feature area according to the fusion weight to obtain the composite gain of each feature area; and processing the high-bit-width image according to the synthesis gain to obtain a low-bit-width output image.
2. The method according to claim 1, wherein calculating the luminance factor of the high-bandwidth image according to the pixel values of the high-bandwidth image at different positions, the white balance coefficient, and the image edge comprises: Obtaining preliminary brightness information of each position of the high-bit-width image according to the product of the white balance coefficient and the pixel value; And window filtering is carried out on the preliminary brightness information of each position in the high-bit wide image to obtain the brightness factor of each position.
3. The method of claim 1, wherein the first characteristic information further comprises a gradient factor and detail information, wherein the gradient factor is calculated according to the luminance factor.
4. A method of adjusting tone mapping of an image according to claim 3, wherein said performing preliminary tone mapping on each of said feature areas of said high bandwidth image according to said first feature information to obtain a mapping gain for pixels in each of said feature areas comprises: establishing a standard tone mapping function and acquiring standard parameter information of the standard tone mapping function; calculating a first intermediate parameter and a second intermediate parameter according to the brightness factor and the gradient factor; calculating characteristic region parameter information of each characteristic region according to the first intermediate parameter and the second intermediate parameter, and obtaining a characteristic tone mapping function corresponding to each characteristic region according to the characteristic region parameter information; and calculating the mapping gain of the pixel points in the characteristic region according to the characteristic tone mapping function and the brightness factor of each characteristic region.
5. The method of adjusting tone mapping of an image according to claim 4, wherein the mapping gain satisfies the following formula: , Wherein, the Representing the luminance factor at the midpoint (x, y) of the feature region, , The feature tone mapping function representing each of the feature regions, wherein A, B, C, D, E represents the feature region parameter information of the feature tone mapping function.
6. The method according to claim 1, wherein calculating the fusion weights of the feature areas according to the second feature information, and compositing the mapping gain of each feature area according to the fusion weights to obtain the composite gain of different positions of the high-bandwidth image, comprises: Constructing a linear function, and calculating fusion weights of each characteristic region based on the linear function and the second characteristic information; and obtaining the synthesized gain according to the mapping gain in the characteristic region and the fusion weight weighting fusion calculation corresponding to the characteristic region.
7. An adjustment system for tone mapping of an image, comprising: The information acquisition module is used for acquiring pixel values and image related information of an input high-bit-width image, and determining first characteristic information of the high-bit-width image according to the pixel values and the image related information, wherein the information acquisition module comprises the steps of acquiring current scene information, exposure information, image brightness information and equipment information of the high-bit-width image, determining a white balance coefficient according to the current scene information, the exposure information, the image brightness information and the equipment information, and calculating brightness factors of the high-bit-width image according to the pixel values, the white balance coefficient and the image edges of the high-bit-width image at different positions; The dividing module is used for dividing the high-bit-width image into a plurality of characteristic areas according to the first characteristic information and carrying out image enhancement processing on the characteristic areas; The mapping gain calculation module is used for carrying out preliminary tone mapping on each characteristic region of the high-bit-width image according to the first characteristic information so as to obtain the mapping gain of the pixel point in each characteristic region; the reconstruction module is used for acquiring brightness information and detail information of each layer of image in the characteristic region and reconstructing second characteristic information based on the original resolution of the high-bit-width image according to the brightness information, the detail information and the division relation of the characteristic region; The synthetic gain calculation module is used for calculating the fusion weight of the characteristic areas according to the second characteristic information, and compositing the mapping gain of each characteristic area according to the fusion weight to obtain the synthetic gain of each characteristic area; And the conversion output module is used for processing the high-bit-width image according to the synthesis gain to obtain a low-bit-width output image.
8. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the method of adjusting the tone mapping of an image according to any one of claims 1 to 6.
9. The terminal is characterized by comprising a processor and a memory; The memory is used for storing a computer program; The processor is configured to execute the computer program stored in the memory, so that the terminal performs the adjustment method of image tone mapping according to any one of claims 1 to 6.

Description

Image tone mapping adjustment method, system, storage medium and terminal Technical Field The present invention relates to the field of image processing technologies, and in particular, to a method, a system, a storage medium, and a terminal for adjusting tone mapping of an image. Background Currently, in computational photography, the performance of a photographing device on areas with different brightness of an image is limited, so that a Low dynamic range (Low DYNAMIC RANGE, LDR) image with different brightness is often obtained by collecting multiple frames of continuous surrounding exposure, and is fused into a High Dynamic Range (HDR) image with high bit width through a high dynamic range (HIGH DYNAMIC RANGE, HDR) fusion algorithm. However, the display range of the display is limited, typically an RGB image of 8 bits, so that an HDR image of high bit width needs to be mapped to a luminance section that the display can display, a process called Tone Mapping (Tone Mapping). After the image is fused by HDR, the bit width of the image is usually higher, in order to map the image to a displayable brightness range and ensure that dark information is not lost, highlight information is not overexposed, and linear compression is obviously not feasible, so a curve is usually used as a mapping function. The curve has small compression ratio in dark and even stretches brightness, and has large compression ratio in bright, namely, the curve has large slope in a low-brightness range and small slope in a high-brightness range. The most widely used in computing photography is the Reinhard tone mapping algorithm, based on tone mapping of bayer format data (RAW). Due to defects of the high-bandwidth image, large brightness range, defects of the Reinhard tone mapping method and the like, negative effects such as poor contrast (partial), high-light edge abnormality, high-light overexposure, darkness and the like often occur in tone mapping results. In real devices and scenes, defects of the image itself and the need for high dynamic range are unavoidable, and therefore in tone mapping, these negative effects are often present, affecting the image imaging quality. Therefore, there is a need to provide a new image tone mapping adjustment method, system, storage medium and terminal to solve the above-mentioned problems in the prior art. Disclosure of Invention The invention aims to provide an adjusting method, an adjusting system, a storage medium and a terminal for tone mapping of an image, which can map an image with high bit width to a displayable brightness range of a display and improve image quality. To achieve the above object, the method for adjusting tone mapping of an image according to the present invention includes: acquiring pixel values and image related information of an input high-bit-width image, and determining first characteristic information of the high-bit-width image according to the pixel values and the image related information; dividing the high-bit wide image into a plurality of characteristic areas according to the first characteristic information, and carrying out image enhancement processing on the characteristic areas; performing preliminary tone mapping on each characteristic region of the high-bit-width image according to the first characteristic information to obtain mapping gain of pixel points in each characteristic region; Acquiring brightness information and detail information of each layer of image in the characteristic region, and reconstructing second characteristic information based on the original resolution of the high-bit-width image according to the brightness information, the detail information and the division relation of the characteristic region; Calculating the fusion weight of the feature areas according to the second feature information, and compositing the mapping gain of each feature area according to the fusion weight to obtain the composite gain of each feature area; and processing the high-bit-width image according to the synthesis gain to obtain a low-bit-width output image. The image tone mapping adjustment method has the beneficial effects that the image tone mapping adjustment method is divided into a plurality of characteristic areas according to the difference of the first characteristic information of the high-bit-width image, so that tone mapping is conveniently and adaptively calculated according to the difference of the characteristic areas to obtain mapping gains of different characteristic areas, the tone mapping can show better effects in different scenes, the generated low-bit-width high-dynamic-range image can ensure effective information of the image, and meanwhile, the high-gloss details of the image are better, the edge appearance is smooth, dark place information is better, the contrast of the area with easy bias of middle brightness is higher, the phenomena of high-gloss overflow, dark place detail loss, bias and the like of the image can be effect