WO2019232834A1

WO2019232834A1 - Face brightness adjustment method and apparatus, computer device and storage medium

Info

Publication number: WO2019232834A1
Application number: PCT/CN2018/092652
Authority: WO
Inventors: 戴磊
Original assignee: 平安科技（深圳）有限公司
Priority date: 2018-06-06
Filing date: 2018-06-25
Publication date: 2019-12-12
Also published as: CN108961175B; CN108961175A

Abstract

A face brightness adjustment method and apparatus, a computer device and a storage medium. The face brightness adjustment method comprises: obtaining a face image, and obtaining a target detection area on the basis of the face image (S10); obtaining a brightness value of each pixel in the target detection area (S20); using an image natural brightness algorithm to perform processing on the brightness value so as to obtain an average natural brightness value of the target detection area (S30); if the average natural brightness value is not within a preset natural brightness range, constructing a brightness adjustment curve on the basis of the average natural brightness value, and using the brightness adjustment curve to adjust the brightness of the face image so as to obtain a target image (S40). The described face brightness adjustment method obtains an average natural brightness value according to a target detection area of a face image, constructs a brightness adjustment curve on the basis of the average natural brightness value and then adjusts the brightness, thereby improving the adjustment speed and reducing distortion.

Description

Face brightness adjustment method and device, computer equipment and storage medium

This application is based on a Chinese patent application filed on June 6, 2018 with the application number 201810573812.3 and entitled "Face Brightness Adjustment Method, Apparatus, Computer Equipment, and Storage Medium" and claims its priority.

Technical field

The present application relates to the field of image processing technology, and in particular, to a method, a device, a computer device, and a storage medium for adjusting human face brightness.

Background technique

Image brightness adjustment is a part of image processing, which refers to adjusting the brightness of pixel values in an image. The brightness of an image is essentially the brightness of each pixel in the image. The brightness of each pixel can be expressed by the size of the RGB value, where R represents the red channel component, G represents the green channel component, and B represents the blue channel component. When the values of R, G, and B are all 0, the pixels are black and the brightness is the darkest; when the values of R, G, and B are all 255, the pixels are white and the brightness is the brightest.

In the existing image brightness adjustment algorithms, the simple addition and subtraction adjustment of pixels on the RGB channel after calculating the arithmetic mean of the pixels are mostly performed. The adjusted image brightness has severe distortion, especially for extremely bright or dark pixels. The brightness distortion is serious, and the layered sense of the image cannot be guaranteed.

Summary of the Invention

Embodiments of the present application provide a method, a device, a computer device, and a storage medium for adjusting the brightness of a face, so as to solve the problem of serious distortion in the current brightness adjustment process of a face image.

A method for adjusting face brightness, including:

Acquiring a face image, and acquiring a target detection area based on the face image;

Obtaining a brightness value of each pixel in the target detection area;

Processing the brightness value by using an image natural brightness algorithm to obtain an average natural brightness value of the target detection area;

If the average natural brightness value is not within a preset natural brightness range, a brightness adjustment curve is constructed based on the average natural brightness value, and the brightness adjustment curve is used to perform brightness adjustment on the face image to obtain a target image.

A face brightness adjusting device includes:

A target detection area acquisition module, configured to obtain a face image, and obtain a target detection area based on the face image;

A pixel brightness obtaining module, configured to obtain a brightness value of each pixel in the target detection area;

An area average brightness acquisition module, configured to process the brightness value by using an image natural brightness algorithm to obtain an average natural brightness value of the target detection area;

A brightness adjustment module configured to construct a brightness adjustment curve based on the average natural brightness value if the average natural brightness value is not within a preset natural brightness range, and use the brightness adjustment curve to perform brightness adjustment on the face image To get the target image.

A computer device includes a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the following steps are implemented:

Obtaining a brightness value of each pixel in the target detection area;

One or more non-volatile readable storage media storing computer-readable instructions, which when executed by one or more processors, cause the one or more processors to perform the following steps:

Obtaining a brightness value of each pixel in the target detection area;

Details of one or more embodiments of the present application are set forth in the accompanying drawings and description below, and other features and advantages of the present application will become apparent from the description, the drawings, and the claims.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to explain the technical solutions of the embodiments of the present application more clearly, the drawings used in the description of the embodiments of the application will be briefly introduced below. Obviously, the drawings in the following description are just some embodiments of the application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without paying creative labor.

FIG. 1 is an application environment diagram of a face brightness adjustment method according to an embodiment of the present application; FIG.

FIG. 2 is a flowchart of a method for adjusting face brightness in an embodiment of the present application; FIG.

FIG. 3 is a flowchart of step S10 in FIG. 2;

4 is a flowchart of step S13 in FIG. 3;

FIG. 5 is a flowchart of step S40 in FIG. 2;

6 is a schematic diagram of a face brightness adjusting device according to an embodiment of the present application;

FIG. 7 is a schematic diagram of a computer device in an embodiment of the present application.

Detailed ways

In the following, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by a person of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

The face brightness adjustment method provided in the embodiment of the present application may be applied in the application environment shown in FIG. 1, that is, the face brightness adjustment method is applied in the brightness adjustment system shown in FIG. 1. The brightness adjustment system includes a server and a server. Clients connected to the server via the network. The client can communicate with the server through any of wireless networks such as WiFi, 3G, 4G, and 5G or wired networks. Client (Client), or client, refers to the program that provides local services to clients corresponding to the server. The client can be installed on, but not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server can be implemented by an independent server or a server cluster composed of multiple servers.

In an embodiment, as shown in FIG. 2, the face brightness adjustment method is applied to the server in FIG. 1 as an example for description. As shown in FIG. 2, the face brightness adjustment method includes the following steps:

S10: Acquire a face image, and obtain a target detection area based on the face image.

A face image refers to an image containing a person's face. It is an unprocessed image. Specifically, it can be an image that the client sends to the server in real time for image processing, or it can be sent to the server in advance by the client and stored by the server. Called image. The image can be an image stored locally by the client or an image captured by the client through a camera. The target detection area refers to the detection area in the face image. The detection area may be an area where only the face is retained after removing the image background and other interference factors.

The server may use a face detection algorithm to identify a face image, and use the detected face area as a target detection area. The target detection area refers to the area that remains after removing the background and other factors that do not belong to the face of the person in the face image. For example, an edge detection algorithm may be used to process a face image, obtain a human face contour, and determine a target detection area containing the human face contour.

S20: Obtain the brightness value of each pixel in the target detection area.

A pixel is the smallest unit in an image. The brightness value of a pixel represents the brightness of the image corresponding to that pixel. The computer uses the position, color, and brightness of each pixel to display the brightness of the entire image. The brightness value of a pixel is composed of three pixel units, R, G, and B. The brightness value represents the brightness of the image color. The brightness of an image can be represented by the brightness values of all the pixel values that make up the image. The brightness value of each pixel is between 0 and 255, where 0 is completely black and 255 is completely white.

In this embodiment, after the server obtains the target detection area according to the face image, the server calculates the brightness value of the target detection area to obtain the brightness value of each pixel. After the server obtains the brightness value of each pixel in the target detection area, it can use the grayscale conversion formula to perform grayscale calculation to obtain the converted brightness value. The grayscale conversion formula is as follows: Gray = R * 0.299 + G * 0.587 + B * 0.114, where Gray is the gray value of the image, that is, the brightness value of the converted image, R represents the red channel component, G represents the green channel component, and B represents the blue channel component. In the process of converting an RGB image into a grayscale image, the grayscale conversion formula used can achieve fast and high-precision calculations, so that the brightness value of each pixel in the target detection area can be quickly obtained for subsequent subsequent grayscale image-based processing. Brightness adjustment helps to increase the rate of brightness adjustment.

S30: Use the image natural brightness algorithm to process the brightness value to obtain the average natural brightness value of the target detection area.

The image natural brightness algorithm refers to an algorithm that calculates the average brightness of an image based on the brightness value of each pixel in the image. The average natural brightness value is an average value indicating how bright the image is.

In this embodiment, after the server obtains the brightness value of each pixel in the target detection area, based on the brightness value of each pixel, the image natural brightness algorithm is used to calculate the average natural brightness value. The formula of the image natural brightness algorithm is as follows :

Among them, N is the number of pixels in the target detection area, Lum _ave is the average natural brightness value of the target detection area, Lum (x, y) is the brightness value corresponding to the pixel points (x, y) of the target detection area, and δ is a constant . δ is a small constant used to prevent the calculation result of the logarithm from going to negative infinity. For example, δ can be 0.0001. For each pixel in the target detection area, calculate the luminance value Lum (x, y) of the pixel, and then find the natural logarithm of the sum of the luminance value Lum (x, y) and a constant δ; then The natural logarithm corresponding to the luminance value Lum (x, y) of all pixels is averaged; then the natural index value of the average is obtained to obtain the average natural luminance value. The average natural brightness value obtained by the image natural brightness algorithm is more in line with the human eye's judgment of the brightness than the average brightness of the average value directly calculated, which can effectively reduce the distortion of subsequent brightness adjustments and help improve the level of brightness of the face image after adjustment sense.

S40: If the average natural brightness value is not within the preset natural brightness range, a brightness adjustment curve is constructed based on the average natural brightness value, and the brightness adjustment curve is used to adjust the brightness of the face image to obtain the target image.

The preset natural brightness range refers to a range of brightness values set in advance to determine whether the average natural brightness value of the image meets the processing requirements. The brightness value range of the image is between [0, 255]. The preset natural brightness range can be based on Demand adjustment. For example, to perform face training on a face image, a suitable natural brightness range (90-190) is set in advance during model training, and 90-190 is a preset natural brightness range. The brightness adjustment curve refers to a brightness curve obtained by curve fitting an image using a smooth curve, and the brightness curve can be used to adjust the brightness of the image. Brightness adjustment refers to the adjustment of the brightness value of each pixel in the face image according to the brightness adjustment curve. The target image refers to a face image whose brightness value is adjusted at each pixel according to the brightness adjustment curve.

In this embodiment, constructing the brightness adjustment curve based on the average natural brightness value can be implemented by a smooth curve fitting method, and performing curve fitting with three coordinate points to obtain a smooth curve, that is, a brightness adjustment curve. Specifically, curve fitting may be performed using a uniform B-spline, a quasi-uniform B-spline, a segmented Bezier, or a non-uniform B-spline curve. Based on the brightness adjustment curve, the target brightness value of each pixel in the face image is obtained, and the brightness value of each pixel in the face image is adjusted, that is, the brightness value of each pixel is adjusted to the corresponding target brightness Value, get the target image after completion.

For example, the server may use a non-linear adjustment method to construct a brightness adjustment curve based on the average natural brightness value, which specifically includes: setting the abscissa and ordinate in the plane coordinate system to each specific brightness value in the range of 0-255. The abscissa and ordinate divide 0-255 into multiple brightness intervals, and use the abscissa as the brightness value of each pixel in the face image before adjustment, and the ordinate as the brightness value of the corresponding pixel in the face image after adjustment. By determining the curve type and the coordinate points of the fitting process, the fitting process is performed based on the average natural brightness value. Further, when performing a fitting process on a face image, the curve type may select a non-uniform B-spline curve, and the coordinate points of the fitting process may be preset, based on the non-uniform B-spline curve and the coordinates of the fitting process Point to obtain the brightness adjustment curve, and fit the face image based on the brightness adjustment curve, which is helpful for the subsequent use of the brightness adjustment curve to quickly adjust the brightness of the face image, and the entire adjustment process is simple.

In the face brightness adjustment method provided in this embodiment, the server obtains a face image, obtains a target detection area based on the face image, and provides a basis for subsequently adjusting the brightness of the entire face image based on the brightness value of the target detection area. In order to reduce the amount of calculation data, the brightness adjustment speed is increased. The server obtains the brightness value of each pixel in the target detection area and uses the image natural brightness algorithm to calculate the average natural brightness value of the target detection area. The calculation process is simple and fast, which helps to improve the brightness adjustment speed. By setting the preset natural brightness range, the server can construct a brightness adjustment curve based on the average natural brightness value when the average natural brightness value is not within the preset natural brightness range, and use the brightness adjustment curve to adjust the brightness of the face image to obtain the target. image. The preset natural brightness range is used as the basis for judging whether to adjust the brightness, and the brightness adjustment curve is constructed based on the average natural brightness value, so that when the brightness adjustment curve is used to obtain the target image, the brightness adjustment is more balanced, and the brightness during the image adjustment is reduced. Distortion, to improve the layered sense of the brightness of the adjusted image.

In an embodiment, as shown in FIG. 3, acquiring the target detection area based on the face image in step S10 specifically includes the following steps:

S11: A facial feature point detection algorithm is used to detect facial feature point data of a face image. The facial feature point data includes coordinate data and local feature values.

Face feature point detection algorithm refers to an algorithm that automatically locates key feature points of the face (such as eyes, nose tip, mouth corners, eyebrows, and contour points of various parts of the face) based on the input face image. The facial feature point data refers to data of all feature points on a person's face. The feature points in this embodiment mainly include eyes, nose tips, mouth corners, eyebrows, and the like. The facial feature point data includes coordinate data and local feature values. The coordinate data refers to data indicating the position information of the facial feature points, and the local feature value is data indicating the feature information of the facial feature points.

In this embodiment, an Adaboost (iterative algorithm) facial feature point detection algorithm based on Harr features can be used, and a Harr feature-based V-Jdetector (detector) that comes with OpenCV (Computer Vision Library) is used to Detect facial feature points of a face image, and obtain facial feature point data. The detection process is as follows: first load the Harr feature detection classifier, load the face image to be detected, call internal functions for identification and labeling, and generate a detection window when a person is detected. Then, each facial feature point in the detection window is detected according to a cascade classifier (i.e., an Adaboost-based classifier) to obtain facial feature point data. The facial feature point data includes coordinate data and local feature values. The center position of the detection window of the facial feature point is used as the coordinate data of the facial feature point, and the Harr feature value extracted according to the detection window is used as a part of the facial feature point. Eigenvalues.

S12: Extract facial features in a face image according to the coordinate data and local feature values.

The facial features area refers to the area containing human facial features (eyebrows, eyes, nose, mouth, and ears), that is, to determine which facial feature points belong to the facial features based on the coordinate data and local feature values of the facial feature points.

In this embodiment, the facial feature point data can be obtained based on the Harr feature-based V-Jdetector that comes with OpenCV, and according to the coordinate data and local feature value data of any facial feature point, which one of the facial feature points is identified Officer characteristics. The process of extracting the facial features area is as follows: load Harr feature facial features detection classifier, load the face image to be detected, call internal functions for identification and labeling, and when the corresponding facial features are detected, generate a detection window to determine the facial features points, Thereby, the facial features region in the face image is determined.

S13: Obtain the target detection area according to the coordinate data of the tip of the nose and the outer corners of the eyes in the facial features area.

Nasal tip refers to the facial features that are located in the center of the human face and are composed of the inner feet of both sides of the wing cartilage and the soft tissues at the tip of the nose. The outer corners of the eyes are the tails of the eyes of a person near the temples.

In this embodiment, the facial features area extracted from facial feature point data obtained by the Harr feature-based V-Jdetector that comes with OpenCV can be used to obtain the target detection area based on the coordinate data of the nose tip and the corners of the eyes outside the eyes. The specific process is as follows: first extract the coordinate data of the tip of the nose and the outer corners of the eyes, and use these three coordinate data to determine the target detection area. There can be multiple methods for determining the first. According to the coordinate data of the three points, each Points are connected with each other to build a triangle area, and this triangle area is used as the target detection area. The second type is to select the coordinate data of any point to make a horizontal straight line based on the coordinate data of the three points, and then make the other two points This horizontal straight line is a vertical line that connects the other two points to build a rectangular area, and this rectangular area is used as the target detection area.

In steps S11-S13, the Adaboost (iterative algorithm) facial feature point detection algorithm based on Harr features is used, and the Harr feature-based V-Jdetector (detector) that comes with OpenCV (Computer Vision Library) is used Detect facial feature point data of a face image, extract facial features from facial feature point data, and determine target detection areas based on the coordinates of the nose clip and the corners of the eyes outside the eyes in the facial features area, which can quickly perform facial feature detection to determine target detection Area in order to use the brightness value of the target detection area to perform brightness adjustment, which helps to improve the speed of face brightness adjustment.

In an embodiment, as shown in FIG. 4, step S13, obtaining the target detection area according to the coordinate data of the nose tip and the outer corners of the eyes in the facial features region, specifically includes the following steps:

S131: Make a horizontal straight line based on the coordinate data of the nose tip.

The horizontal straight line refers to a line parallel to the horizontal plane. In this embodiment, a rectangular coordinate system can be constructed in the face image, with the X-axis direction of the rectangular coordinate system as the horizontal direction and the y-axis direction as the vertical direction. The nose coordinate data is used to obtain the x and y coordinates of the nose tip, and a horizontal straight line is made using the y coordinate of the nose tip.

S132: Construct a rectangular area based on the coordinate data of the outer corners of the eyes and the horizontal straight line, and determine the rectangular area as the target detection area.

A rectangular region refers to a rectangular region. In this embodiment, in the rectangular coordinate system of a face image, a horizontal straight line parallel to the X-axis direction is made with the y coordinate of the nose tip, and then according to the coordinates of the corners of the eyes outside the eyes. In the data, two points of the outer corners of the two eyes are respectively made vertical lines perpendicular to the horizontal straight line, and the two points of the outer corners of the two eyes are connected to construct a rectangular area, and the rectangular area is used as the target detection area.

In steps S131 and S132, the rectangular area constructed by the tip of the nose and the outer corners of both eyes is used as the target detection area, which reduces the area and calculation amount of the detection area, and helps to improve the rate of face brightness adjustment.

In an embodiment, as shown in FIG. 5, the brightness adjustment curve is constructed based on the average natural brightness value in step S40, and the brightness adjustment curve is used to adjust the brightness of the face image to obtain the target image, which specifically includes the following steps:

S41: Obtain a brightness adjustment value drtb based on the average natural brightness value and a preset optimal brightness value.

The optimal brightness value refers to a brightness value that is most suitable as a reference basis for brightness adjustment. For example, a preset natural brightness range is 90-190, and an intermediate value 140 can be taken as an optimal brightness value. The brightness adjustment value drtb refers to an adjustment value for determining an intermediate point of curve fitting, and is a value obtained by subtracting an average natural brightness value from an optimal brightness value. For example, the preset natural brightness range is 90-190. When the average natural brightness value is less than 90, drtb is a positive value. For example, when the average brightness value is 80, drtb = 140-80 = 60; when the average natural brightness value is greater than At 190, drtb is a negative value. For example, when the average natural brightness value is 200, drtb = 140-200 = -60.

S42: Perform smooth curve fitting on the three coordinate points (0, 0), (127, 127 + drtb), and (255, 255) to obtain a brightness adjustment curve.

Smooth curve fitting refers to selecting an appropriate curve type to fit the brightness value of a pixel, and using a fitting function formula to determine the relationship between the brightness value of the pixel before adjustment and the brightness value of the pixel after adjustment. The brightness adjustment curve refers to a curve used to adjust the brightness of an image obtained by smooth curve fitting. The brightness adjustment curve can reflect the relationship between the brightness value of the pixel point before adjustment and the brightness value of the pixel value after adjustment.

In this embodiment, the server may use a uniform B-spline, a quasi-uniform B-spline, a segmented Bezier, or a non-uniform B-spline curve to perform smooth curve fitting. For example, a plane coordinate system is established in a face image. In the plane coordinate system, the abscissa and ordinate values are 0-255, the preset natural brightness range is 90-190, and the average natural brightness value is 80. The brightness value is not within the preset natural brightness range, so brightness adjustment is required. Since the average natural brightness value is 80 and the optimal brightness value is 140, the calculated brightness adjustment value drtb is 60, that is, the three coordinate points determined are (0, 0), (127, 187), and (255, 255), select a curve type, such as fitting the three coordinate points (0, 0), (127, 187), and (255, 255) after a uniform B-spline curve to obtain a smooth curve. The smooth curve is the brightness adjustment curve.

S43: Adjust the brightness value of each pixel in the face image according to the brightness adjustment curve to obtain a target image.

Adjusting the brightness value of each pixel point in the face image according to the brightness adjustment curve specifically includes: finding the adjusted target brightness value corresponding to each pixel point in the face image on the brightness adjustment curve, and judging the brightness value of the pixel point and the target Whether the brightness value is the same. If the brightness value of the pixel is different from the target brightness value, adjust the brightness value of the pixel to the same brightness value as the target brightness value.

In step S41 to step S43, based on the average natural brightness value and the preset optimal brightness value, a brightness adjustment value drtb is obtained, and three (0, 0), (127, 127 + drtb), and (255, 255) are obtained. The coordinate points are subjected to smooth curve fitting to obtain a brightness adjustment curve; the brightness value of each pixel in the face image is adjusted according to the brightness adjustment curve to obtain a target image, and the brightness of the face brightness can be quickly adjusted to reduce the adjusted brightness. The distortion of the brightness of the target image ensures the layered sense of the brightness of the target image.

In an embodiment, step S42, performing smooth curve fitting on the three coordinate points (0, 0), (127, 127 + drtb), and (255, 255) to obtain a brightness adjustment curve, which specifically includes:

A non-uniform B-spline curve is used to fit the three coordinate points (0, 0), (127, 127 + drtb), and (255, 255) to a smooth curve fitting to obtain the brightness adjustment curve;

Among them, the mathematical expression of non-uniform B-spline curve is

Wherein, P (K) is a position vector on the _curve, N _{i, m} (K) is the m-th spline base function, R _i is a weighting factor, P _i of the control points, K is knot vector.

Non-uniform B-spline (NURBS, Non Uniform, Rational B-spline) is a versatile spline curve that can be used not only to describe free curves and surfaces, but also to provide a precise representation of conic curves and surfaces. The unified expression of various geometries, the mathematical expression is:

In the formula, P (K) is a position vector on the _curve, N _{i, m} (K) is the m-th spline base function, R _i is a weighting factor, P _i of the control points, K is knot vector.

The m-th spline basis function is defined by a recursive formula:

The interval interval can be any value, so different mixed function shapes can be obtained in different intervals, which provides greater freedom for freely controlling the shape of the curve. The main difference between uniform and non-uniform is the value of the node vector. If the node vector is set appropriately, an open uniform spline can be generated, which is the intersection of uniform and non-uniform. The node values of the open uniform spline at both ends are repeated d times, and the node spacing is uniform. The position vector is a vector with the origin as the starting point and the point as the end. For example, at any point C (127, 127 + drtb) in the coordinate plane, the vector OC is called the position vector of point C.

In this embodiment, a non-uniform B-spline curve can be used to fit the three coordinate points (0, 0), (127, 127 + drtb), and (255, 255) to obtain a smooth curve as a brightness adjustment curve. In order to adjust the brightness value of each pixel through the brightness adjustment curve, the brightness of the adjusted image is more balanced, and the adjusted target image has serious brightness distortion, and the brightness of the adjusted target image is guaranteed. The sense of gradation, and the brightness adjustment curve enables fast brightness adjustment of the face image.

It should be understood that the size of the sequence numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

In an embodiment, FIG. 6 shows a principle block diagram of a face brightness adjustment device corresponding to the face brightness adjustment method in the above embodiment. As shown in FIG. 6, the face brightness adjustment device includes a target detection area acquisition module 10, a pixel brightness acquisition module 20, an area average brightness acquisition module 30, and a brightness adjustment module 40. Each functional module is described in detail below:

The target detection area obtaining module 10 is configured to obtain a face image, and obtain a target detection area based on the face image.

The pixel brightness obtaining module 20 is configured to obtain a brightness value of each pixel in the target detection area.

The area average brightness obtaining module 30 is configured to process the brightness value by using an image natural brightness algorithm to obtain an average natural brightness value of the target detection area.

A brightness adjustment module 40 configured to construct a brightness adjustment curve based on the average natural brightness value if the average natural brightness value is not within a preset natural brightness range, and use the brightness adjustment curve to brightness the face image Adjust to get the target image.

Specifically, the target detection area acquisition module 10 includes a facial feature point data acquisition unit 11, a facial features area extraction unit 12, and a target detection area determination unit 13.

The facial feature point data acquisition unit 11 is configured to detect facial feature point data of the face image by using a facial feature point detection algorithm, where the facial feature point data includes coordinate data and local feature values.

The facial features region extraction unit 12 is configured to extract facial features in the face image according to the coordinate data and the local feature value.

The target detection area determining unit 13 is configured to obtain the target detection area according to the coordinate data of the nose tip and the outer corners of the eyes in the facial features area.

Specifically, the target detection area determination unit 13 includes a horizontal straight line determination sub-unit 131 and a rectangular area determination sub-unit 132.

The horizontal straight line determining subunit 131 is configured to make a horizontal straight line based on the coordinate data of the nose tip.

The rectangular region determining sub-unit 132 is configured to construct a rectangular region based on the coordinate data of the outer corners of the eyes and the horizontal straight line, and determine the rectangular region as the target detection region.

Specifically, the formula of the image natural brightness algorithm is:

Among them, N is the number of pixels in the target detection area, Lum _ave is the average natural brightness value of the target detection area, Lum (x, y) is the brightness value corresponding to the pixel points (x, y) of the target detection area, and δ is a constant .

Specifically, the brightness adjustment module 40 includes a brightness adjustment value acquisition unit 41, a brightness adjustment curve acquisition unit 42, and a target image acquisition unit 43.

The brightness adjustment value obtaining unit 41 is configured to obtain a brightness adjustment value drtb based on the average natural brightness value and a preset optimal brightness value.

The brightness adjustment curve obtaining unit 42 is configured to perform smooth curve fitting on three coordinate points (0, 0), (127, 127 + drtb), and (255, 255) to obtain a brightness adjustment curve.

A target image obtaining unit 43 is configured to adjust the brightness value of each pixel point in the face image according to the brightness adjustment curve to obtain a target image.

Specifically, the brightness adjustment curve obtaining unit 42 is configured to use a non-uniform B-spline curve to perform smooth curve fitting on the three coordinate points (0, 0), (127, 127 + drtb), and (255, 255). Get the brightness adjustment curve;

Among them, the mathematical expression of non-uniform B-spline curve is

For specific limitations on the face brightness adjustment device, refer to the foregoing limitation on the face brightness adjustment method, and details are not described herein again. Each module in the above-mentioned face brightness adjustment device may be implemented in whole or in part by software, hardware, and a combination thereof. The above-mentioned modules may be embedded in the hardware in or independent of the processor in the computer device, or may be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. 7. The computer device includes a processor, a memory, a network interface, and a database connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer-readable instructions, and a database. The internal memory provides an environment for the operation of the operating system and computer-readable instructions in a non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer-readable instructions are executed by a processor to implement a method for adjusting human face brightness.

In one embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions stored on the memory and executable on the processor. When the processor executes the computer-readable instructions, the processor implements the following steps: obtaining a person A face image, obtaining a target detection area based on the face image; obtaining a brightness value of each pixel in the target detection area; processing the brightness value using an image natural brightness algorithm to obtain an average of the target detection area Natural brightness value; if the average natural brightness value is not within a preset natural brightness range, a brightness adjustment curve is constructed based on the average natural brightness value, and the brightness adjustment curve is used to adjust the brightness of the face image to obtain The target image.

In an embodiment, when the processor executes the computer-readable instructions, the following steps are further implemented: using a facial feature point detection algorithm to detect facial feature point data of the face image, the facial feature point data including coordinate data and local features Value; extracting the facial features region in the face image according to the coordinate data and the local feature value; obtaining the target detection region according to the coordinate data of the nose tip and the corners of the eyes outside the eyes in the facial feature region.

In an embodiment, when the processor executes the computer-readable instructions, the processor further implements the following steps: making a horizontal straight line based on the coordinate data of the tip of the nose; and constructing a rectangular area based on the coordinate data of the outer corners of the eyes and the horizontal straight line, Determining the rectangular area as the target detection area.

In one embodiment, the formula of the image natural brightness algorithm is:

In one embodiment, when the processor executes the computer-readable instructions, the following steps are further implemented: obtaining a brightness adjustment value drtb based on the average natural brightness value and a preset optimal brightness value; for (0,0), (127 (127 + drtb) and (255, 255) coordinate smooth curve fitting to obtain a brightness adjustment curve; adjust the brightness value of each pixel in the face image according to the brightness adjustment curve to obtain a target image .

In one embodiment, when the processor executes the computer-readable instructions, the following steps are further implemented: using non-uniform B-spline curve pairs (0, 0), (127, 127 + drtb), and (255, 255) three coordinate points The smoothing curve fitting is adopted to obtain the brightness adjustment curve. Among them, the mathematical expression of the non-uniform B-spline curve is

In one embodiment, one or more non-volatile readable storage media storing computer-readable instructions are provided, and when the computer-readable instructions are executed by one or more processors, the one or more The processors execute the following steps: acquiring a face image, obtaining a target detection area based on the face image; obtaining a brightness value of each pixel in the target detection area; and processing the brightness value using an image natural brightness algorithm Obtaining an average natural brightness value of the target detection area; if the average natural brightness value is not within a preset natural brightness range, constructing a brightness adjustment curve based on the average natural brightness value, and using the brightness adjustment curve The face image is adjusted for brightness to obtain a target image.

In one embodiment, when the computer-readable instructions are executed by one or more processors, the one or more processors are caused to perform the following steps: using a facial feature point detection algorithm to detect a face of the facial image Feature point data, the facial feature point data including coordinate data and local feature values; extracting facial features in the face image according to the coordinate data and the local feature values; and according to the nose tip and The coordinate data of the corners of the eyes outside the eyes to obtain the target detection area.

In one embodiment, when the computer-readable instructions are executed by one or more processors, the one or more processors are caused to perform the following steps: making a horizontal straight line based on the coordinate data of the nose tip; based on the The coordinate data of the outer corners of the eyes and the horizontal straight line construct a rectangular area, and the rectangular area is determined as the target detection area.

In one embodiment, the formula of the image natural brightness algorithm is:

In one embodiment, when the computer-readable instructions are executed by one or more processors, the one or more processors are caused to perform the following steps: based on the average natural brightness value and a preset optimal brightness value To obtain a brightness adjustment value drtb; perform smooth curve fitting on the three coordinate points (0, 0), (127, 127 + drtb), and (255, 255) to obtain a brightness adjustment curve; The brightness value of each pixel in the face image is described to obtain a target image.

In one embodiment, when the computer-readable instructions are executed by one or more processors, the one or more processors are caused to perform the following steps: using a non-uniform B-spline curve pair (0, 0), ( 127, 127 + drtb) and (255, 255) are used to perform smooth curve fitting to obtain the brightness adjustment curve; where the mathematical expression of the non-uniform B-spline curve is

A person of ordinary skill in the art can understand that all or part of the processes in the methods of the foregoing embodiments can be implemented by using computer-readable instructions to instruct related hardware. The computer-readable instructions can be stored in a non-volatile computer. In the readable storage medium, the computer-readable instructions, when executed, may include the processes of the embodiments of the methods described above. Wherein, any reference to the storage, storage, database, or other media used in the embodiments provided in this application may include non-volatile and / or volatile storage. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous chain Synchlink DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

Those skilled in the art can clearly understand that, for the convenience and brevity of the description, only the above-mentioned division of functional units and modules is used as an example. In practical applications, the above functions can be assigned by different functional units, Module completion, that is, dividing the internal structure of the device into different functional units or modules to complete all or part of the functions described above.

The above-mentioned embodiments are only used to describe the technical solution of the present application, but not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still implement the foregoing implementations. The technical solutions described in the examples are modified, or some of the technical features are equivalently replaced; and these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the application, and should be included in Within the scope of this application.

Claims

A method for adjusting the brightness of a human face, comprising:

Acquiring a face image, and acquiring a target detection area based on the face image;

Obtaining a brightness value of each pixel in the target detection area;

Processing the brightness value by using an image natural brightness algorithm to obtain an average natural brightness value of the target detection area;

If the average natural brightness value is not within a preset natural brightness range, a brightness adjustment curve is constructed based on the average natural brightness value, and the brightness adjustment curve is used to perform brightness adjustment on the face image to obtain a target image.
The method for adjusting the brightness of a face according to claim 1, wherein the acquiring a target detection area based on the face image comprises:

Using a facial feature point detection algorithm to detect facial feature point data of the face image, where the facial feature point data includes coordinate data and local feature values;

Extracting facial features in the face image according to the coordinate data and the local feature value;

The target detection area is acquired according to the coordinate data of the tip of the nose and the corners of the eyes outside the eyes in the facial features area.
The method for adjusting the brightness of a face according to claim 2, wherein the acquiring the target detection area according to the coordinate data of the nose tip and the outer corners of the eyes in the facial features area comprises:

Making a horizontal straight line based on the coordinate data of the nose tip;

A rectangular area is constructed based on the coordinate data of the outer corners of the eyes and the horizontal straight line, and the rectangular area is determined as the target detection area.
The method for adjusting the brightness of a face according to claim 1, wherein a formula of the natural brightness algorithm of the image is:
Among them, N is the number of pixels in the target detection area, Lum ave is the average natural brightness value of the target detection area, Lum (x, y) is the brightness value corresponding to the pixel points (x, y) of the target detection area, and δ is a constant .
The method for adjusting the brightness of a face according to claim 1, wherein the brightness adjustment curve is constructed based on the average natural brightness value, and the brightness adjustment curve is used to adjust the brightness of the face image to obtain a target Images, including:

Obtaining a brightness adjustment value drtb based on the average natural brightness value and a preset optimal brightness value;

Perform smooth curve fitting on the three coordinate points (0,0), (127,127 + drtb) and (255,255) to obtain the brightness adjustment curve;

Adjusting the brightness value of each pixel point in the face image according to the brightness adjustment curve to obtain a target image.
The method for adjusting the brightness of a face according to claim 5, wherein the smoothing curve fitting is performed on three coordinate points (0,0), (127,127 + drtb), and (255,255) to obtain Brightness adjustment curve, including:

A non-uniform B-spline curve is used to fit the three coordinate points (0, 0), (127, 127 + drtb), and (255, 255) to a smooth curve fitting to obtain the brightness adjustment curve;

Among them, the mathematical expression of non-uniform B-spline curve is
Wherein, P (K) is a position vector on the curve, N i, m (K) is the m-th spline base function, R i is a weighting factor, P i of the control points, K is knot vector.
A human face brightness adjustment device, comprising:

A target detection area acquisition module, configured to obtain a face image, and obtain a target detection area based on the face image;

A pixel brightness obtaining module, configured to obtain a brightness value of each pixel in the target detection area;

An area average brightness acquisition module, configured to process the brightness value by using an image natural brightness algorithm to obtain an average natural brightness value of the target detection area;

A brightness adjustment module configured to construct a brightness adjustment curve based on the average natural brightness value if the average natural brightness value is not within a preset natural brightness range, and use the brightness adjustment curve to perform brightness adjustment on the face image To get the target image.
The device for adjusting the brightness of a face according to claim 7, wherein the target detection area acquisition module comprises:

A facial feature point data acquisition unit, configured to detect facial feature point data of the face image by using a facial feature point detection algorithm, where the facial feature point data includes coordinate data and local feature values;

Facial features region extracting unit, configured to extract facial features in the face image according to the coordinate data and the local feature value;

The target detection area determining unit is configured to acquire the target detection area according to the coordinate data of the nose tip and the outer corners of the eyes in the facial features area.
A computer device includes a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, and is characterized in that the processor implements the computer-readable instructions as follows step:

Acquiring a face image, and acquiring a target detection area based on the face image;

Obtaining a brightness value of each pixel in the target detection area;

Processing the brightness value by using an image natural brightness algorithm to obtain an average natural brightness value of the target detection area;

If the average natural brightness value is not within a preset natural brightness range, a brightness adjustment curve is constructed based on the average natural brightness value, and the brightness adjustment curve is used to perform brightness adjustment on the face image to obtain a target image.
The computer device according to claim 9, wherein the acquiring a target detection area based on the face image comprises:

Using a facial feature point detection algorithm to detect facial feature point data of the face image, where the facial feature point data includes coordinate data and local feature values;

Extracting facial features in the face image according to the coordinate data and the local feature value;

The target detection area is acquired according to the coordinate data of the tip of the nose and the corners of the eyes outside the eyes in the facial features area.
The computer device according to claim 10, wherein the acquiring the target detection area according to the coordinate data of the nose tip and the outer corners of the eyes in the facial features area comprises:

Making a horizontal straight line based on the coordinate data of the nose tip;

A rectangular area is constructed based on the coordinate data of the outer corners of the eyes and the horizontal straight line, and the rectangular area is determined as the target detection area.
The computer device according to claim 9, wherein the formula of the image natural brightness algorithm is:
Among them, N is the number of pixels in the target detection area, Lum ave is the average natural brightness value of the target detection area, Lum (x, y) is the brightness value corresponding to the pixel points (x, y) of the target detection area, and δ is a constant .
The computer device according to claim 9, wherein the constructing a brightness adjustment curve based on the average natural brightness value, and using the brightness adjustment curve to perform brightness adjustment on the face image to obtain a target image, comprises: :

Obtaining a brightness adjustment value drtb based on the average natural brightness value and a preset optimal brightness value;

Perform smooth curve fitting on the three coordinate points (0,0), (127,127 + drtb) and (255,255) to obtain the brightness adjustment curve;

Adjusting the brightness value of each pixel point in the face image according to the brightness adjustment curve to obtain a target image.
The computer device according to claim 13, wherein the smoothing curve fitting is performed on three coordinate points (0, 0), (127, 127 + drtb), and (255, 255) to obtain a brightness adjustment curve ,include:

A non-uniform B-spline curve is used to fit the three coordinate points (0, 0), (127, 127 + drtb), and (255, 255) to a smooth curve fitting to obtain the brightness adjustment curve;

Among them, the mathematical expression of non-uniform B-spline curve is
Wherein, P (K) is a position vector on the curve, N i, m (K) is the m-th spline base function, R i is a weighting factor, P i of the control points, K is knot vector.
One or more non-volatile readable storage media storing computer readable instructions, characterized in that when the computer readable instructions are executed by one or more processors, the one or more processors are caused to execute The following steps:

Acquiring a face image, and acquiring a target detection area based on the face image;

Obtaining a brightness value of each pixel in the target detection area;

Processing the brightness value by using an image natural brightness algorithm to obtain an average natural brightness value of the target detection area;

If the average natural brightness value is not within a preset natural brightness range, a brightness adjustment curve is constructed based on the average natural brightness value, and the brightness adjustment curve is used to perform brightness adjustment on the face image to obtain a target image.
The non-volatile readable storage medium according to claim 15, wherein the acquiring a target detection area based on the face image comprises:

Using a facial feature point detection algorithm to detect facial feature point data of the face image, where the facial feature point data includes coordinate data and local feature values;

Extracting facial features in the face image according to the coordinate data and the local feature value;

The target detection area is acquired according to the coordinate data of the tip of the nose and the corners of the eyes outside the eyes in the facial features area.
The non-volatile readable storage medium according to claim 16, wherein the obtaining the target detection area according to the coordinate data of the nose tip and the outer corners of the eyes in the facial features area comprises:

Making a horizontal straight line based on the coordinate data of the nose tip;

A rectangular area is constructed based on the coordinate data of the outer corners of the eyes and the horizontal straight line, and the rectangular area is determined as the target detection area.
The non-volatile readable storage medium according to claim 15, wherein a formula of the image natural brightness algorithm is:
Among them, N is the number of pixels in the target detection area, Lum ave is the average natural brightness value of the target detection area, Lum (x, y) is the brightness value corresponding to the pixel points (x, y) of the target detection area, and δ is a constant .
The non-volatile readable storage medium according to claim 15, wherein the brightness adjustment curve is constructed based on the average natural brightness value, and the brightness adjustment curve is used to perform brightness adjustment on the face image To obtain the target image, including:

Obtaining a brightness adjustment value drtb based on the average natural brightness value and a preset optimal brightness value;

Perform smooth curve fitting on the three coordinate points (0,0), (127,127 + drtb) and (255,255) to obtain the brightness adjustment curve;

Adjusting the brightness value of each pixel point in the face image according to the brightness adjustment curve to obtain a target image.
The non-volatile readable storage medium according to claim 19, wherein the smoothing curve fitting is performed on three coordinate points (0, 0), (127, 127 + drtb), and (255, 255) Combine to get the brightness adjustment curve, including:

A non-uniform B-spline curve is used to fit the three coordinate points (0, 0), (127, 127 + drtb), and (255, 255) to a smooth curve fitting to obtain the brightness adjustment curve;

Among them, the mathematical expression of non-uniform B-spline curve is
Wherein, P (K) is a position vector on the curve, N i, m (K) is the m-th spline base function, R i is a weighting factor, P i of the control points, K is knot vector.