WO2020210954A1 - 用于校准图像的方法、装置和电子设备 - Google Patents

用于校准图像的方法、装置和电子设备 Download PDF

Info

Publication number
WO2020210954A1
WO2020210954A1 PCT/CN2019/082739 CN2019082739W WO2020210954A1 WO 2020210954 A1 WO2020210954 A1 WO 2020210954A1 CN 2019082739 W CN2019082739 W CN 2019082739W WO 2020210954 A1 WO2020210954 A1 WO 2020210954A1
Authority
WO
WIPO (PCT)
Prior art keywords
calibration
iterative learning
image
original image
calibration parameter
Prior art date
Application number
PCT/CN2019/082739
Other languages
English (en)
French (fr)
Inventor
程雷刚
Original Assignee
深圳市汇顶科技股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳市汇顶科技股份有限公司 filed Critical 深圳市汇顶科技股份有限公司
Priority to CN201980000591.3A priority Critical patent/CN110192201B/zh
Priority to PCT/CN2019/082739 priority patent/WO2020210954A1/zh
Publication of WO2020210954A1 publication Critical patent/WO2020210954A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/30—Noise filtering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/12—Fingerprints or palmprints
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/12—Fingerprints or palmprints
    • G06V40/1347—Preprocessing; Feature extraction

Definitions

  • the embodiments of the present application relate to the field of electronics, and more specifically, to methods, devices, and electronic equipment for calibrating images.
  • the signal carrier carrying fingerprint information will be disturbed when it penetrates the obstacle.
  • the quality of the fingerprint image will be further affected.
  • a method, device and electronic equipment for calibrating images are provided, which can effectively calibrate images.
  • a method for calibrating an image including:
  • the calibration parameters obtained after iterative learning of the i+1th original image in the n original images are based on the calibration parameters obtained after iterative learning of the i-th original image in the n original images and the calibration parameters
  • the parameter determined by the pixel value of the i+1th original image, the first calibration parameter is the calibration parameter of the nth original image in the n original images after iterative learning, and n is the first preset value And it is a positive integer, 1 ⁇ i ⁇ n;
  • the target image is calibrated based on the first calibration parameter.
  • an apparatus for calibrating an image including:
  • a determining unit configured to determine the first calibration parameter through iterative learning of n original images
  • the calibration parameters obtained after iterative learning of the i+1th original image in the n original images are based on the calibration parameters obtained after iterative learning of the i-th original image in the n original images and the calibration parameters
  • the parameter determined by the pixel value of the i+1th original image, the first calibration parameter is the calibration parameter of the nth original image in the n original images after iterative learning, and n is the first preset value And it is a positive integer, 1 ⁇ i ⁇ n;
  • the calibration unit is configured to calibrate the target image based on the first calibration parameter.
  • an electronic device including:
  • a fingerprint module is arranged on the surface or inside of the electronic device;
  • the fingerprint module is electrically connected to the device
  • the fingerprint module is configured to receive a fingerprint detection signal returned by reflection or scattering of a human finger above the display screen, and the fingerprint detection signal carries fingerprint information of the finger.
  • the method, device and electronic device for calibrating images in this application are embodiments. They can not rely on or rely on a small amount of prior information, and update the fingerprint image for calibration by continuously learning the original image in the user application process.
  • the calibration parameters can not only simplify the operation process, but also effectively improve the calibration accuracy.
  • the method, device and electronic equipment can filter out the image used to calibrate the fingerprint image from the original image in the user application process (for example, the fingerprint identification process), and the technical solution of the present application can be applied to relatively harsh Under the scenario, it has a wider application scenario and higher fingerprint recognition performance.
  • the electronic device selects original images without abnormalities in the collected original images for iterative learning, or cuts out abnormal parts in the original images, and only performs iterative learning on some areas, which can effectively avoid the occurrence of The effect of the abnormal part of the abnormal original image on the iterative learning, thereby improving the accuracy of the calibration parameters.
  • Fig. 1 is a schematic plan view of an electronic device with a collection area in a display screen to which the present application can be applied.
  • Fig. 2 is a schematic partial cross-sectional view of the electronic device shown in Fig. 1 along A'-A'.
  • Fig. 3 is another schematic structural diagram of an electronic device according to an embodiment of the present application.
  • Fig. 4 is a schematic flowchart of a method for calibrating an image according to an embodiment of the present application.
  • FIG. 5 is another schematic flowchart of a method for calibrating an image according to an embodiment of the present application.
  • Fig. 6 is a schematic block diagram of an apparatus for calibrating an image according to an embodiment of the present application.
  • FIG. 7 is another schematic block diagram of the apparatus for calibrating an image according to an embodiment of the present application.
  • Fig. 8 is another schematic block diagram of an electronic device according to an embodiment of the present application.
  • portable or mobile computing devices such as smartphones, notebook computers, tablet computers, and gaming devices, as well as other electronic devices such as electronic databases, automobiles, and bank automated teller machines (ATM).
  • ATM bank automated teller machines
  • the embodiments of the present application are not limited thereto.
  • biometric technology includes but is not limited to fingerprint recognition, palmprint recognition, iris recognition, face recognition, and living body recognition.
  • biometric identification technology may be a capacitive, optical, ultrasonic or other biometric identification technology.
  • optical fingerprint recognition technology is an example to describe the application scenarios of the embodiments of the present application.
  • Optical fingerprint recognition technology can be used for under-screen fingerprint recognition technology and in-screen fingerprint recognition technology.
  • the under-screen fingerprint recognition technology refers to the installation of the fingerprint recognition module below the display screen, so as to realize the fingerprint recognition operation in the display area of the display screen. There is no need to set a fingerprint collection area on the front of the electronic device except for the display area.
  • the fingerprint recognition module uses light returned from the top surface of the display assembly of the electronic device or an externally added invisible LED light source (such as an infrared LED light source) to perform fingerprint sensing and other sensing operations.
  • This returned light carries information about objects (for example, fingers) in contact with the top surface of the display assembly, and the fingerprint recognition module located below the display assembly collects and detects this returned light to realize fingerprint recognition under the screen.
  • the fingerprint recognition module can be designed to achieve desired optical imaging by appropriately configuring optical elements for collecting and detecting the returned light.
  • in-display fingerprint recognition technology refers to the installation of fingerprint recognition modules or part of fingerprint recognition modules inside the display screen, so as to realize fingerprint recognition operations in the display area of the display screen without the need for electronic
  • the fingerprint collection area is set on the front of the device except the display area.
  • FIG. 1 and 2 show schematic diagrams of an electronic device 100 to which under-screen fingerprint recognition technology can be applied.
  • FIG. 1 is a front schematic diagram of the electronic device 100
  • FIG. 2 is a partial cross-sectional structure diagram of the electronic device 100 shown in FIG.
  • the electronic device 100 may include a display screen 120 and a fingerprint recognition module 140.
  • the display screen 120 may be a self-luminous display, which uses a self-luminous display unit as display pixels.
  • the display screen 120 may be an Organic Light-Emitting Diode (OLED) display screen or a Micro-LED (Micro-LED) display screen.
  • the display screen 120 may also be a liquid crystal display (LCD) or other passive light-emitting display, which is not limited in the embodiment of the present application.
  • the display screen 120 may also be specifically a touch display screen, which can not only perform screen display, but also detect a user's touch or pressing operation, thereby providing a user with a human-computer interaction interface.
  • the electronic device 100 may include a touch sensor, and the touch sensor may specifically be a touch panel (TP), which may be provided on the surface of the display screen 120, or may be partially integrated or The whole is integrated into the display screen 120 to form the touch display screen.
  • TP touch panel
  • the fingerprint recognition module 140 may be an optical fingerprint recognition module, such as an optical fingerprint sensor.
  • the fingerprint identification module 140 may include a fingerprint sensor chip with an optical sensing array (hereinafter also referred to as an optical fingerprint sensor).
  • the optical sensing array includes multiple optical sensing units, and each optical sensing unit may specifically include a photodetector or a photoelectric sensor.
  • the fingerprint identification module 140 may include a photodetector array (or called a photodetector array, a photodetector array), which includes a plurality of photodetectors distributed in an array.
  • the fingerprint recognition module 140 may be arranged in a partial area below the display screen 120, so that the fingerprint collection area (or detection area) 130 of the fingerprint recognition module 140 is at least partially located on the display screen 120. ⁇ display area 102.
  • the fingerprint identification module 140 can also be arranged in other positions, such as the side of the display screen 120 or the non-transparent area of the edge of the electronic device 100.
  • the optical signal of at least part of the display area of the display screen 120 can be guided to the fingerprint recognition module 140 through the optical path design, so that the fingerprint collection area 130 is actually located in the display area of the display screen 120 .
  • the fingerprint recognition module 140 may include only one fingerprint sensor chip. At this time, the fingerprint collection area 130 of the fingerprint recognition module 140 has a small area and a fixed position. Therefore, the user needs to input fingerprints. Press the finger to a specific position of the fingerprint collection area 130, otherwise the fingerprint recognition module 140 may not be able to collect the fingerprint image, resulting in poor user experience.
  • the fingerprint identification module 140 may specifically include a plurality of fingerprint sensor chips; the plurality of fingerprint sensor chips may be arranged side by side under the display screen 120 in a splicing manner, and the plurality of fingerprint sensor chips The sensing areas of the two fingerprint sensor chips together constitute the fingerprint collection area 130 of the fingerprint identification module 140.
  • the fingerprint collection area 130 of the fingerprint identification module 140 may include multiple sub-areas, and each sub-area corresponds to the sensing area of one of the fingerprint sensor chips, so that the fingerprint of the optical fingerprint module 130 is collected
  • the area 130 can be extended to the main area of the lower half of the display screen, that is, to the area where the finger is habitually pressed, so as to realize the blind fingerprint input operation.
  • the fingerprint detection area 130 can also be extended to half of the display area or even the entire display area, thereby realizing half-screen or full-screen fingerprint detection.
  • the multiple fingerprint sensor chips may be individually packaged fingerprint sensor chips, or multiple chips (Die) packaged in the same chip package.
  • the multiple fingerprint sensor chips can also be fabricated on different regions of the same chip (Die) through a semiconductor process.
  • the area or light sensing range of the optical sensing array of the fingerprint identification module 140 corresponds to the fingerprint collection area 130 of the fingerprint identification module 140.
  • the fingerprint collection area 130 of the fingerprint recognition module 140 may be equal to or not equal to the area or the light sensing range of the optical sensing array of the fingerprint recognition module 140, which is not specifically limited in the embodiment of the present application.
  • the fingerprint collection area 130 of the fingerprint identification module 140 can be designed to be substantially the same as the area of the sensing array of the fingerprint identification module 140.
  • the area of the fingerprint collection area 130 of the fingerprint recognition module 140 can be larger than the area of the fingerprint recognition module 140 sensing array through the design of the light path of convergent light or the design of the light path of reflected light.
  • the optical path design of the fingerprint identification module 140 is exemplified below.
  • the optical collimator may be specifically a collimator layer made on a semiconductor silicon wafer. , It has a plurality of collimating units or micro-holes.
  • the collimating unit may be specifically a small hole.
  • the reflected light reflected from the finger the light that is perpendicularly incident on the collimating unit can pass through and be
  • the fingerprint sensor chip receives, and the light whose incident angle is too large is attenuated by multiple reflections inside the collimating unit. Therefore, each fingerprint sensor chip can basically only receive the reflected light reflected by the fingerprint lines directly above it. It can effectively improve the image resolution, thereby improving the fingerprint recognition effect.
  • a collimating unit may be configured for one optical sensor unit in the optical sensor array of each fingerprint sensor chip, and the collimating unit may be attached to the corresponding optical sensor.
  • the multiple optical sensing units can also share one collimating unit, that is, the one collimating unit has an aperture large enough to cover the multiple optical sensing units. Since one collimating unit can correspond to multiple optical sensing units, the correspondence between the spatial period of the display screen 120 and the spatial period of the fingerprint sensor chip is destroyed.
  • the spatial structure of the light-emitting display array of the display screen 120 and the fingerprint sensor chip are The spatial structure of the optical sensor array is similar, and it can also effectively prevent the fingerprint identification module 140 from using the light signal passing through the display screen 120 to perform fingerprint imaging to generate moiré fringes, which effectively improves the fingerprint identification effect of the fingerprint identification module 140.
  • the optical lens may include an optical lens (Lens) layer, which has one or more lens units, such as one or more aspheric lenses.
  • the lens group is used to converge the reflected light reflected from the finger to the sensing array of the fingerprint sensor chip below it, so that the sensing array can perform imaging based on the reflected light, thereby acquiring the fingerprint image of the finger.
  • the optical lens layer may also be formed with a pinhole in the optical path of the lens unit, and the pinhole may cooperate with the optical lens layer to expand the field of view of the fingerprint recognition module 140 to improve the fingerprint recognition module 140 Fingerprint imaging effect.
  • each fingerprint sensor chip may be configured with an optical lens for fingerprint imaging, or multiple fingerprint sensor chips may be configured with an optical lens to achieve light convergence and fingerprint imaging.
  • the fingerprint sensor chip can also be equipped with two or more optical lenses to cooperate with the two sensors. The array or multiple sensing arrays perform optical imaging, thereby reducing the imaging distance and enhancing the imaging effect.
  • the micro-lens layer may have a micro-lens array formed by a plurality of micro-lenses, which may be obtained through a semiconductor growth process or other The process is formed above the sensing array of the fingerprint sensor chip, and each microlens can correspond to one of the sensing units of the sensing array.
  • Other optical film layers may be formed between the microlens layer and the sensing unit, such as a dielectric layer or a passivation layer. More specifically, the microlens layer and the sensing unit may also include micropores.
  • the light blocking layer wherein the micro holes are formed between the corresponding micro lens and the sensing unit, the light blocking layer can block the optical interference between the adjacent micro lens and the sensing unit, and allow light to pass through the micro lens
  • the lens is converged into the microhole and is transmitted to the sensing unit corresponding to the microlens through the microhole to perform optical fingerprint imaging.
  • a microlens layer can be further provided under the collimator layer or the optical lens layer.
  • the collimator layer or the optical lens layer is used in combination with the micro lens layer, its specific laminated structure or optical path may need to be adjusted according to actual needs.
  • the fingerprint identification module 140 can be used to collect user fingerprint information (such as fingerprint image information).
  • the display screen 120 can adopt a display screen with a self-luminous display unit, such as an organic light-emitting diode (Organic Light-Emitting Diode, OLED) display or a micro-LED (Micro-LED) display Screen.
  • the fingerprint recognition module 140 can use the display unit (ie, the OLED light source) of the OLED display screen located in the fingerprint collection area 130 as the excitation light source for optical fingerprint detection.
  • the display screen 120 When a finger touches, presses, or approaches (for ease of description, collectively referred to as pressing in this application) in the fingerprint collection area 130, the display screen 120 emits a beam of light to the finger above the fingerprint collection area 130. The surface is reflected to form reflected light or is scattered inside the finger to form scattered light. In related patent applications, for ease of description, the above-mentioned reflected light and scattered light are collectively referred to as reflected light. Because the ridge and valley of the fingerprint have different light reflection capabilities, the reflected light from the fingerprint ridge and the fingerprint ridge have different light intensities. After the reflected light passes through the display screen 120, it is affected by the fingerprint.
  • the fingerprint sensor chip in the identification module 140 receives and converts it into a corresponding electrical signal, that is, a fingerprint detection signal; fingerprint image data can be obtained based on the fingerprint detection signal, and fingerprint matching verification can be further performed, so that the electronic The device 100 implements an optical fingerprint recognition function.
  • the electronic device 100 adopting the above structure does not need to reserve a special space on the front of the fingerprint button (such as the Home button), so a full screen solution can be adopted. Therefore, the display area 102 of the display screen 120 can be substantially extended to the entire front surface of the electronic device 100.
  • the fingerprint identification module 140 may also use a built-in light source or an external light source to provide an optical signal for fingerprint detection and identification.
  • the fingerprint identification module 140 can be applied not only to self-luminous displays such as OLED displays, but also to non-self-luminous displays, such as liquid crystal displays or other passive light-emitting displays.
  • the optical fingerprint system of the electronic device 100 may also include an excitation light source for optical fingerprint detection.
  • the light source may specifically be an infrared light source or a light source of non-visible light of a specific wavelength, which may be arranged under the backlight module of the liquid crystal display or in the edge area under the protective cover of the electronic device 100, and the fingerprint recognition module 140 may
  • the liquid crystal panel or the protective cover is arranged under the edge area and guided by the light path so that the fingerprint detection light can reach the fingerprint identification module 140; or, the fingerprint identification module 140 can also be arranged under the backlight module, and
  • the backlight module is designed to allow the fingerprint detection light to pass through the liquid crystal panel and the backlight module and reach the fingerprint recognition module 140 by opening holes or other optical designs on the film layers such as the diffusion sheet, the brightness enhancement sheet, and the reflection sheet.
  • the fingerprint identification module 140 adopts a built-in light source or
  • the electronic device 100 may further include a protective cover 110.
  • the cover 110 may be specifically a transparent cover, such as a glass cover or a sapphire cover, which is located above the display screen 120 and covers the front of the electronic device 100, and the surface of the cover 110 may also be provided with a protective layer. Therefore, in the embodiment of the present application, the so-called finger pressing the display screen 120 may actually refer to the finger pressing the cover 110 above the display 120 or covering the surface of the protective layer of the cover 110.
  • a circuit board 150 such as a flexible printed circuit (FPC) (Flexible Printed Circuit, FPC), may also be provided under the fingerprint identification module 140.
  • FPC Flexible Printed Circuit
  • the fingerprint recognition module 140 can be soldered to the circuit board 150 through pads, and realize electrical interconnection and signal transmission with other peripheral circuits or other components of the electronic device 100 through the circuit board 150.
  • the fingerprint recognition module 140 can receive the control signal of the processing unit of the electronic device 100 through the circuit board 150, and can also output the fingerprint detection signal from the fingerprint recognition module 140 to the processing unit of the electronic device 100 through the circuit board 150. Control unit, etc.
  • Fig. 3 is another schematic structural diagram of an electronic device according to an embodiment of the present application.
  • the electronic device 300 may include a display screen 220 and an optical fingerprint sensor 230.
  • the fingerprint detection signal carrying fingerprint information can penetrate the display screen 220 to reach the upper surface of the optical fingerprint sensor 230, so that the optical fingerprint sensor 230 performs fingerprint imaging based on the fingerprint detection signal, and then performs fingerprint recognition.
  • the display screen 220 may include a cover glass 221, a first adhesive layer 222, a polarizer 223, a second adhesive layer 224, and a touch panel (TP) layer 225 from top to bottom.
  • the sealing glass 226 is used for sealing.
  • the substrate glass 228, the sealing glass 226, and the display pixel layer 227 between them cooperate with the display driving circuit to realize the display function.
  • the TP layer 225 above the sealing glass 226 cooperates with the touch driving circuit to realize the touch function.
  • the TP layer 225 may be etched into various patterns.
  • the polarizer 223 is disposed on the TP layer 225 through the second adhesive layer 224, and the polarizer 223 can be used to suppress the reflection of the display screen 220 to ambient light, thereby achieving a higher display contrast.
  • the cover glass 221 is disposed on the polarizer 223 through the first adhesive layer 221 to protect the display screen 220.
  • the optical fingerprint sensor 230 is placed or attached to the bottom of the substrate glass 228, so that the under-screen optical fingerprint recognition can be realized locally or in the full screen in the display area of the display screen.
  • the fingerprint detection signal carrying fingerprint information will be interfered when passing through the display screen 220, thereby affecting the fingerprint image. Image quality.
  • the quality of the fingerprint image will be further affected.
  • This application proposes a method for calibrating an image, which can not rely on prior information or rely on a small amount of prior information.
  • the calibration parameters used to calibrate the fingerprint image can be updated, which can not only simplify The operation process can also effectively improve the calibration accuracy.
  • the technical solution of the present application can be applied to more severe scenarios, making it have a broader application scenario and higher Fingerprint recognition performance.
  • FIG. 4 is a schematic flowchart of a method 300 for calibrating an image according to an embodiment of the present application. It should be understood that the method 300 may be executed by any electronic device with image processing capabilities. For ease of understanding, the method 300 is described below by taking an electronic device as an example.
  • the method 300 may include:
  • S310 The electronic device determines the first calibration parameter through iterative learning of n original images.
  • the calibration parameters obtained after iterative learning of the i+1th original image in the n original images are based on the calibration parameters obtained after iterative learning of the i-th original image in the n original images and the calibration parameters
  • the parameter determined by the pixel value of the i+1th original image, the first calibration parameter is the calibration parameter of the nth original image in the n original images after iterative learning, and n is the first preset value And it is a positive integer, 1 ⁇ i ⁇ n;
  • S320 The electronic device calibrates the target image based on the first calibration parameter.
  • the electronic device after the electronic device acquires the first calibration parameter through iterative learning of n original images, it can calibrate subsequent original images based on the first calibration parameter. For example, suppose the target image is the n+jth original image, and j is a positive integer. When j is 1, the electronic device may directly calibrate the n+1th original image based on the first calibration parameter.
  • the electronic device can directly calibrate the n+jth original image based on the first calibration parameter, or it can be based on the first calibration parameter by comparing the n+1th to n
  • the iterative learning of part or all of the original images in the +j original images first obtains the second calibration parameter, and then calibrates the n+j+1th original image based on the second calibration parameter.
  • the abnormal original image or the abnormal part in the original image may not be used for iterative learning. Specifically, after the electronic device acquires the original image, it first determines whether the acquired original image is abnormal. If there is an abnormality, it is not used for iterative learning; or the abnormal part of the abnormal image is cut off, and the normal part is used for local Regional iterative learning; if there is no abnormality, it is used for iterative learning.
  • the electronic device may determine the first calibration parameter by iterative learning on data other than the abnormal data in the n original images. Calibration parameter; or the electronic device may determine the first calibration parameter through iterative learning of images other than abnormal images in the n original images.
  • the abnormal image may be an original image with abnormal data.
  • the abnormal data includes but is not limited to: incomplete data caused by a part of the user's finger not pressing the electronic device and abnormal data caused by strong light.
  • the electronic device screens out the image used to calibrate the fingerprint image from the original image in the user application process (for example, the fingerprint recognition process), so that the technical solution of this application can be applied to more severe scenarios. , Making it have a wider application scenario and higher fingerprint recognition performance.
  • the electronic device can effectively avoid iterative learning of abnormal original images by selecting original images without abnormalities or normal parts of abnormal images from the collected original images for iterative learning, thereby improving the calibration parameters. Accuracy.
  • the calibration parameter obtained after the i-th original image undergoes iterative learning may correspond to the first weight value, and the pixel value of the i+1-th original image corresponds to the second weight value.
  • the first weight value The sum of the second weight value is 1.
  • the second weight value gradually decreases to a constant value as the number of the original images used for iterative learning increases.
  • FIG. 4 is another schematic flowchart of a method 300 for calibrating an image according to an embodiment of the present application.
  • the method 300 may include:
  • the electronic device initializes the calibration parameters. For example, the electronic device initializes the calibration parameter to zero.
  • the electronic device collects original images and counts only the original images that are learned. Specifically, after the electronic device has learned the m-1 original images, it collects the m-th original image.
  • the electronic device determines whether the number (m-1) of original images that have undergone iterative learning is greater than a first preset value, thereby determining whether the calibration parameters obtained after the iterative learning of the m-1 original images can be used Calibrate the image. Specifically, when m-1 is greater than or equal to the first preset value, the electronic device determines that the calibration parameters obtained after iterative learning of the m-1 original images can be used to calibrate the image, when m-1 When it is less than the first preset value, it is determined that the calibration parameters obtained after the m-1 original images undergo iterative learning cannot be used to calibrate the images.
  • the electronic device determines whether the m-th original image can be calibrated based on the calibration parameters of the m-1 original image by determining whether the number of m-1 meets the preset condition.
  • subsequent operations may also be triggered by other judgment methods.
  • an algorithm can be used to estimate whether the spatial noise of each original image in the m-1 original image is less than a certain value, and then it can be determined whether the m-th original image can be performed based on the calibration parameters of the m-1 original image. calibration.
  • S441 Calibrate the m-th original image based on the calibration parameters obtained after iterative learning of the m-1th original image.
  • the electronic device determines that the calibration parameters obtained after iterative learning of the m-1 original images can be used to calibrate the original image when acquiring the original image, calibrate the calibration parameters based on the calibration parameters obtained after the m-1 original image undergoes iterative learning The mth original image.
  • iterative learning is performed on the m-th original image. Specifically, when it is determined that there is no abnormality in the m-th original image, iterative learning is performed on the m-th original image.
  • the electronic device determines whether the number of original images (m-1) used for iterative learning is greater than or equal to a third threshold, and determines the second weight value corresponding to the pixel value of the m-th original image based on the determination result, and The first weight value corresponding to the calibration parameter obtained after the m-1 original images undergoes iterative learning, and the sum of the first weight value and the second weight value is 1.
  • S470 Determine a second weight value corresponding to the pixel value of the m-th original image according to the value of m-1.
  • the second weight value corresponding to the pixel value of the m-th original image is determined according to the value of m-1 .
  • the electronic device determines that the number (m-1) of the original images used for iterative learning is greater than or equal to the third preset value, it determines the second corresponding to the pixel value of the m-th original image according to the value of m-1. Weights.
  • the second weight value gradually decreases to a constant value as the number of the original images used for iterative learning increases.
  • the second weight value may be a weight value determined by another preset protocol.
  • the preset protocol may be a preset strategy or a preset rule for determining the second weight value.
  • the electronic device may determine the second weight value according to the following preset protocol:
  • the second weight value may be increased, the m-th original image is an abnormal image or there is abnormal data, and the electronic device uses the m-th original image
  • the second weight value can be appropriately adjusted.
  • the electronic device may also learn from scratch intermittently or periodically.
  • the iterative learning of the original image in the embodiment of the present application may be local learning or global learning, which is not specifically limited in this application.
  • the electronic device can learn locally for the original images that meet the learning conditions collected within a specific time, or perform global learning for all the original images that meet the learning conditions.
  • the light received by the optical fingerprint sensor 230 mainly includes leakage light (indicated by PL) and medium light (indicated by PM).
  • the light leakage may include light directly emitted by the light source toward the optical fingerprint sensor 230 and light reflected by the light source toward the optical fingerprint sensor 230 through obstacles such as the display screen 220.
  • the medium light may include reflected light and transmitted light of pressing a medium (for example, a finger).
  • the media light can be further divided into media light that carries fingerprint information (indicated by PMF) and media light that does not carry fingerprint information (indicated by PMD).
  • the light carrying fingerprint information is mainly concentrated on the surface of the cover glass 221.
  • the signal carrier arrives at different heights from the optical fingerprint sensor 230 from the signal source
  • the propagation path of the optical fingerprint sensor 230 is not consistent, so the calibration information required when calibrating the fingerprint image is also not consistent.
  • the light leakage is closer to the fingerprint optical fingerprint sensor 230 than the medium light, so the calibration information required by the two is not completely consistent.
  • the output electrical signals may also be inconsistent.
  • the relationship between the electrical signal output by the nth Pixel of the optical fingerprint sensor 230 and the received light intensity can be expressed by the following formula:
  • V n represents the output electrical signal of the nth Pixel of the optical fingerprint sensor 230
  • b n represents the response difference of the nth Pixel
  • the double underline represents the variable with a mean value of 1
  • P n represents the light intensity of the nth Pixel received light
  • PL n represents the light intensity of the leaked light received by the nth Pixel
  • PMF n represents the light intensity of the medium light that carries fingerprint information received by the nth Pixel
  • PMD n represents the light intensity of the medium light that does not carry fingerprint information received by the nth Pixel
  • the total number of pixels of the optical fingerprint sensor 230 is N .
  • the fingerprint recognition process may be performed in various changing scenarios (for example, the signal source intensity ⁇ changes, the finger reflectivity ⁇ changes, and/or the signal carrier propagation path ⁇ changes, etc.), it is difficult to fully cover with fixed calibration information All changing scenarios.
  • the embodiment of the present application can update the calibration information of the optical fingerprint sensor 230 through continuous learning of fingerprint images, so as to achieve the purpose of eliminating interference in various changing scenes.
  • the propagation path of the leakage light and the propagation path of the medium light are respectively expressed by the following formulas:
  • the signal source also has uneven intensity distribution, that is, the signal source intensity corresponding to each pixel of the optical fingerprint sensor 230 is not consistent.
  • the signal source intensity corresponding to each pixel of the optical fingerprint sensor 230 is quantified by the following formula:
  • ⁇ n represents the intensity of the signal source corresponding to the nth Pixel of the optical fingerprint sensor 230.
  • ⁇ n can be divided into ⁇ and Double underscores indicate variables with a mean of unit 1. During the fingerprint recognition process, It is basically unchanged or changes slowly, while ⁇ will change significantly or rapidly.
  • the light received by the optical fingerprint sensor 230 can be quantified.
  • the leakage light PL n received by the nth Pixel of the optical fingerprint sensor 230 can be determined by the signal source intensity ⁇ and The propagation path ⁇ L n of the leaked light and the leaked light reflectance ⁇ L are represented.
  • the medium light PMF n that carries fingerprint information received by the nth Pixel of the optical fingerprint sensor 230 can use the signal source intensity ⁇ and The propagation path ⁇ M n of the medium light and the fingerprint signal rate ⁇ FP are expressed.
  • the medium light PMD n that does not carry fingerprint information received by the nth Pixel of the optical fingerprint sensor 230 can use the signal source intensity ⁇ and The propagation path ⁇ M n of the medium light and the finger reflectivity ⁇ M are represented.
  • the optical signal received by the nth Pixel of the optical fingerprint sensor 230 can be expressed by the following formula:
  • V n is equal to b n .
  • all V n is subtracted from b n to obtain:
  • the embodiment of the present application only takes the change of any one of the signal source intensity ⁇ , finger reflectivity ⁇ , and propagation path ⁇ as an example for analysis, but it should not be understood as a specific limitation on itself.
  • the analysis process and formula derivation do not consider the influence of secondary factors such as time domain noise.
  • the electronic device may determine the first calibration parameter according to the following iterative learning formula:
  • Klm(i+1) (1-T(i+1))*Klm(i)+T(i+1)*FP/uFP;
  • the Klm(i+1) represents the calibration parameters obtained after the i+1th original image undergoes iterative learning
  • the T(i+1) represents the i+1th original image undergoes iterative learning
  • the weight value of the calibration parameter obtained later the FP represents the pixel value of the (i+1)th original image
  • the uFP represents the average value of the pixel value of the (i+1)th original image.
  • the electronic device After acquiring the first calibration parameter, the electronic device calibrates the target image according to the following calibration formula and the first calibration parameter to obtain the calibration image:
  • the CaliFP represents the correction value of the pixel value of the (i+1)th original image.
  • the following is an analysis of the iterative learning process and calibration effect of the n original images based on the technical solution of the first embodiment when the electronic device is in different application scenarios.
  • the light signal received by the nth pixel of the mth fingerprint image in the M sheets can be expressed by the following formula:
  • the received optical signal can be further expressed as:
  • the calibration method of the embodiment of the present application can eliminate most of the interference, leaving only a small amount of interference, and basically does not affect fingerprint recognition performance.
  • the calibration effect is basically not affected, indicating that the technical solution of this embodiment can be compatible with scenes where the intensity ⁇ of the signal source changes drastically or rapidly.
  • the received optical signal can be further expressed as:
  • the change of the finger reflectivity ⁇ will affect the calibration result, thereby affecting the fingerprint recognition performance.
  • the change of finger reflectivity ⁇ will affect the calibration effect.
  • the technical solutions of the embodiments of the present application can still calibrate fingerprint images with changing finger reflectivity ⁇ , especially for scenes where the finger reflectivity ⁇ changes slightly or slowly, which can reduce the finger reflectivity ⁇ . The impact of changes on fingerprint images.
  • the optical signal can be further expressed as:
  • the propagation path ⁇ of the signal carrier will affect the calibration result, which in turn affects the fingerprint recognition performance.
  • changes in the propagation path ⁇ will affect the calibration effect.
  • the technical solutions of the embodiments of the present application can still calibrate fingerprint images with changes in the propagation path ⁇ , especially for scenarios where the propagation path ⁇ changes slightly or slowly, which can reduce the effect of changes in the propagation path ⁇ on the fingerprint. The impact of the image.
  • the electronic device may determine the first calibration parameter according to the following iterative learning formula:
  • the Blm(i+1) represents the calibration parameters obtained after the i+1th original image undergoes iterative learning
  • the T(i+1) represents the i+1th original image undergoes iterative learning
  • the weight value of the calibration parameter obtained later, the FP represents the pixel value of the i-th original image.
  • the target image may be calibrated according to the following calibration formula and the first calibration parameter to obtain the calibration image:
  • the CaliFP represents the correction value of the pixel value of the (i+1)th original image.
  • the light signal received by the nth pixel of the mth fingerprint image in the M sheets can be expressed by the following formula:
  • the received optical signal can be further expressed as:
  • the average of the M fingerprint images can be obtained:
  • the change of the intensity ⁇ of the signal source will affect the calibration result, and then affect the fingerprint recognition performance.
  • the change of the intensity ⁇ of the signal source will affect the calibration effect.
  • the technical solutions of the embodiments of the present application can still calibrate the fingerprint image whose intensity ⁇ of the signal source changes, especially for scenarios where the intensity ⁇ of the signal source changes slightly or slowly, which can reduce the signal source's intensity. The influence of the change of intensity ⁇ on the fingerprint image.
  • the received optical signal can be further expressed as:
  • the change of the finger reflectivity ⁇ will affect the calibration result, thereby affecting the fingerprint recognition performance.
  • the change of finger reflectivity ⁇ will affect the calibration effect.
  • the technical solutions of the embodiments of the present application can still calibrate fingerprint images with changing finger reflectivity ⁇ , especially for scenes where the finger reflectivity ⁇ changes slightly or slowly, which can reduce the finger reflectivity ⁇ . The impact of changes on fingerprint images.
  • the optical signal can be further expressed as:
  • the propagation path ⁇ of the signal carrier will affect the calibration result, which in turn affects the fingerprint recognition performance.
  • changes in the propagation path ⁇ will affect the calibration effect.
  • the technical solutions of the embodiments of the present application can still calibrate fingerprint images with changes in the propagation path ⁇ , especially for scenarios where the propagation path ⁇ changes slightly or slowly, which can reduce the effect of changes in the propagation path ⁇ on the fingerprint. The impact of the image.
  • the application also provides a device for calibrating images.
  • FIG. 6 is a schematic block diagram of an apparatus 500 for calibrating an image according to an embodiment of the present application.
  • the device 500 may include a determination unit 510 and a calibration unit 520.
  • the determining unit 510 is configured to determine the first calibration parameter through iterative learning of n original images; wherein the calibration parameter obtained after iterative learning of the i+1th original image among the n original images is based on the n
  • the calibration parameters of the i-th original image in the original images obtained after iterative learning and the parameters determined by the pixel value of the i+1-th original image, the first calibration parameter is the value of the n original images
  • Calibration parameters of the nth original image after iterative learning, n is the first preset value and a positive integer, 1 ⁇ i ⁇ n.
  • the calibration unit 520 is configured to calibrate the target image based on the first calibration parameter.
  • the target image is the n+jth original image, and j is a positive integer; wherein, the calibration unit 520 is specifically configured to:
  • the second calibration parameter is obtained through iterative learning of part or all of the original images from the n+1th to n+j original images; +1 original image for calibration.
  • the original image used for iterative learning is an image whose light intensity is less than a second preset value.
  • the calibration parameter obtained after the iterative learning of the i-th original image corresponds to the first weight value
  • the pixel value of the i+1-th original image corresponds to the second weight value
  • the sum of the first weight value and the second weight value is 1.
  • the second weight value gradually decreases to a constant value as the number of the original images used for iterative learning increases.
  • the determining unit 510 is specifically configured to:
  • Klm(i+1) (1-T(i+1))*Klm(i)+T(i+1)*FP/uFP;
  • the Klm(i+1) represents the calibration parameters obtained after the i+1th original image undergoes iterative learning
  • the T(i+1) represents the i+1th original image undergoes iterative learning
  • the weight value of the calibration parameter obtained later the FP represents the pixel value of the (i+1)th original image
  • the uFP represents the average value of the pixel value of the (i+1)th original image.
  • the calibration unit 520 is specifically configured to:
  • the target image is calibrated according to the following calibration formula and the first calibration parameter to obtain the calibration image:
  • the CaliFP represents the correction value of the pixel value of the (i+1)th original image.
  • the determining unit 510 is specifically configured to:
  • the Blm(i+1) represents the calibration parameters obtained after the i+1th original image undergoes iterative learning
  • the T(i+1) represents the i+1th original image undergoes iterative learning
  • the weight value of the calibration parameter obtained later, the FP represents the pixel value of the i-th original image.
  • the calibration unit 520 is specifically configured to:
  • the target image is calibrated according to the following calibration formula and the first calibration parameter to obtain the calibration image:
  • the CaliFP represents the correction value of the pixel value of the (i+1)th original image.
  • the present application also provides an electronic device, which may include a display screen, a fingerprint module, and the device for calibrating an image mentioned above; the fingerprint module is arranged below the display screen or The inside of the display screen; the fingerprint module is electrically connected to the device for calibrating images; wherein the fingerprint module is used to receive fingerprint detection signals returned by the reflection or scattering of a human finger above the display screen , The fingerprint detection signal carries fingerprint information of the finger.
  • an electronic device may include a display screen, a fingerprint module, and the device for calibrating an image mentioned above; the fingerprint module is arranged below the display screen or The inside of the display screen; the fingerprint module is electrically connected to the device for calibrating images; wherein the fingerprint module is used to receive fingerprint detection signals returned by the reflection or scattering of a human finger above the display screen , The fingerprint detection signal carries fingerprint information of the finger.
  • apparatus 500 may correspond to a corresponding main body that executes each method embodiment in FIG. 4 and FIG. 5 according to the present application. For brevity, details are not described herein again.
  • the apparatus for calibrating an image according to an embodiment of the present application is described above from the perspective of functional modules in conjunction with FIG. 6. It should be understood that the functional module can be implemented in the form of hardware, can also be implemented in the form of software instructions, or can be implemented in a combination of hardware and software modules.
  • the steps of the method embodiments in the embodiments of the present application can be completed by hardware integrated logic circuits in the processor and/or instructions in the form of software, and the steps of the methods disclosed in the embodiments of the present application can be directly embodied as hardware.
  • the execution of the decoding processor is completed, or the execution is completed by a combination of hardware and software modules in the decoding processor.
  • the software module may be located in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, and registers.
  • the storage medium is located in the memory, and the processor reads the information in the memory, and completes the steps in the foregoing method embodiments in combination with its hardware.
  • both the determining unit 510 and the calibration unit 520 shown in FIG. 6 may be implemented by a processor, wherein the calibration parameters determined by the determining unit 510 may be stored in a memory.
  • FIG. 7 is a schematic structural diagram of an apparatus 600 for fingerprint identification according to an embodiment of the present application.
  • the device 600 shown in FIG. 7 includes a fingerprint sensor chip 610, a processor 620, and a memory 630.
  • the fingerprint sensor chip 610 can be used to obtain fingerprint information. For example, when the processor 620 determines that the pressing force of the user pressing the collection area in the display screen is greater than or equal to the trigger threshold, it triggers the fingerprint sensor chip 610 to acquire the fingerprint information, that is, triggers the fingerprint sensor chip 610 to pair Fingerprint data collection operation.
  • the memory 630 may be used to store the aforementioned fingerprint information for fingerprint registration or fingerprint identification, and may also be used to store codes and instructions executed by the processor 620. For example, the calibration parameter determined by the processor 620.
  • the processor 620 may call and run a computer program from the memory 630 to implement the method in the embodiment of the present application.
  • the memory 630 may be a separate device independent of the processor 620, or may be integrated in the processor 620.
  • the device 600 may correspond to the device 500 in the embodiment of the present application, and may correspond to the corresponding main body that executes each method embodiment in FIG. 3 and FIG. 4 according to the present application. For brevity, it will not be omitted here. Repeat.
  • the various components in the device 600 are connected by a bus system, where in addition to a data bus, the bus system also includes a power bus, a control bus, and a status signal bus.
  • processor mentioned in the embodiments of the present application may be an integrated circuit chip with signal processing capability, and can implement or execute the methods, steps, and logical block diagrams disclosed in the embodiments of the present application.
  • the above-mentioned processor may be a general-purpose processor, a digital signal processor (digital signal processor, DSP), an application specific integrated circuit (ASIC), a ready-made programmable gate array (field programmable gate array, FPGA), or Other programmable logic devices, transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like.
  • the memory mentioned in the embodiments of the present application may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.
  • the non-volatile memory can be read-only memory (ROM), programmable read-only memory (programmable ROM, PROM), erasable programmable read-only memory (erasable PROM, EPROM), and electronic Erase programmable read-only memory (electrically EPROM, EEPROM) or flash memory.
  • ROM read-only memory
  • PROM programmable read-only memory
  • EPROM erasable programmable read-only memory
  • EPROM erasable PROM
  • EPROM erasable programmable read-only memory
  • electronic Erase programmable read-only memory electrically EPROM, EEPROM
  • flash memory electrically EPROM, EEPROM
  • the volatile memory may be random access memory (RAM), which is used as an external cache.
  • the memory in the embodiment of the present application may also be static random access memory (static RAM, SRAM), dynamic random access memory (dynamic RAM, DRAM), Synchronous dynamic random access memory (synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (double data rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchronous connection Dynamic random access memory (synch link DRAM, SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DR RAM), etc.
  • FIG. 8 is a schematic structural diagram of an electronic device (such as a touch screen mobile phone) 700 to which an embodiment of the present application is applied. As shown in FIG. 8, the electronic device 700 may include:
  • a processor 710 a processor 710, a memory 720, and a touch screen 730.
  • the touch display screen 730 includes a pressure sensor 731, and the pressure sensor 731 is used to sense the pressure of the touch input signal on the touch display screen 730.
  • the processor 710 is configured to receive a pressure signal sensed by the pressure sensor 731, and to process the pressure signal, for example, to trigger an application in the mobile terminal 100 based on the pressure signal.
  • the electronic device 700 may further include a fingerprint sensor chip 780, and the fingerprint sensor chip 780 is used to obtain a fingerprint image (ie, an original image).
  • the fingerprint sensor chip 780 may include a device for fingerprint identification (for example, the device 500 shown in FIG. 6 or the device 600 shown in FIG. 7), which is used to perform image calibration on the fingerprint image.
  • the electronic device 700 may further include an illuminance sensor 790 for determining whether the touch display screen 730 is blocked.
  • the electronic device may also include other components, such as the audio circuit 740, the power supply 750, the WiFi module 760, and the radio frequency circuit 770 as shown in FIG. 1.
  • the power supply 750 may include a visible light source and an infrared light source, wherein the visible light emitted by the visible light source is used for displaying images, and the infrared light emitted by the infrared light source is used for fingerprint identification.
  • FIG. 8 is only an example of this application, and should not be construed as a limitation to this application.
  • the fingerprint sensor chip 780 may be arranged inside the touch display screen 730, or the fingerprint sensor chip 780 and a device for fingerprint identification (for example, as shown in FIG. 6 The device 500 or the device 600 as shown in FIG. 7) may be physically separated.
  • the device shown in FIG. 7 may also be applied to electronic equipment that does not include a display screen.
  • electronic equipment that does not include a display screen.
  • fingerprint access control machine or punch card machine and so on For example, fingerprint access control machine or punch card machine and so on.
  • the technical solutions of the embodiments of the present application can be embodied in the form of software products in essence or the parts that contribute to the prior art or the parts of the technical solutions, and the computer software products are stored in a storage medium.
  • Including several instructions to enable a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in the embodiments of the present application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory, random access memory, magnetic disk or optical disk and other media that can store program codes.
  • the division of units or modules or components in the device embodiments described above is only a logical function division, and there may be other divisions in actual implementation.
  • multiple units or modules or components can be combined or integrated.
  • To another system, or some units or modules or components can be ignored or not executed.
  • the units/modules/components described as separate/display components may or may not be physically separated, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units/modules/components may be selected according to actual needs to achieve the objectives of the embodiments of the present application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Image Input (AREA)

Abstract

一种用于校准图像的方法、装置和电子设备，方法包括：通过对n张原始图像的迭代学习确定第一校准参数；其中，n张原始图像中的第i+1张原始图像经过迭代学习后获取的校准参数是根据n张原始图像中的第i张原始图像经过迭代学习后获取的校准参数和第i+1张原始图像的像素值确定的参数，第一校准参数为n张原始图像中的第n张原始图像经过迭代学习后的校准参数，n为第一预设值且为正整数，1≤i≤n；基于第一校准参数对目标图像进行校准。可以不依托或者依托少量的先验信息，通过不断学习用户应用过程中的原始图像，更新用于校准指纹图像的校准参数，不仅可以简化操作流程，还能够有效提高校准精度。

Description

用于校准图像的方法、装置和电子设备 技术领域
本申请实施例涉及电子领域，并且更具体地，涉及用于校准图像的方法、装置和电子设备。
背景技术
响应于越来越严苛的市场需求，各种电容式、光学式、超声波式等指纹识别方案应运而生。其中，一部分方案需要将指纹传感器(Sensor)裸露在电子设备的表面，另一部分方案可以将指纹Sensor放置在电子设备的内部。信号源(可以指代电信号、光源或超声波发生器等)发出的信号载体到达手指后，携带有指纹信息的信号载体能够穿透障碍物到达指纹Sensor表面，指纹Sensor能够基于所述携带有指纹信息的信号载体进行指纹成像，最后结合软件算法实现指纹识别功能。
但是，由于障碍物叠层的不均匀性，携带有指纹信息的信号载体穿透障碍物时会受到干扰。另外，由于信号源的不均匀性、指纹Sensor响应不一致性等因素，会进一步影响指纹图像的质量。
因此，本领域急需一种能够校准指纹图像的方法和装置。
发明内容
提供了一种用于校准图像的方法、装置和电子设备，能够有效校准图像。
第一方面，提供了一种用于校准图像的方法，包括：
通过对n张原始图像的迭代学习确定第一校准参数；
其中，所述n张原始图像中的第i+1张原始图像经过迭代学习后获取的校准参数是根据所述n张原始图像中的第i张原始图像经过迭代学习后获取的校准参数和所述第i+1张原始图像的像素值确定的参数，所述第一校准参数为所述n张原始图像中的第n张原始图像经过迭代学习后的校准参数，n为第一预设值且为正整数，1≤i≤n；
基于所述第一校准参数对目标图像进行校准。
第二方面，提供了一种用于校准图像的装置，包括：
确定单元，用于通过对n张原始图像的迭代学习确定第一校准参数；
其中，所述n张原始图像中的第i+1张原始图像经过迭代学习后获取的校准参数是根据所述n张原始图像中的第i张原始图像经过迭代学习后获取的校准参数和所述第i+1张原始图像的像素值确定的参数，所述第一校准参数为所述n张原始图像中的第n张原始图像经过迭代学习后的校准参数，n为第一预设值且为正整数，1≤i≤n；
校准单元，用于基于所述第一校准参数对目标图像进行校准。
第三方面，提供了一种电子设备，包括：
指纹模组，所述指纹模组设置在所述电子设备的表面或者内部；
第二方面所述的用于校准图像的装置，所述指纹模组电连接至所述装置；
其中，所述指纹模组用于接收经由所述显示屏上方的人体手指反射或散射而返回的指纹检测信号，所述指纹检测信号携带有所述手指的指纹信息。
基于以上技术方案，本申请是实施例的用于校准图像的方法、装置和电子设备可以不依托或者依托少量的先验信息，通过不断学习用户应用过程中的原始图像，更新用于校准指纹图像的校准参数，不仅可以简化操作流程，还能够有效提高校准精度。
此外，所述方法、装置和电子设备可以通过在用户应用过程(例如指纹识别过程)中的原始图像中筛选出用于校准指纹图像的图像，可以将本申请的技术方案可以应用在较为恶劣的场景下，使得其具有更宽泛的应用场景和更高的指纹识别性能。换句话说，所述电子设备通过在采集的原始图像中选择不存在异常的原始图像用于迭代学习，或者将原始图像中异常部分裁去，只对部分区域进行迭代学习，能够有效避免对发生异常的原始图像的异常部分对迭代学习的影响，进而提高校准参数的准确度。
附图说明
图1是本申请可以适用的显示屏内具有采集区域的电子设备的平面示意图。
图2是图1所示的电子设备沿A’-A’的部分剖面示意图。
图3是本申请实施例的电子设备的另一示意性结构图。
图4是本申请实施例的用于校准图像的方法的示意性流程图。
图5是本申请实施例的用于校准图像的方法的另一示意性流程图。
图6是本申请实施例的用于校准图像的装置的示意性框图。
图7是本申请实施例的用于校准图像的装置的另一示意性框图。
图8是本申请实施例的电子设备的另一示意性框图。
具体实施方式
下面将结合附图，对本申请实施例中的技术方案进行描述。
本申请实施例的技术方案可以应用于各种电子设备。
例如，智能手机、笔记本电脑、平板电脑、游戏设备等便携式或移动计算设备，以及电子数据库、汽车、银行自动柜员机(Automated Teller Machine，ATM)等其他电子设备。但本申请实施例对此并不限定。
本申请实施例的技术方案可以用于任何需要进行图像校准的技术。例如生物特征识别技术。其中，生物特征识别技术包括但不限于指纹识别、掌纹识别、虹膜识别、人脸识别以及活体识别等识别技术。所述生物特征识别技术可以是电容式、光学式、超声波式等生物特征识别技术。为了便于说明，下文以光学指纹识别技术为例对本申请实施例的应用场景进行说明。
光学指纹识别技术可以用于屏下指纹识别技术和屏内指纹识别技术。
屏下指纹识别技术是指将指纹识别模组安装在显示屏下方，从而实现在显示屏的显示区域内进行指纹识别操作，不需要在电子设备正面除显示区域外的区域设置指纹采集区域。具体地，指纹识别模组使用从电子设备的显示组件的顶面返回的光或外部添加的不可见LED光源(例如红外LED光源)来进行指纹感应和其他感应操作。这种返回的光携带与显示组件的顶面接触的物体(例如手指)的信息，位于显示组件下方的指纹识别模组通过采集和检测这种返回的光以实现屏下指纹识别。其中，指纹识别模组的设计可以为通过恰当地配置用于采集和检测返回的光的光学元件来实现期望的光学成像。
相应的，屏内(In-display)指纹识别技术是指将指纹识别模组或者部分指纹识别模组安装在显示屏内部，从而实现在显示屏的显示区域内进行指纹识别操作，不需要在电子设备正面除显示区域外的区域设置指纹采集区域。
图1和图2示出了屏下指纹识别技术可以适用的电子设备100的示意图，其中图1为电子设备100的正面示意图，图2为图1所示的电子设备100的部分剖面结构示意图。
如图1和图2所示，电子设备100可以包括显示屏120和指纹识别模组140。
显示屏120可以为自发光显示屏，其采用具有自发光的显示单元作为显示像素。比如显示屏120可以为有机发光二极管(Organic Light-Emitting Diode，OLED)显示屏或者微型发光二极管(Micro-LED)显示屏。在其他可替代实施例中，显示屏120也可以为液晶显示屏(Liquid Crystal Display,LCD)或者其他被动发光显示屏，本申请实施例对此不做限制。
此外，显示屏120还可以具体为触控显示屏，其不仅可以进行画面显示，还可以检测用户的触摸或者按压操作，从而为用户提供一个人机交互界面。比如，在一种实施例中，电子设备100可以包括触摸传感器，所述触摸传感器可以具体为触控面板(Touch Panel,TP)，其可以设置在所述显示屏120表面，也可以部分集成或者整体集成到所述显示屏120内部，从而形成所述触控显示屏。
指纹识别模组140可以为光学指纹识别模组，比如光学指纹传感器。
具体来说，指纹识别模组140可以包括具有光学感应阵列的指纹传感器芯片(后面也称为光学指纹传感器)。其中，光学感应阵列包括多个光学感应单元，每个光学感应单元可以具体包括光探测器或者光电传感器。或者说，指纹识别模组140可以包括光探测器(Photo detector)阵列(或称为光电探测器阵列、光电传感器阵列)，其包括多个呈阵列式分布的光探测器。
如图1所示，指纹识别模组140可以设置在所述显示屏120的下方的局部区域，从而使得指纹识别模组140的指纹采集区域(或检测区域)130至少部分位于所述显示屏120的显示区域102内。
当然，在其他可替代实施例中，指纹识别模组140也可以设置在其他位置，比如显示屏120的侧面或者电子设备100的边缘非透光区域。这种情况下，可以通过光路设计将显示屏120的至少部分显示区域的光信号导引到指纹识别模组140，从而使得所述指纹采集区域130实际上位于所述显示屏120的显示区域内。
在本申请的一些实施例中，指纹识别模组140可以仅包括一个指纹传感器芯片，此时指纹识别模组140的指纹采集区域130的面积较小且位置固定，因此用户在进行指纹输入时需要将手指按压到所述指纹采集区域130的特定位置，否则指纹识别模组140可能无法采集到指纹图像而造成用户体验不佳。
在本申请的另一些实施例中，指纹识别模组140可以具体包括多个指纹传感器芯片；所述多个指纹传感器芯片可以通过拼接方式并排设置在所述显示屏120的下方，且所述多个指纹传感器芯片的感应区域共同构成所述指纹识别模组140的指纹采集区域130。也即是说，所述指纹识别模组140的指纹采集区域130可以包括多个子区域，每个子区域分别对应于其中一个指纹传感器芯片的感应区域，从而将所述光学指纹模组130的指纹采集区域130可以扩展到所述显示屏的下半部分的主要区域，即扩展到手指惯常按压区域，从而实现盲按式指纹输入操作。可替代地，当所述指纹传感器芯片数量足够时，所述指纹检测区域130还可以扩展到半个显示区域甚至整个显示区域，从而实现半屏或者全屏指纹检测。
应理解，本申请实施例对所述多个指纹传感器芯片的具体形式不做限定。
例如，所述多个指纹传感器芯片可以分别是独立封装的指纹传感器芯片，也可以是封装在同一个芯片封装体内的多个芯片(Die)。
又例如，还可以通过半导体工艺在同一个芯片(Die)的不同区域上制作形成所述多个指纹传感器芯片。
如图2所示，指纹识别模组140的光学感应阵列的所在区域或者光感应范围对应所述指纹识别模组140的指纹采集区域130。其中，指纹识别模组140的指纹采集区域130可以等于或不等于指纹识别模组140的光学感应阵列的所在区域的面积或者光感应范围，本申请实施例对此不做具体限定。
例如，通过光线准直的光路设计，指纹识别模组140的指纹采集区域130可以设计成与所述指纹识别模组140的感应阵列的面积基本一致。
又例如，通过汇聚光线的光路设计或者反射光线的光路设计，可以使得所述指纹识别模组140的指纹采集区域130的面积大于所述指纹识别模组140感应阵列的面积。
下面对指纹识别模组140的光路设计进行示例性说明。
以指纹识别模组140的光路设计采用具有高深宽比的通孔阵列的光学准直器为例，所述光学准直器可以具体为在半导体硅片制作而成的准直器(Collimator)层，其具有多个准直单元或者微孔，所述准直单元可以具体为小孔，从手指反射回来的反射光中，垂直入射到所述准直单元的光线可以穿过并被其下方的指纹传感器芯片接收，而入射角度过大的光线在所述准直 单元内部经过多次反射被衰减掉，因此每一个指纹传感器芯片基本只能接收到其正上方的指纹纹路反射回来的反射光，能够有效提高图像分辨率，进而提高指纹识别效果。
进一步地，当指纹识别模组140包括多个指纹传感器芯片时，可以为每个指纹传感器芯片的光学感应阵列中的一个光学感应单元配置一个准直单元，并贴合设置在其对应的光学感应单元的上方。当然，所述多个光学感应单元也可以共享一个准直单元，即所述一个准直单元具有足够大的孔径以覆盖多个光学感应单元。由于一个准直单元可以对应多个光学感应单元，破坏了显示屏120的空间周期和指纹传感器芯片的空间周期的对应性，因此，即使显示屏120的发光显示阵列的空间结构和指纹传感器芯片的光学感应阵列的空间结构类似，也能够有效避免指纹识别模组140利用经过显示屏120的光信号进行指纹成像生成莫尔条纹，有效提高了指纹识别模组140的指纹识别效果。
以指纹识别模组140的光路设计采用光学镜头的光路设计为例，所述光学镜头可以包括光学透镜(Lens)层，其具有一个或多个透镜单元，比如一个或多个非球面透镜组成的透镜组，其用于将从手指反射回来的反射光汇聚到其下方的指纹传感器芯片的感应阵列，以使得所述感应阵列可以基于所述反射光进行成像，从而获取所述手指的指纹图像。所述光学透镜层在所述透镜单元的光路中还可以形成有针孔，所述针孔可以配合所述光学透镜层扩大指纹识别模组140的视场，以提高所述指纹识别模组140的指纹成像效果。
进一步地，当指纹识别模组140包括多个指纹传感器芯片时，可以为每一个指纹传感器芯片配置一个光学镜头进行指纹成像，或者为多个指纹传感器芯片配置一个光学镜头来实现光线汇聚和指纹成像。甚至于，当一个指纹传感器芯片具有两个感应阵列(Dual Array)或多个感应阵列(Multi-Array)时，也可以为这个指纹传感器芯片配置两个或多个光学镜头配合所述两个感应阵列或多个感应阵列进行光学成像，从而减小成像距离并增强成像效果。
以指纹识别模组140的光路设计采用微透镜(Micro-Lens)层的光路设计为例，所述微透镜层可以具有由多个微透镜形成的微透镜阵列，其可以通过半导体生长工艺或者其他工艺形成在所述指纹传感器芯片的感应阵列上方，并且每一个微透镜可以分别对应于所述感应阵列的其中一个感应单元。所述微透镜层和所述感应单元之间还可以形成其他光学膜层，比如介质层或者钝 化层，更具体地，所述微透镜层和所述感应单元之间还可以包括具有微孔的挡光层，其中所述微孔形成在其对应的微透镜和感应单元之间，所述挡光层可以阻挡相邻微透镜和感应单元之间的光学干扰，并使光线通过所述微透镜汇聚到所述微孔内部并经由所述微孔传输到所述微透镜对应的感应单元，以进行光学指纹成像。
应当理解，上述光路引导结构的几种实现方案可以单独使用也可以结合使用，比如，可以在所述准直器层或者所述光学透镜层下方进一步设置微透镜层。当然，在所述准直器层或者所述光学透镜层与所述微透镜层结合使用时，其具体叠层结构或者光路可能需要按照实际需要进行调整。
指纹识别模组140可以用于采集用户的指纹信息(比如指纹图像信息)。
以显示屏120采用OLED显示屏为例，显示屏120可以采用具有自发光显示单元的显示屏，比如有机发光二极管(Organic Light-Emitting Diode，OLED)显示屏或者微型发光二极管(Micro-LED)显示屏。指纹识别模组140可以利用OLED显示屏的位于指纹采集区域130的显示单元(即OLED光源)来作为光学指纹检测的激励光源。
当手指触摸、按压或者接近(为便于描述，在本申请中统称为按压)在指纹采集区域130时，显示屏120向指纹采集区域130上方的手指发出一束光，这一束光在手指的表面发生反射形成反射光或者经过手指的内部散射后而形成散射光，在相关专利申请中，为便于描述，上述反射光和散射光统称为反射光。由于指纹的嵴(ridge)与峪(vally)对于光的反射能力不同，因此，来自指纹嵴的反射光和来自指纹峪的发生过具有不同的光强，反射光经过显示屏120后，被指纹识别模组140中的指纹传感器芯片所接收并转换为相应的电信号，即指纹检测信号；基于所述指纹检测信号便可以获得指纹图像数据，并且可以进一步进行指纹匹配验证，从而在所述电子设备100实现光学指纹识别功能。
由此可见，用户需要对电子设备100进行指纹解锁或者其他指纹验证的时候，只需要将手指按压在位于显示屏120的指纹采集区域130，便可以实现指纹特征的输入操作。由于指纹特征的采集可以在显示屏120的显示区域102的内部实现，采用上述结构的电子设备100无需其正面专门预留空间来设置指纹按键(比如Home键)，因而可以采用全面屏方案。因此，所述显示屏120的显示区域102可以基本扩展到所述电子设备100的整个正面。
当然，在其他替代方案中，指纹识别模组140也可以采用内置光源或者外置光源来提供用于进行指纹检测识别的光信号。在这种情况下，指纹识别模组140不仅可以适用于如OLED显示屏等自发光显示屏，还可以适用于非自发光显示屏，比如液晶显示屏或者其他的被动发光显示屏。
以应用在具有背光模组和液晶面板的液晶显示屏为例，为支持液晶显示屏的屏下指纹检测，电子设备100的光学指纹***还可以包括用于光学指纹检测的激励光源，所述激励光源可以具体为红外光源或者特定波长非可见光的光源，其可以设置在所述液晶显示屏的背光模组下方或者设置在电子设备100的保护盖板下方的边缘区域，而指纹识别模组140可以设置液晶面板或者保护盖板的边缘区域下方并通过光路引导以使得指纹检测光可以到达所述指纹识别模组140；或者，指纹识别模组140也可以设置在所述背光模组下方，且所述背光模组通过对扩散片、增亮片、反射片等膜层进行开孔或者其他光学设计以允许指纹检测光穿过液晶面板和背光模组并到达指纹识别模组140。当采用所述指纹识别模组140采用内置光源或者外置光源来提供用于进行指纹检测的光信号时，其检测原理可以相同。
如图1所示，电子设备100还可以包括保护盖板110。
盖板110可以具体为透明盖板，比如玻璃盖板或者蓝宝石盖板，其位于显示屏120的上方并覆盖所述电子设备100的正面，且盖板110表面还可以设置有保护层。因此，本申请实施例中，所谓的手指按压显示屏120实际上可以是指手指按压在显示屏120上方的盖板110或者覆盖所述盖板110的保护层表面。
如图1所示，指纹识别模组140的下方还可以设置有电路板150，比如软性电路板(Flexible Printed Circuit，FPC)。
指纹识别模组140可以通过焊盘焊接到电路板150，并通过电路板150实现与其他***电路或者电子设备100的其他元件的电性互连和信号传输。比如，指纹识别模组140可以通过电路板150接收电子设备100的处理单元的控制信号，并且还可以通过电路板150将来自指纹识别模组140的指纹检测信号输出给电子设备100的处理单元或者控制单元等。
图3是本申请实施例的电子设备的另一示意性结构图。
如图3所示，电子设备300可以包括显示屏220和光学指纹传感器230。
光源发出的光信号到达手指并经由所述手指反射或散射后，携带有指纹 信息的指纹检测信号能够穿透所述显示屏220到达所述光学指纹传感器230的上表面，使得所述光学指纹传感器230基于所述指纹检测信号进行指纹成像，进而进行指纹识别。
请继续参见图3，所述显示屏220由上到下依次可以包括盖板玻璃221、第一胶层222、偏光片223、第二胶层224、触控面板(Touch Panel，TP)层225、密封玻璃226、显示像素层227以及基板玻璃228。换句话说，在基板玻璃228上蒸馏各种有机材料形成显示像素层227后，使用密封玻璃226进行密封。基板玻璃228、密封玻璃226及其之间的显示像素层227配合显示驱动电路实现显示功能。在密封玻璃226上方的TP层225配合触控驱动电路实现触控功能。TP层225可以被蚀刻成各种图案。偏光片223通过第二胶层224设置在TP层225上，偏光片223可以用于抑制显示屏220对环境光的反射，进而实现更高的显示对比度。盖板玻璃221通过第一胶层221设置在偏光片223上，用于保护所述显示屏220。光学指纹传感器230放置或者贴合在基板玻璃228的底部，由此可以在显示屏的显示区域中局部实现或全屏实现屏下光学指纹识别。
在图3所示的结构中，由于所述显示屏220的各个叠层具有不均匀性，会使得携带有指纹信息的指纹检测信号穿透所述显示屏220时受到干扰，进而影响指纹图像的成像质量。
此外，考虑到信号源也具有不均匀性以及光学指纹传感器230在不同位置上的响应也不一致等因素，会进一步影响指纹图像的质量。
本申请提出了一种校准图像的方法，其可以不依托先验信息或者依托少量的先验信息，通过不断学习用户应用过程中的原始图像，更新用于校准指纹图像的校准参数，不仅可以简化操作流程，还能够有效提高校准精度。此外，通过在用户应用过程中的原始图像中筛选出用于校准指纹图像的图像，可以将本申请的技术方案可以应用在较为恶劣的场景下，使得其具有更宽泛的应用场景和更高的指纹识别性能。
图4是本申请实施例的用于校准图像的方法300的示意性流程图。应理解，所述方法300可以由任何具有图像处理能力的电子设备执行，为了便于理解，下面以电子设备为例对所述方法300进行说明。
如图4所示，所述方法300可以包括：
S310，电子设备通过对n张原始图像的迭代学习确定第一校准参数。
其中，所述n张原始图像中的第i+1张原始图像经过迭代学习后获取的校准参数是根据所述n张原始图像中的第i张原始图像经过迭代学习后获取的校准参数和所述第i+1张原始图像的像素值确定的参数，所述第一校准参数为所述n张原始图像中的第n张原始图像经过迭代学习后的校准参数，n为第一预设值且为正整数，1≤i≤n；
S320，所述电子设备基于所述第一校准参数对目标图像进行校准。
简而言之，所述电子设备通过对n张原始图像的迭代学习获取第一校准参数后，可以基于所述第一校准参数校准后续的原始图像。例如，假设所述目标图像为第n+j张原始图像，j为正整数。j为1时，所述电子设备可以基于所述第一校准参数直接校准第n+1张原始图像。j为大于1的正整数时，所述电子设备可以基于所述第一校准参数直接校准第n+j张原始图像，也可以基于所述第一校准参数，通过对第n+1张至n+j张原始图像中的部分或全部原始图像的迭代学习先获取第二校准参数，然后基于所述第二校准参数对第n+j+1张原始图像进行校准。
本申请实施例中，可以不依托或者依托少量的先验信息，通过不断学习用户应用过程中的原始图像，更新用于校准指纹图像的校准参数，不仅可以简化操作流程，还能够有效提高校准精度。
在本申请的一些实施例中，异常的原始图像或者原始图像中的异常部分可以不用于迭代学习。具体而言，所述电子设备获取原始图像后，先确定已获取的原始图像是否存在异常，如果存在异常则不用于迭代学习；或者将该异常图像的异常部分裁去，将正常部分用于局部区域的迭代学习；如果不存在异常则用于迭代学习。
以电子设备对上述n张原始图像的迭代学习确定第一校准参数为例，所述电子设备可以通过对所述n张原始图像中的除异常数据之外的数据的迭代学习确定所述第一校准参数；或所述电子设备可以通过对所述n张原始图像中的除异常图像之外的图像的迭代学习确定所述第一校准参数。
其中，异常图像可以是存在异常数据的原始图像。所述异常数据包括但不限于：由于用户的手指的一部分没有按压所述电子设备导致的数据不完整和由于强光导致的发生异常的数据。
本实施例中，所述电子设备通过在用户应用过程(例如指纹识别过程)中的原始图像中筛选出用于校准指纹图像的图像，可以将本申请的技术方案 可以应用在较为恶劣的场景下，使得其具有更宽泛的应用场景和更高的指纹识别性能。换句话说，所述电子设备通过在采集的原始图像中选择不存在异常的原始图像或者异常图像正常部分用于迭代学习，能够有效避免对发生异常的原始图像的迭代学习，进而提高校准参数的准确度。
在迭代学习过程中，第i张原始图像经过迭代学习后获取的校准参数可以对应于第一权重值，第i+1张原始图像的像素值对应于第二权重值，所述第一权重值和所述第二权重值的和为1。可选地，所述第二权重值随着所述用于迭代学习的所述原始图像的数量的增大逐渐减小直至恒定值。
图4是本申请实施例的用于校准图像的方法300的另一示意性流程图。
如图4所示，所述方法300可以包括：
S410，开始。
触发电子设备进行图像校准操作。
S420，初始校准参数。
所述电子设备将校准参数初始化。例如，所述电子设备将校准参数初始化为0。
S430，采集第m张原始图像。
所述电子设备采集原始图像，并对仅学习的原始图像进行计数。具体地，所述电子设备已学习了m-1张原始图像后，采集第m张原始图像。
S440，m-1是否大于第一阈值？
所述电子设备确定已经过迭代学习的原始图像的数量(m-1)是否大于第一预设值，由此确定所述m-1张原始图像经过迭代学习后获取的校准参数是否可以用于校准图像。具体地，当m-1大于或等于所述第一预设值时，所述电子设备确定所述m-1张原始图像经过迭代学习后获取的校准参数可以用于校准图像，当m-1小于所述第一预设值时，确定所述m-1张原始图像经过迭代学习后获取的校准参数不能够用于校准图像。
即所述电子设备通过判断m-1的数量是否满足预设条件，进而确定是否可以基于所述m-1张原始图的校准参数对第m张原始图像进行校准。
但是，应当理解，S440的实现方式仅为本申请的一种示例，不应理解为对本申请的限制。
例如，在其他可替代实施例中，还可以通过其它判断方式触发后续操作。例如，可以通过算法估计所述m-1张原始图像中每张原始图像的空域噪声是 否小于特定值，进而确定是否可以基于所述m-1张原始图像的校准参数对第m张原始图像进行校准。
S441，基于第m-1张原始图像经过迭代学习后获取的校准参数校准所述第m张原始图像。
所述电子设备确定所述m-1张原始图像经过迭代学习后获取的校准参数可以用于校准图像采集原始图像时，基于第m-1张原始图像经过迭代学习后获取的校准参数校准所述第m张原始图像。
S450，判定第m张原始图像是否存在异常？
所述电子设备确定所述m-1张原始图像经过迭代学习后获取的校准参数不能用于校准图像时，对所述第m张原始图像进行迭代学习。具体地，确定所述第m张原始图像不存在异常时，对所述第m张原始图像进行迭代学习。
S460，m-1是否大于或等于第三阈值？
所述电子设备确定用于迭代学习的所述原始图像的数量(m-1)是否大于或等于第三阈值，并基于确定结果确定第m张原始图像的像素值对应的第二权重值，以及所述m-1张原始图像经过迭代学习后获取的校准参数对应的第一权重值，所述第一权重值和所述第二权重值之和为1。
S470，根据m-1的数值确定第m张原始图像的像素值对应的第二权重值。
所述电子设备确定用于迭代学习的所述原始图像的数量(m-1)小于第三预设值时，根据m-1的数值确定第m张原始图像的像素值对应的第二权重值。
S471，将第m张原始图像的像素值对应的第二权重值确定为恒定值。
所述电子设备确定用于迭代学习的所述原始图像的数量(m-1)大于或等于第三预设值时，根据m-1的数值确定第m张原始图像的像素值对应的第二权重值。
简而言之，在S460至S471的实现方式，所述第二权重值随着所述用于迭代学习的所述原始图像的数量的增大逐渐减小直至恒定值。
当然，应当理解，S460至S471的实现方式仅为本申请的一种示例，不应理解为对本申请的限制。
例如，所述第二权重值可以是其它预设协议确定的权重值。所述预设协 议可以是用于确定所述第二权重值的预设策略或预设规则。
例如，所述电子设备可以按照以下预设协议确定所述第二权重值：
所述第m张原始图像不是异常图像时，可以调高所述第二权重值，所述第m张原始图像为异常图像或存在异常数据，且所述电子设备利用所述第m张原始图像中除异常数据之外的数据(即异常图像裁去异常数据所在的区域)进行迭代学习时，可以适当调小所述第二权重值。
又例如，所述电子设备也可以间歇性或周期性的从零开始学习等。
S480，确定基于第m张原始图像经过迭代学习后获取的校准参数。
S490，结束。
应理解，本申请实施例中针对原始图像的迭代学习，可以局部学习也可以全局学习，本申请对此不做具体限定。
例如，电子设备可以针对特定时间内采集的满足学习条件的原始图像局部学习，也可以针对所有满足学习条件的原始图像进行全局学习。
下面结合具体实施例对本申请的迭代学习的实现方式，以及基于校准参数校准图像的实现方式进行说明。
请继续参见图3，当用户的手指按压盖板玻璃221的上表面时，光学指纹传感器230接收的光主要包括漏光(用PL表示)和媒介光(用PM表示)。其中漏光可以包括光源直接射向光学指纹传感器230的光和光源经过障碍物例如显示屏220反射向光学指纹传感器230的光。媒介光可以包括按压媒介(例如手指)的反射光和透射光。媒介光可以进一步拆分成携带有指纹信息的媒介光(用PMF表示)和没有携带指纹信息的媒介光(用PMD表示)。其中携带有指纹信息的光主要集中在盖板玻璃221的表面。由于光学指纹传感器230在不同物距上的分辨率存在差异以及显示屏220的内部叠层的分布具有不均匀性等因素，在距离光学指纹传感器230不同高度的位置上，信号载体从信号源到达光学指纹传感器230的传播路径并不一致，因此校准指纹图像时所需的校准信息也并不一致。例如，漏光比媒介光更靠近指纹光学指纹传感器230，因此两者所需的校准信息并不完全一致。
由于光学指纹传感器230对光信号的响应也存在不均匀性，即光学指纹传感器230中不同的像素(Pixel)接收相同光强的光信号时，输出的电信号也有可能不一致。
本申请实施例中，考虑Pixel输出的电信号与其接收光强成线性关系， 光学指纹传感器230的第n个Pixel输出的电信号与接收光强的关系可以通过如下公式表示：
Figure PCTCN2019082739-appb-000001
Figure PCTCN2019082739-appb-000002
其中，V n表示光学指纹传感器230的第n个Pixel的输出电信号，
Figure PCTCN2019082739-appb-000003
和b n表示第n个Pixel的响应差异，双下划线表示均值为单位1的变量，P n表示第n个Pixel接收光的光强，PL n表示第n个Pixel接收的漏光的光强，PMF n表示第n个Pixel接收的携带有指纹信息的媒介光的光强，PMD n表示第n个Pixel接收的没有携带指纹信息的媒介光的光强，光学指纹传感器230的Pixel的总数量为N。
由于指纹识别过程可能在各种变化的场景中进行(例如信号源强度λ发生变化、手指反射率η发生变化和/或信号载体的传播路径γ发生变化等)，采用固定的校准信息难以完全覆盖所有变化场景。本申请实施例通过对指纹图像不断学习，能够更新光学指纹传感器230的校准信息，以达到在各种变化场景中消除干扰的目的。
在图3所示的结构中，由于显示屏220的内部叠层的分布存在不均匀性，(例如显示像素层227中发光像素点及金属走线分布不均匀，TP层225的图案分布不均匀等)，使得光学指纹传感器230的不同的Pixel接收到的信号载体的传播路径γ不一致。
以漏光和媒介光为例，所述漏光的传播路径和媒介光的传播路径分别通过以下公式表示：
Figure PCTCN2019082739-appb-000004
Figure PCTCN2019082739-appb-000005
其中，
Figure PCTCN2019082739-appb-000006
用于表示光学指纹传感器230的第n个Pixel接收到的漏光的传播路径，
Figure PCTCN2019082739-appb-000007
用于表示媒介光的信号载体传播路径，双下划线表示均值为单位1的变量。
在实际应用中，信号源也存在强度分布不均匀情况，即光学指纹传感器230的各Pixel对应的信号源强度并不一致。本申请实施例通过以下公式量化光学指纹传感器230的各Pixel对应的信号源强度：
Figure PCTCN2019082739-appb-000008
Figure PCTCN2019082739-appb-000009
Figure PCTCN2019082739-appb-000010
其中，λ n表示光学指纹传感器230的第n个Pixel对应的信号源的强度。λ n可以拆分为λ和
Figure PCTCN2019082739-appb-000011
双下划线表示均值为单位1的变量。在指纹识别过程中，
Figure PCTCN2019082739-appb-000012
基本不变或者发生缓慢变化，而λ会发生较大或者快速变化。
结合上述对像素对应的传播路径和信号强度的量化，可以量化光学指纹传感器230接收的光。
具体地，光学指纹传感器230的第n个Pixel接收到的漏光PL n可以用信号源强度λ和
Figure PCTCN2019082739-appb-000013
漏光的传播路径γL n以及漏光反射率ηL表示。光学指纹传感器230的第n个Pixel接收到的携带有指纹信息的媒介光PMF n可以用信号源强度λ和
Figure PCTCN2019082739-appb-000014
媒介光的传播路径γM n以及指纹信号率ηFP表示。光学指纹传感器230的第n个Pixel接收到的没有携带有指纹信息的媒介光PMD n可以用信号源强度λ和
Figure PCTCN2019082739-appb-000015
媒介光的传播路径γM n以及手指反射率ηM表示。
因此，光学指纹传感器230的第n个Pixel接收到的光信号可以通过以下公式表示：
Figure PCTCN2019082739-appb-000016
显然，λ n为零时V n等于b n。后文为便于描述，默认将V n都减去b n得到：
Figure PCTCN2019082739-appb-000017
下面结合上述量化公式分场景介绍本申请的实施例。
应理解，本申请实施例仅以信号源的强度λ、手指反射率η、传播路径γ中的任一项发生变化为例进行分析，但不应理解为对本身的具体限定。此外，为简便起见，分析过程及公式推导不考虑时域噪声等次要因素的影响。
实施例一：
在本申请实施例中，所述电子设备可以根据以下迭代学习公式，确定所述第一校准参数：
Klm(i+1)＝(1-T(i+1))*Klm(i)+T(i+1)*FP/uFP；
其中，所述Klm(i+1)表示所述第i+1张原始图像经过迭代学***均值。
所述电子设备获取所述第一校准参数后，根据以下校准公式和所述第一校准参数对所述目标图像进行校准，以获取所述校准图像：
CaliFP＝FP/Klm，和/或
CaliFP＝FP-uFP*Klm；
其中，所述CaliFP表示所述第i+1张原始图像的像素值的校正值。
下面针对电子设备在不同应用场景下时，基于实施例一的技术方案，对所述n张原始图像的迭代学习过程以及校准效果进行分析。
假设在短时间内用户使用指纹识别功能的过程中采集了M张指纹图像。则所述M张中第m张指纹图像的第n个像素点接收到的光信号可以通过以下公式表示：
Figure PCTCN2019082739-appb-000018
场景一：
假设在此短时间内只有信号源强度λ发生变化，而手指反射率η和信号载体的传播路径γ不发生变化或者只发生了轻微变化，则所述第m张指纹图像的第n个像素点接收到的光信号可以进一步表示为：
Figure PCTCN2019082739-appb-000019
则所述第m张指纹图像的像素值对应的平均值可以表示为：
Figure PCTCN2019082739-appb-000020
将所述第m张指纹图像的第n个像素点的像素值除以所述第m张指纹图像的像素值对应的平均值uFP m，可以得到对所述第n个像素点进行归一化处理后的像素值：
Figure PCTCN2019082739-appb-000021
将所述M张指纹图像的第n个像素点进行归一化处理后的像素值取平均可以得到：
Figure PCTCN2019082739-appb-000022
将所述M张指纹图像平均时，指纹会慢慢消失，即
Figure PCTCN2019082739-appb-000023
平均后变成ηFP， 利用实施例一中的校准公式可以得到：
Figure PCTCN2019082739-appb-000024
Figure PCTCN2019082739-appb-000025
通过上述校准公式可以看出，本申请实施例的校准方法可以消除大部分干扰，只残留极少量干扰，基本不影响指纹识别性能。换句话说，信号源的强度λ发生变化时基本不影响校准效果，说明本实施例的技术方案可以兼容信号源的强度λ剧烈或快速变化场景。
场景二：
假设在此短时间内只有手指反射率η发生变化，而信号源强度λ和信号载体的传播路径γ不发生变化或者只发生轻微变化，则所述第m张指纹图像的第n个像素点接收到的光信号可以进一步表示为：
Figure PCTCN2019082739-appb-000026
则所述第m张指纹图像的像素值对应的平均值可以表示为：
Figure PCTCN2019082739-appb-000027
将所述第m张指纹图像的第n个像素点的像素值除以所述第m张指纹图像的像素值对应的平均值uFP m，可以得到对所述第n个像素点进行归一化处理后的像素值：
Figure PCTCN2019082739-appb-000028
将所述M张指纹图像的第n个像素点进行归一化处理后的像素值取平 均可以得到：
Figure PCTCN2019082739-appb-000029
将所述M张指纹图像平均时，指纹会慢慢消失，即
Figure PCTCN2019082739-appb-000030
平均后变成ηFP，利用实施例一中的校准公式可以得到：
Figure PCTCN2019082739-appb-000031
Figure PCTCN2019082739-appb-000032
通过上述校准公式可以看出，手指反射率η变化会对校准结果产生影响，进而影响指纹识别性能。也就是说，手指反射率η变化会影响校准效果。但需要注意的是，本申请实施例的技术方案仍然可以实现对手指反射率η发生变化的指纹图像进行校准，尤其是针对手指反射率η轻微或者缓慢变化的场景，能够降低手指反射率η的变化对指纹图像的影响。
场景三：
假设在此短时间内只有信号载体传播路径γ发生变化，而信号源强度λ和手指反射率η不发生变化或者只发生轻微变化，则所述第m张指纹图像的第n个像素点接收到的光信号可以进一步表示为：
Figure PCTCN2019082739-appb-000033
则所述第m张指纹图像的像素值对应的平均值可以表示为：
Figure PCTCN2019082739-appb-000034
将所述第m张指纹图像的第n个像素点的像素值除以所述第m张指纹图像的像素值对应的平均值uFP m，可以得到对所述第n个像素点进行归一化 处理后的像素值：
Figure PCTCN2019082739-appb-000035
将所述M张指纹图像的第n个像素点进行归一化处理后的像素值取平均可以得到：
Figure PCTCN2019082739-appb-000036
将所述M张指纹图像平均时，指纹会慢慢消失，即
Figure PCTCN2019082739-appb-000037
平均后变成ηFP，利用实施例一中的校准公式可以得到：
Figure PCTCN2019082739-appb-000038
Figure PCTCN2019082739-appb-000039
通过上述校准公式可以看出，信号载体的传播路径γ会对校准结果产生影响，进而影响指纹识别性能。也就是说，传播路径γ变化会影响校准效果。但需要注意的是，本申请实施例的技术方案仍然可以实现对传播路径γ发生变化的指纹图像进行校准，尤其是针对传播路径γ轻微或者缓慢变化的场景，能够降低传播路径γ的变化对指纹图像的影响。
实施例二：
在本申请实施例中，所述电子设备可以根据以下迭代学习公式，确定所述第一校准参数：
Blm(i+1)＝(1-T(i+1))*Blm(i)+T(i+1)*FP；
其中，所述Blm(i+1)表示所述第i+1张原始图像经过迭代学习后获取的校准参数，所述T(i+1)表示所述第i+1张原始图像经过迭代学习后获取的校准参数的权重值，所述FP表示所述第i张原始图像的像素值。
所述电子设备获取所述第一校准参数后，可以根据以下校准公式和所述第一校准参数对所述目标图像进行校准，以获取所述校准图像：
CaliFP＝FP/Klm，和/或
CaliFP＝FP-Blm；
其中，所述CaliFP表示所述第i+1张原始图像的像素值的校正值。
与实施例一类似，下面针对电子设备在不同应用场景下时，基于实施例二的技术方案，对所述n张原始图像的迭代学习过程以及校准效果进行分析。
假设在短时间内用户使用指纹识别功能的过程中采集了M张指纹图像。则所述M张中第m张指纹图像的第n个像素点接收到的光信号可以通过以下公式表示：
Figure PCTCN2019082739-appb-000040
场景一：
假设在此短时间内只有信号源强度λ发生变化，而手指反射率η和信号载体的传播路径γ不发生变化或者只发生了轻微变化，则所述第m张指纹图像的第n个像素点接收到的光信号可以进一步表示为：
Figure PCTCN2019082739-appb-000041
将所述M张指纹图像平均可以得到：
Figure PCTCN2019082739-appb-000042
所述M张指纹图像平均时，指纹会慢慢消失，即
Figure PCTCN2019082739-appb-000043
平均后变成ηFP，利用实施例二中的校准公式可以得到：
Figure PCTCN2019082739-appb-000044
Figure PCTCN2019082739-appb-000045
通过上述校准公式可以看出，信号源的强度λ变化会对校准结果产生影响，进而影响指纹识别性能。也就是说，信号源的强度λ变化会影响校准效果。但需要注意的是，本申请实施例的技术方案仍然可以实现对信号源的强度λ发生变化的指纹图像进行校准，尤其是针对信号源的强度λ轻微或者缓慢变化的场景，能够降低信号源的强度λ的变化对指纹图像的影响。
场景二：
假设在此短时间内只有手指反射率η发生变化，而信号源强度λ和信号载体的传播路径γ不发生变化或者只发生轻微变化，则所述第m张指纹图像的第n个像素点接收到的光信号可以进一步表示为：
Figure PCTCN2019082739-appb-000046
则所述第m张指纹图像的像素值对应的平均值可以表示为：
Figure PCTCN2019082739-appb-000047
所述M张指纹图像平均时，指纹会慢慢消失，即
Figure PCTCN2019082739-appb-000048
平均后变成ηFP，利用实施例二中的校准公式可以得到：
Figure PCTCN2019082739-appb-000049
Figure PCTCN2019082739-appb-000050
通过上述校准公式可以看出，手指反射率η变化会对校准结果产生影响，进而影响指纹识别性能。也就是说，手指反射率η变化会影响校准效果。但 需要注意的是，本申请实施例的技术方案仍然可以实现对手指反射率η发生变化的指纹图像进行校准，尤其是针对手指反射率η轻微或者缓慢变化的场景，能够降低手指反射率η的变化对指纹图像的影响。
场景三：
假设在此短时间内只有信号载体传播路径γ发生变化，而信号源强度λ和手指反射率η不发生变化或者只发生轻微变化，则所述第m张指纹图像的第n个像素点接收到的光信号可以进一步表示为：
Figure PCTCN2019082739-appb-000051
则所述第m张指纹图像的像素值对应的平均值可以表示为：
Figure PCTCN2019082739-appb-000052
所述M张指纹图像平均时，指纹会慢慢消失，即
Figure PCTCN2019082739-appb-000053
平均后变成ηFP，利用实施例二中的校准公式可以得到：
Figure PCTCN2019082739-appb-000054
Figure PCTCN2019082739-appb-000055
通过上述校准公式可以看出，信号载体的传播路径γ会对校准结果产生影响，进而影响指纹识别性能。也就是说，传播路径γ变化会影响校准效果。但需要注意的是，本申请实施例的技术方案仍然可以实现对传播路径γ发生变化的指纹图像进行校准，尤其是针对传播路径γ轻微或者缓慢变化的场景，能够降低传播路径γ的变化对指纹图像的影响。
本申请还提供了一种用于校准图像的装置。
图6是本申请实施例的用于校准图像的装置500的示意性框图。
如图6所示，所述装置500可以包括确定单元510和校准单元520。
确定单元510用于通过对n张原始图像的迭代学习确定第一校准参数；其中，所述n张原始图像中的第i+1张原始图像经过迭代学习后获取的校准参数是根据所述n张原始图像中的第i张原始图像经过迭代学习后获取的校准参数和所述第i+1张原始图像的像素值确定的参数，所述第一校准参数为所述n张原始图像中的第n张原始图像经过迭代学习后的校准参数，n为第一预设值且为正整数，1≤i≤n。
校准单元520用于基于所述第一校准参数对目标图像进行校准。
在本申请的一些实施例中，所述目标图像为第n+j张原始图像，j为正整数；其中，所述校准单元520具体用于：
基于所述第一校准参数，通过对第n+1张至n+j张原始图像中的部分或全部原始图像的迭代学习获取第二校准参数；基于所述第二校准参数对第n+j+1张原始图像进行校准。
在本申请的一些实施例中，用于迭代学习的所述原始图像为采集图像时光强小于第二预设值的图像。
在本申请的一些实施例中，所述第i张原始图像经过迭代学习后获取的校准参数对应于第一权重值，所述第i+1张原始图像的像素值对应于第二权重值，所述第一权重值和所述第二权重值的和为1。
在本申请的一些实施例中，所述第二权重值随着所述用于迭代学习的所述原始图像的数量的增大逐渐减小直至恒定值。
在本申请的一些实施例中，所述确定单元510具体用于：
根据以下迭代学习公式，确定所述第一校准参数：
Klm(i+1)＝(1-T(i+1))*Klm(i)+T(i+1)*FP/uFP；
其中，所述Klm(i+1)表示所述第i+1张原始图像经过迭代学***均值。
在本申请的一些实施例中，所述校准单元520具体用于：
根据以下校准公式和所述第一校准参数对所述目标图像进行校准，以获取所述校准图像：
CaliFP＝FP/Klm，和/或
CaliFP＝FP-uFP*Klm；
其中，所述CaliFP表示所述第i+1张原始图像的像素值的校正值。
在本申请的一些实施例中，所述确定单元510具体用于：
根据以下迭代学习公式，确定所述第一校准参数：
Blm(i+1)＝(1-T(i+1))*Blm(i)+T(i+1)*FP；
其中，所述Blm(i+1)表示所述第i+1张原始图像经过迭代学习后获取的校准参数，所述T(i+1)表示所述第i+1张原始图像经过迭代学习后获取的校准参数的权重值，所述FP表示所述第i张原始图像的像素值。
在本申请的一些实施例中，所述校准单元520具体用于：
根据以下校准公式和所述第一校准参数对所述目标图像进行校准，以获取所述校准图像：
CaliFP＝FP/Blm，和/或
CaliFP＝FP-Blm；
其中，所述CaliFP表示所述第i+1张原始图像的像素值的校正值。
本申请还提供了一种电子设备，所述电子设备可以包括显示屏、指纹模组和上文涉及的用于校准图像的装置；所述指纹模组设置在所述显示屏的下方或所述显示屏的内部；所述指纹模组电连接至所述用于校准图像的装置；其中，所述指纹模组用于接收经由所述显示屏上方的人体手指反射或散射而返回的指纹检测信号，所述指纹检测信号携带有所述手指的指纹信息。
应理解，上述装置500可以对应于执行根据本申请图4、图5中的各个方法实施例中的相应主体，为了简洁，在此不再赘述。
上文中结合图6从功能模块的角度描述了本申请实施例的用于校准图像的装置。应理解，该功能模块可以通过硬件形式实现，也可以通过软件形式的指令实现，还可以通过硬件和软件模块组合实现。
具体地，本申请实施例中的方法实施例的各步骤可以通过处理器中的硬件的集成逻辑电路和/或软件形式的指令完成，结合本申请实施例公开的方法的步骤可以直接体现为硬件译码处理器执行完成，或者用译码处理器中的硬件及软件模块组合执行完成。
可选地，软件模块可以位于随机存储器，闪存、只读存储器、可编程只读存储器、电可擦写可编程存储器、寄存器等本领域的成熟的存储介质中。该存储介质位于存储器，处理器读取存储器中的信息，结合其硬件完成上述方法实施例中的步骤。
例如，可替代地，图6所示的确定单元510和校准单元520均可以由处理器实现，其中，确定单元510确定的校准参数可由存储器存储。
图7是本申请实施例的用于指纹识别的装置600示意性结构图。图7所示的装置600包括指纹传感芯片610、处理器620以及存储器630。
所述指纹传感芯片610可以用于获取指纹信息。例如，所述处理器620确定用户按压显示屏内的采集区域的按压力度大于或等于触发阈值时，触发所述指纹传感芯片610获取所述指纹信息，即触发所述指纹传感芯片610对指纹数据的采集操作。
存储器630可以用于存储上述涉及的指纹信息，用于指纹注册或指纹识别，还可以用于存储处理器620执行的代码、指令等。例如，所述处理器620确定的校准参数。其中，处理器620可以从存储器630中调用并运行计算机程序，以实现本申请实施例中的方法。
其中，存储器630可以是独立于处理器620的一个单独的器件，也可以集成在处理器620中。
可选地，该装置600可对应于本申请实施例中的装置500，并可以对应于执行根据本申请图3以及图4中的各个方法实施例中的相应主体，为了简洁，在此不再赘述。
应当理解，该装置600中的各个组件通过总线***相连，其中，总线***除包括数据总线之外，还包括电源总线、控制总线和状态信号总线。
还应理解，本申请实施例中提及的处理器可能是一种集成电路芯片，具有信号的处理能力，可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。
例如，上述的处理器可以是通用处理器、数字信号处理器(digital signal processor，DSP)、专用集成电路(application specific integrated circuit，ASIC)、现成可编程门阵列(field programmable gate array，FPGA)或者其他可编程逻辑器件、晶体管逻辑器件、分立硬件组件等等。此外，通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
此外，本申请实施例中提及的存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。
其中，非易失性存储器可以是只读存储器(read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储 器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(random access memory，RAM)，其用作外部高速缓存。
应理解，上述存储器为示例性但不是限制性说明，例如，本申请实施例中的存储器还可以是静态随机存取存储器(static RAM，SRAM)、动态随机存取存储器(dynamic RAM，DRAM)、同步动态随机存取存储器(synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(double data rate SDRAM，DDR SDRAM)、增强型同步动态随机存取存储器(enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(synch link DRAM，SLDRAM)以及直接内存总线随机存取存储器(Direct Rambus RAM，DR RAM)等等。
图8是本申请实施例应用于的一种电子设备(例如触摸屏手机)700的结构示意图。如图8所示，电子设备700可以包括：
处理器710、存储器720与触摸显示屏730。
触摸显示屏730中包括压力传感器731，压力传感器731用于感应触摸显示屏730上的触摸输入信号的压力大小。处理器710用于接收压力传感器731感应的压力信号，并用于处理该压力信号，例如基于该压力信号触发该移动终端100中的某个应用程序。
可选地，如图8所示，电子设备700还可以包括指纹传感器芯片780，所述指纹传感器芯片780用于获取指纹图像(即原始图像)。所述指纹传感器芯片780可以包括用于指纹识别的装置(例如如图6所示的装置500或如图7所述的装置600)，其用于对所述指纹图像进行图像校准。
可选地，如图8所示，电子设备700还可以包括照度传感器790，所述照度传感器790用于确定触摸显示屏730是否被遮挡。
可选地，该电子设备还可以包括其他部件，如图1所示的音频电路740、电源750、WiFi模块760和射频电路770等其部件。应理解，所述电源750可以包括可见光源和红外光源，其中，所述可见光源发出的可见光用于显示图像，所述红外光源发出的红外光用于指纹识别。
应理解，图8仅为本申请的一种示例，不应理解为对本申请的限制。
例如，在其他可替代实施例中，所述指纹传感器芯片780可以设置在所述触摸显示屏730的内部，或者所述指纹传感器芯片780和用于指纹识别的 装置(例如如图6所示的装置500或如图7所述的装置600)可以是物理上分开设置的。
又例如，图7所示的装置还可以应用于不包括显示屏的电子设备。例如，指纹门禁机或打卡机等等。
需要说明的是，在本申请实施例和所附权利要求书中使用的术语是仅仅出于描述特定实施例的目的，而非旨在限制本申请实施例。
例如，在本申请实施例和所附权利要求书中所使用的单数形式的“一种”、“所述”、“上述”和“该”也旨在包括多数形式，除非上下文清楚地表示其他含义。
所属领域的技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请实施例的范围。
如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请实施例的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器、随机存取存储器、磁碟或者光盘等各种可以存储程序代码的介质。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的设备、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请提供的几个实施例中，应该理解到，所揭露的电子设备、装置和方法，可以通过其它的方式实现。
例如，以上所描述的装置实施例中单元或模块或组件的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如，多个单元或模块或组件可以结合或者可以集成到另一个***，或一些单元或模块或组件可以忽略，或不执行。
又例如，上述作为分离/显示部件说明的单元/模块/组件可以是或者也可以不是物理上分开的，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元/模块/组件来实现本申请实施例的目的。
最后，需要说明的是，上文中显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
以上内容，仅为本申请实施例的具体实施方式，但本申请实施例的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请实施例揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请实施例的保护范围之内。因此，本申请实施例的保护范围应以权利要求的保护范围为准。

Claims (19)

  1. 一种用于校准图像的方法，其特征在于，包括：
    通过对n张原始图像的迭代学习确定第一校准参数；
    其中，所述n张原始图像中的第i+1张原始图像经过迭代学习后获取的校准参数是根据所述n张原始图像中的第i张原始图像经过迭代学习后获取的校准参数和所述第i+1张原始图像的像素值确定的参数，所述第一校准参数为所述n张原始图像中的第n张原始图像经过迭代学习后的校准参数，n为第一预设值且为正整数，1≤i≤n；
    基于所述第一校准参数对目标图像进行校准。
  2. 根据权利要求1所述的方法，其特征在于，所述目标图像为第n+j张原始图像，j为正整数；
    其中，所述基于所述第一校准参数对目标图像进行校准，包括：
    基于所述第一校准参数，通过对第n+1张至n+j张原始图像中的部分或全部原始图像的迭代学习获取第二校准参数；
    基于所述第二校准参数对第n+j+1张原始图像进行校准。
  3. 根据权利要求1或2所述的方法，其特征在于，所述通过对n张原始图像的迭代学习确定第一校准参数，包括：
    通过对所述n张原始图像中的除异常数据之外的数据的迭代学习确定所述第一校准参数；或
    通过对所述n张原始图像中的除异常图像之外的图像的迭代学习确定所述第一校准参数。
  4. 根据权利要求1至3中任一项所述的方法，其特征在于，所述第i张原始图像经过迭代学习后获取的校准参数对应于第一权重值，所述第i+1张原始图像的像素值对应于第二权重值，所述第一权重值和所述第二权重值的和为1。
  5. 根据权利要求4所述的方法，其特征在于，所述第二权重值是根据预设协议确定的权重值。
  6. 根据权利要求1至5中任一项所述的方法，其特征在于，所述通过对n张原始图像的迭代学习确定第一校准参数，包括：
    根据以下迭代学习公式，确定所述第一校准参数：
    Klm(i+1)＝(1-T(i+1))*Klm(i)+T(i+1)*FP/uFP；
    其中，所述Klm(i+1)表示所述第i+1张原始图像经过迭代学***均值。
  7. 根据权利要求6所述的方法，其特征在于，所述通过对n张原始图像的迭代学习确定第一校准参数，包括：
    根据以下校准公式和所述第一校准参数对所述目标图像进行校准，以获取所述校准图像：
    CaliFP＝FP/Klm，和/或
    CaliFP＝FP-uFP*Klm；
    其中，所述CaliFP表示所述第i+1张原始图像的像素值的校正值。
  8. 根据权利要求1至5中任一项所述的方法，其特征在于，所述通过对n张原始图像的迭代学习确定第一校准参数，包括：
    根据以下迭代学习公式，确定所述第一校准参数：
    Blm(i+1)＝(1-T(i+1))*Blm(i)+T(i+1)*FP；
    其中，所述Blm(i+1)表示所述第i+1张原始图像经过迭代学习后获取的校准参数，所述T(i+1)表示所述第i+1张原始图像经过迭代学习后获取的校准参数的权重值，所述FP表示所述第i张原始图像的像素值。
  9. 根据权利要求8所述的方法，其特征在于，所述基于所述第一校准参数对目标图像进行校准，以获取校准图像，包括：
    根据以下校准公式和所述第一校准参数对所述目标图像进行校准，以获取所述校准图像：
    CaliFP＝FP/Blm，和/或
    CaliFP＝FP-Blm；
    其中，所述CaliFP表示所述第i+1张原始图像的像素值的校正值。
  10. 一种用于校准图像的装置，其特征在于，包括：
    确定单元，用于通过对n张原始图像的迭代学习确定第一校准参数；
    其中，所述n张原始图像中的第i+1张原始图像经过迭代学习后获取的校准参数是根据所述n张原始图像中的第i张原始图像经过迭代学习后获取的校准参数和所述第i+1张原始图像的像素值确定的参数，所述第一校准参 数为所述n张原始图像中的第n张原始图像经过迭代学习后的校准参数，n为第一预设值且为正整数，1≤i≤n；
    校准单元，用于基于所述第一校准参数对目标图像进行校准。
  11. 根据权利要求10所述的装置，其特征在于，所述目标图像为第n+j张原始图像，j为正整数；
    其中，所述校准单元具体用于：
    基于所述第一校准参数，通过对第n+1张至n+j张原始图像中的部分或全部原始图像的迭代学习获取第二校准参数；
    基于所述第二校准参数对第n+j+1张原始图像进行校准。
  12. 根据权利要求10或11所述的装置，其特征在于，所述确定单元具体用于：
    通过对所述n张原始图像中的除异常数据之外的数据的迭代学习确定所述第一校准参数；或
    通过对所述n张原始图像中的除异常图像之外的图像的迭代学习确定所述第一校准参数。
  13. 根据权利要求10至12中任一项所述的装置，其特征在于，所述第i张原始图像经过迭代学习后获取的校准参数对应于第一权重值，所述第i+1张原始图像的像素值对应于第二权重值，所述第一权重值和所述第二权重值的和为1。
  14. 根据权利要求13所述的装置，其特征在于，所述第二权重值是根据预设协议确定的权重值。
  15. 根据权利要求10至14中任一项所述的装置，其特征在于，所述确定单元具体用于：
    根据以下迭代学习公式，确定所述第一校准参数：
    Klm(i+1)＝(1-T(i+1))*Klm(i)+T(i+1)*FP/uFP；
    其中，所述Klm(i+1)表示所述第i+1张原始图像经过迭代学***均值。
  16. 根据权利要求15所述的装置，其特征在于，所述校准单元具体用于：
    根据以下校准公式和所述第一校准参数对所述目标图像进行校准，以获 取所述校准图像：
    CaliFP＝FP/Klm，和/或
    CaliFP＝FP-uFP*Klm；
    其中，所述CaliFP表示所述第i+1张原始图像的像素值的校正值。
  17. 根据权利要求10至14中任一项所述的装置，其特征在于，所述确定单元具体用于：
    根据以下迭代学习公式，确定所述第一校准参数：
    Blm(i+1)＝(1-T(i+1))*Blm(i)+T(i+1)*FP；
    其中，所述Blm(i+1)表示所述第i+1张原始图像经过迭代学习后获取的校准参数，所述T(i+1)表示所述第i+1张原始图像经过迭代学习后获取的校准参数的权重值，所述FP表示所述第i张原始图像的像素值。
  18. 根据权利要求17所述的装置，其特征在于，所述校准单元具体用于：
    根据以下校准公式和所述第一校准参数对所述目标图像进行校准，以获取所述校准图像：
    CaliFP＝FP/Blm，和/或
    CaliFP＝FP-Blm；
    其中，所述CaliFP表示所述第i+1张原始图像的像素值的校正值。
  19. 一种电子设备，其特征在于，包括：
    指纹模组，所述指纹模组设置在所述电子设备的表面或者内部；
    权利要求10至18中任一项所述的用于校准图像的装置，所述指纹模组电连接至所述装置；
    其中，所述指纹模组用于接收经由人体手指反射或散射而返回的指纹检测信号，所述指纹检测信号携带有所述手指的指纹信息。
PCT/CN2019/082739 2019-04-15 2019-04-15 用于校准图像的方法、装置和电子设备 WO2020210954A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201980000591.3A CN110192201B (zh) 2019-04-15 2019-04-15 用于校准图像的方法、装置和电子设备
PCT/CN2019/082739 WO2020210954A1 (zh) 2019-04-15 2019-04-15 用于校准图像的方法、装置和电子设备

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/082739 WO2020210954A1 (zh) 2019-04-15 2019-04-15 用于校准图像的方法、装置和电子设备

Publications (1)

Publication Number Publication Date
WO2020210954A1 true WO2020210954A1 (zh) 2020-10-22

Family

ID=67725912

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/082739 WO2020210954A1 (zh) 2019-04-15 2019-04-15 用于校准图像的方法、装置和电子设备

Country Status (2)

Country Link
CN (1) CN110192201B (zh)
WO (1) WO2020210954A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7476057B2 (ja) * 2020-09-11 2024-04-30 キオクシア株式会社 欠陥検査装置
CN116311396B (zh) * 2022-08-18 2023-12-12 荣耀终端有限公司 用于指纹识别的方法和装置

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8666162B1 (en) * 2010-12-20 2014-03-04 Csr Technology Inc. Advanced sensor binning correction
CN104182772A (zh) * 2014-08-19 2014-12-03 大连理工大学 一种基于深度学习的手势识别方法
CN108460356A (zh) * 2018-03-13 2018-08-28 上海海事大学 一种基于监控***的人脸图像自动处理***
CN108496184A (zh) * 2018-04-17 2018-09-04 深圳市汇顶科技股份有限公司 图像处理方法、装置和电子设备
CN108513667A (zh) * 2018-04-17 2018-09-07 深圳市汇顶科技股份有限公司 图像处理方法、装置和电子设备

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017161501A1 (zh) * 2016-03-22 2017-09-28 深圳市汇顶科技股份有限公司 指纹图像的校正方法、装置和终端
CN106650614B (zh) * 2016-11-07 2020-05-22 普道(上海)信息科技有限公司 一种动态校准方法和装置
CN107657240B (zh) * 2017-10-09 2020-11-24 上海天马微电子有限公司 一种显示装置及其指纹识别校准方法、以及电子设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8666162B1 (en) * 2010-12-20 2014-03-04 Csr Technology Inc. Advanced sensor binning correction
CN104182772A (zh) * 2014-08-19 2014-12-03 大连理工大学 一种基于深度学习的手势识别方法
CN108460356A (zh) * 2018-03-13 2018-08-28 上海海事大学 一种基于监控***的人脸图像自动处理***
CN108496184A (zh) * 2018-04-17 2018-09-04 深圳市汇顶科技股份有限公司 图像处理方法、装置和电子设备
CN108513667A (zh) * 2018-04-17 2018-09-07 深圳市汇顶科技股份有限公司 图像处理方法、装置和电子设备

Also Published As

Publication number Publication date
CN110192201A (zh) 2019-08-30
CN110192201B (zh) 2022-05-17

Similar Documents

Publication Publication Date Title
US11455823B2 (en) Under-screen fingerprint identification apparatus and electronic device
WO2020151158A1 (zh) 生物特征识别的装置
US11514709B2 (en) Biometric identification device using a light detection apparatus with light blocking layer/diaphragm
US11200400B2 (en) Fingerprint identification apparatus and electronic device
US11917763B2 (en) Fingerprint identification apparatus and electronic device
US20200218920A1 (en) Fingerprint identification apparatus and electronic device
US20200097699A1 (en) Fingerprint identification apparatus and electronic device
CN110235143B (zh) 屏下指纹识别装置和电子设备
EP3706036A1 (en) Fingerprint recognition apparatus and electronic device
KR102374723B1 (ko) 광학 지문 장치 및 전자 기기
US11928885B2 (en) Fingerprint identification method, fingerprint identification apparatus and electronic device
CN211319247U (zh) 指纹识别装置、背光模组、液晶显示屏和电子设备
WO2020168495A1 (zh) 用于指纹识别的方法、装置和终端设备
CN111095275B (zh) 指纹识别的装置、方法和电子设备
WO2020168496A1 (zh) 用于指纹识别的方法、装置和终端设备
WO2020186415A1 (zh) 指纹识别的装置、方法和电子设备
WO2021007964A1 (zh) 指纹检测装置和电子设备
US20210117644A1 (en) Optical sensing systems and devices including apertures supplanting photodiodes for increased light throughput
WO2021174423A1 (zh) 指纹识别装置、显示屏和电子设备
CN112528953A (zh) 指纹识别装置、电子设备和指纹识别的方法
WO2020210954A1 (zh) 用于校准图像的方法、装置和电子设备
WO2021056318A1 (zh) 指纹识别的方法、装置和电子设备
WO2021056392A1 (zh) 光学指纹装置，电子设备和测量距离的方法
CN211087274U (zh) 指纹检测装置和电子设备
WO2022134079A1 (zh) 指纹识别装置、电子设备和指纹识别的方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19924964

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19924964

Country of ref document: EP

Kind code of ref document: A1