• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    A method of processing an image comprising a set of pixels, each pixel being associated with a gray level, the method comprising a step of segmenting the image to generate a modified image containing only regions of the image. the image having an alternation of light areas and dark areas at a frequency higher than a minimum frequency, said segmentation step comprising: ○ assigning, to each pixel of the image, a corresponding frequency response level, at a frequency of alternations of light areas and dark areas in the neighborhood of the pixel, ○ the definition of image regions by grouping neighboring pixels with the same frequency response level, ○ the determination of a threshold frequency response level , and ○ the generation of an image comprising only regions whose pixels have a frequency response level greater than or equal to the response level is frequency threshold.
    公开号:FR3048540A1
    申请号:FR1651822
    申请日:2016-03-04
    公开日:2017-09-08
    发明作者:Laurent Kazdaghli；Cedric Thuillier；Lauriane Couturier
    申请人:Morpho SA；
    IPC主号:
    专利说明:

    FIELD OF THE INVENTION The invention relates to a method for processing an image comprising a set of pixels associated with gray levels. The invention applies in particular to the processing of fingerprint images, and in particular acquired by thin film transistor type sensors.
    STATE OF THE ART
    New types of sensors are being developed for the acquisition of fingerprint images, based in particular on the direct view of the finger. This is the case, for example, of sensors based on Thin Film Transistor (TFT) technology.
    These sensors can be less bulky and faster to use than conventional sensors hitherto used, which are sensors based on the principle of total reflection frustrated light on an interface on which a user puts the finger.
    Indeed, these sensors can for example take a series of images of the hand of an individual approaching a contact surface of the sensor, and exploit images of the fingers even in the absence of contact with the contact surface , or in the absence of significant pressure exerted by the fingers on the contact surface.
    In contrast, the direct view sensors produce a fingerprint image in which the contrast is generally much lower than the images obtained by the sensors based on the frustrated total reflection principle. In addition, stray areas may exist in the images, such as shadows worn in the case where an image was acquired in the absence of contact of the fingers on the contact surface of the sensor.
    As a result, the fingerprint images obtained by sensors based on the direct view are not at this stage directly exploitable by the algorithms implemented in systems for identification or automatic authentication from fingerprints.
    In order to make these images compatible, that is to say to ensure that a fingerprint acquired with a direct light sensor can be recognized using the same algorithms as the fingerprints acquired with the conventional technologies based on the reflection total frustrated, it is necessary to offer appropriate treatment.
    This treatment must take into account the great variability of the images obtained with sensors based on the direct view. For example, shadows worn on the images can have very variable sizes and positions depending on the number and position of the light sources illuminating the finger, and the position of the hand during the acquisition.
    DESCRIPTION OF THE INVENTION The purpose of the invention is to overcome the need identified above, by proposing an image processing method that makes it possible to adapt a fingerprint image acquired with a sensor in direct view, for example TFT type, so that the image can be exploited in a system of identification or automatic authentication by fingerprint.
    Another object of the invention is to propose a sufficiently fast processing to be implemented in real time during the acquisition of images by a sensor.
    Another object of the invention is to propose a processing that makes it possible to generate a fingerprint image exploitable by an identification or authentication system that is generated from a single initial image, and not synthesized from several shots, to avoid the risk of errors related to a movement between two shots. In this regard, the subject of the invention is a method for processing an image comprising a set of pixels, each pixel being associated with a gray level, the method comprising a step of segmenting the image to generate a modified image containing only regions of the image having alternating light areas and dark areas at a frequency greater than a minimum frequency, said segmenting step comprising: assigning, at each pixel of the image, a frequency response level, corresponding to a frequency of alternations of light areas and dark areas in the vicinity of the pixel, O the definition of regions of the image by grouping neighboring pixels of the same frequency response level. ο the determination of a threshold frequency response level, and O the generation of an image comprising only regions whose pixels have a frequency response level greater than or equal to the threshold frequency response level.
    Advantageously, but optionally, the method according to the invention may further comprise at least one of the following features: the regions are advantageously structured as a topological tree. the regions of the image are advantageously structured according to a parent-daughter relation defined as follows, for each region i of the image: O initially all the regions close to the region i on the image are considered as potential parents. O if the number of potential parents equals 1, neighbor j is assigned as parent to region i, and region i is removed from the list of potential parents in region j. O Otherwise, each parent region of region i is selected, from the list of potential parents having a frequency response level lower than that of region i, such as that having the level closest to that of region i, and region i is removed from the list of potential parents in its parent region, and O if all potential parents in region i have a higher frequency response than region i, each parent region in region i is selected as the one with the frequency response level closest to that of region i, and region i is removed from the list of potential parents of its parent region. the determination of the threshold frequency response level can comprise the implementation of the following steps: O for a fixed frequency response level N, definition of a set of so-called level N macro-regions, such as each level macro-region N comprises a parent region of pixels of frequency response level less than or equal to N and the set of daughter regions of this region,, ο for each frequency response level value from an initial value of frequency response level in the image, calculating the relative surface variation of macro-regions of a level N "relative to the macro-regions of the previous level Nn-i, and O the minimum frequency response level is determined as the level Ni for which the relative surface variation of N-level macro-regions | compared to the previous level Nm is minimal. the segmentation step comprises, before the step of defining the regions, the implementation of the steps of: erosion and then morphological dilation of the image as a function of the gray level values of the pixels, O generation of a differential image by subtracting, from the initial image, the image having undergone erosion and morphological dilation, O application to the differential image of a median filter on the value of the frequency response level, and in which the step of defining the regions is implemented on the image resulting from the application of the median filter on the differential image.
    The method may further comprise, after the segmentation step, a step of raising the gray levels associated with the pixels of the modified image, the degree of enhancement of the pixels of a region being a function of the frequency response level of the pixels. of the region. Advantageously, the greater the frequency response level of the pixels of a region of the modified image is important, and the higher the degree of enhancement of the gray levels associated with the pixels of the region is important. the processed image is preferably an image of fingers comprising at least one fingerprint. The invention also relates to a computer program product, comprising code instructions for implementing the method according to the preceding description, when it is executed by a processor. The subject of the invention is also an image processing system, comprising a processing unit of processing means adapted to implement the method according to the preceding description.
    Advantageously, but optionally, the image processing system further comprises an image acquisition means, said image acquisition means being a thin film transistor type fingerprint sensor.
    The proposed method comprises a particular segmentation step which makes it possible to keep the initial image only the zones of the image having a frequency of variations between light areas and large dark areas, which corresponds, for an image of fingerprint, to the exploitable part for identification or authentication corresponding to the peaks and valleys of the footprint.
    In particular, uniformly bright areas (areas of the image where there is no finger) and evenly dark areas (drop shadows) are removed from the image.
    The formation of the zones is achieved by the implementation of a "rising water" algorithm, which makes it possible to keep related regions of the images and to avoid the presence of holes in the areas corresponding to the fingerprints.
    The method may comprise an enhancement of the gray levels of the pixels as a function of the frequency of variations between light areas and dark areas: in other words, the larger the area of the image corresponds to a usable area of peaks and valleys, and the more the contrast of this area is increased.
    In addition, the proposed method can be implemented in real time during the acquisition of the image because the segmentation process requires only to browse the image once to define a set of regions, then to process regions. in blocks, this step requiring to process only a number of regions much less than the number of pixels of the image.
    DESCRIPTION OF THE FIGURES Other characteristics, objects and advantages of the present invention will appear on reading the detailed description which follows, with reference to the appended figures, given by way of non-limiting examples and in which:
    FIG. 1 represents the main steps of an image processing method,
    Figure 2 schematically shows an image processing system. Figures 3a to 3c show schematically the implementation of the different steps of the method on an exemplary image.
    DETAILED DESCRIPTION OF AT LEAST ONE EMBODIMENT OF THE INVENTION
    With reference to FIG. 1, the main steps of an image processing method are represented. The processed image comprises a set of pixels, each pixel being associated with a gray level, typically between 0 and 255. The processed image is advantageously an image of one or more fingers, on the side of the palm of the hand. , and the end of the finger or fingers on which are fingerprints, or the palm of the hand. More preferably, the processed image is an image acquired from a direct view type fingerprint sensor, such as a sensor based on Thin Film Transistor (TFT) technology. For example, documents US 20020054394 or US 20020000915 may be referred to.
    As shown diagrammatically in FIG. 2, the image processing method is implemented by an image processing system 1 comprising a processing unit 10, for example a computer. The processing unit 10 comprises processing means 11, for example a processor. Image processing can be implemented by a suitable computer algorithm. The processing means 11 are then adapted to execute code instructions for implementing the image processing algorithm.
    The image processing system 1 also advantageously comprises an image sensor 20, adapted to communicate with the processing unit 10 to transmit the acquired images. Advantageously, the image sensor is a fingerprint sensor in direct view, for example of the thin film transistor type.
    The image sensor 20 may be remote from the processing unit, and connected thereto by a wireless connection, for example of the wifi type, etc. The processing unit 10 comprises a memory 12 and a communication interface 13 with the image sensor 20. In a variant, the image processing system may also comprise a database (not represented) of images, from which the processing unit can recover images to be processed, these images having been obtained by an image sensor.
    FIG. 3a shows an example of an image acquired from a fingerprint sensor in direct view. As can be seen, this image is not usable in the state to carry out identification or authentication processing on fingerprints, for example by extraction and comparison of minutiae. It includes shadows (around the fingers), and areas of the fingers extending beyond the fingerprints and devoid of interest for identification or authentication processing.
    Returning to FIG. 1, the image processing method comprises a first step 100 of segmentation of the image. This segmentation step is designed so as to keep the initial image only the areas having a high frequency of alternation between light areas and dark areas. This makes it possible, in the case where the processed image is an image of the fingertips carrying the fingerprints, to keep only the useful area of the fingerprints themselves.
    To do this, step 100 of segmentation comprises a first step 110 of assigning, at each pixel, a frequency response level, which corresponds to a frequency of alternations between light areas and dark areas in the vicinity of the pixel.
    The level can be determined by positioning around each pixel a window of determined size, and by evaluating the variabilities of the gray levels of the pixels contained in the window. The window may for example be a square of the order of 10 to 20 pixels side.
    In other words, by noting Y the number of pixels contained in the window, and pixel (i) the gray level of the pixel i, the frequency response level can be calculated as follows:
    Where pixel (i) is the gray level of pixel i. From this step, the values processed in each pixel are no longer gray levels but levels of frequency response. The terminology "pixel" is preserved.
    Then the segmentation step 100 advantageously comprises a step of morphological closure of the image, which comprises a step 121 of morphological erosion followed by a step 122 of morphological dilation of the image. Morphological erosion is a process of assigning to a pixel the value of the lowest frequency response level of a pixel window encompassing it. For example, the pixel window may be a square window, for example of the order of 10 pixels or less, like 3 pixels on the side.
    Morphological dilation is a process of assigning a pixel the value of the highest frequency response level of a pixel window that encompasses it. Here again the pixel window may for example be a square window, for example of the order of 10 side pixels, or less, such as 3 pixels apart.
    These two steps make it possible to average the image of which the high frequencies of alternations between light areas and dark areas are suppressed. The morphological closing step is followed by a step 130 of generating a differential image, which is obtained by subtracting from the initial image the averaged image resulting from the processing of steps 121 and 122. Thus, the differential image has more than the areas of interest in the image, that is to say the areas having a high frequency of alternation between light areas and dark areas, as shown in Figure 3b.
    Then the segmentation step 100 comprises a step 140 of applying a median filter to the differential image obtained at the end of the step 130, the median filter being implemented as a function of the response level values. frequency of pixels.
    The segmentation 100 then comprises a step 150 of defining, in the image, regions grouping neighboring pixels of the same frequency response level. These regions are advantageously structured in a topological tree, or tree of related components.
    To do this the regions are defined and structured as follows.
    A region is first defined by the set of neighboring pixels with the same frequency response level. For each region, a list of neighboring regions is defined, i.e. regions having pixels in contact with pixels of the region of interest.
    Then a parent relationship between regions, or in other words a daughter-parent relationship, is defined as follows:
    For a region i called the daughter region, initially all the neighboring regions of region i are considered as potential parents. - If the number of potential parents equals 1, that is, region i has only one neighbor j, neighbor j is assigned as parent to region i, and the region i is removed from the list of potential parents in region j.
    Otherwise, the list of potential parents with a frequency response level lower than that of region i is selected from the list closest to that of region i. There can therefore be several parents for the same region i, who are then not neighbors. When region j is assigned as parent to region i, region i is removed from the list of potential parents in region j.
    If all potential parents have a higher frequency response level than region i, the region (s) with the frequency response level closest to that of region i (the level being then superior to that of region i). When region j is assigned as parent to region i, region i is removed from the list of potential parents in region j.
    The segmentation will then comprise a selection of the regions defined above, which have a frequency response level higher than a determined threshold level.
    However, if by trying to keep only the areas of interest corresponding to fingerprints, the threshold is set too high, there is a risk of recovering from the initial image only fragmented regions, possibly with holes which, although corresponding to zones of lower frequency response, may be relevant for fingerprinting.
    To avoid this phenomenon, the segmentation step 100 comprises a step 160 of determining the frequency response level threshold comprising a first sub-step 161 for defining macro-regions from the regions and parent-daughter relations defined below. before between regions.
    Macro regions are defined for a fixed N response level value. For this fixed value N, a macro-region comprises a region of frequency response level less than or equal to N and all the daughter regions of this region.
    The definition of such a macro-region is performed by determining, for each region of the image, the parent region having the highest level of frequency response which is lower than the level N. Then from this parent, the macro-region as grouping all the daughter regions of this parent.
    Given the construction of the parent-daughter relationship defined above, the daughter regions do not necessarily have a higher frequency response level than the parent region. In some cases, there are some isolated but lesser daughter regions for a parent region (for example, in which the daughter region has only one potential parent). Because of this construction, the fact of integrating all the daughter regions in the macro-region makes it possible to avoid the appearance, in the macro-regions, of empty zones corresponding to areas of the image which are generally darker or more clear and devoid of alternations of light and dark areas.
    Each macro-region therefore comprises a region of pixels of frequency response level less than or equal to N, and a set of neighboring regions defined as daughter regions by the definition given above.
    The definition of a macro-region therefore varies as a function of the level of frequency response N. Consequently, the area of the image occupied by the macro-region also varies. This appears in Figure 3c, which represents macro-regions defined for several frequency response levels.
    In particular, the higher the level N increases, and the more the number of daughter regions in a macro-region increases, and therefore the more the area of the macro-region increases. The segmented image at the end of step 100 comprises the macro-regions thus defined, for a particular value of frequency response level N.
    To determine this value, step 160 comprises an incremental process 162 comprising incrementing the value of the frequency response level from an initial level, and advantageously to the highest level value in the image. For each increment of the value of the frequency response level, the area of the image covered by all macro-regions of the corresponding level is measured, and the relative variation of the area occupied by said macro-regions relative to the level previous is calculated.
    Noting Surf (Nn) the area occupied by the macro-regions of frequency response level Nn, we calculate: R = (Surf (Nn) -Surf (Nn-i)) / Surf (Nn-i).
    The threshold frequency response level Ns determined for the segmentation of the image as being the level at which the relative variation of the area occupied by the macro-regions between a level and the previous level is minimal, that is to say when R is the lowest, which corresponds to the level for which the surface of the macro-regions is the most stable.
    Once the threshold level has been determined, the segmentation step 100 comprises a step 170 for generating the segmented image, in which only the macros-level region Ns are preserved from the image resulting from the step 140. In addition, macro-regions overlapping areas are removed: when two macro-regions have pixels in common, the definition of the parent-daughter relationship implies that one macro-region is necessarily encompassed in the other. Then the macro-area overlap areas are removed by removing macro-regions enclosed in larger macro-regions.
    Then the image processing method comprises a second step 200 of raising the image obtained at the end of step 100. The raising is an operation consisting in modifying the gray level values of the pixels of the image to improve the contrast. For example, if in an image the pixels have gray levels between 100 and 200, the enhancement is to give the pixels with a gray level of 200 a new level to 255, those with a gray level to 100 a new level to 0, and to distribute the values of the other pixels between 0 and 255 according to a particular law.
    In this case, in step 200, the law of reassignment of the value of the gray level of a pixel is chosen so as to be a function of the value of the frequency response level of this pixel. In other words, the higher the frequency response level of a pixel, the higher the enhancement is, that is to say the higher the contrast is important for the pixels considered.
    An example of a law of reassignment of the value of the gray level of the pixel is the following law:
    With:
    Where a is the frequency response level of the pixel x, normalized to be reduced between 0 and 1. With a low value of a, corresponding to a low level of frequency response, the value of the gray level of the pixel will be almost unchanged , and with a high value, the value of the gray level of the pixel will be more modified.
    This makes it possible to improve the contrast in the areas of the image corresponding to the fingerprints, and which are therefore the richest in information for future exploitation by an identification or authentication system. The image obtained at the end of the processing method is therefore exploitable to implement authentication or identification of individuals by comparison of fingerprints.
    权利要求:
    Claims (11)
    [1" id="c-fr-0001]
    A method of processing an image comprising a set of pixels, each pixel being associated with a gray level, the method comprising a step (100) of segmenting the image to generate a modified image containing only regions of the image. the image having an alternation of light areas and dark areas at a frequency higher than a minimum frequency, said segmenting step (100) comprising: allocating (110), at each pixel of the image, a frequency response level, corresponding to a frequency of alternations of bright areas and dark areas in the vicinity of the pixel, O definition of regions (120) of the image by grouping neighboring pixels of the same frequency response level, O la determining a threshold frequency response level (160), and O generating an image (170) comprising only regions whose pixels have a higher frequency response level o u equal to the threshold frequency response level.
    [2" id="c-fr-0002]
    The method of claim 1, wherein the regions are structured as a topological tree.
    [3" id="c-fr-0003]
    3. Treatment method according to one of claims 1 or 2, wherein the regions of the image are structured according to a parent-daughter relationship defined as follows, for each region i of the image: initially all neighboring regions of region i on the image are considered as potential parents, if the number of potential parents equals 1, neighbor j is assigned as parent to region i, and region i is removed from the list of potential parents from region j, if not, each parent region in region i is selected from the list of potential parents with a frequency response level lower than that of region i, such as that having the level closest to that of the region i, and region i is removed from the list of potential parents in its parent area, and if all potential parents in region i have a higher frequency response than region i, each The parent region of region i is determined to be the one with the frequency response level closest to that of region i and region i is removed from the list of potential parents in its parent region.
    [4" id="c-fr-0004]
    4. Processing method according to one of claims 2 or 3, wherein the determination of the threshold frequency response level (160) comprises the implementation of the following steps: for a fixed frequency response level N, definition (161) ) of a set of so-called N-level macroregions, such that each N-level macroregion comprises a parent region of frequency response level pixels less than or equal to N and the set of daughter regions of this region, - for each frequency response level value from an initial value of frequency response level in the image, calculating (162) the relative surface variation of the macro-regions of a level N "with respect to the macro-regions from the previous level Nn-i, and the minimum frequency response level is determined as the level Ni for which the relative surface variation of the level macros-regions Ni with respect to the preceeding level. Edent Nm is minimal.
    [5" id="c-fr-0005]
    5. Treatment method according to one of the preceding claims, wherein the segmentation step (100) comprises, before the step of defining the regions, the implementation of the steps of: erosion (121) and then dilation (122). ) of the image as a function of the gray level values of the pixels, generation (130) of a differential image by subtracting, from the initial image, the image having undergone erosion and morphological dilation, application to the differential image of a median filter (140) on the value of the frequency response level, and in which the region definition step (150) is implemented on the image resulting from the application of the filter median on the differential image.
    [6" id="c-fr-0006]
    6. Processing method according to one of the preceding claims, further comprising, after the segmentation step (100), a step of raising (200) the gray levels associated with the pixels of the modified image, the degree of raising the pixels of a region depending on the frequency response level of the pixels of the region.
    [7" id="c-fr-0007]
    The processing method according to claim 6, wherein the higher the frequency response level of the pixels of a region of the modified image, and the greater the degree of gray level enhancement associated with the pixels of the region is important. .
    [8" id="c-fr-0008]
    8. Method according to one of the preceding claims, wherein the processed image is an image of fingers comprising at least one fingerprint.
    [9" id="c-fr-0009]
    9. Computer program product, comprising code instructions for implementing the method according to one of the preceding claims, when executed by a processor (11).
    [10" id="c-fr-0010]
    An image processing system (1), comprising a processing unit (10) of processing means (11) adapted to implement the method according to one of claims 1 to 8.
    [11" id="c-fr-0011]
    An image processing system (1) according to claim 10, further comprising image acquisition means (20), said image acquisition means (20) being a fingerprint sensor of transistor type in thin layers.
    类似技术:
    公开号 | 公开日 | 专利标题
    FR3048540B1|2019-09-13|IMAGE ENHANCEMENT METHOD APPLICABLE TO DIGITAL IMPRESSION IMAGES
    EP3608834A1|2020-02-12|Method for analysing a fingerprint
    Jan et al.2014|A dynamic non-circular iris localization technique for non-ideal data
    EP2754088B1|2014-12-17|Identification through iris recognition
    EP2988249B1|2017-05-24|Method for determining, in an image, at least one area likely to represent at least one finger of an individual
    EP3608835A1|2020-02-12|Method for processing fingerprint images
    EP3271869A1|2018-01-24|Method for processing an asynchronous signal
    EP3709049B1|2021-06-23|Radar processing system and related noise reduction method
    EP3620970B1|2021-10-27|Method for extracting characteristics of a fingerprint represented by an input image
    EP2943935B1|2017-03-08|Estimation of the movement of an image
    EP2526526B1|2014-06-04|Method for segmenting images, computer program, and corresponding computer system
    EP3032463A1|2016-06-15|Method and device for tracking characters on a plurality of images of a video stream of a text
    EP3832535A1|2021-06-09|Method for detecting at least one visible element of interest in an input image by means of a convolutional neural network
    EP3608836B1|2021-01-06|Method for obtaining a digital fingerprint image
    FR3066633A1|2018-11-23|METHOD FOR DEFLOWING AN IMAGE
    FR3088755A1|2020-05-22|METHOD FOR UNFLOUTING AN IMAGE
    WO2011036046A1|2011-03-31|Method for detecting a fake finger for fingerprint acquisition software
    FR3082341A1|2019-12-13|PROCESS FOR PROCESSING A PAPILLARY FOOTPRINT IMAGE
    EP2767929A2|2014-08-20|Method for allocating a first value or a second value to the pixels of an image
    WO2012107696A1|2012-08-16|Methods, device and computer programs for recognising shapes, in real time, by means of an appliance including limited resources
    EP1684210A1|2006-07-26|Red-eye detection based on skin region detection
    WO2007068748A1|2007-06-21|Method of characterizing a region of interest in an image, signal representative of an image, method of comparing images, corresponding devices and computer program
    FR3057690A1|2018-04-20|METHOD OF COMPLETING AN IMAGE OF A SCENE UTILIZING THE SPECTRAL CHARACTERISTICS OF THE SCENE
    EP1146434A1|2001-10-17|Method for identifying geometric patterns on a curve
    FR3030835A1|2016-06-24|METHOD FOR CLASSIFYING A DOCUMENT FROM ITS VISUAL ASPECT
    同族专利:
    公开号 | 公开日
    KR102335907B1|2021-12-06|
    US10216973B2|2019-02-26|
    EP3214601A1|2017-09-06|
    EP3214601B1|2020-10-07|
    KR20170103703A|2017-09-13|
    US20170255807A1|2017-09-07|
    FR3048540B1|2019-09-13|
    CA2960247A1|2017-09-04|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US6784413B2|1998-03-12|2004-08-31|Casio Computer Co., Ltd.|Reading apparatus for reading fingerprint|
    KR100381048B1|2000-06-23|2003-04-18|엘지.필립스 엘시디 주식회사|Thin Film Transistor Type Finger Print Sensor|
    US8953848B2|2009-07-17|2015-02-10|University Of Maryland, College Park|Method and apparatus for authenticating biometric scanners|
    WO2017091663A1|2015-11-23|2017-06-01|Jensen Eric Dean|Fingerprint reader|CN108121946B|2017-11-15|2021-08-03|大唐微电子技术有限公司|Fingerprint image preprocessing method and device|
    FR3081245B1|2018-05-17|2020-06-19|Idemia Identity & Security France|CHARACTER RECOGNITION PROCESS|
    FR3084944B1|2018-08-10|2021-05-07|Idemia Identity & Security France|IMPRESSION IMAGE PROCESSING PROCESS|
    KR20200070878A|2018-12-10|2020-06-18|삼성전자주식회사|Method and apparatus for preprocessing fingerprint image|
    法律状态:
    2017-02-22| PLFP| Fee payment|Year of fee payment: 2 |
    2017-09-08| PLSC| Publication of the preliminary search report|Effective date: 20170908 |
    2018-02-20| PLFP| Fee payment|Year of fee payment: 3 |
    2019-02-20| PLFP| Fee payment|Year of fee payment: 4 |
    2020-02-20| PLFP| Fee payment|Year of fee payment: 5 |
    2020-04-24| CD| Change of name or company name|Owner name: IDEMIA IDENTITY AND SECURITY, FR Effective date: 20200317 |
    2020-04-24| CA| Change of address|Effective date: 20200317 |
    2021-02-19| PLFP| Fee payment|Year of fee payment: 6 |
    2022-02-18| PLFP| Fee payment|Year of fee payment: 7 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1651822|2016-03-04|
    FR1651822A|FR3048540B1|2016-03-04|2016-03-04|IMAGE ENHANCEMENT METHOD APPLICABLE TO DIGITAL IMPRESSION IMAGES|FR1651822A| FR3048540B1|2016-03-04|2016-03-04|IMAGE ENHANCEMENT METHOD APPLICABLE TO DIGITAL IMPRESSION IMAGES|
    US15/451,246| US10216973B2|2016-03-04|2017-03-06|Method for improving images applicable to fingerprint images|
    KR1020170028459A| KR102335907B1|2016-03-04|2017-03-06|Method for improving images applicable to fingerprint images|
    EP17305233.3A| EP3214601B1|2016-03-04|2017-03-06|Image enhancement method applicable to fingerprint images|
    CA2960247A| CA2960247A1|2016-03-04|2017-03-06|Image improvement device applicable to fingerprint images|
    [返回顶部]