WO1999046737A1 - Verfahren zum verifizieren der authentizität eines im rahmen einer personenerkennung aufgenommenen bildes - Google Patents
Verfahren zum verifizieren der authentizität eines im rahmen einer personenerkennung aufgenommenen bildes Download PDFInfo
- Publication number
- WO1999046737A1 WO1999046737A1 PCT/EP1999/001600 EP9901600W WO9946737A1 WO 1999046737 A1 WO1999046737 A1 WO 1999046737A1 EP 9901600 W EP9901600 W EP 9901600W WO 9946737 A1 WO9946737 A1 WO 9946737A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- head
- determined
- difference image
- function
- individual images
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C9/00—Individual registration on entry or exit
- G07C9/30—Individual registration on entry or exit not involving the use of a pass
- G07C9/32—Individual registration on entry or exit not involving the use of a pass in combination with an identity check
- G07C9/37—Individual registration on entry or exit not involving the use of a pass in combination with an identity check using biometric data, e.g. fingerprints, iris scans or voice recognition
-
- 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
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
Definitions
- the invention relates to a method for verifying the authenticity of an image of a person to be subjected to person identification, which is recorded as part of a person identification.
- German utility model 2 950660 to restrict the spectral range of light used for person recognition to an area in which a photo or a mask-like image has reflectance characteristics that differ from those of an actually recorded one, by means of an IR filter disk Differentiate facial.
- a disadvantage of such an IR filter is, however, that it lowers the image quality and thus increases the demands on the quality of the detection method.
- the three-dimensional structure of a face that is completely or partially covered are laser triangulation or the so-called shape-from-shading methods, as disclosed in Attick, Griffin and Redlich in Neural Computation 8, 1321-1340, 1996, which allow conclusions to be drawn about depth information of the object under consideration from the gray value curve for a given direction of illumination.
- these methods can be used to detect delusions in which an image is held planar in front of the recording system, delusions caused by a cylindrically curved image or by a three-dimensional mask cannot be recognized as a delusion.
- thermography processes methods are known in the prior art in which the images of the person to be recognized are recorded by means of thermography processes, see for example the German patent 4 009 051. In this case, the heat radiation emitted by the person to be recognized is detected and identified with a predetermined reference compared.
- a disadvantage of these methods is that the devices for carrying out these methods, in particular the thermographic cameras for recording the person to be recognized, are very expensive. Accordingly, thermography processes cannot be used in access controls that are to be used in many places, such as, for example, in automated teller machines for withdrawing money and / or carrying out banking transactions.
- the object underlying the invention is to create a method for verifying the authenticity of an image of a person to be subjected to person recognition, which is recorded as part of a person identification, in which on the one hand the reliability verification of authenticity is increased compared to the prior art and thus the security to be achieved by an access control is increased, and which, on the other hand, can be implemented inexpensively.
- a method for verifying the authenticity of an image recorded as part of a person identification of a person to be subjected to person identification which comprises the steps of recording a sequence of chronologically successive individual pictures of the person, and determining the authenticity of the picture taken, if from at least two successive individual images of the sequence of intrinsic movements of the person are determined. Since in the method according to the invention intrinsic movements of a person to be recognized, i.e. movements which each and thus a person to be recognized consciously and unconsciously executes without changing their focus, are used to verify the authenticity of the images taken, the authenticity is only established if the consecutively recorded individual images actually come from a person who carries out intrinsic movements, that is to say from a living person. Accordingly, it is no longer possible to deceive an access control using the method according to the invention by holding a cylindrically bent photo or a mask-like image of the person to be recognized, as a result of which the security guaranteed by the access control is considerably increased.
- the method can be implemented relatively inexpensively. Consequently, the method according to the invention for verifying the authenticity of a recorded image is also suitable for use in access controls which are to be used in many places, such as in automated teller machines for withdrawing money and / or for carrying out banking transactions.
- At least one difference image of two successive individual images of the sequence can be evaluated to determine the intrinsic movements.
- the amount of data to be processed in the method according to the invention can be considerably reduced, as a result of which the method can be carried out more quickly.
- a difference image can be evaluated to determine the intrinsic movements, which results from an AND operation of two successive difference images.
- each difference image can be binarized before the evaluation.
- This measure in turn can greatly reduce the data to be processed, which leads to an increase in the speed of the method and thus to lower hardware requirements with the same performance.
- the binarization can be carried out, for example, by means of a threshold value which is determined by evaluating the background of one of the difference images, for example by averaging the intensities of those pixels in the difference image which are in the area of the static background.
- At least one section of increased movement can be selected in each difference image, and intrinsic movements can be determined if this at least one section in the difference image is consistent with a predetermined range in which intrinsic movements are to be expected.
- This configuration of the method according to the invention can in turn reduce the amount of data to be processed, which leads to the advantages already discussed above.
- the section of increased movement can be defined by a vertical and a horizontal extent, the vertical extent being essentially a peak of the function of the products from the horizontal is determined with the horizontal variances of the difference image, and the horizontal extent is essentially determined by a peak of the function of the products from the vertical projections with the vertical variances of the difference image.
- each function can advantageously be smoothed with a low-pass filter before the corresponding peak is determined.
- This measure avoids high-frequency noise in the difference images, which under certain circumstances leads to a poorer determination of the peaks and thus of the section of increased movement. Overall, this results in a more robust determination of the section of increased movement.
- sections can be selected from the difference image, which are expected to have a symmetry relationship to one another, and intrinsic movements can be determined if the symmetry relationship is verified by a correlation analysis of the corresponding sections. Since intrinsic movements, for example in the face of a person to be recognized, are generally symmetrical, this configuration provides a check as to whether the intrinsic movement detected in a specific area is actually an intrinsic movement or just another type of movement, a noise or the like.
- a further check can be carried out to determine whether the increased movement in the area of increased movement is distributed unevenly over this area.
- a function dependent on the vertical position which for a given vertical position is determined by the product of the horizontal projection with the horizontal variance of the difference image, and a function dependent on the horizontal position, can also be used to determine intrinsic movements 6
- each function can preferably be smoothed with a low-pass filter before the corresponding peak is determined, in order to filter high-frequency noise from the difference images.
- At least one peak can be determined in the function dependent on the vertical position and at least one peak in the function dependent on the horizontal position can be determined for evaluating the functions described above, and intrinsic movements can finally be determined if these Peaks in the vertical or horizontal direction lie within predetermined limits, which are given by the vertical or the horizontal extent of at least one area in which intrinsic movements are to be expected.
- sections can be evaluated which are expected to be in a symmetrical relationship to one another, and intrinsic movements can ultimately be determined if by a correlation analysis of the corresponding sections the symmetry relationship is verified.
- the sections can be normalized with respect to one another prior to the correlation analysis.
- the function dependent on the vertical position and / or the function dependent on the horizontal position can be evaluated, and intrinsic movements can be determined if it is determined by a correlation analysis that the values of the vertical position 7
- intrinsic movements in the head region of the person to be recognized can be determined to verify the authenticity of the images taken.
- the head area contains most and the most dominant intrinsic movements. Accordingly, the method according to the invention can be carried out most effectively in the head region of a person to be recognized.
- the mouth region and / or the cheek region and / or the nose region and / or the eye region can be cited as areas with particularly pronounced intrinsic movements in the head region.
- the intrinsic movements to be determined can be intrinsic movements in at least two mutually symmetrical regions in the head region of the person to be recognized, and these at least two mutually symmetrical regions can be used to determine the axis of symmetry of the head region.
- an image that can be used for person recognition can be generated, in which a head region rotated in frontoparallel is transformed into a head region with a rotated axis of symmetry.
- the symmetry axis found can also be used to generate an image of the person to be recognized which can be used for person recognition and which is composed of the left head area and the mirrored left head area, 8th
- images can also be used in the case of image recognition in which the head region of the person to be recognized is inclined with respect to the vertical.
- images can also be used for personal identification in which one half of the image is outshone and which would therefore be unusable for personal identification.
- the areas in the individual images in which intrinsic movements are expected can be extracted from the individual images before the authenticity is determined.
- This development has the advantage that the images required to determine the authenticity can be extracted from any video sequences.
- At least one difference image of two successive individual images of the sequence can advantageously be evaluated.
- each difference image can be binarized before the evaluation and the binarization can be carried out here, for example, by means of a threshold value which is determined by evaluating the background of one of the difference images, for example by averaging the intensities of those pixels in the difference image which are in the area of the static background .
- the head region of the person to be recognized can advantageously be extracted.
- the extracted head area can be transformed to a predetermined standard size for person recognition.
- size variations can be taken into account, which can lead to problems in the complex face recognition algorithms, in particular with regard to the computing time.
- At least two head boundaries can be determined in the corresponding individual images or difference images, on the basis of which the head region is extracted from the corresponding individual images or the difference images.
- the determination of the head boundary in the corresponding individual images or difference images can include the determination of the upper and the left head boundary.
- the right head boundary can also be determined.
- each head boundary may be defined by a head boundary line that is such that the outline of the head is substantially entirely within the head boundary lines.
- the function of the vertical projections of the binarized difference image can be determined to determine the upper head limit, and the upper head limit can be defined by the first maximum of the amount of the first derivative of this function that lies above a predetermined threshold value.
- the function of the vertical projections can be smoothed with a low-pass filter before defining the head boundary.
- the function of the horizontal projections of the binarized difference image can be determined, and the left head boundary can be defined by the first maximum of the amount of the first derivative of this function, which lies above a predetermined threshold value.
- the difference image can also be divided into a plurality of stripes which follow one another in the vertical direction, the first stripe following the determined upper head boundary in the vertical direction, in each stripe the function of the horizontal projections of the binarized difference image can then be performed can then be determined, the amounts of the first derivatives of the plurality of functions of the horizontal projections obtained in this way can be formed, the sum of the plurality of amounts formed in this way can be added, and finally the left head boundary can be the first maximum of this sum, that lies above a predetermined threshold value.
- the function or the functions of the horizontal projections can optionally be smoothed with a low-pass filter before the head boundary is defined.
- the function of the horizontal projections of the binarized difference image can be determined to determine the right head boundary, and the right head boundary by the last maximum of the Amount of the first derivative of this function, which lies above a predetermined threshold value.
- the difference image can be divided into a plurality of stripes successively in the vertical direction in order to determine the right head boundary, the first stripe being in the vertical direction Direction downwards to the determined upper head limit, the function of the horizontal projections of the binarized difference image are determined in each stripe, the amounts of the first derivatives of the plurality of functions of the horizontal projections obtained in this way are formed, the sum of the plurality formed in this way Amounts are added, and the right head boundary is defined as the last maximum of this sum, which is above a predetermined threshold.
- determining the head boundaries when determining the head boundaries, it can be provided that two successive individual images or differential images obtained therefrom are only used to determine the head boundaries if there is a change between the two successive individual images in a predetermined range.
- a movement strength can be calculated to determine the change between two successive individual images.
- the movement intensity can essentially be calculated by the sum of the gray levels of the difference image.
- the amount of movement can expediently be calculated as the sum of the 1 pixels or O pixels of the binarized difference image.
- predetermined head boundaries can be used to extract the head region.
- predetermined head boundaries can be used to extract the head region if one of the determined head boundaries is not within predetermined boundaries.
- the step for determining the head boundaries can be carried out with subsequent difference images.
- the limits determined in the method can also be checked for plausibility. In other words, it is checked whether the head of a person would actually be expected in the area defined by these head boundaries with regard to the arrangement of the device. If the determined head boundaries are not plausible, predetermined head boundaries are finally assumed. This additional check can further increase the quality of the determination of the head boundary and thus ultimately the quality of the verification of the authenticity.
- the lower or the right and lower head limits can expediently be determined in the configurations described above for determining the head boundaries in such a way that a square head region is extracted from the corresponding individual images or difference images.
- This training can be used independently of and as an alternative to the extraction process for areas in which intrinsic movements are expected. Accordingly, all the advantages mentioned in connection with the determination of the head limits can be mentioned here as an advantage.
- this development can also be used to increase the accuracy of the areas determined during the extraction with intrinsic movements.
- the stabilization can be carried out by means of a correlation correction.
- the correlation correction can be carried out by template matching, which comprises calculating the difference image, the first image used to form the difference image being shifted from the second image used to form the difference image when calculating the difference image, that a correlation function between the two individual images is maximal.
- two successive individual images or differential images obtained therefrom are used in the previously described methods for determining the intrinsic movements only if the change between the two successive individual images lies in a predetermined range.
- the change between two successive individual images can preferably be calculated by means of a movement strength.
- the amount of movement for a grayscale image can be calculated from the sum of the gray levels.
- the movement strength can be expediently calculated by the sum of the 1 pixels or O pixels of the binarized difference image.
- FIG. 1 shows a flowchart of a first embodiment of the method for verifying the authenticity of an image recorded as part of a person recognition according to the present invention
- FIG. 2 shows a flowchart of a second embodiment of the method for verifying the authenticity according to the present invention
- FIG. 3 shows a flowchart of a third embodiment of the method for verifying the authenticity according to the present invention
- 4 shows a flowchart of a fourth embodiment of the method for verifying the authenticity according to the present invention
- 5 shows a flowchart of a fifth embodiment of the method for verifying the authenticity according to the present invention
- FIG. 6 shows a flowchart of a sixth embodiment of the method for verifying the authenticity according to the present invention
- FIG. 7 shows a flowchart for performing a correlation analysis which can be used in the fourth or fifth embodiment of the method for verifying the authenticity according to the present invention
- FIG. 8 shows a flowchart of a seventh embodiment of the method for verifying the authenticity according to the present invention.
- FIG. 9 shows a flowchart of an eighth embodiment of the method for verifying the authenticity according to the present invention.
- FIG. 1 shows a flowchart of a method according to the invention for verifying the authenticity of an image of a person to be subjected to person identification as part of a person identification.
- video cameras known from the prior art which record the sequence in digitized form, are expediently used.
- Individual images that are recorded with such a video camera are generally in the form of pixels, each individual image comprising c x r pixels if c denotes the number of columns and r the number of lines of the relevant single image.
- K (t) therefore denotes the entire individual image at time t and Kc r (t) the pixel in the c-th column and the r-th row.
- the individual images are recorded at a speed of 8-12 images / second. 16
- step S120 two chronologically successive individual images K (t-1) and K (t) are selected from the recorded sequence of the individual images.
- the individual images K (t-1) and K (t) are compared with one another in order to determine whether the person subjected to the person recognition has carried out intrinsic movements between the two individual images K (t-1) and K (t).
- the two individual images K (t-1) and K (t) can be compared with one another pixel by pixel, and the change resulting from this pixel comparison can be compared with a predetermined threshold value. Accordingly, if the amount of change is above the predetermined threshold value, the person has carried out an intrinsic movement between the two individual images K (t-1) and K (t), but if the amount of change is below the threshold value, no intrinsic movements of the person could be determined become.
- This pixel-by-pixel comparison leads to satisfactory results if no additional movements are superimposed on the intrinsic movements of the person, i.e. if the person does not move or only moves minimally with respect to the receiving device.
- the amount of the difference between the gray levels of two corresponding pixels in the individual image can be K (t-1) and K (t) as a measure of the change in a pixel, and the sum of the amounts of these differences as a measure for the change between the two individual images K (t-1) and K (t), which is finally compared with the predetermined threshold value.
- the change between two individual images results from the sum of the pixels corresponding to one another in the individual images K (t-1) and K (t) differentiate from each other. In the case of binarized individual images, this change must therefore be compared with a predefined threshold value.
- the binarization of grayscale images is described in detail in connection with the second embodiment (step S240). If the comparison in step S130 finally shows that intrinsic movements of the person between the two individual images have taken place, the authenticity of the recorded individual images K (t-1) and K (t) is verified in step S170.
- step S180 If, on the other hand, the comparison shows that the person has not carried out any intrinsic movements between the individual images K (t-1) and K (t), the authenticity of the recorded individual images K (t-1) and K (t) cannot be verified in step S180 become.
- step S190 there is a possibility to cancel the procedure, and if this procedure is used as part of an access control, access can be refused.
- step S190 one can jump back to step S120 and the method for verifying the authenticity can be carried out again with two further individual images, for example K (t) and K (t + 1).
- the loop represented by step S190 may be run j times, where j is a predetermined number that is at most n-1.
- a difference image D (t) K (t) - K (t-1) can be formed from the two individual images to reduce the amount of data to be processed, and finally the difference image D (t) can be examined directly to determine whether the person has carried out intrinsic movements between the two individual images K (t-1) and K (t).
- a UN D linkage of two successive difference images D (t) and D (t-1) can be evaluated to determine the intrinsic movements.
- This alternative is particularly suitable when there are strong changes in two successive individual images due to the release of a dominant background, for example an illumination source, in two successive individual images and are reflected in the difference image. Accordingly, this alternative can be used to free the difference image that is ultimately to be assessed from changes that do not result from intrinsic movements. lo
- FIG. 2 shows a flow chart of a second embodiment of the method according to the invention.
- This second embodiment is essentially an alternative to step S130 in the first embodiment.
- steps S110 and S120 are also first carried out, i.e. a sequence of consecutive individual images is recorded and two consecutive individual images are selected from this sequence.
- step S120 the difference image D (t) is determined in step S130 from the individual images K (t) and K (t-1).
- this difference image can be binarized according to the following equation:
- ⁇ represents a predetermined threshold
- the threshold value ⁇ can be predefined or determined by evaluating the background.
- the threshold value ⁇ is the maximum value found in the top m lines of the difference image, ie
- At least one area of increased movement is determined in the difference image D (t) or in the binarized difference image B (t).
- a frame of any size can be specified and guided over the corresponding difference image, with the movement, i.e. the number of the sum of the pixels with the value 1 or alternatively the pixels with the value 0 is calculated.
- the frame of the shift with the largest sum finally defines the area of increased movement.
- step S170 the authenticity of the recorded individual images K (t-1) and K (t) on which the difference image D (t) and the binarized difference image B (t) are based can be verified in step S170.
- step S180 If the area of increased movement does not match the predetermined area, the authenticity of the recorded individual images cannot be verified in step S180, and the method can be terminated or repeated in accordance with step S190 (see FIG. 1).
- the area where most of a person's intrinsic movements are expected is the head area.
- the head area as such, most intrinsic movements in the mouth region, the cheek region, the nose region or the eye region can be expected.
- the second embodiment in FIG. therefore at least one determined area of increased movement compared with the mouth region, the cheek region, the nose region and / or the eye region.
- step S250 shows a flowchart of a third embodiment of the method according to the invention, which specifies an alternative to step S250, that is to say the determination of the area of increased movement, in the second embodiment and otherwise corresponds to the second embodiment.
- difference image D (t) or, alternatively, a binarized difference image B (t) is assumed.
- the difference image D (t) and the binarized difference image B (t) are referred to as difference image K in the following.
- step S350 the function Y of the products from the horizontal projections P H and the horizontal variances V H is determined from the difference image K, in formulas:
- step S352 the functions Y v and Y H obtained according to the formulas above can be smoothed using a suitable low-pass filter TPF ⁇ .
- Peaks are identified in the functions Y v and Y H or the smoothed functions TPF ⁇ Y ⁇ and TPF ⁇ Y H ⁇ in step S353.
- the peaks of Y v and Y H define the vertical and horizontal extent and position of the area of increased movement.
- the position of a peak can be specified by the maximum value and the extent of the corresponding area by its half-value width.
- the extent can also be indicated by the turning points of the peak.
- the extent can also be specified by the limits which lead to a predetermined area under the peak.
- the methods customary in the prior art can be used to find the individual peaks. These methods are based essentially on searching for the sections of a function that exceed a predetermined threshold.
- a further area of increased movement which is expected to be in a symmetrical relationship with the first area of increased movement, is determined in step S461.
- step S250 second embodiment
- steps S350 to S353 third embodiment
- the first and second areas in the fourth embodiment are, for example, the mouth region in the left and right half of the face, the cheek region in the left and right half of the face, the nose region in the left and right half of the face as well as the left and right eyes.
- a correlation analysis can be used to verify whether the symmetry relationship is fulfilled or not.
- the fourth embodiment thus provides a plausibility check of the determined areas of increased movement. If it is verified in step S462 by a correlation analysis that the symmetry relationship is fulfilled, the authenticity of the recorded individual images can be verified in step S170.
- step S180 the authenticity cannot be verified (see step S180), and accordingly the method can be terminated with this result, or - as already discussed - carried out repeatedly.
- FIG. 5 shows the flowchart of a fifth embodiment of the method according to the invention.
- the steps S530 to S565 shown in FIG. 5 represent a preferred embodiment of step S130 according to the first embodiment.
- a sequence of chronologically consecutive individual images of a person is first recorded, and then two consecutive individual images are selected from this sequence. This is represented in FIG. 5 by step S120.
- step S530 a difference image is determined in step S530, which can optionally be binarized in step S540 (steps S530 and S540 correspond to steps S230 and S240 in the second embodiment, so that a description and the corresponding description are omitted here of steps S230 and S240).
- the vertical position and the extent as well as the horizontal position and the extent of the mouth region of a person to be recognized are determined from the difference image in step S550.
- step S550 in the difference 24 In contrast to step S250 in the second embodiment, in which areas of increased movements are determined in the difference image and it is checked whether these areas can be assigned to specific regions, in step S550 in the difference 24
- border image for a specific area in which intrinsic movements are expected, namely for the mouth region.
- step S560 It is then checked in step S560 whether or not the determined mouth region lies within predetermined limits. If the mouth region is within predetermined limits, the authenticity of the captured individual images can be verified in step S170; if the determined mouth region is not within predetermined limits, the verification of the authenticity of the individual images must be denied.
- step S565 as an alternative to step S560, a correlation analysis can be carried out with the determined mouth region to determine intrinsic movements.
- step S170 the authenticity of the individual images is only determined if the determined mouth region lies within predetermined limits and a presence of intrinsic movements follows from a correlation analysis of the mouth region.
- the fifth embodiment specifically relates to the mouth region, it is of course possible to carry out the method with other regions in which intrinsic movements are expected or with combinations of such regions.
- FIG. 6 shows the flowchart of a sixth embodiment of the method according to the invention.
- This sixth embodiment represents a preferred embodiment of step S550 in the fifth embodiment for determining the vertical position and extent and the horizontal position and extent of the mouth region from the difference image. Accordingly, it is assumed that according to step 530 or a difference image D (t) or a binarized difference image B (t) is present according to S540. 25th
- the function Y v can optionally be smoothed by a low-pass filter.
- the vertical position and the vertical extension of the mouth region are determined from this function or the smoothed function in step S651.
- the vertical mouth position is defined here as the position of the last (lowest) maximum of Y, which is greater than a predetermined threshold value. In the case of the mouth region, the double mean of Y has been found to be a suitable threshold.
- the width of the peak at which the peak has dropped to a predetermined percentage of its maximum value can be specified as the vertical extent of the mouth area.
- 50% have proven to be a favorable value.
- the vertical extent of the mouth area can also be specified as the width of the peak, which results from the distance between the two closest turning points to the left and right of the peak position.
- the plausibility of the values obtained for the vertical position and the vertical extent can be checked.
- a minimum mouth position and a maximum mouth position can be specified and it can be checked whether the determined mouth position lies within these two extreme values.
- the same procedure can alternatively or cumulatively be carried out for the vertical expansion of the mouth area.
- step 652 the function Y H; t , b of the products from the vertical projections P ; t , b and the vertical variances V ; t, b in a horizontal stripe that extends from the upper mouth limit t determined in step 651 to the lower mouth limit b, that is 26
- the function Y H; tb can also be optionally smoothed.
- the left horizontal boundary of the mouth region is then defined as the first maximum of b H , which is greater than a predetermined threshold.
- the right horizontal border of the mouth region is accordingly defined as the last maximum of b H that is greater than this threshold.
- the mean value of b H was used as the threshold value.
- a plausibility check can be carried out in such a way that the determined mouth region must lie within a predetermined maximum range so that the area of increased movement is regarded as the mouth area.
- FIG. 7 shows a flowchart of a correlation analysis, as can be carried out, for example, in step S462 according to the fourth embodiment or in step S565 according to the fifth embodiment. 27
- step S766 the function Y v and / or Y H is first broken down into sections which are expected to be in a symmetrical relationship to one another.
- step S767 a correlation analysis is used to check whether the symmetry relationship assumed in step S766 is fulfilled.
- Steps S766 and S767 can of course also be carried out for several symmetry relationships.
- step S170 If it can be verified with the correlation analysis that the symmetry relationship is fulfilled or the symmetry relationships are fulfilled, the authenticity of the recorded individual images can be verified in accordance with step S170.
- step S767 For the case of determining intrinsic movements in the mouth region already discussed above, the following correlations can be checked in step S767.
- the curve Y H is expediently transformed to standardized variables before the correlation analysis, so that the function Y H has the mean value 0 and the standard deviation.
- n c (Y H ) for indices c ⁇ 0 and c> n c is continued with 0.
- the correlation of the image section from the difference image of the mirrored mouth with the mouth itself can also be checked. If this correlation is less than a predetermined value, it can be assumed that it was not an actual intrinsic movement that took place, but that the movement determined is due to movements, noise or the like.
- step S768 the function Y and / or Y H can be broken down into predetermined sections in step S768. However, this step is only intended for the case that certain areas of the corresponding function are to be examined. If the function Y v and / or Y H is to be examined in its entirety, step S768 can accordingly be omitted.
- step S769 a correlation analysis is used to check whether the values of the corresponding function are substantially non-uniform in their definition area (or the predetermined sections, if step S768 is carried out).
- step S769 In the case of the mouth area, correlations between the vertical projections and vertical variances for the right and left sides of the mouth can be checked.
- the detected movement is an intrinsic movement.
- the mouth area is shown as an exemplary embodiment of the method according to the invention.
- the method according to the invention is not restricted to the mouth region.
- any areas in which intrinsic movements are to be expected such as the cheek region, the nose region and the eye region or any combination of these regions, are possible.
- FIG 8 shows the flowchart of an eighth embodiment of the method according to the invention.
- the steps of the eighth embodiment serve to optimize the individual images after the recording for the determination of the intrinsic movements.
- step S120 as shown in FIG. 8, after a sequence of time-sequential frames has been captured in step S110, two time-sequential frames are selected.
- step S825 it is first checked whether the change between the two successive individual images lies in a predetermined range or not.
- the fact that the change between two successive individual images must be less than a predetermined value ensures that differential images, in which too much movement is reflected, such as, for example, due to wobbling of the recording device or extremely rapid movement of the person to be recognized are not used to determine the intrinsic movement, since in this case the separation between an actual movement and an intrinsic movement can no longer be carried out with sufficient certainty.
- Step S825 can be carried out by calculating a movement strength to determine the change between two successive frames.
- the amount of movement can essentially be calculated by the sum of the gray levels of the difference image.
- the movement strength is expediently calculated by the sum of the 1-pixels or O-pixels of the binarized difference image.
- step S826 which can be carried out as an alternative to or cumulatively with step S825, the areas in the individual images in which intrinsic movements are expected are extracted from the corresponding individual images before the individual images are compared with one another. This can significantly reduce the amount of data to be processed.
- a predetermined section can be extracted for the extraction in the individual images, in which experience has shown that the head of the person to be recognized comes to rest.
- the head area of the person to be recognized is determined by determining the head boundaries from a difference image. With these determined head boundaries, the head region can then finally be extracted from the respective individual images.
- step S827 the individual images can be stabilized in step S827.
- two or more individual images can be stabilized with respect to one another in step S827, or in other words, aligned with one another.
- FIG. 9 is a flowchart of the alternative for determining the head limits already mentioned, by means of which the head region in which the most intrinsic movements are expected can be extracted from individual images K (t-1) and K (t).
- the difference image D (t) is first determined in step S921.
- step S922 the upper head limit is then determined from the difference image D (t).
- the horizontal projection s H (t) is first followed by
- the function s H (t) can be smoothed by filtering with a suitable low-pass filter TPF.
- the amount of the first derivative is then formed from the function s H (t) or the smoothed function TPF ⁇ s H (t) ⁇
- the position of the first maximum of b H which is greater than a predetermined threshold value, here half the mean value of b H , is now determined as the upper head limit.
- the calculation of the upper head limit can be repeated or set to a predetermined value, which has been determined, for example, for the individual images K (t-2) and K (t-1). 34
- the left and right head boundaries are determined in step S923.
- the difference image can be divided into q successive strips that extend downwards in the vertical direction, starting from the upper head boundary t.
- the function s ⁇ q) (t) can be smoothed by filtering with a suitable low-pass filter TPF.
- the amount of the first derivative of the function s ( ⁇ (t) or the smoothed function TPF ⁇ s ⁇ q) (t) ⁇ is formed and then summed up via q:
- This function b v determines all maxima that are greater than a predetermined threshold value, in the present case half the mean value b.
- the first maximum (the leftmost maximum) is defined as the position of the left head boundary.
- the last maximum (the rightmost maximum) is defined as the right head boundary.
- the determination for the left and the right head boundary can be determined for the next difference image.
- the left and right head boundaries can also be replaced by predetermined values which have resulted, for example, from the head boundary determination with the aid of the difference image D (t-1).
- the lower head limit is finally determined with the aid of the upper left and right head limits. For this purpose, the width of the extracted area is determined from the difference between the right head boundary and the left head boundary. The lower head limit can then be set so that there is an overall square area.
- the lower and the right head boundaries can then be calculated by a predefined width and a predefined height, which typically corresponds to the head region in the individual images.
- the head boundaries can also be determined from a difference image which has been obtained as an AND operation from two successive difference images.
- step S825 Since the same procedures as explained in connection with step S825 can be used for this purpose, there is no need for a description here and reference is only made to the description in connection with step S825.
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
- Collating Specific Patterns (AREA)
Abstract
Description
Claims
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP99911778A EP1062640B1 (de) | 1998-03-12 | 1999-03-11 | Verfahren zum verifizieren der authentizität eines im rahmen einer personenerkennung aufgenommenen bildes |
US09/623,985 US6922478B1 (en) | 1998-03-12 | 1999-03-11 | Method for verifying the authenticity of an image recorded in a person identifying process |
DE59911699T DE59911699D1 (de) | 1998-03-12 | 1999-03-11 | Verfahren zum verifizieren der authentizität eines im rahmen einer personenerkennung aufgenommenen bildes |
JP2000536043A JP2002507035A (ja) | 1998-03-12 | 1999-03-11 | 個人識別のために記録された画像の正当性を認証する方法 |
CA002323867A CA2323867A1 (en) | 1998-03-12 | 1999-03-11 | Method for verifying the authenticity of an image recorded in a person identifying process |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE19810792.7 | 1998-03-12 | ||
DE19810792A DE19810792A1 (de) | 1998-03-12 | 1998-03-12 | Verfahren zum Verifizieren der Authentizität eines im Rahmen einer Personenerkennung aufgenommenen Bildes |
Publications (1)
Publication Number | Publication Date |
---|---|
WO1999046737A1 true WO1999046737A1 (de) | 1999-09-16 |
Family
ID=7860678
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/EP1999/001600 WO1999046737A1 (de) | 1998-03-12 | 1999-03-11 | Verfahren zum verifizieren der authentizität eines im rahmen einer personenerkennung aufgenommenen bildes |
Country Status (6)
Country | Link |
---|---|
US (1) | US6922478B1 (de) |
EP (1) | EP1062640B1 (de) |
JP (1) | JP2002507035A (de) |
CA (1) | CA2323867A1 (de) |
DE (2) | DE19810792A1 (de) |
WO (1) | WO1999046737A1 (de) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2003010719A2 (en) * | 2001-07-24 | 2003-02-06 | Memco Limited | Door or access control system |
Families Citing this family (27)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
IL152756A0 (en) * | 2000-05-16 | 2003-06-24 | Swisscom Mobile Ag | Biometric identification and authentication method |
US7039224B2 (en) * | 2002-04-29 | 2006-05-02 | Activcard Ireland Limited | Method and device for preventing false acceptance of latent fingerprint images |
JP4013684B2 (ja) * | 2002-07-23 | 2007-11-28 | オムロン株式会社 | 個人認証システムにおける不正登録防止装置 |
KR100480781B1 (ko) * | 2002-12-28 | 2005-04-06 | 삼성전자주식회사 | 치아영상으로부터 치아영역 추출방법 및 치아영상을이용한 신원확인방법 및 장치 |
CN1860490A (zh) * | 2003-09-30 | 2006-11-08 | 皇家飞利浦电子股份有限公司 | 自适应设定生物统计测量阈值的系统和方法 |
US7503488B2 (en) * | 2003-10-17 | 2009-03-17 | Davis Bruce L | Fraud prevention in issuance of identification credentials |
US7225977B2 (en) * | 2003-10-17 | 2007-06-05 | Digimarc Corporation | Fraud deterrence in connection with identity documents |
JP4085959B2 (ja) * | 2003-11-14 | 2008-05-14 | コニカミノルタホールディングス株式会社 | 物体検出装置、物体検出方法、および記録媒体 |
US7356165B2 (en) * | 2004-03-09 | 2008-04-08 | Matsushita Electric Industrial Co., Ltd. | Imaging apparatus performing a high reliability data analysis |
US7386369B1 (en) * | 2004-08-09 | 2008-06-10 | Graftech International Holdings Inc. | Digital electrode observation |
JP4516516B2 (ja) * | 2005-12-07 | 2010-08-04 | 本田技研工業株式会社 | 人物検出装置、人物検出方法及び人物検出プログラム |
JP4924603B2 (ja) * | 2006-03-01 | 2012-04-25 | 日本電気株式会社 | 顔認証装置、顔認証方法およびプログラム |
JP4816321B2 (ja) * | 2006-08-14 | 2011-11-16 | ソニー株式会社 | 認証装置及び認証方法並びにプログラム |
JP2009088801A (ja) * | 2007-09-28 | 2009-04-23 | Hitachi Software Eng Co Ltd | 撮影対象物の静止判定方法および装置 |
JP5020920B2 (ja) * | 2008-10-29 | 2012-09-05 | 株式会社豊田中央研究所 | 歩行者検出装置及びプログラム |
DE102008055884A1 (de) | 2008-11-03 | 2010-05-06 | Cross Match Technologies Gmbh | Verfahren und Vorrichtung zur Detektion einer zweidimensionalen Darstellung des Gesichtes einer Person |
WO2010137157A1 (ja) * | 2009-05-28 | 2010-12-02 | 株式会社東芝 | 画像処理装置、方法、プログラム |
JP4609805B2 (ja) * | 2009-05-28 | 2011-01-12 | Necカシオモバイルコミュニケーションズ株式会社 | 端末装置及びプログラム |
US20120076368A1 (en) * | 2010-09-27 | 2012-03-29 | David Staudacher | Face identification based on facial feature changes |
US9665763B2 (en) * | 2014-08-31 | 2017-05-30 | Qualcomm Incorporated | Finger/non-finger determination for biometric sensors |
US9582705B2 (en) | 2014-08-31 | 2017-02-28 | Qualcomm Incorporated | Layered filtering for biometric sensors |
US9195879B1 (en) | 2014-08-31 | 2015-11-24 | Qualcomm Incorporated | Air/object determination for biometric sensors |
US9619723B1 (en) | 2016-02-17 | 2017-04-11 | Hong Kong Applied Science and Technology Research Institute Company Limited | Method and system of identification and authentication using facial expression |
US10839251B2 (en) | 2017-06-26 | 2020-11-17 | Rank One Computing Corporation | Method and system for implementing image authentication for authenticating persons or items |
US10061996B1 (en) * | 2017-10-09 | 2018-08-28 | Hampen Technology Corporation Limited | Face recognition method and system for personal identification and authentication |
US11166077B2 (en) | 2018-12-20 | 2021-11-02 | Rovi Guides, Inc. | Systems and methods for displaying subjects of a video portion of content |
US20220350928A1 (en) * | 2019-10-17 | 2022-11-03 | Sony Group Corporation | Information processing system, information processing method, program, and user interface |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE2950660A1 (de) | 1979-12-15 | 1981-07-02 | Alfred Teves Gmbh, 6000 Frankfurt | Bremssattel fuer eine teilbelag-scheibenbremse |
DE4009051A1 (de) | 1990-03-21 | 1991-09-26 | Diehl Gmbh & Co | Biometrisches identifizierungs- und zugangs-kontrollsystem |
EP0582989A2 (de) * | 1992-08-11 | 1994-02-16 | Istituto Trentino Di Cultura | Erkennungssystem, insbesonders zum Erkennen von Personen |
DE4406020C1 (de) | 1994-02-24 | 1995-06-29 | Zentrum Fuer Neuroinformatik G | Verfahren zur automatisierten Erkennung von Objekten |
DE4413788C1 (de) * | 1994-03-15 | 1995-10-12 | Fraunhofer Ges Forschung | Personenidentifikation mit Bewegungsinformation |
US5570434A (en) * | 1990-09-05 | 1996-10-29 | U.S. Philips Corporation | Circuit arrangement for recognizing a human face |
US5625704A (en) * | 1994-11-10 | 1997-04-29 | Ricoh Corporation | Speaker recognition using spatiotemporal cues |
US5719951A (en) * | 1990-07-17 | 1998-02-17 | British Telecommunications Public Limited Company | Normalized image feature processing |
Family Cites Families (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE3633360A1 (de) * | 1986-10-01 | 1988-04-14 | Heinrich Prof Dr Ing Reents | Verfahren mit den dazu gehoerigen vorrichtungen zur absicherung unbefugter benutzung von zahlungskonten, raeumlichkeiten und geraeten |
US4843377A (en) * | 1987-04-21 | 1989-06-27 | Guardian Technologies, Inc. | Remote confinement system |
DE3807727A1 (de) * | 1988-03-09 | 1989-09-21 | Robot Foto Electr Kg | Ueberwachungsvorrichtung zum sichern von automaten |
US5150432A (en) * | 1990-03-26 | 1992-09-22 | Kabushiki Kaisha Toshiba | Apparatus for encoding/decoding video signals to improve quality of a specific region |
JPH0411466A (ja) * | 1990-04-28 | 1992-01-16 | Sony Corp | ノイズリデューサ |
JPH0546743A (ja) * | 1991-08-09 | 1993-02-26 | Matsushita Electric Ind Co Ltd | 個人識別装置 |
US6181805B1 (en) * | 1993-08-11 | 2001-01-30 | Nippon Telegraph & Telephone Corporation | Object image detecting method and system |
US6101264A (en) | 1994-03-15 | 2000-08-08 | Fraunhofer Gesellschaft Fuer Angewandte Forschung E.V. Et Al | Person identification based on movement information |
ES2110841T5 (es) | 1994-03-24 | 2005-12-16 | Minnesota Mining And Manufacturing Company | Sistema de autenticacion personal biometrica. |
US5995153A (en) * | 1995-11-02 | 1999-11-30 | Prime Image, Inc. | Video processing system with real time program duration compression and expansion |
US5761329A (en) * | 1995-12-15 | 1998-06-02 | Chen; Tsuhan | Method and apparatus employing audio and video data from an individual for authentication purposes |
DE19610066C1 (de) * | 1996-03-14 | 1997-09-18 | Siemens Nixdorf Advanced Techn | Verfahren zur Erfassung gesichtsbezogener Personendaten und deren Verwendung zur Identifizierung oder Verifikation von Personen |
DE19641000B8 (de) * | 1996-10-04 | 2005-08-25 | Viisage Technology Ag | Verfahren und Anordnung zur automatischen Erkennung der Anzahl von Personen in einer Personenschleuse |
US6678393B1 (en) * | 1997-12-23 | 2004-01-13 | Intel Corporation | Image selection based on image content |
US6421453B1 (en) * | 1998-05-15 | 2002-07-16 | International Business Machines Corporation | Apparatus and methods for user recognition employing behavioral passwords |
-
1998
- 1998-03-12 DE DE19810792A patent/DE19810792A1/de not_active Withdrawn
-
1999
- 1999-03-11 JP JP2000536043A patent/JP2002507035A/ja active Pending
- 1999-03-11 US US09/623,985 patent/US6922478B1/en not_active Expired - Lifetime
- 1999-03-11 CA CA002323867A patent/CA2323867A1/en not_active Abandoned
- 1999-03-11 DE DE59911699T patent/DE59911699D1/de not_active Expired - Lifetime
- 1999-03-11 WO PCT/EP1999/001600 patent/WO1999046737A1/de active IP Right Grant
- 1999-03-11 EP EP99911778A patent/EP1062640B1/de not_active Expired - Lifetime
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE2950660A1 (de) | 1979-12-15 | 1981-07-02 | Alfred Teves Gmbh, 6000 Frankfurt | Bremssattel fuer eine teilbelag-scheibenbremse |
DE4009051A1 (de) | 1990-03-21 | 1991-09-26 | Diehl Gmbh & Co | Biometrisches identifizierungs- und zugangs-kontrollsystem |
US5719951A (en) * | 1990-07-17 | 1998-02-17 | British Telecommunications Public Limited Company | Normalized image feature processing |
US5570434A (en) * | 1990-09-05 | 1996-10-29 | U.S. Philips Corporation | Circuit arrangement for recognizing a human face |
EP0582989A2 (de) * | 1992-08-11 | 1994-02-16 | Istituto Trentino Di Cultura | Erkennungssystem, insbesonders zum Erkennen von Personen |
DE4406020C1 (de) | 1994-02-24 | 1995-06-29 | Zentrum Fuer Neuroinformatik G | Verfahren zur automatisierten Erkennung von Objekten |
DE4413788C1 (de) * | 1994-03-15 | 1995-10-12 | Fraunhofer Ges Forschung | Personenidentifikation mit Bewegungsinformation |
US5625704A (en) * | 1994-11-10 | 1997-04-29 | Ricoh Corporation | Speaker recognition using spatiotemporal cues |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2003010719A2 (en) * | 2001-07-24 | 2003-02-06 | Memco Limited | Door or access control system |
WO2003010719A3 (en) * | 2001-07-24 | 2003-05-22 | Memco Ltd | Door or access control system |
Also Published As
Publication number | Publication date |
---|---|
EP1062640A1 (de) | 2000-12-27 |
EP1062640B1 (de) | 2005-03-02 |
US6922478B1 (en) | 2005-07-26 |
JP2002507035A (ja) | 2002-03-05 |
DE59911699D1 (de) | 2005-04-07 |
CA2323867A1 (en) | 1999-09-16 |
DE19810792A1 (de) | 1999-09-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP1062640B1 (de) | Verfahren zum verifizieren der authentizität eines im rahmen einer personenerkennung aufgenommenen bildes | |
DE60312427T2 (de) | Verfahren zum Überprüfen eines Benutzers und zum Aktualisieren einer Datenbank, und dieses Verfahren benutzendes Gesichtsprüfungssystem | |
DE19609455C2 (de) | Gesichtsbild-Verarbeitungssystem | |
EP1395945B1 (de) | Verfahren zur faelschungserkennung bei der fingerabdruckerkennung unter verwendung einer texturklassifikation von grauwertdifferenzbildern | |
DE60317025T2 (de) | Vorrichtung und Verfahren zur Gesichtserkennung | |
DE19614975C2 (de) | Gesichtsbildverarbeitungssystem | |
DE60016589T2 (de) | Verfahren und vorrichtung zur erzeugung eines zusammengesetzten fingerabdruckbildes | |
DE19613614C2 (de) | Vorrichtung zur Bearbeitung eines Bildes eines Gesichts | |
EP2584493B1 (de) | Verfahren zur Unterscheidung zwischen einem realen Gesicht und einer zweidimensionalen Abbildung des Gesichts in einem biometrischen Erfassungsprozess | |
EP1413972B1 (de) | Prüfung von Bildaufnahmen von Personen | |
DE102006045752B4 (de) | Vorrichtung zur Gesichtsbildauthentifizierung | |
DE112009000949T5 (de) | Detektion eines freien Fahrpfads für ein Fahrzeug | |
DE102014117102B4 (de) | Spurwechselwarnsystem und Verfahren zum Steuern des Spurwechselwarnsystems | |
WO2009019250A2 (de) | Verfahren und einrichtung zur objekterkennung in einem bild | |
EP3287951A1 (de) | Verfahren für die urheberschaftsbewertung eines gemäldes sowie eine entsprechende verwendung | |
WO2006133974A1 (de) | Verfahren und vorrichtung zur erkennung einer münze unter verwendung ihres prägebildes | |
EP2603905B1 (de) | Verfahren und vorrichtung zum erkennen und verifizieren von manipulationsversuchen an einem selbstbedienungsterminal | |
CH716082A1 (de) | Verfahren zur Benutzeridentifikation. | |
DE102019102051A1 (de) | Verfahren und system zur überwachung eines raumbereichs in einer personenschleuse | |
DE102011011931A1 (de) | Verfahren zum Auswerten einer Mehrzahl zeitlich versetzter Bilder, Vorrichtung zur Auswertung von Bildern, Überwachungssystem | |
DE10118485B4 (de) | Automatische Identifikation und/oder Verifikation linienhafter Texturen wie Fingerabdrücke | |
EP3629232A1 (de) | Vorrichtung zum erkennen einer bildmanipulation in einem digitalen referenzgesichtsbild einer person | |
EP3820145B1 (de) | Türkommunikationsanlage mit einer eine kamera aufweisenden türstation | |
DE102020211007A1 (de) | Bildverarbeitungseinrichtung und Bildverarbeitungsverfahren, und Bildverarbeitungsprogramm | |
DE10123367B4 (de) | Verfahren zur Fingerabdruckerkennung unter Verwendung von Bildsequenzen |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): CA JP US |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
ENP | Entry into the national phase |
Ref document number: 2323867 Country of ref document: CA Ref country code: CA Ref document number: 2323867 Kind code of ref document: A Format of ref document f/p: F |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1999911778 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 1999911778 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 09623985 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: CA |
|
WWG | Wipo information: grant in national office |
Ref document number: 1999911778 Country of ref document: EP |