WO2022059064A1 - フォーカス判定装置、虹彩認証装置、フォーカス判定方法、及び、記録媒体 - Google Patents
フォーカス判定装置、虹彩認証装置、フォーカス判定方法、及び、記録媒体 Download PDFInfo
- Publication number
- WO2022059064A1 WO2022059064A1 PCT/JP2020/034887 JP2020034887W WO2022059064A1 WO 2022059064 A1 WO2022059064 A1 WO 2022059064A1 JP 2020034887 W JP2020034887 W JP 2020034887W WO 2022059064 A1 WO2022059064 A1 WO 2022059064A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- iris
- image
- shape
- focus
- illumination
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims description 45
- 238000005286 illumination Methods 0.000 claims abstract description 126
- 238000001914 filtration Methods 0.000 claims abstract description 8
- 238000001514 detection method Methods 0.000 claims description 33
- 238000000605 extraction Methods 0.000 claims description 17
- 230000010354 integration Effects 0.000 description 27
- 238000010586 diagram Methods 0.000 description 7
- 230000015654 memory Effects 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 3
- 208000016339 iris pattern Diseases 0.000 description 3
- 239000000284 extract Substances 0.000 description 2
- 210000001747 pupil Anatomy 0.000 description 2
- 238000010187 selection method Methods 0.000 description 2
- 239000011521 glass Substances 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000001678 irradiating effect Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000003936 working memory Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/98—Detection or correction of errors, e.g. by rescanning the pattern or by human intervention; Evaluation of the quality of the acquired patterns
- G06V10/993—Evaluation of the quality of the acquired pattern
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/60—Extraction of image or video features relating to illumination properties, e.g. using a reflectance or lighting model
-
- 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
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/19—Sensors therefor
-
- 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
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/193—Preprocessing; Feature extraction
-
- 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
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/197—Matching; Classification
Definitions
- This disclosure relates to the processing of iris images.
- Iris recognition is known as one of the biometrics.
- iris authentication a user's iris pattern is registered in a database, and the iris pattern acquired from the user at the time of authentication is collated with the iris pattern registered in the database for authentication.
- Patent Document 1 describes a method of determining whether or not an iris image is in focus by using the edge intensity of the iris region.
- the focus is determined by using the edge of the iris region of the captured image, but if the captured image is not an image that is in focus to some extent, the edge of the iris region is correctly detected. And the focus cannot be determined.
- One object of this disclosure is to provide a focus determination device capable of accurately performing focus determination on a plurality of iris images.
- the focus determination device An acquisition method for acquiring an iris image of a user's iris, A filtering means for filtering the iris image by applying an image processing filter having a shape determined based on the shape of the illumination to be applied to the user. A focus determining means for determining whether or not the iris image is an in-focus image based on the filtered iris image is provided.
- the focus determination method is: Acquire an iris image of the user's iris, The iris image is filtered by applying an image processing filter having a shape determined based on the shape of the illumination to be applied to the user. Based on the filtered iris image, it is determined whether or not the iris image is an in-focus image.
- the recording medium is: Acquire an iris image of the user's iris, The iris image is filtered by applying an image processing filter having a shape determined based on the shape of the illumination to be applied to the user.
- a program for causing a computer to execute a process of determining whether or not the iris image is an in-focus image based on the filtered iris image is recorded.
- An example of an iris image photographing device is shown.
- An example of the focus determination device according to the first embodiment is shown.
- the hardware configuration of the focus determination device according to the first embodiment is shown.
- the functional configuration of the focus determination apparatus which concerns on 1st Embodiment is shown.
- An example of illumination and an example of a reflection area of illumination in an iris image are shown.
- An example of an image processing filter is shown. It is a flowchart of the focus determination process of 1st Embodiment.
- the functional configuration of the focus determination apparatus which concerns on 9th Embodiment is shown. It is a flowchart of the focus determination process of 9th Embodiment.
- the functional configuration of the iris recognition apparatus according to the tenth embodiment is shown.
- the functional configuration of the iris recognition apparatus according to the eleventh embodiment is shown. An example of selecting an iris image in the eleventh and twelfth embodiments is shown. The functional configuration of the iris recognition apparatus according to the thirteenth embodiment is shown. The functional configuration of the focus determination apparatus which concerns on 14th Embodiment is shown. It is a flowchart of the focus determination process of 14th Embodiment.
- FIG. 1 shows an example of an iris image photographing device.
- FIG. 1A is a side view of the user X and the photographing device 1
- FIG. 1B is a view of the user X and the photographing device 1 from above.
- the photographing device 1 includes a camera 2 and a lighting 3, and photographs an iris image of the user X.
- the illumination 3 is arranged in the vicinity of the camera 2.
- the illumination 3 is arranged on the front side of the camera 2.
- the illumination 3 uses a linear light source, but the shape of the illumination 3 is not limited to the linear shape. Further, the lighting 3 may be integrated with the camera 2 or may be installed independently of the camera 2.
- user X stands in front of camera 2. With the face of the user X illuminated by the illumination 3, the camera 2 takes a picture of the vicinity of the user's face.
- the shooting range of the camera 2 may be the entire face of the user X or only the vicinity of the eyes of the user X.
- the image of the eye region of the user X is used as the iris image. Therefore, when the shooting range of the camera 2 is wider than the eye region of the user X, the eye region of the captured image is cut out and used as an iris image.
- FIG. 2 shows an example of a focus determination device.
- the focus determination device 100 is connected to the camera 2 and acquires an iris image from the camera 2.
- the focus determination device 100 determines whether or not the acquired iris image is in focus (hereinafter, referred to as “focus determination”).
- focus determination The state in which the subject is in focus is also referred to as a "focused state", and the state in which the subject is out of focus is also referred to as a "non-focused state”.
- FIG. 3 is a block diagram showing a hardware configuration of the focus determination device 100.
- the focus determination device 100 includes an interface (IF) 11, a processor 12, a memory 13, a recording medium 14, and a database (DB) 15.
- IF interface
- DB database
- IF11 inputs and outputs data to and from an external device. Specifically, the iris image of the user X is input to the focus determination device 100 via the IF 11. Further, the determination result by the focus determination device 100 is output to the outside through the IF 11.
- the processor 12 is a computer such as a CPU (Central Processing Unit), and controls the entire focus determination device 100 by executing a program prepared in advance.
- the processor 12 may use one of a CPU, a GPU (Graphics Processing Unit), an FPGA (Field-Programmable Gate Array), a DSP (Demand-Side Platform), and an ASIC (Application Specific Circuit). A plurality of them may be used in parallel. Specifically, the processor 12 executes the focus determination process described later.
- the memory 13 is composed of a ROM (Read Only Memory), a RAM (Random Access Memory), and the like.
- the memory 13 is also used as a working memory during execution of various processes by the processor 12.
- the recording medium 14 is a non-volatile, non-temporary recording medium such as a disk-shaped recording medium or a semiconductor memory, and is configured to be removable from the focus determination device 100.
- the recording medium 14 records various programs executed by the processor 12. When the focus determination device 100 executes the focus determination process, the program recorded on the recording medium 14 is loaded into the memory 13 and executed by the processor 12.
- the database 15 temporarily stores the iris image input through the IF 11, the focus determination result by the focus determination device 100, and the like. If the memory 13 is sufficient for storing these data, the database 15 may be omitted. Further, the focus determination device 100 may include an input unit such as a keyboard and a mouse for an administrator or the like to give an instruction or input, and a display unit such as a liquid crystal display.
- FIG. 4 is a block diagram showing a functional configuration of the focus determination device 100.
- the focus determination device 100 functionally includes an iris image acquisition unit 101, a filter processing unit 102, a filter output integration unit 103, and a focus determination unit 104.
- the iris image acquisition unit 101 acquires an iris image from the camera 2 or the like shown in FIG.
- the iris image acquisition unit 101 may acquire the iris image from a database or recording medium that stores the captured iris image, instead of acquiring the iris image directly from the camera 2.
- the filter processing unit 102 executes the filter processing for the focus determination. Specifically, the filter processing unit 102 applies an image processing filter having a shape determined based on the shape of the illumination 3 illustrated in FIG. 1 to the iris image to perform filter processing.
- the filter processing unit 102 applies an image processing filter having a shape determined based on the shape of the illumination 3 illustrated in FIG. 1 to the iris image to perform filter processing.
- the user's iris is illuminated by the illumination 3 as shown in the photographing apparatus 1 of FIG. Visible light or near-infrared light is used as the illumination light.
- the reflection of the illumination 3 is reflected in the captured iris image. Therefore, in the present embodiment, the focus of the iris image is determined by utilizing the reflection of the illumination 3 reflected in the iris image.
- an image processing filter having a shape determined based on the shape of the illumination 3 is prepared in advance.
- the image processing filter prepared in advance is a general-purpose image processing filter used for a plurality of iris images.
- the filter processing unit 102 applies this image processing filter to the iris image to perform filter processing, and outputs the filtered iris image to the filter output integration unit 103.
- FIG. 5A shows an example of the illumination 3.
- the illumination 3 is a linear light source having a vertically long rectangular shape.
- FIG. 5B shows an example of the reflection region of the illumination reflected in the iris image.
- reference numeral 51 indicates an iris
- reference numeral 52 indicates a pupil.
- a reflection region 53 in which the illumination 3 is reflected appears inside the pupil 52.
- the reflection region 53 is formed by reflecting the illumination light on the spherical surface of the eye.
- FIG. 6A shows an example of an image processing filter in the case of FIGS. 5A and 5B.
- the coefficient of the vertically long region C1 in the center is set to be larger than that of the other regions, corresponding to the shape of the illumination 3 shown in FIG. 5A.
- the shape of the image processing filter is determined based on the shape of the illumination 3
- the shape of the image processing filter is determined in advance based on the shape of the illumination 3 actually used.
- the image processing filter thus determined in advance is used for the focus determination of the iris image of the plurality of users X.
- the "filter shape” is a concept including the size of the filter and the coefficient value of each segment of the filter.
- the shape of the image processing filter F1 of FIG. 6A includes a filter size of 8 ⁇ 8 pixels and a coefficient value of each segment constituting the filter F1.
- the focus determination device 100 makes a focus determination using the reflection region 53 of the illumination.
- the difference between the pixel values of the reflection region 53 of the illumination and the region around it becomes large. Therefore, the more the focus is, the higher the pixel value of the reflection region 53 of the illumination in the filtered iris image. Therefore, in the filtered iris image, if there is a region having a large pixel value corresponding to the reflection region 53 of the illumination, it can be determined that the iris image is in focus.
- the size of the image processing filter itself may be a shape based on the shape of the illumination 3.
- the coefficient of the vertically long region in the center indicated by the reference numeral C2 is set to be large according to the shape of the illumination 3.
- the filter output integration unit 103 takes an iris image after filtering using an image processing filter (hereinafter, also referred to as “filtered image”) as an input, and performs integration processing using the pixel values of the filtered image.
- the value of the result of integration is output as the filter output integration value P.
- the filter output integration unit 103 outputs the average value of all the pixel values of the filtered image as the filter output integration value P.
- the filter output integration unit 103 may divide the filtered image into a plurality of local regions, calculate the average value of the pixel values for each local region, and output the image as the filter output integration value P.
- the filter output integrated value P is a value indicating the degree of focus of the iris image, that is, the degree of focusing.
- the filter output integration unit 103 outputs the filter output integration value P to the focus determination unit 104.
- the focus determination unit 104 compares the filter output integrated value P with a predetermined threshold value Q, and determines the focus of the iris image. Specifically, when the filter output integrated value P is equal to or greater than the threshold value Q, the focus determination unit 104 determines that the iris image is in focus. On the other hand, when the filter output integrated value P is less than the threshold value Q, the focus determination unit 104 determines that the iris image is out of focus.
- the filter output integration unit 103 outputs the filter output integration value for each local region
- the focus determination unit 104 determines that the number of local regions having the filter output integration value equal to or higher than the threshold value Q is a predetermined number or more. It may be determined that the iris image is in focus, and if the number of local regions having a filter output integrated value equal to or higher than the threshold value Q is less than a predetermined number, it may be determined that the iris image is out of focus. ..
- the iris image acquisition unit 101 is an example of acquisition means
- the filter processing unit 102 is an example of filter processing means
- the filter output integration unit 103 and the focus determination unit 104 are examples of focus determination means.
- FIG. 7 is a flowchart of the focus determination process by the focus determination device 100 of the first embodiment. This process is realized by the processor 12 shown in FIG. 3 executing a program prepared in advance and operating as each element shown in FIG.
- the iris image acquisition unit 101 acquires the iris image input from the camera 2 or the like and outputs it to the filter processing unit 102 (step S11).
- the filter processing unit 102 applies the image processing filter prepared in advance to the iris image, executes the filter processing, and outputs the filtered image to the filter output integration unit 103 (step S12).
- the filter output integration unit 103 performs integration processing using the pixel values of the image after the filter processing, and outputs the filter output integration value P to the focus determination unit 104 (step S13).
- the focus determination unit 104 compares the filter output integrated value P with a predetermined threshold value Q to perform focus determination (step S14), and outputs the determination result (step S15). Then, the focus determination process ends.
- the filter processing is performed using the image processing filter having the shape determined based on the shape of the illumination used when the iris image is taken, and the focus is determined.
- the shape of the lighting actually used as a reference it is possible to perform the focus determination with high accuracy.
- the shape of the image processing filter is determined based on the shape of the illumination 3.
- the shape of the image processing filter is predetermined based on the shape of the reflection region of the illumination in the region of the eye of the iris image. As a result, an image processing filter having a shape corresponding to the reflection region of the illumination can be obtained, so that the focus determination can be performed with high accuracy.
- the second embodiment is the same as the first embodiment.
- the shape of the image processing filter is the ratio of the horizontal and vertical aspect ratios (aspect ratios) of the illumination 3 to the horizontal and vertical magnitudes of the spatial frequency components of the image processing filter. Is determined to match. As a result, an image processing filter having a shape having the same aspect ratio as the illumination can be obtained, so that the focus can be determined with high accuracy. Except for this point, the third embodiment is the same as the first embodiment.
- the shape of the image processing filter is the horizontal and vertical aspect ratios (aspect ratios) of the reflection region 53 of the illumination, and the horizontal and vertical magnitudes of the spatial frequency components of the image processing filter. It is determined to match the ratio of the space.
- the shape of the reflection region of the illumination is detected using the image of the region of the eye taken experimentally by irradiating the illumination light. Then, the shape of the image processing filter may be determined based on the shape. As a result, an image processing filter having a shape having the same aspect ratio as the reflection region of the illumination can be obtained, so that the focus determination can be performed with high accuracy.
- the second embodiment is the same as the first embodiment.
- the shape of the image processing filter is the ratio of the magnitudes of the horizontal spatial frequency component and the vertical spatial frequency component of the illumination 3, and the horizontal spatial frequency component and the vertical direction of the image processing filter. It is determined so that the ratio of the frequency components matches. As a result, an image processing filter having a shape having the same aspect ratio as the illumination can be obtained, so that the focus can be determined with high accuracy. Except for this point, the fifth embodiment is the same as the first embodiment.
- the shape of the image processing filter is the ratio of the magnitude of the horizontal spatial frequency component to the vertical spatial frequency component of the reflection region 53 of the illumination and the lateral spatial frequency component of the image processing filter. And the ratio of the longitudinal frequency components are determined to match. As a result, an image processing filter having a shape having the same aspect ratio as the reflection region of the illumination can be obtained, so that the focus determination can be performed with high accuracy. Except for this point, the sixth embodiment is the same as the first embodiment.
- the shape of the image processing filter is the ratio of the size of the horizontal spatial frequency component to the vertical spatial frequency component of the shape of the illumination 3 or the reflection region 53 of the illumination, and the above-mentioned. It is determined that the ratio of the horizontal spatial frequency component and the vertical frequency component of the image processing filter matches. However, the ratio of the spatial frequency components in the vertical direction and the horizontal direction of the image processing filter does not have to exactly match the aspect ratio of the shape of the illumination 3 or the shape of the reflection region 53 of the illumination.
- the shape of the image processing filter is the size of the horizontal and vertical lengths of the shape of the illumination 3, and the horizontal size and the vertical direction of the spatial frequency component of the image processing filter. It is determined that the size matches the size. As a result, an image processing filter having a shape in which the magnitude relationship between the illumination and the vertical and horizontal directions matches can be obtained, so that the focus determination can be performed using a highly versatile image processing filter. Except for this point, the seventh embodiment is the same as the first embodiment.
- the shape of the image processing filter is the size of the horizontal and vertical lengths of the shape of the illumination 3, and the size of the spatial frequency component of the image processing filter in the horizontal direction and the size in the vertical direction. Is determined to match.
- the shape of the image processing filter is determined by the size of the horizontal and vertical lengths of the shape of the reflection region 53 of the illumination and the lateral size of the spatial frequency component of the image processing filter. It may be determined so that the size in the vertical direction matches the size.
- an image processing filter having a shape in which the vertical and horizontal magnitude relations match the reflection region of the illumination can be obtained, so that the focus determination can be performed using a highly versatile image processing filter.
- the eighth embodiment is the same as the first embodiment.
- filter processing is performed using an image processing filter prepared in advance based on the shape of the illumination or the shape of the reflection region of the illumination. That is, in the first to eighth embodiments, a general-purpose image processing filter prepared in advance is applied to the plurality of input iris images.
- the reflection region of the illumination in the input iris image is detected, the shape of the illumination is estimated, and the shape of the filter is determined based on the estimated shape.
- the hardware configuration of the focus determination device of the ninth embodiment is the same as that of the focus determination device 100 of the first embodiment shown in FIG.
- FIG. 8 is a block diagram showing a functional configuration of the focus determination device 100X according to the ninth embodiment.
- the focus determination device 100X includes an iris image acquisition unit 101, a filter processing unit 102, a filter output integration unit 103, and a focus determination unit 104, as in the first embodiment, as well as an illumination reflection detection unit 105 and an illumination shape estimation unit 106. And a filter generation unit 107.
- the illumination reflection detection unit 105 detects an illumination reflection region as illustrated in the reflection region 53 of FIG. 5B from the iris image acquired by the iris image acquisition unit 101.
- the illumination reflection detection unit 105 detects, for example, a region of pixels having a brightness value (brightness) equal to or higher than a predetermined value among the pixels of the iris image as a reflection region. Then, the illumination reflection detection unit 105 outputs the size of the detected reflection region in the vertical direction and the horizontal direction to the illumination shape estimation unit 106.
- the illumination shape estimation unit 106 estimates the shape of the illumination based on the vertical and horizontal sizes of the reflection region detected by the illumination reflection detection unit 105, and outputs the conditions defining the illumination shape to the filter generation unit 107. do.
- the condition that defines the shape of the lighting (hereinafter, also referred to as “lighting shape condition”) may be, for example, the ratio of the vertical and horizontal lengths of the lighting (that is, the aspect ratio), and is based on the magnitude relation of the vertical and horizontal lengths of the lighting. There may be.
- the condition that defines the shape of the lighting may be a condition that defines the shape of the lighting, such as a circle or a square.
- the filter generation unit 107 determines the shape of the image processing filter used in the filter processing unit 102 based on the lighting shape conditions input from the lighting shape estimation unit 106. That is, the filter generation unit 107 determines the shape of the image processing filter for extracting the shape defined by the input illumination shape condition, and generates the image processing filter. As described above, the shape of the filter includes the size of the filter and the coefficient value of each segment of the filter.
- the filter generation unit 107 determines the shape of the image processing filter by the same method as in any of the first to eighth embodiments. That is, in the filter generation unit 107, the horizontal and vertical aspect ratios of the illumination defined by the illumination shape conditions and the ratio of the horizontal and vertical magnitudes of the spatial frequency components of the image processing filter match.
- the shape of the image processing filter may be determined as described above.
- the filter generation unit 107 determines the ratio of the magnitudes of the horizontal spatial frequency component and the vertical spatial frequency component of the illumination defined by the illumination shape condition, and the horizontal spatial frequency component and the vertical frequency of the image processing filter.
- the shape of the image processing filter may be determined so that the ratio of the components matches.
- the filter generation unit 107 has a large / small horizontal and vertical length of the lighting shape defined by the lighting shape condition, and a large / small horizontal size and vertical size of the spatial frequency component of the image processing filter.
- the shape of the image processing filter may be determined so as to match with.
- the filter generation unit 107 outputs the generated image processing filter to the filter processing unit 102.
- the operations of the filter processing unit 102, the filter output integration unit 103, and the focus determination unit 104 are the same as those in the first embodiment.
- the filter processing unit 102 applies the image processing filter generated by the filter generation unit 107 to the iris image input from the iris image acquisition unit 101 to perform filter processing.
- the image processing is individually generated based on the reflection region of the illumination in the iris image to be the focus determination, instead of the general-purpose image processing filter used for the plurality of iris images. Filter processing is performed using the filter, and focus determination is performed. Therefore, it is possible to perform focus determination with higher accuracy for each iris image.
- the illumination reflection detection unit 105 is an example of the detection means
- the illumination shape estimation unit 106 is an example of the estimation means
- the filter generation unit 107 is an example of the filter generation means.
- FIG. 9 is a flowchart of the focus determination process by the focus determination device 100X of the ninth embodiment. This process is realized by the processor 12 shown in FIG. 3 executing a program prepared in advance and operating as each element shown in FIG.
- the iris image acquisition unit 101 acquires the iris image input from the camera 2 or the like and outputs it to the filter processing unit 102 and the illumination reflection detection unit 105 (step S21).
- the illumination reflection detection unit 105 detects the illumination reflection region on the ocular spherical surface from the input iris image, and outputs the vertical and horizontal lengths of the detected illumination reflection region to the illumination shape estimation unit 106 (.
- Step S22) The illumination shape estimation unit 106 estimates the shape of the illumination based on the length of the input reflection region in the vertical and horizontal directions, and outputs the illumination shape condition defining the shape of the illumination to the filter generation unit 107 (step).
- the filter generation unit 107 determines the shape of the image processing filter based on the input illumination shape condition, generates an image processing filter, and outputs the image processing filter to the filter processing unit 102 (step S24).
- the filter processing unit 102 applies the image processing filter generated by the filter generation unit 107 to the iris image input from the iris image acquisition unit 101 to perform filter processing (step S25). Subsequent processing is basically the same as in the first embodiment. That is, the filter output integration unit 103 performs integration processing using the pixel values of the image after the filter processing, and outputs the filter output integration value P to the focus determination unit 104 (step S26). Next, the focus determination unit 104 compares the filter output integrated value P with a predetermined threshold value Q to perform focus determination (step S27), and outputs the determination result (step S28). Then, the focus determination process ends.
- the image processing filter used for the focus determination of a certain iris image is generated based on the shape of the illumination detected from the iris image. Therefore, the focus determination can be performed with high accuracy by using an image processing filter having a shape close to the reflection shape of the illumination included in each iris image.
- a tenth embodiment relates to an iris recognition device including the focus determination device according to any one of the first to ninth embodiments.
- iris recognition it is important to use an iris image that is in focus in order to capture the fine patterns of the iris.
- a method of capturing a plurality of images by changing the focus position at the time of capturing the iris image is used.
- the focus determination device is used to efficiently perform iris recognition.
- the hardware configuration of the iris recognition device according to the tenth embodiment is basically the same as the configuration of the focus determination device 100 of the first embodiment shown in FIG. However, it is assumed that the iris image to be authenticated for iris and the feature amount of the iris (hereinafter, referred to as "iris feature amount”) are registered in advance in DB15.
- FIG. 10 is a block diagram showing a functional configuration of the iris recognition device 200 according to the tenth embodiment.
- the iris recognition device 200 includes the focus determination device 100 according to any one of the first to eighth embodiments, the focus determination device 100X according to the ninth embodiment, the focus image selection unit 201, the iris detection unit 202, and the iris feature. It includes an extraction unit 203, an iris authentication unit 204, and an iris DB 205.
- a plurality of iris images having different focus states for the same person are input to the focus determination device 100 or 100X.
- the focus determination device 100 or 100X determines the focus of the input iris image, and outputs the determination results for the plurality of input iris images to the focus image selection unit 201.
- the focus image selection unit 201 outputs an iris image (hereinafter, also referred to as a “focused iris image”) determined to be in focus from a plurality of iris images to the iris detection unit 202 based on the determination result. ..
- the iris detection unit 202 detects the iris region from the focused iris image and outputs it to the iris feature extraction unit 203.
- the iris feature extraction unit 203 extracts the iris feature amount from the iris region detected by the iris detection unit 202 and outputs the iris feature amount to the iris authentication unit 204.
- the iris feature extraction unit 203 extracts the iris feature amount using a Daugman algorithm or any other algorithm.
- the iris DB 205 stores an iris image and an iris feature amount for a person whose iris information has been registered.
- the iris recognition unit 204 collates the iris feature amount input from the iris feature extraction unit 203 with the iris feature amount of the person registered in the iris DB 205, and performs iris recognition. For example, in the iris recognition unit 204, when the collation score between the iris feature extraction unit 203 and the iris feature amount of a person registered in the iris DB 205 is equal to or higher than a predetermined score, the person in the iris image is a registered person. Judge that there is.
- the focus determination device 100 or 100X detects the reflection region of the illumination by performing a filter process, but usually, the pixel value of the reflection region of the illumination is overexposed, which is not suitable for the authentication of the feature amount. Appropriate. Therefore, it is preferable that the iris recognition unit 204 does not use the reflection area of the illumination for authentication, or lowers the weight in the authentication to make it difficult for the reflection area of the illumination to be reflected in the score. Then, the iris recognition unit 204 outputs the authentication result.
- the focus image selection unit 201 is an example of the image selection means
- the iris detection unit 202 is an example of the iris detection means
- the iris feature extraction unit 203 is an example of the feature amount extraction means
- the iris recognition unit. 204 is an example of an iris recognition means.
- the iris authentication device 200 of the tenth embodiment it is possible to extract an in-focus iris image from a plurality of iris images of the same person by a focus determination process and perform the iris authentication process. can.
- the iris detection process, the iris feature extraction process, and the iris authentication process can be omitted, and the calculation amount and the processing time in the iris authentication process can be reduced.
- the out-of-focus image it is possible to prevent the out-of-focus iris image, which is inappropriate for authentication, from being used in the iris authentication process, and it is possible to reduce the occurrence of erroneous authentication. It becomes.
- the eleventh embodiment also relates to an iris recognition device including the focus determination device according to any one of the first to ninth embodiments.
- an iris recognition device including the focus determination device according to any one of the first to ninth embodiments.
- a method has been proposed in which a camera is provided at a security gate or the like, an iris image of a walking user is taken, and iris authentication is performed.
- This method is called walk-through iris recognition or the like.
- walk-through iris recognition the shape of the reflection area of the illumination changes over time because the user moves.
- the focus determination processing of the first to ninth embodiments is applied to all the iris images taken in time series, the processing amount becomes enormous. Therefore, in the eleventh embodiment, the iris image in the focused state is efficiently extracted from a large number of iris images, and the iris authentication is performed.
- the hardware configuration of the iris recognition device according to the eleventh embodiment is basically the same as the configuration of the focus determination device 100 of the first embodiment shown in FIG. However, it is assumed that the iris image and the iris feature amount are stored in advance in the DB 15 for the registered person.
- FIG. 11 is a block diagram showing a functional configuration of the iris recognition device 200X according to the eleventh embodiment.
- the iris recognition device 200X includes an iris image selection unit 210 in addition to the configuration of the iris recognition device 200 according to the third embodiment.
- a large number of iris images taken about the same person are input to the iris image selection unit 210 by a walk-through iris authentication system or the like.
- the iris image selection unit 210 selects a part of the iris images out of a large number of input iris images as the determination target image, and outputs the iris image to the focus determination device 100 or 100X.
- the focus determination device 100 or 100X performs a focus determination on the input determination target image, and outputs the determination result to the focus image selection unit 201.
- the focus image selection unit 201 selects an in-focus iris image based on the determination result for the determination target image, and outputs the iris image to the iris detection unit 202.
- the focus image selection unit 201 has a predetermined number or more of determination target images whose filter output integrated value P exceeds a predetermined threshold value Q, the focus determination process is applied again to those iris images, and the focus state is changed. You may narrow down to a good predetermined number of iris images and output them to the iris detection unit 202. Since the processing of the iris detection unit 202, the iris feature extraction unit 203, and the iris authentication unit 204 is the same as that of the third embodiment, the description thereof will be omitted.
- FIG. 12A shows an example of an image selection method by the iris image selection unit 210 of the eleventh embodiment.
- FIG. 12A schematically shows a plurality of iris images P1 to P16 taken in time series.
- the iris image selection unit 210 selects an arbitrary plurality of images from the plurality of iris images P1 to P16.
- the iris image selection unit 210 selects the three iris images P6, P11, and P15 in FIG. 12A and outputs them to the focus determination device 100 or 100X.
- the focus determination device 100 or 100X determines the focus of the iris images P6, P11 and P15, and outputs the determination result to the focus image selection unit 201.
- the focus image selection unit 201 selects the iris image P10 as the in-focus iris image and outputs it to the iris detection unit 202. Instead of determining the iris image P10 as the in-focus iris image, the focus image selection unit 201 inputs the iris images P7 to P10 to the focus determination device 100 or 100X to perform the focus determination again, and the determination result is used.
- the in-focus iris image may be determined.
- the iris image selection unit 210 determines that the size of the reflection region of the illumination or the size of the portion of the reflection region that is covered by the iris is a predetermined value or more. If it is larger than the specified value, it is the focus of the images before and after the image that can be used for authentication even if it is not in focus, and the reflection area of the illumination. An image with a small size may be reselected as an image to be used for authentication.
- the above method is a method of selecting an iris image that has a good focus and is suitable for authentication from a plurality of iris images after the iris image of a moving user has been taken.
- filter processing and focus determination processing are performed on a plurality of iris images taken when the user is at a certain distance, and if the iris images that are in focus and suitable for authentication are obtained, the processing is performed. finish.
- the iris recognition device 200X of the eleventh embodiment even when a large number of iris images are taken by the walk-through iris recognition system or the like, the focus is determined only for a part of them and the iris is in focus. By selecting the iris image, iris recognition can be performed efficiently.
- FIG. 12B shows an example of an image selection method by the iris image selection unit 210 according to the tenth embodiment.
- FIG. 12B schematically shows a plurality of iris images P1 to P16 taken in time series.
- the iris image selection unit 210 selects an arbitrary plurality of images from the plurality of iris images P1 to P16.
- the iris image selection unit 210 selects the two iris images P6 and P15 in FIG. 12B and outputs them to the focus determination device 100 or 100X.
- the focus determination device 100 or 100X determines the focus of the iris images P6 and P15, and outputs the determination result to the focus image selection unit 201.
- the focus image selection unit 201 inputs the iris images P2, P3, P9, and P10 to the focus determination device 100 or 100X to perform the focus determination again, and determines the focused iris image from the determination result.
- the iris recognition device 200X of the twelfth embodiment even when a large number of iris images are taken by the walk-through iris recognition system or the like, the focus is determined only for a part of them and the iris is in focus. By selecting the iris image, iris recognition can be performed efficiently.
- the thirteenth embodiment also relates to an iris recognition device including the focus determination device according to any one of the first to ninth embodiments.
- an iris recognition device including the focus determination device according to any one of the first to ninth embodiments.
- the user's iris is usually illuminated with light.
- the reflection from the lens or frame of the spectacles is reflected in the captured image. If you use an iris image that reflects the reflections from the lens or frame of the spectacles, the iris will be affected by the edges of the reflections from the lens or frame of the spectacles when the focus is determined, even though the iris is not actually in focus. May be erroneously determined to be in focus.
- the iris image (hereinafter referred to as “eyewear-wearing image”) taken while wearing a device covering the eyes such as eyeglasses (hereinafter referred to as “eyewear”) is referred to as “eyewear-wearing image”.
- eyewear-wearing image Is performed by the focus determination device 100 or 100X, and the iris image (hereinafter referred to as “eyewear-free image”) taken without eyewear is not focused. I will do it.
- eyewear includes sunglasses, goggles, and the like.
- the hardware configuration of the iris recognition device according to the thirteenth embodiment is basically the same as the configuration of the focus determination device 100 of the first embodiment shown in FIG. However, it is assumed that the iris image and the iris feature amount thereof are stored in advance in the DB 15 for the registered person.
- FIG. 13 is a block diagram showing a functional configuration of the iris recognition device 200Y according to the thirteenth embodiment.
- the iris recognition device 200Y includes an eyewear-worn image detection unit 215 in addition to the configuration of the iris recognition device 200 according to the third embodiment.
- a plurality of iris images are input to the eyewear wearing image detection unit 215.
- the eyewear-wearing image detection unit 215 detects the eyewear-wearing image from the input iris image.
- the eyewear-wearing image detection unit 215 may detect the eyewear-wearing image based on the color of the iris image or the like. Further, when the iris image includes the entire eyewear, the eyewear itself may be detected by image processing.
- the eyewear-mounted image detection unit 215 outputs the eyewear-wearing image to the focus determination device 100 or 100X, and outputs the non-eyewear-wearing image to the iris detection unit 202.
- the focus determination device 100 or 100X determines the focus of the eyewear-worn image, and the focus image selection unit 201 selects the iris image in the in-focus state from the eyewear-wearing images and outputs the iris image to the iris detection unit 202. In this way, the iris image in the focused state and the iris non-wearing image among the eyewear-wearing images are input to the iris detection unit 202. Since the processing of the iris detection unit 202, the iris feature extraction unit 203, and the iris authentication unit 204 is the same as that of the third embodiment, the description thereof will be omitted.
- the focus determination is not performed on the iris images of all users, but the focus determination is performed only on the iris images of the users wearing eyewear, so that the focus determination and the focus determination can be performed efficiently. Iris recognition can be performed.
- FIG. 14 is a block diagram showing a functional configuration of the focus determination device according to the 14th embodiment.
- the focus determination device 70 includes an acquisition unit 71, a filter processing unit 72, and a focus determination unit 73.
- the acquisition means 71 acquires an iris image obtained by photographing the user's iris.
- the filter processing means 72 applies an image processing filter having a shape determined based on the shape of the illumination to be applied to the user to the iris image to perform filter processing.
- the focus determination means 73 determines whether or not the iris image is an in-focus image based on the filtered iris image.
- FIG. 15 is a flowchart of the focus determination process of the 14th embodiment.
- the acquisition means 71 acquires an iris image obtained by photographing the user's iris (step S71).
- the filter processing means 72 applies an image processing filter having a shape determined based on the shape of the illumination to be applied to the user to the iris image to perform filter processing (step S72).
- the focus determination means 73 determines whether or not the iris image is an in-focus image based on the filtered iris image (step S73).
- the focus state of the iris image is determined by using the shape of the illumination used when photographing the iris, it is possible to accurately determine the focus state based on the known shape. ..
- An acquisition method for acquiring an iris image of a user's iris A filtering means for filtering the iris image by applying an image processing filter having a shape determined based on the shape of the illumination to be applied to the user.
- a focus determination means for determining whether or not the iris image is an in-focus image based on the filtered iris image.
- a focus determination device comprising.
- Appendix 2 The focus determination device according to Appendix 1, wherein the shape of the image processing filter is predetermined based on the shape of the illumination.
- the shape of the image processing filter is the ratio of the magnitudes of the horizontal spatial frequency component and the vertical spatial frequency component of the illumination, and the ratio of the horizontal spatial frequency component and the vertical frequency component of the image processing filter.
- the shape of the image processing filter is such that the size of the horizontal and vertical lengths of the illumination matches the size of the horizontal size and the vertical size of the spatial frequency component of the image processing filter.
- the focus determination device according to Appendix 2 which has been determined.
- Appendix 6 The focus determination device according to Appendix 1, wherein the shape of the image processing filter is predetermined based on the shape of the reflection region of the illumination in the iris image.
- the shape of the image processing filter is such that the horizontal and vertical aspect ratios of the shape of the reflection region match the ratio of the horizontal magnitude to the vertical magnitude of the spatial frequency component of the image processing filter.
- the shape of the image processing filter is the ratio of the magnitude of the horizontal spatial frequency component and the vertical spatial frequency component of the shape of the reflection region, and the horizontal spatial frequency component and the vertical frequency component of the image processing filter.
- the shape of the image processing filter matches the size of the horizontal and vertical lengths of the shape of the reflection region with the size of the spatial frequency component of the image processing filter in the horizontal direction and the size in the vertical direction.
- Appendix 10 A detection means for detecting the reflection region of the illumination from the iris image, An estimation means for estimating the shape of the illumination based on the shape of the reflection region, and A filter generation means for determining the shape of the image processing filter based on the estimated shape of the illumination, and The focus determination device according to Appendix 1.
- An image selection means for selecting an in-focus iris image from a plurality of iris images by using the focus determination device according to any one of Supplementary note 1 to 10.
- An iris detecting means for detecting an iris region from the focused iris image
- a feature amount extraction means for extracting a feature amount from the detected iris region
- An iris recognition means that performs iris recognition using the extracted features
- the image selection means selects a part of the iris images among the plurality of iris images as the determination target image, performs focus determination on the determination target image using the focus determination device, and the determination result is used as the basis for the determination result.
- the iris authentication device according to Appendix 11, which selects an iris image in focus.
- the image selection means selects a part of the iris images among the plurality of iris images as the determination target image, performs focus determination on the determination target image using the focus determination device, and next based on the determination result.
- the iris authentication device according to Appendix 11 which determines the image to be determined and selects the focused iris image based on the determination result for the next image to be determined.
- An eyewear-worn image detecting means for detecting an iris image taken with the eyewear worn from the plurality of iris images, An image selection means for selecting an in-focus iris image from a plurality of iris images taken with eyewear attached using the focus determination device according to any one of Supplementary note 1 to 10.
- An iris detecting means for detecting an iris region from an iris image taken without wearing eyewear and the focused iris image selected by the image selecting means.
- a feature amount extraction means for extracting a feature amount from the detected iris region, An iris recognition means that performs iris recognition using the extracted features, An iris recognition device equipped with.
- (Appendix 15) Acquire an iris image of the user's iris, The iris image is filtered by applying an image processing filter having a shape determined based on the shape of the illumination to be applied to the user. A focus determination method for determining whether or not the iris image is an in-focus image based on the filtered iris image.
- a recording medium recording a program that causes a computer to execute a process of determining whether or not the iris image is an in-focus image based on the filtered iris image.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Ophthalmology & Optometry (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Software Systems (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Image Input (AREA)
Abstract
Description
利用者の虹彩を撮影した虹彩画像を取得する取得手段と、
前記虹彩画像に対して、利用者に照射する照明の形状に基づいて決定された形状の画像処理フィルタを適用してフィルタ処理を行うフィルタ処理手段と、
前記フィルタ処理後の虹彩画像に基づいて、前記虹彩画像がフォーカスの合った画像であるか否かを判定するフォーカス判定手段と、を備える。
利用者の虹彩を撮影した虹彩画像を取得し、
前記虹彩画像に対して、利用者に照射する照明の形状に基づいて決定された形状の画像処理フィルタを適用してフィルタ処理を行い、
前記フィルタ処理後の虹彩画像に基づいて、前記虹彩画像がフォーカスの合った画像であるか否かを判定する。
利用者の虹彩を撮影した虹彩画像を取得し、
前記虹彩画像に対して、利用者に照射する照明の形状に基づいて決定された形状の画像処理フィルタを適用してフィルタ処理を行い、
前記フィルタ処理後の虹彩画像に基づいて、前記虹彩画像がフォーカスの合った画像であるか否かを判定する処理をコンピュータに実行させるプログラムを記録する。
<第1実施形態>
[撮影装置]
図1は、虹彩画像の撮影装置の例を示す。図1(A)は利用者X及び撮影装置1を側方から見た図であり、図1(B)は利用者X及び撮影装置1を上方から見た図である。撮影装置1は、カメラ2と照明3を備え、利用者Xの虹彩画像を撮影する。照明3は、カメラ2の近傍に配置されている。なお、図1(A)では、照明3はカメラ2の手前側に配置されている。図1の例では、照明3は線状光源を用いているが、照明3の形状は線状には限定されない。また、照明3は、カメラ2と一体でもよく、カメラ2と独立して設置されてもよい。
図2は、フォーカス判定装置の例を示す。フォーカス判定装置100は、カメラ2に接続され、カメラ2から虹彩画像を取得する。フォーカス判定装置100は、取得した虹彩画像のフォーカスが合っているか否かの判定(以下、「フォーカス判定」と呼ぶ。)を行う。なお、フォーカスが合っている状態を「合焦状態」とも呼び、フォーカスが合っていない状態を「非合焦状態」とも呼ぶ。
図3は、フォーカス判定装置100のハードウェア構成を示すブロック図である。図示のように、フォーカス判定装置100は、インタフェース(IF)11と、プロセッサ12と、メモリ13と、記録媒体14と、データベース(DB)15とを備える。
図4は、フォーカス判定装置100の機能構成を示すブロック図である。フォーカス判定装置100は、機能的には、虹彩画像取得部101と、フィルタ処理部102と、フィルタ出力統合部103と、フォーカス判定部104とを備える。
図7は、第1実施形態のフォーカス判定装置100によるフォーカス判定処理のフローチャートである。この処理は、図3に示すプロセッサ12が、予め用意されたプログラムを実行し、図4に示す各要素として動作することにより実現される。
第1実施形態では、画像処理フィルタの形状は、照明3の形状に基づいて決定されている。これに対し、第2実施形態では、画像処理フィルタの形状は、虹彩画像の眼の領域における照明の反射領域の形状に基づいて予め決定される。これにより、照明の反射領域に対応する形状の画像処理フィルタが得られるので、精度よくフォーカス判定を行うことができる。なお、この点以外は、第2実施形態は第1実施形態と同様である。
第3実施形態では、画像処理フィルタの形状は、照明3の水平方向と垂直方向のアスペクト比(縦横比)と、画像処理フィルタの空間周波数成分の横方向の大きさと縦方向の大きさの比とが一致するように決定される。これにより、照明と同一の縦横比を有する形状の画像処理フィルタが得られるので、精度よくフォーカス判定を行うことができる。なお、この点以外は、第3実施形態は第1実施形態と同様である。
第4実施形態では、画像処理フィルタの形状は、照明の反射領域53の水平方向と垂直方向のアスペクト比(縦横比)と、画像処理フィルタの空間周波数成分の横方向の大きさと縦方向の大きさの比とが一致するように決定される。なお、照明の反射領域の形状に基づいて画像処理フィルタの形状を決定する場合には、実験的に照明光を照射して撮影した眼の領域の画像を用いて照明の反射領域の形状を検出し、その形状に基づいて画像処理フィルタの形状を決定すればよい。これにより、照明の反射領域と同一の縦横比を有する形状の画像処理フィルタが得られるので、精度よくフォーカス判定を行うことができる。なお、この点以外は、第2実施形態は第1実施形態と同様である。
第5実施形態では、画像処理フィルタの形状は、照明3の水平方向の空間周波数成分と垂直方向の空間周波数成分の大きさの比と、前記画像処理フィルタの横方向の空間周波数成分と縦方向周波数成分の比とが一致するように決定される。これにより、照明と同一の縦横比を有する形状の画像処理フィルタが得られるので、精度よくフォーカス判定を行うことができる。なお、この点以外は、第5実施形態は第1実施形態と同様である。
第6実施形態では、画像処理フィルタの形状は、照明の反射領域53の水平方向の空間周波数成分と垂直方向の空間周波数成分の大きさの比と、前記画像処理フィルタの横方向の空間周波数成分と縦方向周波数成分の比とが一致するように決定される。これにより、照明の反射領域と同一の縦横比を有する形状の画像処理フィルタが得られるので、精度よくフォーカス判定を行うことができる。なお、この点以外は、第6実施形態は第1実施形態と同様である。
上記の第5~第6実施形態では、画像処理フィルタの形状は、照明3の形状又は照明の反射領域53の水平方向の空間周波数成分と垂直方向の空間周波数成分の大きさの比と、前記画像処理フィルタの横方向の空間周波数成分と縦方向周波数成分の比とが一致するように決定されている。しかし、画像処理フィルタの縦方向と横方向の空間周波数成分の比は、照明3の形状又は照明の反射領域53の形状のアスペクト比と厳密に一致しなくてもよい。この点から、第7実施形態では、画像処理フィルタの形状は、照明3の形状の水平方向と垂直方向の長さの大小と、画像処理フィルタの空間周波数成分の横方向の大きさと縦方向の大きさの大小とが一致するように決定される。これにより、照明と縦横の大小関係が一致する形状の画像処理フィルタが得られるので、汎用性の高い画像処理フィルタを用いてフォーカス判定を行うことができる。この点以外は、第7実施形態は第1実施形態と同様である。
第7実施形態では、画像処理フィルタの形状は、照明3の形状の水平方向と垂直方向の長さの大小と、画像処理フィルタの空間周波数成分の横方向の大きさと縦方向の大きさの大小とが一致するように決定されている。その代わりに、第8実施形態では、画像処理フィルタの形状は、照明の反射領域53の形状の水平方向と垂直方向の長さの大小と、画像処理フィルタの空間周波数成分の横方向の大きさと縦方向の大きさの大小とが一致するように決定すればよい。これにより、照明の反射領域と縦横の大小関係が一致する形状の画像処理フィルタが得られるので、汎用性の高い画像処理フィルタを用いてフォーカス判定を行うことができる。この点以外は、第8実施形態は第1実施形態と同様である。
第1~第8実施形態では、照明の形状又は照明の反射領域の形状に基づいて予め用意された画像処理フィルタを用いてフィルタ処理を行っている。即ち、第1~第8実施形態では、入力される複数の虹彩画像について、予め用意された汎用の画像処理フィルタを適用している。これに対し、第9実施形態では、入力された虹彩画像における照明の反射領域を検出して照明の形状を推定し、推定した形状に基づいてフィルタの形状を決定する。
第9実施形態のフォーカス判定装置のハードウェア構成は、図3に示す第1実施形態のフォーカス判定装置100と同様である。
図8は、第9実施形態に係るフォーカス判定装置100Xの機能構成を示すブロック図である。フォーカス判定装置100Xは、第1実施形態と同様の虹彩画像取得部101、フィルタ処理部102、フィルタ出力統合部103及びフォーカス判定部104に加えて、照明反射検出部105と、照明形状推定部106と、フィルタ生成部107とを備える。
図9は、第9実施形態のフォーカス判定装置100Xによるフォーカス判定処理のフローチャートである。この処理は、図3に示すプロセッサ12が、予め用意されたプログラムを実行し、図8に示す各要素として動作することにより実現される。
第10実施形態は、第1~第9実施形態のいずれかのフォーカス判定装置を備える虹彩認証装置に関する。虹彩認証においては、虹彩の細かな模様をとらえるため、虹彩にフォーカスが合っている虹彩画像を使用することが重要である。一般的に、虹彩にフォーカスの合った画像を撮影するため、虹彩画像の撮影時に、フォーカス位置を変えて複数の画像を撮影する方法が用いられている。しかし、撮影された複数の虹彩画像を用いて虹彩認証を行う場合、認証処理を適用する画像枚数が増加することで、認証処理に時間がかかるという課題がある。本実施形態は、第1~第9実施形態のいずれかのフォーカス判定装置を利用して、効率よく虹彩認証を行う。
第10実施形態に係る虹彩認証装置のハードウェア構成は、基本的に図3に示す第1実施形態のフォーカス判定装置100の構成と同様である。但し、DB15には、虹彩認証の対象となる虹彩画像及びその虹彩の特徴量(以下、「虹彩特徴量」と呼ぶ。)が予め登録されているものとする。
図10は、第10実施形態に係る虹彩認証装置200の機能構成を示すブロック図である。虹彩認証装置200は、第1~第8実施形態のいずれかに係るフォーカス判定装置100又は第9実施形態に係るフォーカス判定装置100Xと、フォーカス画像選択部201と、虹彩検出部202と、虹彩特徴抽出部203と、虹彩認証部204と、虹彩DB205とを備える。
第11実施形態も、第1~第9実施形態のいずれかのフォーカス判定装置を備える虹彩認証装置に関する。通常、虹彩画像の撮影時には、図1に例示するように、利用者が撮影装置の前で停止する。これに対し、セキュリティーゲートなどにカメラを設け、歩いてくる利用者の虹彩画像を撮影し、虹彩認証を行う手法が提案されている。この手法は、ウォークスルー虹彩認証などと呼ばれる。ウォークスルー虹彩認証の場合、利用者は移動するため、照明の反射領域の形状は時系列で変化する。この場合、時系列に撮影した全ての虹彩画像について第1~第9実施形態のフォーカス判定処理を適用すると、処理量が膨大となってしまう。そこで、第11実施形態では、多数の虹彩画像から合焦状態にある虹彩画像を効率的に抽出し、虹彩認証を行う。
第11実施形態に係る虹彩認証装置のハードウェア構成は、基本的に図3に示す第1実施形態のフォーカス判定装置100の構成と同様である。但し、DB15には、登録済みの人物について、虹彩画像及び虹彩特徴量が予め記憶されているものとする。
図11は、第11実施形態に係る虹彩認証装置200Xの機能構成を示すブロック図である。虹彩認証装置200Xは、第3実施形態に係る虹彩認証装置200の構成に加えて、虹彩画像選択部210を備える。
第12実施形態は、第11実施形態と同様に、ウォークスルー虹彩認証システムなどにより多数の虹彩画像が撮影された場合に適用される。但し、第12実施形態は、虹彩画像選択部210による虹彩画像の選択方法が、第9実施形態と異なる。図12(B)は、第10実施形態の虹彩画像選択部210による画像の選択方法の例を示す。図12(B)は、時系列に撮影された複数の虹彩画像P1~P16を模式的に示す。虹彩画像選択部210は、複数の虹彩画像P1~P16から、任意の複数の画像を選択する。例えば、虹彩画像選択部210は、図12(B)における2枚の虹彩画像P6及びP15を選択してフォーカス判定装置100又は100Xへ出力する。フォーカス判定装置100又は100Xは、虹彩画像P6及びP15のフォーカス判定を行い、判定結果をフォーカス画像選択部201へ出力する。
第13実施形態も、第1~第9実施形態のいずれかのフォーカス判定装置を備える虹彩認証装置に関する。虹彩画像を撮影する際には、通常、利用者の虹彩に光を照射する。しかし、虹彩を撮影される利用者が眼鏡等の眼を覆うデバイスを装着している場合、眼鏡のレンズやフレームでの反射が撮影画像に映り込む。眼鏡のレンズやフレームでの反射が映り込んだ虹彩画像を使用すると、フォーカス判定の際、実際には虹彩にフォーカスが合っていないにもかかわらず、眼鏡のレンズやフレームでの反射のエッジによって虹彩にフォーカスが合っていると誤判定されることがある。そこで、第11実施形態では、眼鏡などの眼を覆うデバイス(以下、「アイウェア」と呼ぶ。)を装着している状態で撮影された虹彩画像(以下、「アイウェア装着画像」と呼ぶ。)についてはフォーカス判定装置100又は100Xによるフォーカス判定を行い、アイウェアを装着していない状態で撮影された虹彩画像(以下、「アイウェア非装着画像」と呼ぶ。)についてはフォーカス判定を行わないこととする。なお、アイウェアには、メガネの他、サングラス、ゴーグルなどが含まれる。
第13実施形態に係る虹彩認証装置のハードウェア構成は、基本的に図3に示す第1実施形態のフォーカス判定装置100の構成と同様である。但し、DB15には、登録済みの人物について、虹彩画像及びその虹彩特徴量が予め記憶されているものとする。
図13は、第13実施形態に係る虹彩認証装置200Yの機能構成を示すブロック図である。虹彩認証装置200Yは、第3実施形態に係る虹彩認証装置200の構成に加えて、アイウェア装着画像検出部215を備える。
次に、この開示の第14実施形態について説明する。図14は、第14実施形態に係るフォーカス判定装置の機能構成を示すブロック図である。フォーカス判定装置70は、取得手段71と、フィルタ処理手段72と、フォーカス判定手段73とを備える。取得手段71は、利用者の虹彩を撮影した虹彩画像を取得する。フィルタ処理手段72は、虹彩画像に対して、利用者に照射する照明の形状に基づいて決定された形状の画像処理フィルタを適用してフィルタ処理を行う。フォーカス判定手段73は、フィルタ処理後の虹彩画像に基づいて、虹彩画像がフォーカスの合った画像であるか否かを判定する。
利用者の虹彩を撮影した虹彩画像を取得する取得手段と、
前記虹彩画像に対して、利用者に照射する照明の形状に基づいて決定された形状の画像処理フィルタを適用してフィルタ処理を行うフィルタ処理手段と、
前記フィルタ処理後の虹彩画像に基づいて、前記虹彩画像がフォーカスの合った画像であるか否かを判定するフォーカス判定手段と、
を備えるフォーカス判定装置。
前記画像処理フィルタの形状は、前記照明の形状に基づいて予め決定されている付記1に記載のフォーカス判定装置。
前記画像処理フィルタの形状は、前記照明の水平方向と垂直方向のアスペクト比と、前記画像処理フィルタの空間周波数成分の横方向の大きさと縦方向の大きさの比とが一致するように決定されている付記2に記載のフォーカス判定装置。
前記画像処理フィルタの形状は、前記照明の水平方向の空間周波数成分と垂直方向の空間周波数成分の大きさの比と、前記画像処理フィルタの横方向の空間周波数成分と縦方向周波数成分の比とが一致するように決定されている付記2に記載のフォーカス判定装置。
前記画像処理フィルタの形状は、前記照明の水平方向と垂直方向の長さの大小と、前記画像処理フィルタの空間周波数成分の横方向の大きさと縦方向の大きさの大小とが一致するように決定されている付記2に記載のフォーカス判定装置。
前記画像処理フィルタの形状は、前記虹彩画像における照明の反射領域の形状に基づいて予め決定されている付記1に記載のフォーカス判定装置。
前記画像処理フィルタの形状は、前記反射領域の形状の水平方向と垂直方向のアスペクト比と、前記画像処理フィルタの空間周波数成分の横方向の大きさと縦方向の大きさの比とが一致するように決定されている付記6に記載のフォーカス判定装置。
前記画像処理フィルタの形状は、前記反射領域の形状の水平方向の空間周波数成分と垂直方向の空間周波数成分の大きさの比と、前記画像処理フィルタの横方向の空間周波数成分と縦方向周波数成分の比とが一致するように決定されている付記6に記載のフォーカス判定装置。
前記画像処理フィルタの形状は、前記反射領域の形状の水平方向と垂直方向の長さの大小と、前記画像処理フィルタの空間周波数成分の横方向の大きさと縦方向の大きさの大小とが一致するように決定されている付記6に記載のフォーカス判定装置。
前記虹彩画像から、前記照明の反射領域を検出する検出手段と、
前記反射領域の形状に基づいて、前記照明の形状を推定する推定手段と、
推定された照明の形状に基づいて、前記画像処理フィルタの形状を決定するフィルタ生成手段と、
を備える付記1に記載のフォーカス判定装置。
付記1乃至10のいずれか一項に記載のフォーカス判定装置を用いて、複数の虹彩画像からフォーカスの合った虹彩画像を選択する画像選択手段と、
前記フォーカスの合った虹彩画像から虹彩領域を検出する虹彩検出手段と、
検出された虹彩領域から特徴量を抽出する特徴量抽出手段と、
抽出された特徴量を用いて、虹彩認証を行う虹彩認証手段と、
を備える虹彩認証装置。
前記画像選択手段は、前記複数の虹彩画像のうちの一部の虹彩画像を判定対象画像として選択し、前記判定対象画像について前記フォーカス判定装置を用いてフォーカス判定を行い、判定結果に基づいて前記フォーカスの合った虹彩画像を選択する付記11に記載の虹彩認証装置。
前記画像選択手段は、前記複数の虹彩画像のうちの一部の虹彩画像を判定対象画像として選択し、前記判定対象画像について前記フォーカス判定装置を用いてフォーカス判定を行い、判定結果に基づいて次の判定対象画像を決定し、前記次の判定対象画像に対する判定結果に基づいて、前記フォーカスの合った虹彩画像を選択する付記11に記載の虹彩認証装置。
前記複数の虹彩画像から、アイウェアを装着した状態で撮影された虹彩画像を検出するアイウェア装着画像検出手段と、
付記1乃至10のいずれか一項に記載のフォーカス判定装置を用いて、アイウェアを装着した状態で撮影された複数の虹彩画像から、フォーカスの合った虹彩画像を選択する画像選択手段と、
アイウェアを装着していない状態で撮影された虹彩画像、及び、前記画像選択手段により選択された前記フォーカスの合った虹彩画像から虹彩領域を検出する虹彩検出手段と、
検出された虹彩領域から特徴量を抽出する特徴量抽出手段と、
抽出された特徴量を用いて、虹彩認証を行う虹彩認証手段と、
を備える虹彩認証装置。
利用者の虹彩を撮影した虹彩画像を取得し、
前記虹彩画像に対して、利用者に照射する照明の形状に基づいて決定された形状の画像処理フィルタを適用してフィルタ処理を行い、
前記フィルタ処理後の虹彩画像に基づいて、前記虹彩画像がフォーカスの合った画像であるか否かを判定するフォーカス判定方法。
利用者の虹彩を撮影した虹彩画像を取得し、
前記虹彩画像に対して、利用者に照射する照明の形状に基づいて決定された形状の画像処理フィルタを適用してフィルタ処理を行い、
前記フィルタ処理後の虹彩画像に基づいて、前記虹彩画像がフォーカスの合った画像であるか否かを判定する処理をコンピュータに実行させるプログラムを記録した記録媒体。
2 カメラ
3 照明
12 プロセッサ
100、100X フォーカス判定装置
101 虹彩画像取得部
102 フィルタ処理部
103 フィルタ出力統合部
104 フォーカス判定部
105 照明反射検出部
106 照明形状推定部
107 フィルタ生成部
200、200X、200Y 虹彩認証装置
201 フォーカス画像選択部
202 虹彩検出部
203 虹彩特徴抽出部
204 虹彩認証部
205 虹彩DB
Claims (16)
- 利用者の虹彩を撮影した虹彩画像を取得する取得手段と、
前記虹彩画像に対して、利用者に照射する照明の形状に基づいて決定された形状の画像処理フィルタを適用してフィルタ処理を行うフィルタ処理手段と、
前記フィルタ処理後の虹彩画像に基づいて、前記虹彩画像がフォーカスの合った画像であるか否かを判定するフォーカス判定手段と、
を備えるフォーカス判定装置。 - 前記画像処理フィルタの形状は、前記照明の形状に基づいて予め決定されている請求項1に記載のフォーカス判定装置。
- 前記画像処理フィルタの形状は、前記照明の水平方向と垂直方向のアスペクト比と、前記画像処理フィルタの空間周波数成分の横方向の大きさと縦方向の大きさの比とが一致するように決定されている請求項2に記載のフォーカス判定装置。
- 前記画像処理フィルタの形状は、前記照明の水平方向の空間周波数成分と垂直方向の空間周波数成分の大きさの比と、前記画像処理フィルタの横方向の空間周波数成分と縦方向周波数成分の比とが一致するように決定されている請求項2に記載のフォーカス判定装置。
- 前記画像処理フィルタの形状は、前記照明の水平方向と垂直方向の長さの大小と、前記画像処理フィルタの空間周波数成分の横方向の大きさと縦方向の大きさの大小とが一致するように決定されている請求項2に記載のフォーカス判定装置。
- 前記画像処理フィルタの形状は、前記虹彩画像における照明の反射領域の形状に基づいて予め決定されている請求項1に記載のフォーカス判定装置。
- 前記画像処理フィルタの形状は、前記反射領域の形状の水平方向と垂直方向のアスペクト比と、前記画像処理フィルタの空間周波数成分の横方向の大きさと縦方向の大きさの比とが一致するように決定されている請求項6に記載のフォーカス判定装置。
- 前記画像処理フィルタの形状は、前記反射領域の形状の水平方向の空間周波数成分と垂直方向の空間周波数成分の大きさの比と、前記画像処理フィルタの横方向の空間周波数成分と縦方向周波数成分の比とが一致するように決定されている請求項6に記載のフォーカス判定装置。
- 前記画像処理フィルタの形状は、前記反射領域の形状の水平方向と垂直方向の長さの大小と、前記画像処理フィルタの空間周波数成分の横方向の大きさと縦方向の大きさの大小とが一致するように決定されている請求項6に記載のフォーカス判定装置。
- 前記虹彩画像から、前記照明の反射領域を検出する検出手段と、
前記反射領域の形状に基づいて、前記照明の形状を推定する推定手段と、
推定された照明の形状に基づいて、前記画像処理フィルタの形状を決定するフィルタ生成手段と、
を備える請求項1に記載のフォーカス判定装置。 - 請求項1乃至10のいずれか一項に記載のフォーカス判定装置を用いて、複数の虹彩画像からフォーカスの合った虹彩画像を選択する画像選択手段と、
前記フォーカスの合った虹彩画像から虹彩領域を検出する虹彩検出手段と、
検出された虹彩領域から特徴量を抽出する特徴量抽出手段と、
抽出された特徴量を用いて、虹彩認証を行う虹彩認証手段と、
を備える虹彩認証装置。 - 前記画像選択手段は、前記複数の虹彩画像のうちの一部の虹彩画像を判定対象画像として選択し、前記判定対象画像について前記フォーカス判定装置を用いてフォーカス判定を行い、判定結果に基づいて前記フォーカスの合った虹彩画像を選択する請求項11に記載の虹彩認証装置。
- 前記画像選択手段は、前記複数の虹彩画像のうちの一部の虹彩画像を判定対象画像として選択し、前記判定対象画像について前記フォーカス判定装置を用いてフォーカス判定を行い、判定結果に基づいて次の判定対象画像を決定し、前記次の判定対象画像に対する判定結果に基づいて、前記フォーカスの合った虹彩画像を選択する請求項11に記載の虹彩認証装置。
- 前記複数の虹彩画像から、アイウェアを装着した状態で撮影された虹彩画像を検出するアイウェア装着画像検出手段と、
請求項1乃至10のいずれか一項に記載のフォーカス判定装置を用いて、アイウェアを装着した状態で撮影された複数の虹彩画像から、フォーカスの合った虹彩画像を選択する画像選択手段と、
アイウェアを装着していない状態で撮影された虹彩画像、及び、前記画像選択手段により選択された前記フォーカスの合った虹彩画像から虹彩領域を検出する虹彩検出手段と、
検出された虹彩領域から特徴量を抽出する特徴量抽出手段と、
抽出された特徴量を用いて、虹彩認証を行う虹彩認証手段と、
を備える虹彩認証装置。 - 利用者の虹彩を撮影した虹彩画像を取得し、
前記虹彩画像に対して、利用者に照射する照明の形状に基づいて決定された形状の画像処理フィルタを適用してフィルタ処理を行い、
前記フィルタ処理後の虹彩画像に基づいて、前記虹彩画像がフォーカスの合った画像であるか否かを判定するフォーカス判定方法。 - 利用者の虹彩を撮影した虹彩画像を取得し、
前記虹彩画像に対して、利用者に照射する照明の形状に基づいて決定された形状の画像処理フィルタを適用してフィルタ処理を行い、
前記フィルタ処理後の虹彩画像に基づいて、前記虹彩画像がフォーカスの合った画像であるか否かを判定する処理をコンピュータに実行させるプログラムを記録した記録媒体。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2020/034887 WO2022059064A1 (ja) | 2020-09-15 | 2020-09-15 | フォーカス判定装置、虹彩認証装置、フォーカス判定方法、及び、記録媒体 |
JP2022550067A JP7452677B2 (ja) | 2020-09-15 | 2020-09-15 | フォーカス判定装置、虹彩認証装置、フォーカス判定方法、及び、プログラム |
US17/768,562 US20240104923A1 (en) | 2020-09-15 | 2020-09-15 | Focus determination device, iris authentication device, focus determination method, and recording medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2020/034887 WO2022059064A1 (ja) | 2020-09-15 | 2020-09-15 | フォーカス判定装置、虹彩認証装置、フォーカス判定方法、及び、記録媒体 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2022059064A1 true WO2022059064A1 (ja) | 2022-03-24 |
Family
ID=80776832
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2020/034887 WO2022059064A1 (ja) | 2020-09-15 | 2020-09-15 | フォーカス判定装置、虹彩認証装置、フォーカス判定方法、及び、記録媒体 |
Country Status (3)
Country | Link |
---|---|
US (1) | US20240104923A1 (ja) |
JP (1) | JP7452677B2 (ja) |
WO (1) | WO2022059064A1 (ja) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH10508107A (ja) * | 1995-06-07 | 1998-08-04 | ザ トラスティース オブ コロンビア ユニヴァーシティー イン ザ シティー オブ ニューヨーク | 能動型照明及びデフォーカスに起因する画像中の相対的なぼけを用いる物体の3次元形状を決定する装置及び方法 |
JP2003263629A (ja) * | 2002-03-07 | 2003-09-19 | Matsushita Electric Ind Co Ltd | 虹彩認証装置及び携帯端末装置 |
JP2005071009A (ja) * | 2003-08-22 | 2005-03-17 | Matsushita Electric Ind Co Ltd | 画像入力装置およびそれを用いた認証装置 |
-
2020
- 2020-09-15 US US17/768,562 patent/US20240104923A1/en active Pending
- 2020-09-15 WO PCT/JP2020/034887 patent/WO2022059064A1/ja active Application Filing
- 2020-09-15 JP JP2022550067A patent/JP7452677B2/ja active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH10508107A (ja) * | 1995-06-07 | 1998-08-04 | ザ トラスティース オブ コロンビア ユニヴァーシティー イン ザ シティー オブ ニューヨーク | 能動型照明及びデフォーカスに起因する画像中の相対的なぼけを用いる物体の3次元形状を決定する装置及び方法 |
JP2003263629A (ja) * | 2002-03-07 | 2003-09-19 | Matsushita Electric Ind Co Ltd | 虹彩認証装置及び携帯端末装置 |
JP2005071009A (ja) * | 2003-08-22 | 2005-03-17 | Matsushita Electric Ind Co Ltd | 画像入力装置およびそれを用いた認証装置 |
Also Published As
Publication number | Publication date |
---|---|
JP7452677B2 (ja) | 2024-03-19 |
US20240104923A1 (en) | 2024-03-28 |
JPWO2022059064A1 (ja) | 2022-03-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10452936B2 (en) | Systems and methods of biometric analysis with a spectral discriminator | |
JP6416438B2 (ja) | 眼球血管および顔認識のための画像および特徴品質、画像強化および特徴抽出、ならびにバイオメトリックシステムのための眼球血管と顔領域および/またはサブ顔領域との融合 | |
KR101495430B1 (ko) | 바이오메트릭 인증을 위한 품질 메트릭 | |
JP4845698B2 (ja) | 眼部検出装置、眼部検出方法及びプログラム | |
JP4469476B2 (ja) | 眼位置検出方法および眼位置検出装置 | |
JP4307496B2 (ja) | 顔部位検出装置及びプログラム | |
JP5107045B2 (ja) | 目に関して取得される画像中の虹彩を表す画素を特定する方法 | |
Kang et al. | Real-time image restoration for iris recognition systems | |
JP6651074B2 (ja) | 眼球の運動を検出する方法、そのプログラム、そのプログラムの記憶媒体、及び、眼球の運動を検出する装置 | |
KR101373274B1 (ko) | 안경 제거를 통한 얼굴 인식 방법 및 상기 안경 제거를 통한 얼굴 인식 방법을 이용한 얼굴 인식 장치 | |
JP2007188504A (ja) | 画像中の画素強度をフィルタリングする方法 | |
EP1536375A1 (en) | Image input device and authentication device using the same | |
WO2019163066A1 (ja) | なりすまし検知装置、なりすまし検知方法、及びコンピュータ読み取り可能な記録媒体 | |
CN111353404A (zh) | 一种人脸识别方法、装置及设备 | |
JP2007159610A (ja) | 登録装置、認証装置、登録認証装置、登録方法、認証方法、登録プログラムおよび認証プログラム | |
EP3961495B1 (en) | System and method for finding an area of an eye from a facial image | |
JP2006318374A (ja) | 眼鏡判別装置および認証装置ならびに眼鏡判別方法 | |
WO2022059064A1 (ja) | フォーカス判定装置、虹彩認証装置、フォーカス判定方法、及び、記録媒体 | |
KR101122513B1 (ko) | 3차원 위치정보를 이용한 안구위치 추정시스템 및 안구위치추정방법 | |
KR101794727B1 (ko) | 홍채 검증 방법 | |
KR20200049936A (ko) | 생체 인식 장치 및 방법 | |
KR101280439B1 (ko) | 현금인출기 카메라에서 취득된 얼굴 영상에 대한 얼굴 인식 가능성 판단 방법 | |
KR101704717B1 (ko) | 홍채 인식 장치 및 그 동작 방법 | |
JP4996943B2 (ja) | 顔特徴点検出装置、顔特徴点検出方法及びプログラム | |
CN113627263A (zh) | 一种基于人脸检测的曝光方法、装置以及设备 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 17768562 Country of ref document: US |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20954050 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2022550067 Country of ref document: JP Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20954050 Country of ref document: EP Kind code of ref document: A1 |