Information processing device, information processing method, and program

Abstract

[Problem] To reduce a processing load while ensuring security when carrying out an authentication using sensor information. [Solution] Provided is an information processing device equipped with an authentication unit for carrying out a user authentication on the basis of collected sensor information, wherein the authentication unit carries out a user authentication on the basis of a single feature amount extracted on the basis of sensor information collected by at least two sensors, and a single registered feature amount extracted on the basis of sensor information collected in advance by the at least two sensors.

Description

[Technical Field]

[0001] The present disclosure relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] In recent years, various authentication technologies have been developed that authenticate users based on acquired sensor information. Many proposals have also been made to improve the accuracy of authentication. For example, Patent Document 1 discloses a combined authentication system that authenticates users based on multiple pieces of biometric information. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-40684 Summary of the Invention [Problem to be solved by the invention]

[0004] However, as disclosed in Patent Document 1, when authentication is performed based on a plurality of pieces of information, the processing becomes complicated and the burden on the device may increase. [Means for solving the problem]

[0005] According to one aspect of the present disclosure, there is provided an information processing device including an authentication unit that authenticates a user based on collected sensor information, wherein the authentication unit authenticates a user based on a single feature extracted based on sensor information collected by at least two sensors and a single registered feature extracted in advance based on sensor information collected by the at least two sensors.

[0006] According to another aspect of the present disclosure, there is provided an information processing method including: a processor authenticating a user based on collected sensor information; and the authenticating further includes authenticating the user based on a single feature extracted based on sensor information collected by at least two sensors and a single registered feature extracted in advance based on sensor information collected by the at least two sensors.

[0007] According to another aspect of the present disclosure, there is provided a program for causing a computer to function as an information processing device, comprising an authentication unit that authenticates a user based on collected sensor information, the authentication unit authenticating a user based on a single feature extracted based on sensor information collected by at least two sensors and a single registered feature extracted in advance based on sensor information collected by the at least two sensors. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a block diagram showing an example of a functional configuration of an information processing device 10 according to an embodiment of the present disclosure. [Figure 2] 10 is a diagram for explaining the arrangement of at least two sensors 110 according to the embodiment. FIG. [Figure 3] 10 is a diagram for explaining the arrangement of at least two sensors 110 according to the embodiment. FIG. [Figure 4] 10 is a diagram illustrating generation of integrated sensor information by an integration unit 120 according to the embodiment. FIG. [Figure 5] 10 is a diagram illustrating an authentication process performed by an authentication unit 130 according to the embodiment. FIG. [Figure 6] 10 is a flowchart showing an example of a processing flow of the information processing device 10 in a registration phase according to the embodiment. [Figure 7] 10 is a flowchart showing an example of a processing flow of the information processing device 10 in an authentication phase according to the embodiment. [Figure 8]10 is a diagram showing an example of the arrangement of sensors 110 when the information processing device 10 according to the embodiment is realized as a smartphone. FIG. [Figure 9] 10 is a diagram showing an example of the arrangement of sensors 110 when the information processing device 10 according to the embodiment is realized as a head-mounted display. FIG. [Figure 10] 2 is a block diagram showing an example of a hardware configuration of an information processing device 10 according to the embodiment. FIG. DETAILED DESCRIPTION OF THE INVENTION

[0009] Preferred embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functional configurations are designated by the same reference numerals, and redundant description will be omitted.

[0010] The explanation will be given in the following order. 1. Embodiment 1.1. Background 1.2. Example of functional configuration of information processing device 10 1.3.Details 1.4.Processing flow 1.5. Variations 2. Hardware configuration example 3. Summary

[0011] <1. Embodiment> <<1.1.Background>> As described above, in recent years, various authentication techniques have been developed that authenticate users based on acquired sensor information.

[0012] The above authentication techniques include, for example, techniques for acquiring biometric information such as a fingerprint or vein pattern of a user and performing authentication using the biometric information.

[0013] Here, when authentication is performed based on a single input, in order to avoid a situation where a third party is mistakenly identified as a legitimate user, it is necessary to set strict judgment criteria (for example, a judgment threshold regarding the similarity between the acquired fingerprint and a registered fingerprint).

[0014] However, for example, when a fingerprint image is acquired and used for authentication, the acquired image is likely to differ due to differences in the condition of the user's finger, the photographing position, and other photographing environments.

[0015] Therefore, in an environment where variations in the acquired sensor information are likely to occur, it may not be possible to capture sufficient characteristics to determine whether a user is a legitimate user, and the authenticity of the user may not be recognized. In addition, there are many situations where multiple attempts are required before the user is determined to be a legitimate user, which reduces usability.

[0016] In order to avoid the above-mentioned situation, it is conceivable to perform authentication in a composite manner based on acquired information from a plurality of sensors, for example.

[0017] In this case, by combining multiple authentication methods, such as fingerprint authentication and vein authentication, it is possible to relax the criteria for each authentication.

[0018] However, in this case, since multiple authentications are required to satisfy the respective criteria, if low-level sensor information is used in any of the authentications, it may be difficult to establish authentication (affirm the authenticity of the user).

[0019] In view of the above situation, the technology disclosed in Patent Document 1 makes a final determination by comprehensively evaluating the results of one authentication and the results of other authentications if the authenticity of the user is not confirmed in any of the authentications.

[0020] According to the technology disclosed in Patent Document 1, it is possible to increase the probability of successful authentication, and it is expected to have the effect of improving usability.

[0021] On the other hand, even when the technology disclosed in Patent Document 1 is used, if a combination of multiple authentications is performed, the processing becomes complicated and the load on the device may increase.

[0022] The technical concept of the present disclosure was conceived with the above points in mind, and makes it possible to achieve both reduced processing load and guaranteed security in authentication using sensor information.

[0023] For this purpose, the information processing device 10 according to an embodiment of the present disclosure includes an authentication unit 130 that performs authentication based on collected sensor information. Also, one of the features of the authentication unit 130 according to an embodiment of the present disclosure is that it performs user authentication based on a single feature extracted based on sensor information collected by at least two sensors 110 and a single registered feature extracted in advance based on sensor information collected by at least two sensors 110.

[0024] An example of the functional configuration of the information processing device 10 having the above-described characteristics will be described in detail below.

[0025] <<1.2. Example of functional configuration of information processing device 10>> 1 is a block diagram showing an example of a functional configuration of an information processing device 10 according to an embodiment of the present disclosure. The information processing device 10 according to this embodiment may be any of various devices that authenticate users based on sensor information.

[0026] The information processing device 10 according to the present embodiment may be realized as, for example, a smartphone, a tablet, a PC, a wearable device, etc. An example of a wearable device is an earlable device (hearable device) that is worn on the ear of a user.

[0027] As shown in FIG. 1, the information processing device 10 according to this embodiment may include at least two sensors 110, an integration unit 120, an authentication unit 130, a storage unit 140, and a notification unit 150.

[0028] (Sensor 110) The sensor 110 according to this embodiment acquires sensor information relating to the physical characteristics and behavioral characteristics of the user. The sensor information according to this embodiment includes, for example, biometric information of the user. Examples of biometric information include images of fingerprints, veins, irises, face, posture, etc., and recorded voice.

[0029] 1 illustrates an example in which the information processing device 10 according to the present embodiment includes two sensors 110a and 110b. However, the number of sensors 110 according to the present embodiment is not limited to this example. The information processing device 10 according to the present embodiment may include three or more sensors 110.

[0030] Each of the at least two sensors 110 according to this embodiment may acquire, for example, the same type of biometric information of the user.

[0031] As an example, each of the sensors 110a and 110b according to this embodiment may be an imaging sensor that captures a fingerprint of a user.

[0032] On the other hand, the at least two sensors 110 according to this embodiment may acquire different types of biological information, for example.

[0033] As an example, the sensor 110a according to this embodiment may be an image sensor that captures an image of the user's iris, and the sensor 110b may be a microphone that collects the user's voice.

[0034] (Integration Section 120) The integrating unit 120 according to this embodiment integrates the sensor information acquired by at least two sensors 110 to generate integrated sensor information.

[0035] In this case, the integrating unit 120 according to this embodiment may integrate the partial sensor information extracted from each of the pieces of sensor information acquired by at least two sensors to generate integrated sensor information.

[0036] The functions of the integrating unit 120 according to this embodiment are realized by various processors. Details of the functions of the integrating unit 120 according to this embodiment will be described separately.

[0037] (Authentication unit 130) The authentication unit 130 according to this embodiment authenticates a user based on collected sensor information. One of the features of the authentication unit 130 according to this embodiment is that it authenticates a user based on a single feature extracted based on sensor information collected by at least two sensors 110 and a single registered feature extracted in advance based on sensor information collected by at least two sensors 110.

[0038] At this time, the authentication unit 130 according to this embodiment may authenticate the user based on a single feature extracted from the integrated sensor information generated by the integration unit 120 and the registered feature.

[0039] The functions of the authentication unit 130 according to this embodiment are realized by various processors. Details of the functions of the authentication unit 130 according to this embodiment will be explained separately.

[0040] (Storage unit 140) The storage unit 140 according to this embodiment stores various types of information used by the information processing device 10. For example, the storage unit 140 according to this embodiment stores programs used by the integration unit 120, the authentication unit 130, and the notification unit 150.

[0041] Furthermore, for example, the storage unit 140 according to this embodiment stores the above-mentioned registered feature amounts, integrated sensor information, and the like.

[0042] (Notification section 150) The notification unit 150 according to this embodiment notifies the user based on the result of user authentication by the authentication unit 130.

[0043] For example, the notification unit 150 according to this embodiment provides feedback indicating that authentication has been successful or feedback requesting re-entry of biometric information.

[0044] For this purpose, the notification unit 150 according to this embodiment includes various displays for displaying visual information, speakers and amplifiers for outputting sound, and the like.

[0045] The above has described an example of the functional configuration of the information processing device 10 according to this embodiment. Note that the functional configuration described above using Fig. 1 is merely an example, and the functional configuration of the information processing device 10 according to this embodiment is not limited to this example.

[0046] For example, the information processing device 10 according to the present embodiment may include a configuration other than the configuration shown in Fig. 1. For example, the information processing device 10 according to the present embodiment may further include a processing unit that executes various processes based on the result of authentication by the authentication unit 130.

[0047] Furthermore, the above-described functions may be realized by cooperation between multiple devices. For example, the sensor 110 according to the present embodiment may be provided in a sensor device separate from the information processing device 10. In this case, the information processing device 10 may perform each process based on sensor information received from the sensor device via a network.

[0048] The functional configuration of the information processing device 10 according to this embodiment can be flexibly modified according to the specifications and operation.

[0049] <<1.3.Details>> Next, the features of the functions of the information processing device 10 according to this embodiment will be described in detail. First, the features relating to the arrangement of the sensors 110 included in the information processing device 10 according to this embodiment will be described.

[0050] When biometric authentication is performed, the user is typically required to perform an action to input the biometric information into the device. However, this action can become a burden on the user as the number of authentication attempts and device usage increases.

[0051] For this reason, the at least two sensors 110 according to this embodiment may be arranged at positions where there is a high possibility that biometric information of the user can be acquired when the user wears or uses the information processing device 10.

[0052] In the following, an example will be described in which the information processing device 10 according to this embodiment is an earphone that is worn on the user's ear. Also, in the following, a user's fingerprint will be mainly used as an example of biometric information used for authentication.

[0053] 2 and 3 are diagrams for explaining the arrangement of at least two sensors 110 according to this embodiment.

[0054] FIG. 2 shows a scene in which a user pinches the information processing device 10, which is an earlable device, with their fingers when wearing the information processing device 10 in their ear.

[0055] In the above-described operations, the area where the user's finger comes into contact with the information processing device 10 is likely to be naturally limited depending on the shape of the information processing device 10. In the case of a small-sized device such as an earphone device, the area may be particularly limited.

[0056] In view of the above, two or more sensors 110 according to this embodiment may be arranged in accordance with an area where the user's finger is expected to come into contact with the information processing device 10.

[0057] 3, the sensor 110a is placed at a position where the finger F1 is expected to come into contact when the user wears the information processing device 10. The sensor 110b is placed at a position where the finger F2 is expected to come into contact.

[0058] The above-described arrangement of the sensor 110 allows the user to capture fingerprints effortlessly through natural movements when wearing or using the information processing device 10, effectively reducing the burden on the user associated with inputting biometric information, such as visually checking the position of the sensor 110 each time.

[0059] Furthermore, by arranging the sensor 110 as described above, it is expected that variations in the angle and position of the finger F that comes into contact with the sensor 110 will be reduced, thereby improving the authentication accuracy.

[0060] Next, generation of the integrated sensor information according to this embodiment will be described. As described above, the authentication unit 130 according to this embodiment may authenticate a user based on integrated sensor information obtained by integrating the sensor information acquired by the sensors 110.

[0061] According to the above-described process, authentication based on a plurality of pieces of sensor information can be achieved by a single process, and the load on the information processing device 10 can be effectively reduced.

[0062] 4 is a diagram for explaining generation of integrated sensor information by the integrating unit 120 according to this embodiment. Fig. 4 shows a process for generating integrated sensor information UI from images I1 and I2 acquired by two sensors 110a and 110b, respectively.

[0063] When generating the integrated sensor information UI, the integration unit 120 according to this embodiment may first identify a high confidence region CR in each of the sensor information acquired by at least two sensors 110, which has a high confidence level as the user's biometric information.

[0064] In the example shown in FIG. 4, the integration unit 120 identifies a high confidence region CR1, which is a region corresponding to the user's fingerprint, in the image I1 acquired by the sensor 110a and has a high confidence region CR1.

[0065] Similarly, the integration unit 120 identifies a high confidence region CR2, which is a region corresponding to the user's fingerprint, in the image I2 acquired by the sensor 110b.

[0066] To identify the high confidence region CR as described above, a neural network that has been trained to recognize biometric information used for authentication and other information and calculate the confidence level of the biometric information may be used.

[0067] For example, in the case of fingerprint authentication, a recognition task may be set to distinguish between areas in the acquired image that can be estimated as fingerprint valleys and areas that cannot be estimated as fingerprint valleys, and learning may be performed based on the criterion of whether or not the area has an estimation error that is below a certain constant.

[0068] The integration unit 120 according to this embodiment may calculate the confidence level using a neural network that has undergone the above-described learning, and may identify the high confidence level region CR based on the confidence level.

[0069] Furthermore, the integration unit 120 according to this embodiment extracts partial sensor information PI from each of the identified confidence regions CR, and integrates the extracted partial sensor information PI to generate integrated sensor information UI.

[0070] According to the above-described processing, only the areas of the acquired sensor information that are highly certain to be the target biometric information can be treated as input, which is expected to improve authentication accuracy and effectively reduce the number of authentication attempts.

[0071] In the example shown in FIG. 4, the integration unit 120 extracts partial sensor information PI1 from a high confidence region CR1 identified in an image I1 acquired by the sensor 110a.

[0072] Similarly, the integration unit 120 extracts partial sensor information PI2 from the high confidence region CR2 identified in the image I2 acquired by the sensor 110b.

[0073] Furthermore, the integration unit 120 integrates the extracted partial sensor information PI1 and partial sensor information PI2 to generate integrated sensor information UI.

[0074] In this way, the integration unit 120 according to this embodiment may partially crop sensor information in areas that are highly reliable as biometric information to be used for authentication, and synthesize the cropped sensor information to match the size of the subsequent processing by the authentication unit 130.

[0075] In this case, the size of the images I1 and I2 acquired by the sensors 110a and 110b may be equal to or larger than the size of the images to be processed in the subsequent stage.

[0076] For example, assume that the ratio between high confidence region CP1 in image I1 and high confidence region CP2 in image I2 is α:β (α+β=1). In this case, the integration unit 120 may crop and combine the partial sensor information PI1 and the partial sensor information P2 so that the ratio between the partial sensor information PI1 and the partial sensor information P2 in the generated integrated sensor information UI is α:β.

[0077] On the other hand, the integration unit 120 may use the calculated confidence factor only to determine the position when cropping the partial sensor information PI, and may crop the partial sensor information at a fixed ratio (1:1) of the same area.

[0078] The integration unit may also superimpose multiple images as a vector sum weighted by the output of a certain probability distribution (e.g., a beta distribution Beta(α,β)) so that the ratio of the partial sensor information PI is α:β (or in the same ratio of 1:1).

[0079] Next, extraction of features from the integrated sensor information UI according to this embodiment will be described. Prior to the authentication phase, the features extracted from the integrated sensor information UI are stored in the storage unit 140 to be used as reference information for the user in the authentication phase. In this disclosure, these features are referred to as registered features.

[0080] In addition, in the authentication phase, a single feature is extracted from the integrated sensor information UI generated based on the sensor information acquired in real time, and is used for comparison with registered features.

[0081] A machine learning-based technique may be used to extract features from the integrated sensor information UI according to this embodiment.

[0082] The authentication unit 130 according to this embodiment may extract features using, for example, an extractor that converts the entire integrated sensor information UI into a single feature (feature vector).

[0083] As an example, the extractor may be generated by learning based on a loss criterion such that the similarity between features extracted from the integrated sensor information UI originating from the same user is high and the similarity between features extracted from the integrated sensor information UI originating from different users is low.

[0084] The extractor may also be used in common with the classifier that calculates the similarity between feature amounts by the authentication unit 130.

[0085] Next, a detailed description will be given of the authentication process by the authentication unit 130 according to this embodiment. Fig. 5 is a diagram for explaining the authentication process by the authentication unit 130 according to this embodiment.

[0086] The authentication unit 130 in this embodiment may recognize the authenticity of a user when the similarity between a single feature V1 extracted from integrated sensor information UI1 generated by the integration unit 120 based on sensor information acquired in real time and a registered feature V2 stored in the memory unit 140 exceeds a threshold.

[0087] In this case, the authentication unit 130 according to this embodiment may authenticate the user using a classifier 135 that calculates the similarity based on a single feature extracted from the input integrated sensor information UI1 and the registered feature UI2.

[0088] When multiple pieces of information are mixed in the input as in this embodiment, it is difficult to use them as indicators for identifying the individual simply by responding to them rule-based at the granularity of general biometric information features (for example, minutiae in a fingerprint).

[0089] For this reason, it is effective to use a neural network to convert the image into a feature vector so that both the overall integrated features of the image and ultra-local features can be handled.

[0090] As an example, the classifier 135 according to this embodiment is generated based on a deep metric learning approach that enables learning of a feature extraction mechanism based on a neural network, which can separate an input group originating from the person himself / herself and an input group originating from others (other than the person himself / herself) so that the similarity between the features of the input group originating from the person himself / herself is sufficiently low at the feature (feature vector) layer, and also makes the similarity between the features of the input group originating from the person himself / herself sufficiently high.

[0091] The above can be achieved by optimizing the loss function for distance learning in a feature extraction network that can extract complex image features by combining a filter representation using image convolution or pooling at a higher level.

[0092] In addition, in the above learning, input with the same specifications as the integrated sensor information UI, or input that has been augmented in a data augmentation manner from an unintegrated image set when creating input for learning, may be used.

[0093] Furthermore, since the ratio of the partial sensor information PI in the integrated sensor information UI according to this embodiment is expected to change for each trial, in the learning for generating the classifier 135, inputs may be prepared so that the above ratios vary widely.

[0094] According to the classifier 135 generated by learning using the above input, it is defined that if the integrated sensor information UI contains sufficient feature information, it can identify users with high accuracy, almost regardless of the above ratio.

[0095] Furthermore, by using an integrated sensor information UI that integrates information from multiple sensors as in this embodiment, it is expected that more robust authentication can be achieved even when a person inputs an input that has a feature value that is quite similar to a single feature point (for example, a minutia in a fingerprint).

[0096] <<1.4. Processing Flow>> Next, a detailed description will be given of the flow of processing by the information processing device 10 according to this embodiment. The processing by the information processing device 10 according to this embodiment is mainly divided into a registration phase in which registered feature amounts are saved, and an authentication phase in which authentication is performed by the authentication unit 130.

[0097] First, the flow of processing in the registration phase according to this embodiment will be described. Fig. 6 is a flowchart showing an example of the flow of processing by the information processing device 10 in the registration phase according to this embodiment.

[0098] First, at least two sensors 110 acquire sensor information (S102). As described above, the sensor information includes biometric information of the user.

[0099] Next, the integration unit 120 identifies high confidence regions in each piece of sensor information acquired in step S102 (S104).

[0100] Next, the integration unit 120 extracts partial sensor information from the high confidence region identified in step S104 (S106).

[0101] It should be noted that the identification of high confidence regions in step S104 and the extraction of partial sensor information in step S106 do not have to be clearly separated processes. For example, the identification of high confidence regions in step S104 and the extraction of partial sensor information in step S106 may be realized as a series of processes using a trained neural network.

[0102] Next, the integration unit 120 integrates the partial sensor information extracted in step S106 to generate integrated sensor information (S108).

[0103] Next, the authentication unit 130 determines whether the feature amount extracted from the integrated sensor information generated in step S108 satisfies the registration criteria (S110).

[0104] Here, if the feature quantity satisfies the registration criteria (S110: Yes), the feature quantity is stored in the storage unit 140 as a registered feature quantity (S112).

[0105] Furthermore, the notification unit 150 provides the user with feedback indicating that the registration has been completed (S114).

[0106] On the other hand, if the extracted feature does not satisfy the registration criteria (S110: No), the notification unit 150 presents feedback to the user prompting re-entry (S116), and the information processing device 10 returns to step S102.

[0107] Next, a description will be given of the flow of processing by the information processing device 10 in the authentication phase according to this embodiment. Fig. 7 is a flowchart showing an example of the flow of processing by the information processing device 10 in the authentication phase according to this embodiment.

[0108] The processing in steps S202 to S208 shown in FIG. 7 may be substantially the same as the processing in steps S102 to S108 shown in FIG. 6, and therefore detailed description thereof will be omitted.

[0109] When the integrated sensor information is generated in step S208, the authentication unit 130 calculates the similarity between the feature amount of the integrated sensor information and the registered feature amount stored in the storage unit 140 (S210).

[0110] The authentication unit 130 may use the above-mentioned classifier 135 to perform the similarity calculation.

[0111] Next, the authentication unit 130 determines whether the similarity calculated in step S210 exceeds a threshold value (S212).

[0112] If the similarity exceeds the threshold (S212: Yes), the authentication unit 130 recognizes the authenticity of the user, and the notification unit 150 provides the user with feedback indicating that authentication has been established (S214).

[0113] On the other hand, if the similarity does not exceed the threshold (S212: No), the authentication unit 130 does not recognize the authenticity of the user, and the notification unit 150 presents the user with feedback prompting them to re-enter the information (S216).

[0114] Alternatively, if the similarity does not exceed the threshold (S212: No), authentication may be performed using a conventional method in which each piece of sensor information acquired in step S202 is collated.

[0115] <<1.5. Variations>> As described above, according to the information processing device 10 of this embodiment, it is possible to simplify the authentication process based on multiple sensor information, thereby making it possible to reduce the processing load while ensuring security.

[0116] It should be noted that the processes and features of the information processing device 10 described above are merely examples, and the information processing device 10 according to this embodiment can be flexibly modified according to specifications and operations.

[0117] For example, although the information processing device 10 according to this embodiment has been described above as a main example in which the information processing device 10 is an earlable device, the information processing device 10 according to this embodiment may also be a different type of device such as a smartphone.

[0118] FIG. 8 is a diagram showing an example of the arrangement of the sensors 110 when the information processing device 10 according to this embodiment is realized as a smartphone.

[0119] When the information processing device 10 according to this embodiment is realized as a smartphone, at least two sensors 110 may be arranged on the back of the touch panel, such as sensors 110a and 110b in the figure, and may capture the fingerprints of the user's fingers F1 and F2, respectively, that touch the touch panel.

[0120] Furthermore, at least two sensors 110 may be arranged on the sides of the information processing device 10, which is a smartphone, like sensors 110c and 110d in the figure. In this case, it is possible to efficiently capture a fingerprint of a user holding the information processing device 10 in one hand without the user being aware of it.

[0121] Furthermore, the information processing device 10 according to the present embodiment may be, for example, a wearable device such as a bed-mounted display.

[0122] FIG. 9 is a diagram showing an example of the arrangement of the sensors 110 when the information processing device 10 according to this embodiment is realized as a head-mounted display.

[0123] In the example shown in FIG. 9, the sensor 110a may be an imaging sensor that captures an image of the user's iris.

[0124] The sensor 110b may also be an image sensor that captures an image of the user's fingerprint when it comes into contact with the sensor, or a microphone that collects the user's voice.

[0125] In this way, at least two sensors 110 according to this embodiment may acquire different types of biological information.

[0126] If the sensor 110b is a microphone that collects the user's voice, the integration unit 120 may convert the collected voice into a spectral image using a short-time Fourier transform or the like, and extract partial sensor information from the spectral image.

[0127] Furthermore, when integrating different types of biometric information, if it is difficult to calculate the confidence level, the integration unit 120 may convert multiple images into feature vectors using a neural network, rather than generating integrated information at the image layer, and integrate the information as a weighted sum of each vector at the vectorized layer (generally, the layer before substitution into the loss function in learning).

[0128] <2. Hardware configuration example> Next, a description will be given of an example of a hardware configuration of the information processing device 10 according to an embodiment of the present disclosure. Fig. 10 is a block diagram showing an example of a hardware configuration of the information processing device 10 according to an embodiment of the present disclosure.

[0129] 10, the information processing device 10 includes, for example, a processor 871, a ROM 872, a RAM 873, a host bus 874, a bridge 875, an external bus 876, an interface 877, an input device 878, an output device 879, a storage 880, a drive 881, a connection port 882, and a communication device 883. Note that the hardware configuration shown here is an example, and some of the components may be omitted. Furthermore, the information processing device 10 may further include components other than those shown here.

[0130] (Processor 871) The processor 871 functions, for example, as an arithmetic processing device or control device, and controls the overall operation of each component or part of it based on various programs recorded in the ROM 872, RAM 873, storage 880, or removable storage medium 901.

[0131] (ROM872, RAM873) The ROM 872 is a means for storing programs to be read into the processor 871, data to be used for calculations, etc. The RAM 873 temporarily or permanently stores, for example, programs to be read into the processor 871, and various parameters that change as appropriate when the programs are executed.

[0132] (Host bus 874, bridge 875, external bus 876, interface 877) The processor 871, ROM 872, and RAM 873 are connected to one another via, for example, a host bus 874 that is capable of high-speed data transmission. On the other hand, the host bus 874 is connected to, for example, an external bus 876 that has a relatively low data transmission speed via a bridge 875. In addition, the external bus 876 is connected to various components via an interface 877.

[0133] (Input Device 878) The input device 878 may be, for example, a mouse, keyboard, touch panel, button, switch, lever, etc. Furthermore, a remote controller (hereinafter referred to as a remote control) capable of transmitting control signals using infrared rays or other radio waves may also be used as the input device 878. The input device 878 may also include an audio input device such as a microphone.

[0134] (Output Device 879) The output device 879 is a device capable of visually or audibly notifying the user of acquired information, such as a display device such as a CRT (Cathode Ray Tube), LCD, or organic EL, an audio output device such as a speaker or headphones, a printer, a mobile phone, a facsimile, etc. The output device 879 according to the present disclosure also includes various vibration devices capable of outputting tactile stimuli.

[0135] (Storage 880) The storage 880 is a device for storing various types of data. For example, a magnetic storage device such as a hard disk drive (HDD), a semiconductor storage device, an optical storage device, or a magneto-optical storage device may be used as the storage 880.

[0136] (Drive 881) The drive 881 is a device that reads information recorded on a removable storage medium 901 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, or writes information to the removable storage medium 901 .

[0137] (Removable storage medium 901) The removable storage medium 901 is, for example, a DVD medium, a Blu-ray (registered trademark) medium, an HD DVD medium, various semiconductor storage media, etc. Of course, the removable storage medium 901 may also be, for example, an IC card equipped with a contactless IC chip, an electronic device, etc.

[0138] (Connection port 882) The connection port 882 is a port for connecting an external device 902, such as a USB (Universal Serial Bus) port, an IEEE1394 port, a SCSI (Small Computer System Interface), an RS-232C port, or an optical audio terminal.

[0139] (External connection device 902) The externally connected device 902 is, for example, a printer, a portable music player, a digital camera, a digital video camera, or an IC recorder.

[0140] (Communication Device 883) The communication device 883 is a communication device for connecting to a network, such as a communication card for wired or wireless LAN, Bluetooth (registered trademark), or WUSB (Wireless USB), a router for optical communication, a router for ADSL (Asymmetric Digital Subscriber Line), or a modem for various types of communication.

[0141] <3. Summary> As described above, the information processing device 10 according to an embodiment of the present disclosure includes the authentication unit 130 that performs authentication based on collected sensor information. One of the features of the authentication unit 130 according to an embodiment of the present disclosure is that it performs user authentication based on a single feature extracted based on sensor information collected by at least two sensors 110 and a single registered feature extracted in advance based on sensor information collected by at least two sensors 110.

[0142] According to the above configuration, it is possible to reduce the processing load and ensure security in authentication using sensor information.

[0143] Although the preferred embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings, the technical scope of the present disclosure is not limited to such examples. It is clear that a person skilled in the art of the present disclosure can conceive of various modified or altered examples within the scope of the technical idea described in the claims, and it is understood that these also naturally fall within the technical scope of the present disclosure.

[0144] Furthermore, the steps of the processes described in this specification do not necessarily have to be processed in chronological order according to the order shown in the flowcharts or sequence diagrams. For example, the steps of the processes of each device may be processed in an order different from the order shown, or may be processed in parallel.

[0145] Furthermore, the series of processes performed by each device described in this specification may be realized using software, hardware, or a combination of software and hardware. The programs constituting the software are stored in advance, for example, in a storage medium (non-transitory medium) provided inside or outside each device. Then, each program is loaded into RAM when executed by a computer, for example, and executed by various processors. The storage medium is, for example, a magnetic disk, an optical disk, a magneto-optical disk, a flash memory, etc. Furthermore, the computer programs may be distributed, for example, via a network, without using a storage medium.

[0146] Furthermore, the effects described herein are merely descriptive or exemplary and are not limiting. In other words, the technology according to the present disclosure may achieve other effects that are apparent to those skilled in the art from the description of this specification, in addition to or in place of the above-described effects.

[0147] The following configurations also fall within the technical scope of the present disclosure. (1) an authentication unit that authenticates users based on the collected sensor information; Equipped with the authentication unit authenticates the user based on a single feature extracted based on sensor information collected by at least two sensors and a single registered feature extracted in advance based on sensor information collected by the at least two sensors. Information processing device. (2) the authentication unit authenticates the user based on a single feature extracted from integrated sensor information obtained by integrating the sensor information acquired from the at least two sensors and the registered feature. The information processing device according to (1) above. (3) the authentication unit recognizes the authenticity of the user when a similarity between a single feature extracted from the integrated sensor information and the registered feature exceeds a threshold. The information processing device according to (2) above. (4) the authentication unit performs user authentication using a classifier that outputs the similarity based on a single feature extracted from the input integrated sensor information and the registered feature. The information processing device according to (3) above. (5) the authentication unit extracts a single feature from the integrated sensor information using an extractor generated by learning based on a loss criterion that increases a similarity between feature values ​​extracted from the integrated sensor information derived from the same user; The information processing device according to any one of (2) to (4). (6) an integration unit that generates the integrated sensor information; Further provided with The information processing device according to any one of (2) to (5) above. (7) the integration unit integrates partial sensor information extracted from each of the pieces of sensor information acquired by the at least two sensors to generate the integrated sensor information. The information processing device according to (6) above. (8) the integrating unit identifies a high confidence region in each of the sensor information acquired by the at least two sensors, the high confidence region being biometric information of the user, and extracts the partial sensor information from each of the high confidence regions. The information processing device according to (7) above. (9) the at least two sensors; Further provided with The information processing device according to any one of (1) to (8). (10) each of the at least two sensors acquires the same type of biometric information; The information processing device according to (9) above. (11) The at least two sensors each acquire different types of biological information. The information processing device according to (9) above. (12) the at least two sensors include an imaging sensor; The information processing device according to any one of (9) to (11) above. (13) the at least two sensors capture a user's fingerprint; The information processing device according to (12) above. (14) each of the at least two sensors is disposed at a position where there is a high possibility that biometric information of the user can be acquired when the user wears or uses the information processing device; The information processing device according to any one of (9) to (13) above. (15) It is a wearable device, The information processing device according to any one of (1) to (14). (16) It is an inearable device worn in the user's ear, The information processing device according to (15) above. (17) the processor authenticating the user based on the collected sensor information; Including, performing the authentication includes authenticating the user based on a single feature extracted based on sensor information collected by at least two sensors and a single registered feature extracted in advance based on sensor information collected by the at least two sensors; further comprising: Information processing methods. (18) Computer, an authentication unit that authenticates users based on the collected sensor information; Equipped with the authentication unit authenticates the user based on a single feature extracted based on sensor information collected by at least two sensors and a single registered feature extracted in advance based on sensor information collected by the at least two sensors. information processing device, A program to function as a [Explanation of symbols]

[0148] 10. Information processing equipment 110 Sensors 120 Integration Department 130 Authentication Department 140 Storage section 150 Notification Department

Claims

1. An integration unit that integrates partial sensor information extracted from each of the sensor information acquired by at least two sensors and generates integrated sensor information; an authentication unit that authenticates a user based on a single feature extracted from the integrated sensor information and a single registered feature extracted in advance based on sensor information collected by the at least two sensors; Equipped with the integration unit identifies a high confidence region in each piece of sensor information acquired by the at least two sensors that has a high confidence that the sensor information is biometric information of the user, crops the partial sensor information from each of the high confidence regions in accordance with an area ratio of the plurality of high confidence regions, and generates the integrated sensor information by combining the plurality of cropped pieces of partial sensor information; Each of the sensor information collected by the at least two sensors is an image of a fingerprint of a different finger. Information processing device.

2. the authentication unit recognizes the authenticity of the user when a similarity between a single feature extracted from the integrated sensor information and the registered feature exceeds a threshold. The information processing device according to claim 1 .

3. the authentication unit performs user authentication using a classifier that outputs the similarity based on a single feature extracted from the input integrated sensor information and the registered feature. The information processing device according to claim 2 .

4. the authentication unit extracts a single feature from the integrated sensor information using an extractor generated by learning based on a loss criterion that increases a similarity between feature values ​​extracted from the integrated sensor information derived from the same user; The information processing device according to claim 1 .

5. the at least two sensors; Further provided with The information processing device according to claim 1 .

6. each of the at least two sensors is disposed at a position where there is a high possibility of acquiring an image of a fingerprint of the user when the user wears or uses the information processing device; The information processing device according to claim 5 .

7. It is a wearable device, The information processing device according to claim 1 .

8. It is an inearable device worn in the user's ear, The information processing device according to claim 7 .

9. The processor: Integrating partial sensor information extracted from each of the pieces of sensor information acquired by at least two sensors to generate integrated sensor information; performing user authentication based on a single feature extracted from the integrated sensor information and a single registered feature extracted in advance based on sensor information collected by the at least two sensors; Including, identifying a high confidence region in each of the sensor information acquired by the at least two sensors, which has a high confidence that the sensor information is biometric information of the user; cropping the partial sensor information from each of the high confidence regions in accordance with an area ratio of the plurality of high confidence regions; and generating the integrated sensor information by combining the plurality of cropped partial sensor information; Each of the sensor information collected by the at least two sensors is an image of a fingerprint of a different finger. Information processing methods.

10. Computer, an integration unit that integrates partial sensor information extracted from each of the pieces of sensor information acquired by at least two sensors to generate integrated sensor information; an authentication unit that authenticates a user based on a single feature extracted from the integrated sensor information and a single registered feature extracted in advance based on sensor information collected by the at least two sensors; Equipped with the integration unit identifies a high confidence region in each piece of sensor information acquired by the at least two sensors that has a high confidence that the sensor information is biometric information of the user, crops the partial sensor information from each of the high confidence regions in accordance with an area ratio of the plurality of high confidence regions, and generates the integrated sensor information by combining the plurality of cropped pieces of partial sensor information; Each of the sensor information collected by the at least two sensors is an image of a fingerprint of a different finger. information processing device, A program to function as a

Images